Cpu moving average logger

Catatan Rilis 8 September 2016 Perangkat Tambahan Menambahkan tabel pos pemeriksaan isi ulang untuk loader inti blaze. Hal ini memungkinkan loader untuk lebih efisien meneruskan mengisi data dengan membatasi tanggal yang lebih rendah yang harus dicari saat query data. Pos pemeriksaan harus memiliki delta baru yang diterapkan (1276). Diperbarui VagrantFile untuk menyertakan semua persyaratan dev dan menggunakan gambar yang lebih baru (1310). Izinkan korelasi dan regresi yang akan dihitung antara dua faktor 2D dengan melakukan penghitungan aset-bijaksana (1307). Filter telah dibuat windowsafe secara default. Sekarang mereka bisa dilewatkan sebagai argumen pada Filter, Faktor dan Klasifikasi lainnya (1338). Menambahkan parameter groupby opsional ke peringkat (). puncak(). Dan bawah (). (1349). Ditambahkan filter pipa baru, Semua dan Any. Yang mengambil filter dan pengembalian yang lain Benar jika aset menghasilkan True untuk setiap hari di windowlength days sebelumnya (1358). Ditambahkan filter pipa baru AtLeastN. Yang mengambil filter lain dan int N dan mengembalikan True jika sebuah aset menghasilkan True on N atau lebih hari di windowlength days sebelumnya (1367). Gunakan kalkulator perpustakaan eksternal untuk perhitungan risiko. Empiris menyatukan perhitungan metrik risiko antara pyfolio dan zipline. Empiris menambahkan pilihan anualisasi kustom untuk mengembalikan frekuensi khusus. (855) Tambahkan faktor Aroon. (1258) Tambahkan faktor osilator stochastic cepat. (1255) Tambahkan Dockerfile. (1254) Kalender perdagangan baru yang mendukung sesi yang berlangsung sepanjang malam, mis. 24 jam 6:01 PM-6:00PM sesi untuk perdagangan berjangka. Zipline. utils. tradingcalendar sekarang tidak berlaku lagi. (1138) (1312) Izinkan mengiris kolom tunggal dari FactorFilterClassifier. (1267) Sediakan faktor Ichimoku Cloud (1263) Izinkan parameter default pada istilah Pipeline. (1263) Memberikan tingkat perubahan persentase faktor. (1324) Memberikan faktor bobot bergerak tertimbang linear. (1325) Tambahkan NotNullFilter. (1345) Izinkan perubahan modal didefinisikan oleh nilai target. (1337) Tambahkan faktor TrueRange. (1348) Tambahkan pencarian waktu ke aset. db. (1361) Buat kantin tahu tentang pertukaran aset. (1346) Tambahkan metode undian ke semua persyaratan yang dapat dihitung. (1394) Tambahkan QuantopianUSFuturesCalendar. (1414) Aktifkan penerbitan versi asset. db lama. (1430) Aktifkan schedulefunction untuk kalender perdagangan Futures. (1442) Larang regresi dengan panjang 1. (1466) Percobaan Menambahkan dukungan untuk jendela sejarah Future and Equity yang akan datang, dan memungkinkan akses data Masa Depan lainnya melalui portal data. (1435) (1432) Perbaikan Bug Perubahan Faktor AverageDollarVolume built-in untuk mengobati nilai tutup atau volume yang hilang sebagai 0. Sebelumnya, NaNs hanya dibuang sebelum rata-rata, sehingga nilai sisanya terlalu banyak (1309). Hapus rasio bebas risiko dari perhitungan rasio sharpe. Rasio sekarang rata-rata pengembalian disesuaikan risiko atas pelanggaran pengembalian disesuaikan. (853) Sortino ratio akan mengembalikan perhitungan bukan np. nan bila tingkat pengembalian yang diminta sama dengan nol. Rasio ini sekarang mengembalikan rata-rata tingkat pengembalian yang disesuaikan risiko atas risiko penurunan. Fixed mislabeled API dengan mengubah mar menjadi downsiderisk. (747) Risiko kebelakang sekarang mengembalikan akar kuadrat rata-rata jatah perbedaan sisi bawah. (747) Rasio informasi diperbaharui dengan rata-rata pengembalian risiko yang disesuaikan dengan standar deviasi pengembalian disesuaikan risiko. (1322) Rasio alfa dan sharpe sekarang disetahunkan. (1322) Memperbaiki unit saat membaca dan menulis atribut firsttradingday bar pertama. (1245) Modul pengiriman opsional, bila hilang, tidak lagi menyebabkan NameError. (1246) Perlakukan argumen schedulefunction sebagai aturan waktu bila aturan waktu, namun tidak ada aturan tanggal yang diberikan. (1221) Melindungi terhadap kondisi batas pada awal dan akhir hari perdagangan dalam fungsi jadwal. (1226) Terapkan penyesuaian pada hari sebelumnya bila menggunakan riwayat dengan frekuensi 1d. (1256) Gagal cepat pada kolom pipa tidak benar, alih-alih mencoba mengakses kolom yang tidak ada. (1280) Perbaiki penanganan CaNN AverageDollarVolume. (1309) Peningkatan kinerja blaze core loader. (1227) Izinkan kuncian api bersamaan. (1323) Mencegah hilangnya data bcolz menit terdahulu dari melakukan pencarian berulang kali yang tidak perlu. (1451) Cache pencarian rantai masa depan. (1455) Pemeliharaan dan Refactorings Dihapus sisa menyebutkan addhistory. (1287) Dokumentasi Tambahkan perlengkapan uji yang menghasilkan data harga harian dari perlengkapan data harga menit. (1243) Format Data Perubahan BcolzDailyBarReader dan BcolzDailyBarWriter menggunakan contoh kalender perdagangan, bukan hari dagang yang disambung ke JSON. (1330) Mengubah format aset. db untuk mendukung pencarian dalam waktu tertentu. (1361) Ubah BcolzMinuteBarReader dan BcolzMinuteBarWriter untuk mendukung berbagai ukuran kutu. (1428) Versi 1.0.1 Ini adalah rilis bug fix minor dari 1.0.0 dan mencakup sejumlah kecil perbaikan bug dan perbaikan dokumentasi. Enhancements Menambahkan dukungan untuk model komisi yang ditentukan pengguna. Lihat kelas zipline. financemissionmissionModel untuk rincian lebih lanjut tentang penerapan model komisioning. (1213) Menambahkan dukungan untuk kolom non-float ke dataset Pipeline yang didukung Blaze (1201). Ditambahkan zipline. pipeline. slice. Slice. Istilah pipa baru yang dirancang untuk mengekstrak satu kolom dari istilah lain. Irisan dapat dibuat dengan cara mengindeks ke dalam sebuah istilah, dikodekan oleh aset. (1267) Perbaikan Bug Memperbaiki bug di mana pemuat pipa tidak diinisialisasi dengan benar oleh zipline. runalgorithm (). Hal ini juga mempengaruhi seruan zipline yang dijalankan dari CLI. Memperbaiki bug yang menyebabkan sel saraf zipline IPython gagal (533233fae43c7ff74abfb0044f046978817cb4e4). Memperbaiki bug dalam model komisi PerTrade dimana komisi salah diterapkan pada setiap pengisian parsial pesanan dan bukan pada pesanan itu sendiri, sehingga algoritma dibebankan terlalu banyak dalam komisi saat menempatkan pesanan besar. PerTrade sekarang menerapkan komisi secara per-order secara benar (1213). Akses atribut pada CustomFactor yang menentukan beberapa keluaran sekarang akan mengembalikan irisan output dengan benar saat output juga merupakan nama metode Factor (1214). Menggantikan penggunaan pandas. io. data yang tidak berlaku lagi dengan pandasdatareader (1218). Memperbaiki masalah di mana file berhala. pyi untuk zipline. api secara tidak sengaja dikecualikan dari distribusi sumber PyPI. Pengguna Conda seharusnya tidak terpengaruh (1230). Dokumentasi Menambahkan contoh baru, zipline. examples. momentumpipeline. Yang melatih Pipeline API (1230). Highlight Zipline 1.0 Rewrite (1105) Kami telah menulis ulang banyak Zipline dan konsep dasarnya untuk meningkatkan kinerja runtime. Pada saat bersamaan, kami memperkenalkan beberapa API baru. Pada tingkat tinggi, versi sebelumnya dari simulasi Zipline ditarik dari sumber data multiplexing, yang digabungkan melalui heapq. Aliran ini diumpankan ke loop simulasi utama, menggerakkan jam ke depan. Ketergantungan yang kuat pada membaca semua data membuat sulit untuk mengoptimalkan kinerja simulasi karena tidak ada hubungan antara jumlah data yang kita isi dan jumlah data yang sebenarnya digunakan oleh algoritma. Sekarang, kita hanya mengambil data saat algoritma membutuhkannya. Kelas baru, DataPortal. Mengirimkan permintaan data ke berbagai sumber data dan mengembalikan nilai yang diminta. Hal ini membuat runtime skala simulasi lebih dekat dengan kompleksitas algoritma, bukan dengan jumlah aset yang disediakan oleh sumber data. Alih-alih arus data yang menggerakkan jam, sekarang simak iterasi melalui rangkaian cap waktu hari atau menit yang telah dihitung sebelumnya. Cap waktu dipancarkan oleh MinuteSimulationClock dan DailySimulationClock. Dan dikonsumsi oleh lingkaran utama dalam transformasi (). Kami memasangkan API dataid (N) dan sejarah, menggantinya dengan beberapa metode pada objek BarData: current (). sejarah(). Kriket (). Dan isstale (). API lama akan terus bekerja untuk saat ini, namun akan mengeluarkan peringatan yang tidak berlaku lagi. Anda sekarang dapat mengirimkan sumber penyesuaian ke DataPortal. Dan kami akan menerapkan penyesuaian pada data harga saat melihat ke belakang pada data. Harga dan volume untuk eksekusi dan disajikan ke algoritma dalam data. Saat ini adalah nilai aset yang diperdagangkan. New Entry Points (1173 dan 1178) Agar lebih mudah menggunakan zipline, kami telah memperbarui titik masuk untuk backtest. Tiga cara yang didukung untuk menjalankan backtest sekarang: zipline. runalgo () zipline running zipline (sihir IPython) Kumpulan Data (1173 dan 1178) 1.0.0 mengenalkan kumpulan data. Kumpulan data adalah kelompok data yang harus dimuat dan digunakan untuk menjalankan backtests nanti. Hal ini memungkinkan pengguna untuk tidak perlu menentukan ticker mana yang mereka minati setiap kali menjalankan algoritma. Ini juga memungkinkan kita untuk menyimpan data antar jalan. Secara default, bundel quantopian-quandl akan digunakan yang menarik data dari cermin Quantopian8217 dari dataset WIKI quandl. Bundel baru dapat didaftarkan dengan zipline. data. bundles. register () seperti: Fungsi ini harus mengambil data yang dibutuhkannya dan kemudian menggunakan penulis yang telah lulus untuk menulis data tersebut ke cakram di lokasi yang dapat ditemukan zipline nanti. Data ini dapat digunakan dalam backtests dengan melewatkan nama sebagai argumen - b --bundle ke zipline run atau sebagai argumen bundle ke zipline. runalgorithm (). Untuk informasi lebih lanjut, lihat Kumpulan Data untuk informasi lebih lanjut. Dukungan String di Pipeline (1174) Menambahkan dukungan untuk data string di Pipeline. Zipline. pipeline. data. Column sekarang menerima objek sebagai dtype, yang menandakan bahwa pemuat untuk kolom tersebut harus memancarkan iterator berjendela melewati kelas LabelArray yang baru. Beberapa metode Classifier baru juga telah ditambahkan untuk membangun contoh Filter berdasarkan operasi string. Metode baru adalah: elementof didefinisikan untuk semua classifier. Metode yang tersisa hanya didefinisikan untuk pengkodean string-dtype. Perangkat tambahan Membuat kelas pemuatan data memiliki antarmuka yang lebih konsisten. Ini termasuk penulis bar ekuitas, penulis penyesuaian, dan penulis aset db. Antarmuka baru adalah bahwa sumber daya yang akan ditulis dikirimkan pada waktu konstruksi dan data yang akan ditulis diberikan kemudian ke metode penulisan sebagai dataframes atau beberapa iterator dataframes. Model ini memungkinkan kita untuk melewati objek penulis ini sekitar sebagai sumber daya untuk kelas dan fungsi lain untuk dikonsumsi (1109 dan 1149). Ditambahkan masking ke zipline. pipeline. CustomFactor. Faktor kustom sekarang dapat melewati Filter berdasarkan instantiasi. Ini memberitahu faktor untuk hanya menghitung lebih dari saham dimana filter mengembalikan True, daripada selalu menghitung seluruh jagad saham. (1095) Ditambahkan zipline. utils. cache. ExpiringCache. Cache yang membungkus entri dalam zipline. utils. cache. CachedObject. Yang mengelola kadaluarsa entri berdasarkan dt yang disertakan pada metode get. (1130) Melaksanakan zipline. pipeline. factors. RecarrayField. Istilah pipa baru yang dirancang untuk menjadi tipe keluaran CustomFactor dengan beberapa keluaran. (1119) Menambahkan parameter output opsional ke zipline. pipeline. CustomFactor. Faktor kustom sekarang mampu menghitung dan mengembalikan beberapa keluaran, yang masing-masing merupakan Faktor. (1119) Menambahkan dukungan untuk kolom pipa string-dtype. Loader untuk kolom thse harus menghasilkan contoh zipline. lib. labelarray. LabelArray saat dilalui. Terbaru () pada kolom string menghasilkan string-dtype zipline. pipeline. Classifier. (1174) Menambahkan beberapa metode untuk mengubah Classifiers menjadi Filter. Metode baru adalah: - elementof () - beginswith () - endswith () - hassubstring () - match () elementof didefinisikan untuk semua classifier. Metode yang tersisa hanya ditentukan untuk string. (1174) Fetcher telah dipindahkan dari kode internal Quantopian ke Zipline (1105). Fitur Eksperimental Fitur eksperimental dapat berubah sewaktu-waktu. Menambahkan kelas zipline. lib. labelarray. LabelArray baru untuk mewakili dan menghitung secara efisien pada data string dengan numpy. Kelas ini secara konseptual mirip dengan panda. Kategori. Dalam hal itu merupakan array string sebagai array dari indeks ke dalam array (lebih kecil) dari nilai string yang unik. (1174) Perbaikan Bug Ikhtisar Menambahkan dataset EarningsCalendar baru untuk digunakan dalam Pipeline API. (905). Pemanggil AssetFinder (830 dan 817). Dukungan yang lebih baik untuk dtype non-float di Pipeline. Yang paling menonjol, sekarang kita mendukung datetime64 dan int64 dtypes for Factor. Dan BoundColumn. latest sekarang mengembalikan objek Filter yang tepat saat kolomnya adalah tipe dtype bool. Zipline sekarang mendukung 1.85 numpy, panda 0,17, dan scipy 0,16 (969). Batch transforms telah ditinggalkan dan akan dihapus dalam rilis masa depan. Menggunakan riwayat direkomendasikan sebagai alternatif. Perangkat tambahan Menambahkan cara bagi pengguna untuk menyediakan manajer konteks yang akan digunakan saat menjalankan fungsi terjadwal (termasuk penanganan). Manajer konteks ini akan melewati objek BarData untuk bar dan akan digunakan untuk durasi semua fungsi yang dijadwalkan untuk dijalankan. Ini bisa dilewatkan ke TradingAlgorithm dengan kata kunci argumen createeventcontext (828). Ditambahkan dukungan untuk zipline. pipeline. factors. Faktor contoh dengan datetime64ns dtypes. (905) Menambahkan dataset EarningsCalendar baru untuk digunakan dalam Pipeline API. Dataset ini menyediakan antarmuka abstrak untuk menambahkan data pengumuman penghasilan ke algoritma baru. Implementasi referensi berbasis panda untuk dataset ini dapat ditemukan di zipline. pipeline. loaders. earnings. Dan implementasi berbasis blaze eksperimental dapat ditemukan di zipline. pipeline. loaders. blaze. earnings. (905). Menambahkan faktor built-in baru, zipline. pipeline. factors. BusinessDaysUntilNextEarnings dan zipline. pipeline. factors. BusinessDaysSincePreviousEarnings. Faktor-faktor ini menggunakan dataset EarningsCalendar baru. (905). Ditambahkan isnan (). Metode notnan () dan isfinite () ke zipline. pipeline. factors. Factor (861). Ditambahkan zipline. pipeline. factors. Returns. Sebuah faktor built-in yang menghitung perubahan persen dalam harga penutupan di atas windowlength yang diberikan. (884). Menambahkan faktor built-in baru: AverageDollarVolume. (927). Ditambahkan Faktor ExponentialWeightedMovingAverage dan ExponentialWeightedMovingStdDev. (910). Izinkan kelas DataSet dikelompokkan subclass mana subclass mewarisi semua kolom dari induk. Kolom ini akan menjadi sentinel baru sehingga Anda bisa mendaftarkannya loader ubahsuaian (924). Ditambahkan memaksa () untuk memaksa masukan dari satu jenis ke yang lain sebelum meneruskannya ke fungsi (948). Ditambahkan opsional () untuk membungkus fungsi preprocessor lainnya untuk secara eksplisit mengizinkan None (947). Ditambahkan ensuretimezone () untuk memungkinkan argumen string dikonversi menjadi objek datetime. tzinfo. Ini juga memungkinkan objek tzinfo dilewati secara langsung (947). Ditambahkan dua argumen opsional, dataquerytime dan dataquerytz ke BlazeLoader dan BlazeEarningsCalendarLoader. Argumen ini memungkinkan pengguna menentukan waktu cutoff untuk data saat memuat dari sumber. Sebagai contoh, jika saya ingin mensimulasikan pelaksanaan fungsi beforetradingstart saya di 8:45 USEastern maka saya bisa melewati datetime. time (8, 45) dan USEastern ke loader. Ini berarti data yang timestamped pada atau setelah 8:45 tidak akan terlihat pada hari itu dalam simulasi. Data akan tersedia pada hari berikutnya (947). BoundColumn. latest sekarang mengembalikan Filter untuk kolom dtype bool (962). Ditambahkan dukungan untuk contoh Faktor dengan int64 dtype. Kolom sekarang membutuhkan missingvalue bila dtype adalah integral. (962) Sekarang juga mungkin untuk menentukan nilai missingvalue khusus untuk float. tanggal Waktu. Dan istilah pipa bool. (962) Menambahkan dukungan auto-close untuk ekuitas. Setiap posisi yang dimiliki dalam ekuitas yang mencapai autoclosedate akan dilikuidasi untuk uang tunai sesuai dengan harga jual terakhir ekuitas. Selanjutnya, setiap perintah terbuka untuk ekuitas itu akan dibatalkan. Kedua futures dan ekuitas sekarang ditutup secara otomatis pada pagi hari di autoclosedate mereka. Segera sebelum beforetradingstart. (982) Fitur Eksperimental Fitur eksperimental dapat berubah sewaktu-waktu. Ditambahkan dukungan untuk parameterized Faktor subclass. Faktor dapat menentukan params sebagai atribut tingkat kelas yang berisi tuple dari nama parameter. Nilai ini kemudian diterima oleh konstruktor dan diteruskan dengan nama ke fungsi penghitungan faktor8217. API ini eksperimental, dan mungkin berubah dalam rilis mendatang. Perbaikan Bug Memperbaiki masalah yang akan menyebabkan caching metode harian untuk mengubah len objek SIDData. Hal ini akan menyebabkan kita berpikir bahwa benda itu tidak kosong bahkan pada saat itu (826). Memperbaiki kesalahan yang timbul saat menghitung beta saat data datanya jarang. Sebaliknya numpy. nan dikembalikan (859). Memperbaiki masalah pengirim sentinel () benda (872). Peringatan palsu yang tetap pada unduhan pertama data treasury (: issue 922). Mengoreksi pesan kesalahan untuk setcommission () dan setslippage () bila digunakan di luar fungsi inisialisasi. Kesalahan ini disebut override fungsi dan bukan set. Ini juga berganti nama menjadi tipe pengecualian yang diangkat dari OverrideSlippagePostInit dan OverrideCommissionPostInit ke SetSlippagePostInit dan SetCommissionPostInit (923). Memperbaiki masalah di CLI yang akan menyebabkan aset ditambahkan dua kali. Ini akan memetakan simbol yang sama ke dua soket yang berbeda (942). Memperbaiki masalah di mana PerformancePeriod salah melaporkan nilai totalposisi saat membuat Akun (950). Masalah tetap seputar KeyErrors berasal dari sejarah dan BarData pada python 32-bit, di mana Aset tidak dapat dibandingkan dengan int64s (959). Memperbaiki bug di mana operator boolean tidak diterapkan dengan benar pada Filter (991). Pemasangan zipline tidak lagi downgrade numpy menjadi 1,9,2 diam-diam dan tanpa syarat (969). Kinerja Mempercepat lookupsymbol () dengan menambahkan ekstensi, AssetFinderCachedEquities. Yang memuat ekuitas ke dalam kamus dan kemudian mengarahkan lookupsymbol () ke kamus ini untuk menemukan persamaan yang sesuai (830). Peningkatan kinerja lookupsymbol () dengan melakukan query batched. (817). Pemeliharaan dan Refactorings Database aset sekarang berisi informasi versi untuk memastikan kompatibilitas dengan versi Zipline saat ini (815). Versi permintaan upgrade ke 2.9.1 (2ee40db) Upgrade versi logbook menjadi 0.12.5 (11465d9). Upgrade versi Cython menjadi 0.23.4 (5f49fa2). Membuat persyaratan pemasangan zipline lebih fleksibel (825). Gunakan verseer untuk mengelola versi proyek dan versi setup. py (829). Integrasi sambungan tetap pada travis build (840). Membangun konda tetap, yang sekarang menggunakan sumber git sebagai sumbernya dan membaca persyaratan menggunakan setup. py, alih-alih menyalinnya dan membiarkannya tidak sinkron (937). Membutuhkan setuptools gt 18.0 (951). Dokumen Dokumentasi proses pelepasan untuk pengembang (835). Dokumen referensi tambahan untuk Pipeline API. (864). Ditambahkan referensi docs untuk Asset Metadata APIs. (864). Dokumentasi yang dihasilkan sekarang menyertakan tautan ke kode sumber untuk banyak kelas dan fungsinya. (864). Ditambahkan platform-dokumentasi khusus yang menjelaskan bagaimana menemukan dependensi biner. (883). Miscellaneous Menambahkan metode showgraph () untuk membuat Pipeline sebagai gambar (836). Menambahkan subtest () dekorator untuk membuat subtest tanpa noseparameterized. expand () yang membungkuk keluaran tes (833). Batas waktu melaporkan hasil tes menjadi 15 tes terpanjang (838). Treasury dan benchmark downloads sekarang akan menunggu hingga satu jam untuk mendownload lagi jika data yang dikembalikan dari sumber remote tidak sampai ke tanggal yang diharapkan. (841). Menambahkan alat untuk menurunkan aset db ke versi sebelumnya (941). Rilis 0.8.3 Ikhtisar Sistem dokumentasi baru dengan situs web baru di zipline. io Peningkatan kinerja utama. Sejarah dinamis Metode yang ditetapkan pengguna baru: beforetradingstart. Fungsi api baru: schedulefunction (). Fungsi api baru: getenvironment (). Fungsi api baru: setmaxleverage (). Fungsi api baru: setdonotorderlist (). Pipeline API. Dukungan untuk perdagangan berjangka. Perangkat tambahan Objek akun: Menambahkan objek akun ke konteks untuk melacak informasi tentang akun perdagangan. Contoh: Mengembalikan nilai tunai yang tersimpan yang tersimpan pada objek akun. Nilai ini diperbarui sesuai dengan algoritma yang dijalankan (396). HistoryContainer sekarang bisa tumbuh secara dinamis. Panggilan ke sejarah () sekarang akan dapat meningkatkan ukuran atau mengubah bentuk wadah sejarah untuk dapat melayani panggilan. Addhistory () sekarang bertindak sebagai petunjuk preformance untuk mengalokasikan ruang yang cukup dalam wadah. Perubahan ini kompatibel dengan sejarah. Semua algoritma yang ada harus terus bekerja sebagaimana mestinya (412). Transformasi sederhana berasal dari sejarah quantopian dan penggunaan. SIDData sekarang memiliki metode untuk: Metode ini, kecuali untuk pengembalian. Terima beberapa hari Jika Anda menjalankan dengan data menit, maka ini akan menghitung jumlah notulen pada masa itu, yang memperhitungkan penutupan awal dan waktu sekarang dan menerapkan transformasi selama beberapa menit. Return tidak mengambil parameter dan akan mengembalikan return harian dari aset yang diberikan. Contoh: Bidang baru di Periode Kinerja. Periode Kinerja memiliki bidang baru yang dapat diakses dengan nilai pengembalian todict. - leverage kotor - leverage bersih - eksposur pendek - eksposur panjang - jumlah celana pendek - hitung panjang (464). Biarkan orderpercent () bekerja dengan berbagai nilai pasar (oleh Jeremiah Lowin). Saat ini, orderpercent () dan ordertargetpercent () keduanya beroperasi sebagai persentase dari self. portfolio. portfoliovalue. PR ini memungkinkan mereka beroperasi sebagai persentase MV penting lainnya. Juga menambahkan context. getmarketvalue (). Yang memungkinkan fungsi ini. Sebagai contoh: Opsi baris perintah untuk mencetak algo ke stdout (oleh Andrea D8217Amore) (545). Fungsi baru yang ditetapkan pengguna beforetradingstart. Fungsi ini bisa ditimpa oleh pengguna untuk dipanggil sekali sebelum pasar dibuka setiap hari (389). Penjadwal fungsi api baru (). Fungsi ini memungkinkan pengguna menjadwalkan fungsi yang dipanggil berdasarkan peraturan yang lebih rumit mengenai tanggal dan waktu. Misalnya, panggil fungsi 15 menit sebelum penutupan pasar dengan hormat pada penutupan awal (411). Fungsi api baru setdonotorderlist (). Fungsi ini menerima daftar aset dan menambahkan penjaga perdagangan yang mencegah algoritma untuk memperdagangkannya. Menambahkan daftar poin dalam daftar waktu ETF leveraged yang orang mungkin ingin tandai sebagai bukan trade8217 (478). Menambahkan kelas untuk mewakili sekuritas. Order () dan fungsi pesanan lainnya sekarang memerlukan sebuah instance dari Security, bukan sebuah int atau string (520). Generalisasikan kelas Keamanan ke Aset. Ini adalah persiapan untuk menambahkan dukungan untuk jenis aset lainnya (535). Api baru fungsi getenvironment (). Fungsi ini secara default mengembalikan zipline string. Ini digunakan agar algoritma dapat memiliki perilaku yang berbeda pada Quantopian dan zipline lokal (384). Memperluas lingkungan getar () untuk mengekspos lebih banyak lingkungan ke algoritma. Fungsi sekarang menerima argumen yang merupakan field untuk kembali. Secara default, ini adalah platform yang mengembalikan nilai lama zipline namun bidang baru berikut dapat diminta: arena. Apakah ini live trading atau backtesting datafrequency. Apakah mode menit ini atau mode mulai hari ini. Tanggal mulai simulasi akhir. Simulasi tanggal akhir Modal dasar Modal awal untuk simulasi. peron. Platform yang menjalankan algoritma ini. . Kamus berisi semua bidang ini. Fungsi api baru setmaxleveraged (). Metode ini menambahkan penjaga perdagangan yang mencegah algoritma Anda untuk lebih memanfaatkannya sendiri (552). Fitur Eksperimental Fitur eksperimental dapat berubah sewaktu-waktu. Menambahkan API Pipeline baru. API pipa adalah API deklaratif tingkat tinggi untuk merepresentasikan perhitungan trailing window pada dataset besar (630). Menambahkan dukungan untuk perdagangan berjangka (637). Menambahkan pemuat pipa untuk ekspresi blaze. Hal ini memungkinkan pengguna untuk menarik data dari berbagai format blaze mengerti dan menggunakannya di Pipeline API. (775). Perbaikan Bug Perbaiki bug di mana pengembalian yang dilaporkan dapat dipangkas tajam untuk periode waktu yang acak (378). Perbaiki bug yang mencegah debugger untuk menyelesaikan file algoritma (431). Argumen maju dengan tepat untuk fungsi inisialisasi yang ditetapkan pengguna (687). Perbaiki bug yang akan menyebabkan data treasury diunduh ulang setiap backtest antara EST tengah malam dan saat data perbendaharaan tersedia (793). Perbaiki bug yang akan menyebabkan pengguna mendefinisikan fungsi analisis agar tidak dipanggil jika dilewatkan sebagai argumen kata kunci ke TradingAlgorithm (819). Kinerja Peningkatan kinerja utama pada sejarah (oleh Dale Jung) (488). Pemeliharaan dan Refactorings Hapus kode transformasi sederhana. Ini tersedia sebagai metode SIDData (550). Sorot antarmuka baris perintah untuk menjalankan algoritma secara langsung. IPython Magic zipline yang menjalankan algoritma yang didefinisikan dalam sel notebook IPython. Metode API untuk membangun pengamanan terhadap perintah pelarian dan posisi pendek yang tidak diinginkan. Sejarah baru () berfungsi untuk mendapatkan DataFrame data pasar masa lalu yang bergerak (menggantikan BatchTransform). Tutorial pemula yang baru. Perangkat tambahan CLI: Menambahkan sihir CLI dan IPython untuk zipline. Contoh: Meraih data dari yahoo finance, menjalankan file dualmovingavg. py (dan mencari dualmovingavganalyze. py yang, jika ditemukan, akan dieksekusi setelah algoritma dijalankan), dan menampilkan dataFrame ke dma. pickle (325) . Perintah sulap IPython (di bagian atas sel notebook IPython). Contoh: Sama seperti di atas kecuali daripada mengeksekusi file mencari algoritma di dalam sel dan sebagai pengganti keluaran ke file, buatlah sebuah variabel dalam namespace yang disebut perf (325). Menambahkan Kontrol Perdagangan ke algoritma API. Fungsi berikut sekarang tersedia di TradingAlgorithm dan untuk skrip algo: setmaxordersize (self, sidNone, maxsharesNone, maxnotionalNone) Tetapkan batas pada besaran absolut, dalam andor dan nilai total dolar, dari setiap perintah yang diberikan oleh algoritma ini untuk sid yang diberikan . Jika sid tidak ada, maka aturan tersebut diterapkan pada perintah yang ditempatkan oleh algoritma. Contoh: setmaxpositionsize (self, sidNone, maxsharesNone, maxnotionalNone) - Mengatur batas pada besaran absolut, baik dalam nilai saham atau dolar, dari posisi yang dipegang oleh algoritma untuk sid yang diberikan. Jika sid None, maka aturan tersebut diterapkan ke posisi yang dipegang oleh algoritma. Contoh: setlongonly (self) Tetapkan aturan yang menentukan bahwa algoritma mungkin tidak memegang posisi pendek. Contoh: Menambahkan metode kelas allapimethod pada TradingAlgorithm yang mengembalikan daftar semua metode API TradingAlgorithm (333). Expanded record () fungsionalitas untuk penamaan dinamis. Fungsi record () sekarang bisa mengambil argumen posisi sebelum kwargs. Semua penggunaan dan fungsionalitas yang asli sama, namun sekarang penggunaan ekstra ini akan berhasil: Persyaratannya hanya karena argumentasi poritional hanya terjadi sebelum kwargs (355). Sejarah () telah dikirim dari Quantopian ke Zipline dan menyediakan data pasar jendela bergerak. History () menggantikan BatchTransform. Ini lebih cepat, bekerja untuk data tingkat menit dan memiliki antarmuka yang superior. Untuk menggunakannya, panggil addhistory () di dalam inisialisasi () dan kemudian menerima panda DataFrame dengan memanggil history () dari dalam managedata (). Simak tutorial dan contohnya. (345 dan 357). History () sekarang mendukung panjang jendela 1m (345). Perbaikan Bug Perbaiki penyelarasan hari perdagangan dan buka dan tutup di lingkungan perdagangan (331). RollingPanel memperbaiki saat menambahkan field baru (349). Performance Maintenance and Refactorings Menghapus data sumber HDF5 dan CSV yang tidak terdokumentasi dan belum teruji (267). Refactor simparams (352). Refactoring sejarah (340). Ketergantungan berikut telah diperbarui (zipline mungkin bekerja dengan versi lain juga): Ikhtisar Perbaikan utama pada perhitungan risiko, lihat bagian Perbaikan Bug. Port of history () function, lihat bagian Enhancements Mulai dukungan untuk skrip sintaks algoritma Quantopian, lihat bagian ENH. Dukungan manajer paket conda, lihat bagian Build. Perangkat tambahan Selalu memproses perintah baru yaitu pada batang tempat penanganan ditangani, tapi ada 8216clock8217 data mis. Patokan yang konsisten, perintah proses. Posisi kosong sekarang disaring dari kontainer portofolio. Untuk membantu mencegah algoritme beroperasi pada posisi yang tidak berada dalam jagad saham yang ada. Dahulu, iterasi atas posisi akan mengembalikan posisi saham yang memiliki zero shares held. (Dimana cek eksplisit dalam kode algoritma untuk pos. amount 0 dapat mencegah penggunaan posisi yang tidak ada.) Tambahkan kalender perdagangan untuk BMFampBovespa. Tambahkan permulaan dukungan algo script. Mulai di jalan paritas dengan sintaks naskah di Quantopian8217s IDE on quantopian Contoh: Tambahkan sumber HDF5 dan CSV. Batasi penanganan terhadap waktu dengan data pasar. Untuk mencegah kasus di mana tipe data khusus memiliki cap waktu yang tidak sama, hanya memanggil ditanganiata saat data pasar berhasil melewatinya. Data khusus yang datang sebelum data pasar masih akan mengupdate data bar. Namun penanganan data itu hanya akan dilakukan bila ada data pasar yang bisa ditindaklanjuti. Metode Extended Commission PerShare memungkinkan biaya minimum per transaksi. Menambahkan fungsi simbol api Sebuah simbol () fitur pencarian ditambahkan ke Quantopian. Dengan menambahkan fungsi API yang sama ke zipline, kita bisa membuat salinan dari Zipline algo ke Quantopian lebih mudah. Tambahkan sumber perdagangan acak simulasi. Menambahkan sumber data baru yang memancarkan acara dengan frekuensi tertentu yang ditentukan pengguna (menit atau setiap hari). Hal ini memungkinkan pengguna untuk backtest dan debug algoritma dalam mode menit untuk memberikan jalan yang bersih menuju Quantopian. Hapus ketergantungan pada patokan untuk kalender hari perdagangan. Alih-alih indeks benchmark1617, kalender perdagangan sekarang digunakan untuk mengisi hari perdagangan environment8217s. Hapus bidang ekstradat, karena tidak seperti daftar tolok ukur, kalender perdagangan dapat menghasilkan tanggal di masa depan, jadi tanggal untuk perdagangan hari ini tidak perlu ditambahkan. Motivasi: Sumber untuk kalender tutup dan closeear yang dekat dan kalender hari perdagangan sekarang sama, yang seharusnya membantu mencegah kemungkinan masalah karena misalignment. Memungkinkan konfigurasi dimana tolok ukur disediakan sebagai sumber data berbasis generator agar perlu menyediakan daftar benchmark kedua hanya untuk mengisi tanggal. Sejarah pelabuhan () Metode API dari Quantopian. Membuka inti dari fungsi history () yang sebelumnya hanya tersedia di platform Quantopian. Metode sejarah analogis dengan functiondecorator batchtransform, namun dengan spesifikasi dan frekuensi yang diharapkan dari frekuensi bar sebelumnya yang ditangkap. Contoh penggunaan: N. B. Versi sejarah ini tidak memiliki kemampuan backfilling yang memungkinkan mengembalikan DataFrame lengkap di bar pertama. Perbaikan Bug Menyesuaikan acara benchmark agar sesuai dengan jam pasar (241). Sebelumnya, acara benchmark dipancarkan pada pukul 0:00 pada hari patokan yang terkait dengan: dalam mode emisi 8216minute8217 ini berarti tolok ukur dipancarkan sebelum perdagangan intra-hari diproses. Pastikan statistik perf dihasilkan untuk semua hari Saat menjalankan dengan sedikit simulasi, simulator akan melaporkan kepada pengguna bahwa simulasi mensimulasikan 8216n - 18217 hari (di mana n adalah jumlah hari yang ditentukan dalam params simulasi). Sekarang jumlah hari perdagangan yang benar dilaporkan sebagai simulasi. Perbaiki repr untuk metrik risiko kumulatif. Repr untuk RiskMetricsCumulative mengacu pada struktur kelas yang lebih tua, menyebabkan pengecualian saat dicetak. Juga, sekarang cetak nilai terakhir dalam metrik DataFrame. Cegah emisi menit dari benturan pada akhir data yang tersedia. Perhitungan hari berikutnya menyebabkan kesalahan ketika algoritma emisi menit mencapai akhir data yang tersedia. Instead of a generic exception when available data is reached, raise and catch a named exception so that the tradesimulation loop can skip over, since the next market close is not needed at the end. Fix pandas indexing in trading calendar. This could alternatively be filed under Performance. Index using loc instead of the inefficient index-ing of day, then time. Prevent crash in vwap transform due to non-existent member. The WrongDataForTransform was referencing a self. fields member, which did not exist. Add a self. fields member set to price and volume and use it to iterate over during the check. Fix max drawdown calculation. The input into max drawdown was incorrect, causing the bad results. i. e. the compoundedlogreturns were not values representative of the algorithms total return at a given time, though calculatemaxdrawdown was treating the values as if they were. Instead, the algorithmperiodreturns series is now used, which does provide the total return. Fix cost basis calculation. Cost basis calculation now takes direction of txn into account. Closing a long position or covering a short shouldn8217t affect the cost basis. Fix floating point error in order(). Where order amounts that were near an integer could accidentally be floored or ceilinged (depending on being postive or negative) to the wrong integer. misalnya an amount stored internally as -27.99999 was converted to -27 instead of -28. Update perf period state when positions are changed by splits. Otherwise, self. positionamounts will be out of sync with position. amount, etc. Fix misalignment of downside series calc when using exact dates. An oddity that was exposed while working on making the return series passed to the risk module more exact, the series comparison between the returns and mean returns was unbalanced, because the mean returns were not masked down to the downside data points however, in most, if not all cases this was papered over by the call to. valid() which was removed in this change set. Check that self. logger exists before using it. self. logger is initialized as None and there is no guarantee that users have set it, so check that it exists before trying to pass messages to it. Prevent out of sync market closes in performance tracker. In situations where the performance tracker has been reset or patched to handle state juggling with warming up live data, the marketclose member of the performance tracker could end up out of sync with the current algo time as determined by the performance tracker. The symptom was dividends never triggering, because the end of day checks would not match the current time. Fix by having the tradesimulation loop be responsible, in minuteminute mode, for advancing the market close and passing that value to the performance tracker, instead of having the market close advanced by the performance tracker as well. Fix numerous cumulative and period risk calculations. The calculations that are expected to change are: cumulative. beta cumulative. alpha cumulative. information cumulative. sharpe period. sortino How Risk Calculations Are Changing Risk Fixes for Both Period and Cumulative Use sample instead of population for standard deviation. Add a rounding factor, so that if the two values are close for a given dt, that they do not count as a downside value, which would throw off the denominator of the standard deviation of the downside diffs. Standard Deviation Type Across the board the standard deviation has been standardized to using a 8216sample8217 calculation, whereas before cumulative risk was mostly using 8216population8217. Using ddof1 with np. std calculates as if the values are a sample. Cumulative Risk Fixes Use the daily algorithm returns and benchmarks instead of annualized mean returns. Use sample instead of population with standard deviation. The volatility is an input to other calculations so this change affects Sharpe and Information ratio calculations. The benchmark returns input is changed from annualized benchmark returns to the annualized mean returns. The benchmark returns input is changed from annualized benchmark returns to the annualized mean returns. Period Risk Fixes Now uses the downside risk of the daily return vs. the mean algorithm returns for the minimum acceptable return instead of the treasury return. The above required adding the calculation of the mean algorithm returns for period risk. Also, uses algorithmperiodreturns and tresauryperiodreturn as the cumulative Sortino does, instead of using algorithm returns for both inputs into the Sortino calculation. Performance Removed aliasdt transform in favor of property on SIDData. Adding a copy of the Event8217s dt field as datetime via the aliasdt generator, so that the API was forgiving and allowed both datetime and dt on a SIDData object, was creating noticeable overhead, even on an noop algorithms. Instead of incurring the cost of copying the datetime value and assigning it to the Event object on every event that is passed through the system, add a property to SIDData which acts as an alias datetime to dt. Eventually support for datafoo. datetime may be removed, and could be considered deprecated. Remove the drop of 8216null return8217 from cumulative returns. The check of existence of the null return key, and the drop of said return on every single bar was adding unneeded CPU time when an algorithm was run with minute emissions. Instead, add the 0.0 return with an index of the trading day before the start date. The removal of the null return was mainly in place so that the period calculation was not crashing on a non-date index value with the index as a date, the period return can also approximate volatility (even though the that volatility has high noise-to-signal strength because it uses only two values as an input.) Maintenance and Refactorings Allow simparams to provide data frequency for the algorithm. In the case that datafrequency of the algorithm is None, allow the simparams to provide the datafrequency . Also, defer to the algorithms data frequency, if provided. Added support for building and releasing via conda For those who prefer building with conda. pydata. org to compiling locally with pip. The following should install Zipline on many systems. Slideshare uses cookies to improve functionality and performance, and to provide you with relevant advertising. Jika Anda terus browsing situs, Anda setuju dengan penggunaan cookies di situs ini. Lihat Perjanjian Pengguna dan Kebijakan Privasi kami. Slideshare menggunakan cookies untuk meningkatkan fungsionalitas dan kinerja, dan memberi Anda iklan yang relevan. Jika Anda terus browsing situs, Anda setuju dengan penggunaan cookies di situs ini. Lihat Kebijakan Privasi dan Perjanjian Pengguna kami untuk rinciannya. Explore all your favorite topics in the SlideShare app Get the SlideShare app to Save for Later even offline Continue to the mobile site Upload Login Signup Double tap to zoom out Connected Medical Devices in the Internet of Things Real-Time Innovations (RTI) Share this SlideShare LinkedIn Corporation copy 2017Last updated: 2016-12-07 - R820T2 Register Descriptions RTL-SDR and GNU Radio with Realtek RTL2832U Elonics E4000Raphael Micro R820T software defined radio receivers. Originally meant for television reception and streaming the discovery and exploitation of the separate raw mode used in FM reception was perhaps first noticed by Eric Fry in March of 2010 and then expanded upon by Antti Palosaari in Feb 2012 who found that these devices can output unsigned 8bit IQ samples at high rates. Or not. Who knows A lot of people other people have helped build it up from there. rtlsdr as we know it today was created by the osmocom people in the form of rtl-sdr and osmoSDR , librtlsdr - contains the major part of the driver rtlsdr, rtltcp, rtltest, etc - command line capture tools and utilities gr-osmosdr - gnuradio compatible module and a bunch of other stuff. keenerd is the author of many other rtl tools: rtlfm, rtlpower (heatmap. py), rtladsb and code changes accepted into the mainline. patchvonbraun is the author and maintainer of the build-gnuradio script that made it easy for me, and multitudes of others, to get started with rtlsdr under GNU Radio. rtl-sdr has the latest news and tutorials. rtlsdr. org has a clear introduction too. RF, DSP, and USB details The dongles with an E4000 tuner can range between 54-2147 MHz (in my experience) with a gap over 1100-1250 MHz in general. The R820T and R820T2 go from 24-1760 MHz (but with reduced performance above 1500 MHz). The R820T dongles use a 3.57 MHz or 4.57 MHz intermediate frequency (IF) while the E4000s use a Zero-IF. For both kinds the tuner error is 30 -20 PPM, relatively stable once warmed up, and stable from day to day for a given dongle. All of the antenna inputs are 75 Ohm impedance. The dynamic range for most dongles is around 45 dB. The sensitivity is somewhere around -110 dBm typically. The highest safe sample rate is 2.56 MSs but in some situations up to 3.2 MSs works without USB dropping samples (RTL2832U drops them internally). Because the devices use complex sampling (IQ) the sample rate is equal to the bandwidth instead of just half of it. For the data transfer mode USB 2 is required, 1.1 wont work. Antti Palosaaris measurements show the R820T use 300mA of 5v USB power while the E4000 devices use only 170mA. You can cut the leads to the LED to drop usage The rtlsdr dongles use a phased locked loop based synthesizer to produce the local oscillator required by the quadrature mixer. The quadrature mixer produces a complex-baseband output where the signal spans from - bandwidth2 to bandwidth2 and bandwidth is the analog bandwidth of the mixer output stages. (Datasheets. general ref: Quadrature Signals: Complex, But Not Complicated by Richard Lyons) This is complex-sampled (I and Q) by the ADC. The Sigma-Delta ADC samples at some high rate but low precision. From this a 28.8 Msps stream at 8 bits is produced. That can be resampled inside the RTL2832U to present whatever sample rate is desired to the host PC. This resampled output can be up to 3.2 MSs but 2.56 MSs is the max recommended to avoid losing samples. The minimum resampled output is 0.5 MSs. Check this reddit thread for caveats and details. The actual output is interleaved so one byte I, then one byte Q with no header or metadata (timestamps). The samples themselves are unsigned and you subtract 127 from them to get their actual value. Youll almost certainly notice a stable spike around DC. Its from either the 1f noise of the electronics or if its a Zero-IF tuner (E4000) the LO beating with itself in the mixer. Popular software My favorite way to explore the spectrum is using rtlpower to do very wideband multi-day surveys. For general use SDR is probably the best application for windows with secondary mono-based linux and Mac support. I normally use Gqrx but it requires GNU Radio dependencies. Luckily there are Linux and OS X native binaries packages with all dependencies (ie, gnuradio) these days. For doing diagnostic and low signal level work Linrad is full featured and fast. osmocomfft comes with GNU Radio module gr-osmosdr and is the natural and best way to use gr-fosphor a GPU accelerated display. multimode has a very full and configurable GUI (it works great with GPU accelerated displays like gr-fosphor). For command line and low power devices try keenerds rtlfm. Assuming youre on linux, but applicable in general, do not use the OS DVB drivers. Those are for the DVB-T mode and not the debug mode that outputs raw samples. Linux 3.x kernel should check with lsmod grep dvbusbrtl28xxu and if found at least sudo modprobe - r dvbusbrtl28xxu to unload it. While the sampling bandwidth is only 2.56 MHz the frequency can be re-tuned up to 40 times a second. With frequency hopping you can survey very large bandwidths. See tholins annotated 24 hour rtlpower spectrogram. Below is a zoomable 3720031008 pixel 5 day long spectrogram I made using rtlpowers FFT mode and heatmap. py. It starts very far zoomed out. It might load a bit slow too. (view full window ) This page is mostly just notes to myself on how to use rtlsdrs core applications, 3rd party stuff using librtlsdr and wrappers for it, and lots on using the gr-osmosdr source in GNU Radio and GNU Radio Companion. This isnt a blog, dont read it sequentially, just search for terms of interest or use the topics menu. For realtime support on the same topics try Freenode IRCs rtlsdr and reddits rrtlsdr. These days for most people doing most things you want to get an dongle with an R820T2 tuner. Theyll come with MCX coaxial connectors. On sites like eBay shipping from China the average price is about 10 shipped. These work fine for most things. At a bit higher price of 20 some come with improvements like SMA or F connectors, metal cases and heatsinks on the tuner for stability above 1500 MHz, temperature controlled crystal oscillators, extra breakouts on the PCB, and the like. I bought two E4000 based rtlsdr usb dongles for 20 each in early 2012. Then many months later I bought two more R820T tuner based dongles for 11 each. Theres photos of the E4ks up at the top of the page and of an R820T based dongle in the mini format off to the left (most minis do not have eeproms for device ID). It and the Newsky E4k dongle up top are MCX. Back in 2012 some of the cheaper dongles occasionally miss protection diodes but that is no longer an issue. The antenna connector on the E4k ezcap up top is IEC-169-2, Belling-Lee. I usually replace it with an F-connector or use a PAL Male to F-Connector Female. F to MCX for the other style dongles. The default design has the tuner taking 75 Ohm so thats what they all are except SMA. Only three tuners are very desirable at this time. The Elonics E4000 and the Raphael Micro R820TR820T2. In general they are of equal performance but the sticks with R820T2 chips are easy to find, cheaper ( 10 USD), and they have a smaller DC spike due to the use of a non-zero intermediate frequency but must have cooling for the tuner to PLL lock above 1500 MHz or so. The E4K is better for high end (gt1.7GHz) while the R820T can tune down to 13 MHz without any hardware mods (mutabilitys driver). The tuners themselves are set up and retuned with I2C commands. E4000 tuners used to re-tune twice as fast as R820T tuners, but this was fixed in keenerds experimental branch where R820T actually tune a tiny bit faster than the E4Ks. These changes were later adopted by the main rtlsdr. Re-tune speed But that was the old days when rtlsdr sticks re-tuned relatively slowly. As time passed re-tuning speed has been increased by clean-ups in code and specifically keenerds changes so the tuner doesnt wait nearly as long for the pll to settle. More recently tejeezs mod made the re-tuning even faster by updating all the changed registers for a re-tune in one r82xxwrite I2C call. With this done you can re-tune at rates of up to 41() hops per second a 2x improvement over then-existing drivers. Since then all of these re-tuning changes have been incorporated into the main rtlsdr. Further massive speed-ups can be had at the cost of pretty much all reliability. By not waiting for PLL lock at all and always leaving the i2c repeater register enabled tejeez reports retuning speeds of up to 300 jumps per second are possible. Tuning range As of Aug. 2014 a handful of people have found ways to extend the r820t frequency range as well. Initially thought to top out at 1700 MHZ the R820T driver has has re-written to tune from 22 to 1870() MHz. While efforts have been made to extend the lower range as well, with the PLL seeming to lock down to 8 MHz in some cases, this range turns out to be full of images and repeats of the higher frequency range. A later effort with the addition of driver tweaks to the RTL2832 downconverter pushed the low end down to After tejeez worked out the no-mod HF reception a couple people have noted that the tuners with fc0013 receive HF even better than the R820T board designs. So if you have one of those laying around you might want to try HF with it. Gain settings The E4K has settings for LNA (-5..25dB), mixer (4 or 12dB) and total of 6 IF gain stages with various gains allowing for 1dB steps between 3 and 57dB. The software only deals with LNA and mixer gain and not independently. IF gain can be set through the API. R820T also has LNA, mixer and IF gain settings - the exact steps are not known. The numbers in the library code are through measuring the gain at a fixed frequency. That gave 0..33dB for the LNA, 0..16dB for the mixer and -4.7..40.8dB for the IF gain. The current library does not expose these settings through an API, only LNA and mixer are set through some algorithm. IF gain is set to a fixed value. bofh gives more detail about the R820T step size, The mixer gain step is 1dB (matches the empirical data passably, but not great) and the IFVGA gain step is 3.5dB (matches mine basically dead-on). LNA gain step is not mentioned, all it says is 1111 - max, 0000 - min Frequency error All of the dongles have significant frequency offsets from reality that can be measured and corrected at runtime. My ezcap with E4000 tuner has a frequency offset of about 57 PPM from reality as determined by checking against a local 751 Mhz LTE cell using LTE Cell Scanner. Heres a plot of frequency offsets in PPM over a week. The major component of variation in time is ambient temperature. With the R820T tuner dongle after correctly for I have has a -17 Khz offset at GSM frequencies or -35 ppm absolute after applying a 50 ppm initial error correction. When using kalibrate for this the initial frequency error is often too large and the FCCH peak might be outside the sampled 200 KHz bandwidth. This requires passing an initial ppm error parameter (from LTE scanner) - e. Another tool for checking frequency corrections is keenerds version of rtltest which uses (I think) ntp and system clock to estimate it rather than cell phone basestation broadcasts. Also very cool is the MIT Haystack people switching to rtlsdr dongles (pdf) for their SRT and VSRT telescope designs, Use of DVB-T RTL2832U dongle with Rafael R820T tuner (pdf). The first of these characterizes the drift of the R820T clock and gain over time as well as a calibration routine. As of 2015 there are a number of SDR-enthusiast targeting dongles produced with temperature controlled oscillators (TXCO) that run at less than 1 PPM with no start-up drift. R820T2 variant I recently (06-15-2014) found out from prog (of SDR and airspy) that there are actually two different versions of the R820T tuner. The normal one and the R820T2. The T2 has different intermediate frequency filters allowing for wider IF bandwidths and apparently slightly better sensitivity (a few dB lower noise floor). For rtlsdr dongles this difference in IF filter bandwidth usually doesnt matter much since all of them are larger than the RTL2832Us debugSDR mode bandwidth of 3 MHz. But there are certain situations where a larger tuner bandwidth is advantageous: such as when using Jowetts HF tuning mod. As of Sept. 2014 some of the new R820T2 have been showing up in Terratec E4000 upgrade model sticks. But dont count on it. I bought one from ebay seller smallpartsbigdifference which had a photo showing an R820T2 and it was just an R820T. Since 2015 R820T2 have become far more available. Heres a pdf with the R820T2 Register Descriptions . R820T2 IF Filter Settings In Feburary 2015 Leif sm5bsz (of linrad) relased a modified librtlsdr with changes to the rtlsdr R820T tuner code to allow for finer grained control over IF filter settings. The IF filter which actually is a low pass filter and a high pass filter can be set for a bandwidth of 300 kHz. Dynamic range increases by something like 30 dB for the second next channel 400 kHz away. It is also possible to get some more improvement by changing the gain distribution. Following this gat3ways patched gr-osmosdr and Vasilirus SDR driver were released. gat3way made the IF filter width variable from within gqrx by presenting it as a gain value. Vasilis rtlsdr SDR driver also moves the SDR decimation normally applied during demodulation to the front of the IQ stream. This gives better dynamic range for the visual FFT but demodulated quality is not changed. So far this is all experimental but expect it to be brought mainline on both sides soon. keenerds experimental branch automatically set IF filter width based on sample rate but had not exposed them as manually set values. R828D variant In late 2013 Astrometra DVB-T2 dongles with the R828D tuner ( pic ) paired RTL2832U have begun to appear (2 ). The DVB-T2 stuff is done by a separate Panasonic chip on the same I2C bus. merbanan wrote a set of patches, rtl-astrometa. for librtlsdr has better support these tuners. The performance hasnt been characterized but it at least works for broadcast wide FM via SDR. stevems preliminary testing suggests bad performance in the form of the crystal for the DVB-T2 demodulator leaking fixed spurs 25 dB above noise floor in the IF at approximately 196 and -820 KHz. He was able to mitigate these with the hardware mod of removing the crystal for the DVB-T2 chip (ref ). Official support was added to the rtl-sdr on Nov 5th while testing support was added on Nov 4th . Double FC0013 tuner PCI DVB card randomsdr reported on Freenode rtlsdr IRC on 2015-09-03 that the Leadtek Winfast DTV2000DS PLUS pci card has 2x FC0013 tuners and 2x rtl2832u chips like 2 normal rtlsdr dongles. Performance is not good but tools like rtlfm work if the VIDPID is added to the rtlsdr driver table and udev rules set. It isnt recommended except as a novelty. E4000 datasheet All the ones that are documented in the DS are explainedin the driver header file. And the rest, the datasheet call them Ctrl2: Write 0x20 there and no more details R820T original, support, etc 2012-09-07: Experimental support for dongles with the Rafael Micro R820T tuner that started appearing in May has been added to rtl-sdr source base by stevem. These tuners cover 24 MHz to 1766 MHz. They also dont have the DC spike caused by the IQ imbalance since they use a different, non-zero, IF. On the other hand, they might have image aliasing due to being superheterodine receivers. See stevems tuner comparisons. On 2012-09-20 the R820T datasheet was leaked to the ultra-cheap-sdr mailing list. The R820T2 Register Description pdf was provided by luigi tarenga to the ultra cheap sdr mailing list after he received it from RafaelMicro. The official range is 42-1002 Mhz with a 3.5 dB noise figure. On 2012-10-04 my order arrived. Im liking this tuner very much since it actually works well, locking down to 24 Mhz or so without direct sampling mode. Heres a rough gnuplot spectral map of 24 to 1700 Mhz over 3 days I made with some custom perl and python scripts. Dont judge the r820t on the quality of that graph, it is just to show the range. You can see what I think is either front-end mixer filters not attenuating enough or actual intermodulation as RFI. I do almost no processing of the signal (ie, no IQ correction), dont clear the buffer between samples (LSB probably bad), and use a hacky way to display timeseries data in gluplot. Real SDR software like SDR shows them to be equal or better in quality to E4ks. stevem did gain measurement tests with a few dongles using some equipment he had to transmit a GSM FCCH peak, which is a pure tone. This includes the E4000 and R820T tuners. In addition he measured the mixer. IF and LNA for the R820T. High Frequency (0-30Mhz) Direct Sampling Mod 30Mhz() by using the 28.8 MHz RTL2832U ADCs for RF sampling and aliasing to do the conversion. In practice you only get DC-14.4MHz in the first Nyquist zone but the upper could be had by using a 14.4 MHz to 28.8 MHz bandpass filter. In the stereotypical ezcap boards you can test this by connecting an appropriately long wire antenna to the right side of capacitor 17 (on EzTV668 1.1, at least) that goes to pin 1 of the RTL2832U. Thats the one by the dot on the chip surface. Apparently even pressing a wet finger onto the capacitor can pick up strong AM stations. This bypasses static protection among other things so theres a chance of destroying your dongle. For gr-osmosdr the parameter directsamp1 or directsamp2 gives you the two I or two Q inputs. No hardware change, software mod direct sampling It has recently become possible to use direct sampling with no hardware modifications at all. It is still very experimental and performance is bad. In Oct 2012 Anonofish on the rrtlsdr subreddit had discovered the PLL would lock for a small 3686.6 MHz - 3730 MHz range far outside the normal tuning range and there seemed to be signals there. In January 2014 rtlsdr IRC channel user tejeez figured out this bypassed the tuner (mixer leakage) and implemented a set of register settings (R820T IF frequency, IF filter bandwidths, r82xxwriteregmask(priv, 0x12, val, 0x08) replaced with r82xxwriteregmask(priv, 0x12, val0x10, 0x18)) that would exploit this to enable HF reception. Shortly thereafter keenerd assembled everything into a relatively easy to use patch-set. If you want to give HF listening a try with no risk keenerd has added these changes rtlfm and rtlpower in his experimental rtlsdr repository. To use the no mode mode with rtl tools append the argument, - E no-mod. To use the no-mod direct sampling in something that uses gr-osmosdr, like gqrx or GRC flowgraphs, add the following to the the device string parameters: ie directsamp3. Plug your HF antenna into the normal connector, no hardware mods needed. Differential input Ive been told my pin numbering doesnt correspond to the datasheets, so take that with salt. The relative positions are correct regardless of the numbering. Since then the direct sampling branch has been integrated into main and a number of people have also done balun stuff to use both RTL2832U ADC inputs (usually the two I) in direct sampling mode. Dekar has a page showing how to use an ADSL transformer to generate signal for the ADCs differential input using pin 1 (I) and 2(-I) on the RTL2832. mikig has a useful pdf schematic with part numbers for using wide band transformers or toroids for winding your own. Heres a series of posts from bh5ea20tb showing how to use a FT37-43 ferrite core. And another example from IW6OVD Fernando. PY4ZBZ as well. The ADC has a differentialbalanced input so this is done mainly for the unbalanced-gtbalanced conversion. But the ADC input pins also have a DC offset so you cant just connect one to GND for that. Impedance matching can be done as well but the impedance isnt known. A recent study suggested it was near 3 KOhm but 200 Ohms seems reasonable and is mentioned in some of the tuner datasheets. 4:1 baluns that are used for cable-tv might also work, depending on the impedance of your antenna. Tom Berger (K1TRB) used multiple core materials with trifilar wire and performed tests using his N2PK virtual network analyzer on May 19th (2013). Hams love type 43 ferrite, but for almost every application, there is a better choice. For broadband HF transformers Steward 35T is generally a better choice. Therefore, I wound a couple transformers and did the comparison. Type 43 and 35T Transformer Material Compared For my tests with direct sampling mode I ordered a couple wideband transformers from coilcraft. The PWB-2-ALB and PWB-4-ALB to be specific. I sampled the PWB-4-ALB for free and ordered 4 of the PWB-2-ALB for 10 shipped. Both seem to work fine though I have no means of comparative testing. If youre particularly interested in HF work then an upconverter would be better than the HF mod. With the mod there will be aliases() for any frequency over 14.4 Mhz (12 the 28.8 clock rate). So youd want a 14 MHz lowpass for the low end or a 14-28 MHz bandpass for the high end. And probably other little idiosyncracies. A lot of people chose to just use an upconverter instead. KF7LE wrote up short summaries comparing 16 popular upconverters . Another alternative is to make a diplexer so that you get both HF via direct sampling and VHFetc without any switches. G8JNJ has a detailed guide with annotated photos on how to build the appropriate circuit and modify the latest R820T2 type dongles with it. He reports being able to receive from 15 KHz to 1.8 GHz with this mod. Noise, shielding, cables, and why is that FM signal there When you see something weird, like commercial FM broadcasts at 27 MHz, what you are seeing incomplete filtering of mixing products. Its the harmonics of the square wave driving the mixers combined with insufficient rf filtering to suppress the response. You can tell if it is a local oscillator mixer harmonic leakage by sweeping the frequency and seeing how fast the ghost signal moves relative to this look for linear relationships (ie, 2x the speed, 14 the speed). Sometimes local signals can be powerful (ie, pagers) or close enough to make the preamplifier behave non-linearly resulting in intermodulation. For this kind of RFI turning down the gain helps. The tuners all have a certain amount of intrinsic noise too. keenerd had done tests with an R820T rtlsdr terminated to a resistor inside of a metal box. For these tests rtlpower gain was set to max (49.6dB) and a frequency sweep was done through the entire tuner range, r820t Background Noise. The 28.8 MHz spikes from the clock frequency can be seen among other abberations. But not everything is a ghost from hardware design problems. Depending on your computer setup and local electronics there could be a lot of real noise LCD monitors are a common culprit for VHF noise spikes distributed across wide ranges. It is best to shield and put ferrites on everything if you can. To solve the commercial FM mixing problems an FM trap can be used. Commercial ones work fine typically. But for non-commercial FM RFI like emergency services and pagers custom filters must be made or ordered. Adam-9A4QV has a detailed write-up on making FM trap with a very high upper passband (all the way to 1.7 GHz) with links to design for other low VHF bands. tejeez shared his VHF bandstop design on IRC. Like Adams it has the unique feature of not also wiping out harmonics of the FM band: fm-notch. jpg fm-notchschematic. png. This means you can use it and still do wideband frequency hopping (unlike, say, a 14th wave coaxial stub). For more information on this general type of coaxial cable notch filter check out Ed Lorangers write up on VHF Notch filters (photo ). For my powerful 461 MHz RFI that can be received without an antenna I use a custom 3 cavity notch filter from Par Electronics. Acinonyx describes one way to doing this using a single strip of aluminum tape combined with a spring to connect it to the dongle ground. Akos Czermann at the sdrformariners blog made a somewhat confusing but definitely empirical comparison of noise levels compared to different hardware mods like disconnecting the USB ground from the rtlsdr ground. Quite a few people have had success with that and scotch tape around the USB connector works to test it. Some others bond the enclosure to both the antenna and the USB shield and this works reliably and well. Martin from g8jnj finds the most effective mod to reduce USB and DC-converter noise is shielding the antenna input area with metal soldered to the pcb ground, The noise seems to be coupled directly between components on the topside of the PCB. You can find it if you scroll about halfway down the page linked. Additional noise comes from the switching power supply in the RTL2832U that runs at 1.024MHz. This drops the supplied 3.3v down to the 1.2v needed for internal use. ttrftech has successfully disconnected this switching supply replaced it with 3 diodes to drop the 5v line down to 1.2v. In the example linked above ttrftech uses power form the far side of the board but the eeproms power rail would also work. This decreases spurs in HF significantly. It will increase power usage though something to watch out for when R820T dongles start out at Laidukass Mods and performance of R820T2 based RTL SDR receiver covers replacing all the power rails with external linear regulators, increasing the amount of bypass capacitance on power lines, adding extra chip filtering for the USB 5v line, cutting off the IR receiver part of the PCB, wiring in a TCXO 28.8MHz oscillator, creating a shield with kapton tape and copper foil soldered extensively to the PCB ground, and a new heavy metal case and connectors. To reduce signal loss over long distances and get away from computer RFI I like to run long USB active extension cable with hubs at the end and ferrites added instead of coaxial cable. Around this USB cable I clip on 5 or 6 ferrites at each end. Active extensionrepeater USB2 cables of up to 25m in length can be used. Using External Clocks and coherent sampling in general. Multiple coherent dongles The most exciting development in rtlsdr that has happened recently are Juha Vierinens discuss-gnuradio mailing list and blog posts about a simple and inexpensive method to distribute the clock signal from one dongle to multiple others for coherent operation. I recently came up with a trivial hack to build a receiver with multiple coherent channels using the RTL dongles. I do this basically by unsoldering the quartz clock on the slave units and cable the clock from the master RTL dongle to the input of the buffer amplifier (Xtalin) in the slave units (Ive attached some pictures). Since Ive seen a lot of people asking, the dongles he used were Newsky TV28T v2 wR820T tuners. Ben Silverwood later replicated this technique with his Low cost RTL-SDR passive multistatic DAB radar. implementation in matlab. The youtube video description has links to photos of the setup . Also, theres a Japanese seller with high precision SMD 28.8 MHz crystals. And an ebay seller with high precision 28.8 MHz oscillators for around Things again became exciting in June of 2014. Going beyond simple clock sharing and its max of 3 dongles, YO3IIU put up a great post his build of a 4 dongle RTL2832u based coherent multichannel receiver using a CDCLVC1310-EVM dev board from TI for clock distribution. His post shows the results of a gnuradio block he coded that does all the correlation math to align the samples from each receiver (which are out of step due to the way USB works). Unfortunately the software was never released. stevems experiments were the first I heard about back in 2011. He used his 13MHz cell-phone clock as a reference for a PLL to generate 28.8MHz. He said he used 1v peak to peak. He also related it was possible to not even use the PLL and just the 13 MHz clock if wE4000 tuners if you dont care about sample rate offset. The Green Bay Public Packet Radio guys have written up an interesting article on using 14.4 MHz temperature controlled crystal oscillators sent through a passive (two diode) frequency doubler followed by crystal filters made out of the old rtlsdr clock crystals to provide a low PPM error clock for rtlsdr devices. Since their mirror was missing images I cut them out of the Zine pdf and made a mirror here . I first heard about the GBPPR article from patchvonbraun who implemented one and performed tests which he posted about on the Society for Amateur Radio Astronomy list. It turns out that even with a good distributed clock the 2x R820t rtlsdr dongles still have large phase error for some reason, see: Phase-coherence experiments with RTLSDR dongles and the photo post: Progress towards using RTLSDR dongles for interferometry . Alex Paha has also done clock distribution but unlike the others he used E4000 tuner based receivers for his dual coherent receiver. He also seems to be using only half the IQ pairs. This post is in Russian. Actually maintaining coherence over re-tunes and USB2 latency rtlcoherent In October 2015 teejez uploaded his rtlcoherent code for maintaining multi-dongle coherence using external antenna switches to disconnect the antennas and connect all to a common noise source for correlation calibration. Heres a video of him using it to make a 3 dongle direction finder . Each dither-disabled rtl-sdr is fed from the same reference clock. They still have unknown phase shifts and sampling time differences relative to each other. This is calibrated by disconnecting them from antennas and connecting every receiver to the same noise source. Cross correlation of the noise gives their time and phase differences so that it can be corrected. Currently the signal is received and processed in short blocks with each block starting with a burst of calibration noise. As I understand it the switch chips are sa630 that look for dongle i2c traffic. There are controlled by two RC delay circuits so that every time you change frequency (causing i2c traffic) it disconnects antennas, waits for some time, feeds a pulse (just one edge from the logic chip) into all dongles, waits a bit more and connects the antennas back. You can see the evolution of his setup from this earlier prototype to this later prototype and finally the version used in his direction finder. Every time you tune any two (or more) dongles to a new frequency there will be a tiny difference in the frequency each actually tuned to. The offset must corrected before trying to correlate them. If you dont itll look like theres a constantly varying phase shift. Also dont forget to let the dongles warm up to equilibrium otherwise this additional temperature related frequency shift will cause changes even larger than relative tuning offset and youll get the random phase shift again. As of 2016 Piotr Krysiks Multi-RTL (github ) has made maintaining coherence of multiple dongles accessible even to the amateur. His GNU Radio block handles all the complex details of keeping multiple rtlsdr coherent even when theyre tuned to different frequencies and over re-tunes. It requires no external circuitry. You just have to distribute the clock signal with cable. PLL Dithering and you. On the clock coherencey side Michele Bavaros has explored, tweaked, and replaced, librtlsdrs pll setting code, intermediate frequency, and PLL dithering settings, such that the math, and results, work out cleaner. Using this modified driver he was able to minimize frequency setting errors and improve his GPS carrier following code. This is written up with code examples at his blog in, GNSS carrier phase, RTLSDR, and fractional PLLs (the necessary evil). Without dithering you can only tune to increments of 439.45 Hz. With dithering, you can tune to aproximately anything. tejeez from the rtlsdr IRC relates that this can be done in r82xxsetpll by changing r82xxwriteregmask(priv, 0x12, val, 0x08) to r82xxwriteregmask(priv, 0x12, val0x10, 0x18). This has been implemented as an option in rtlsdr, - N, in keenerds experimental branch. In the absence of any useful information about the RTL2832U clock heres some information about the R820Ts clock system. Crystal parallel capacitors are recommended when a default crystal frequency of 16 MHz is implemented. Please contact Rafael Micro application engineering for crystal parallel capacitors using other crystal frequencies. For cost sensitive project, the R820T can share crystal with backend demodulators or baseband ICs to reduce component count. The recommended reference design for crystal loading capacitors and share crystal is shown as below . Antenna, but particularly broadband antenna When I want to do some scanning that takes advantage of the tuners very wide ranges I use five types of antenna: discone, spiral, dual planar disks, vivaldi (tapered slot), and horns (TEM and pyramidal). Discone, dual planar disk. and archimedian spiral antenna can omnidirectionally cover almost the full range of the E4000 tuner but things get a bit too large to go all the way to the 24 Mhz of the R820T. You can refer to the seperate spiral antenna page for construction and technical details. To build my discone I followed Roklobstas D. I.Y. Discone for RTLSDR. With just a discone and rtlpower its possible to see lots of LEO satellite carrier frequencies doppler across the spectrum. To get an idea of how much you can see with a discone heres a directory where I produce 2 to 4 day long 70 to 1000 MHz range 25KHz resolution 45k10k pixel spectrograms. They each have a javascript zoomable interface to load small tiles progressively. An example. With just a discone and rtlpower its possible to see lots of LEO satellite carrier frequencies doppler across the spectrum. But with a band specific helix in a cone reflector (helicone) many more satellites can be picked up. The previous is a link to a zoomable spectrogram of 2 days of the 1616-1626 MHz satellite band that Iridium satellites use. No LNA was used. Theres plenty of RFIEMI even through a 1 GHz high pass but the satellite doppler passes are clearly there in numbers if you zoom in far enough. When using such broadband antenna, or even a band specific helix, it is possible to pick up powerful out of band signals due to overloading or incomplete mixer filtering. Its important to identify any extraordinarily powerful transmitters nearbye and filter them out. In my case I have a 50w transmitter at 461 MHz across the street always going full power. I bought a custom tuned 3 cavity notch filter from PAR Electronics. This limits the upper frequency range to 1GHz but does at least solve the RFI problem. Usually the spectra are much cleaner when using directional and resonant antenna instead of wideband omnidirectionals. But many directional antenna like helix and log periodic dipoles have very large out of band sidebands on low frequencies not in the designed range. Chipset docs, GNU Radio, DSP, and Antenna Links Page Sections RTL-SDR Links Warning: Im learning as I go along. There are errors. Refer to the proper documentation and original sources first. GNU Radio and RTL-SDR Setup You dont need GNU Radio to use the rtlsdr dongles in sdr mode, but there are many useful apps that depend on it. patchvonbraun has made setting up and compiling GNU Radio and RTLSDR with all the right options very simple on Ubuntu and Fedora. It automates grabbing the latest of everything from git and compiling. It will also uninstall any packages providing GNU Radio already installed first. Simply run, sbrac. orgfilesbuild-gnuradio. and itll automate downloading and compiling of prequisites, libraries, correct git branches, udev settings, and more. I had no problems using Ubuntu 10.04, 12.04, or 14.04. These days (2015) pybombs is slowly taking over for build-gnuradio but for now this works best. If youre thinking about trying this in a virtual machine: dont. If you do get it partially working itll still suck. As an aside: If youre an OSX user then you can use the MacPorts version of GNU Radio (including gqrx, etc) maintained by Michael Dickens. Install 3.7. Most gnu radio projects have been ported to it as default. Only a few old things will require 3.6. An (re )install looks like this. It might be useful to save the log output for future reference. Then test it. The test output below is from a very old version of rtltest with an E4K dongle. Newer versions, and R820T tuners will output slightly different text. Once GNU Radio is installed the Known Apps list at the rtl-sdr wiki is a good place to start. Try running a third party receiver, a python file or start up GNU Radio Companion (gnuradio-companion) and load the GRC flowcharts. If youre having Failed to open rtlsdr device 0 errors make sure something like etcudevrules. d15-rtl-sdr. rules exists and youve rebooted. When updating you can just repeat the install instructions which is simple but long. The advantage to repeating the full process is mainly if there are major changes in the gr-osmosdr as well as rtl-sdr. Itll do things like ldconfig for you. Just compileinstalling rtl-sdr If you dont have the patience for a full recompile and there havent been major gnu radio or gr-osmosdr changes its much faster just to recompile rtl-sdr by itself. The instructions to do so are at the osmosdr page. Itll only take a few minutes even on slow machines. Once you have the latest git clone it is like most cmake projects: rtl-sdr supporting receivers, associated tools keenerds rtl-sdr branch This experimental branch contains a number of useful low processing power utilities, expansions of the original rtl tools, and improvements to the R820T driver re-tuning speed. A lot of them have already been merged into the librtlsdr master but rtlfm and rtlpower fixes, features and bugs appear here first. rtlfm is for scanning, listening, and decoding (and not just FM), rtladbs for plane watching with an external ads-b viewer, rtleeprom for checking and setting serial numbers and related data if your dongle has an eeprom. And as of 2013-09-20, rtlpower, a total power frequency scanner. These tools are very good for slow machines or when you want to do command line automation. Just build it like the osmocom rtlsdr page does for the vanilla install. Use these on the raspberry pi. Most people use rtlpower for smaller total bandwidths (lt200 MHz) and higher spectral resolution using the default FFT mode. This is visualized with keenerds heatmap. py and can result in some really impressive plots when done with 25 crop mode. Just refer to the - h help in rtlpower for instruction. There is also an rtlpower guide at keenerds website. For RMS average power mode, which kicks in automatically for FFT bin sizes 1 MHz and larger, I do visualization of the resulting. csv file with gnuplot. Because the entire bandwidth is summed and saved as one value the the data rate to disk, and spectrogram dimensions are much lower than FFT mode. If you do a large number of frequency hops, (hundreds) then the time adds up. On my two computers the R820T tuner dongles average about 55 milliseconds per retune and sample cycle. I sometimes have dongles thatll fail to lock pll and go into a loop. The - e parameter sets a time limit for a run. Combining this time limit with a bash while loop results in pretty low downtime with resiliance to rtlsdr and USB failures. To combine the results from multiple dongles just cat the files together. But on gnuplots end each new. csv filename requires you to manually edit the gnuplot format. Additionally you need to set the output spectrogram filename and a pixel width. I find for 1000 Mhz 1 MHz that approximately 1000px per 100 MB of file size is required to cover all gaps. And that pops out a png . For rtlfm stuff refer to keenerds sites Rtlfm Guide . Spektrum: an rtlpower GUI frontend. There are a lot of rtlpower GUI frontends but the most useful for me is Spektrum. It uses a modified rtlpower with a Processing GUI front-end. Its available for linux and windows. One of its best features is the relative mode for use in measuring changes in antennas and filters. patchvonbraun (Marcus Leech)s multimode : AM, FM, USB, LSB. WFM. TV-FM, PAL-FM. Very nice, easy to use (screenshots: main. scanning ). It has an automated scanning and spectral zoom features with callbacks to click on the spectrogram or panorama to tune to the frequency of interest. Theres a toggle for active gain control too. The way to get it is, then instead of using GRC, just run the multimode. py as is. If you run it outside of the svn created directory you might need to append bin to pythonpath to find the helper script. If you used build-gnuradio itll tell you what this is at the end of the install. Alternately set it in your. bashrc. If you do the below make sure to reload in the terminal by source When setting the sample rate it is rounded-down to a multiple of 200 Ksps so the decimation math works out. If you have overruns like OOOOoo. then try reducing the sample rate or pausing the waterfall or spectrum displays. The audio subsystem uses a as the identifier, and UHD uses u. With RTLSDR, itll issue O when it experiences an overrun. Which means that your machine isnt keeping up with the data stream. Sometimes buffering helps, but only if your machine is right on the edge of working properly. If it really cant, on average keep up, no amount of buffering will help. If you have overruns like aUaUaUaUa or just aaa then the audio system is asking for samples at a higher rate than the DSP flow can provide (44vs48Khz, etc). Use aplay - l to get a list of the devices on your system. The hw:X, Y comes from this mapping of your hardware -- in this case, X is the card number, while Y is the device number. Or you can use pulse for pulseaudio. Try specifying, gqrx : Written by Alexandru Csete OZ9AEC gqrx is an experimental AM, FM and SSB software defined receiver. The original version did not have librtlsdr support so changes were made by a number of others to add it. A couple weeks later Csete added gr-osmosdr support to the original. Dekar established a non-pulseaudio port of gqrx for Mac OSX. GNU Radio 3.7 has recently been released and it is not exactly backwards compatible. patchvonbrauns build-gnuradio. sh pulls 3.6.5 3.7.x by default. As of August 9th 2013 Gqrx 2.2.0 has been released. This upgraded version can now be installed as binaries with all of its dependencies pre-packaged on both Ubuntu linux (a custom PPA. no 10.04 packages) and Mac OS X That includes all the GNU Radio stuff. So this is an all-in-one alternative to building GNU Radio from source. I think this persons guide is better than mine. rtlsdr wGqrx on N900 phones xes provides pre-compiled packages of Gqrx and the GNU Radio dependencies for N900 linux cell phones. SDR. Written by prog (Youssef) for Windows. It is probably the best general purpose software for rtlsdr devices. Mono is slow and ugly on linux but if you restrict the sample rate it works fine. Its probably the easiest program to use, has the most diverse plugin ecosystem (example: Vasilirus plugins ), and has the best DSP and features for dealing with the quirks of the rtlsdr dongles. As of 2015-09-14 the changes to Mono 4 allow SDR to be viable to run on linux again. Make sure you have the latest Mono 4 though. This still requires soft linking in your system rtlsdr and portaudio library to the sdrshape. exe dir like below, Just make sure you link your actual system rtlsdr and libportaudio, not my example path above. On debianubuntu find it by using locate, Update: As of 2015-10-15 ADBS is no more. ADBS is another easy to use application by prog, but specifically for plotting aviation transponders like gr-air-modes does. The distributed binaries also runs under linux with mono (or native in windows) and output virtualradar compatible data on 127.0.0.1:47806. If your antenna condition is crappy, try using filter 1. gr-fosphor. gr-fosphor is an amazingly fast and information dense spectrogram and waterfall visualization using OpenCL hardware acceleration. It surpasses the Wx widget elements in performance, and so usability, by far. With this visualization you can easily skip through 1 GHz of spectrum very quickly and actually notice transient signals as they pass. Right now it is not very configurable, just arrow keys for scale. But expect this to be the preferred visualization block in the future. I have written up an barebones guide to installing gr-fosphor on Ubuntu 12.04. Modern gr-fosphor requires OpenCl 2. If you only have OpenCl 1.2 installed use this commit . gr-air-modes. A decoder of aviation transponder Mode S including ads-b reports near 1090 Mhz. It can be coupled to software to show plane positions in near real time (ex: VirtualRadar ). This works under mono on Ubuntu 12.04 but not 10.04. Originally written by Nick Foster (bistromath) and adapted to rtlsdr devices first by Steve Markgraf (stevem), bistromath later added rtlsdr support. Heres an example of basic decoding done with the stock antenna on the early version by stevem. Nowdays its better to use bistromaths. As of July 23, 2013 there was a major update to gr-air-modes which now includes a nice google maps overlay and works on gnu radio 3.7 branch only . Heres an example of install process and first run looks like. To use with virtual radar output add the below - P switch. Then open up virtualradar with mono and go to tools-gtoptions-gtbasestation and put in the IP of the computer running uhdmodes. There are not many compatible planes in the USA so far so even if you are seeing lots of Mode-S broadcast in uhdmodes you might not see anything in virtualradar. Sometimes my server is running at superkuh:81VirtualRadarGoogleMap. htm. Dump1090. Dump 1090 is a Mode S decoder specifically designed for RTLSDR devices. Antirezs ADS-B program is really slick. It does not depend on GNU Radio, has a number of interactive modes, and it even optionally runs its own HTTP server with googlemaps overlay of discovered planes no virtualradar needed. It uses very little CPU and has impressive error correction. This is your best choice to play with plane tracking quickly. I tried various bits blindly and found a setting that eliminates the AGC in the RTL2832 chip. That is a significant part of the performance improvement. This is an LTE cell searcher that scans a set of downlink frequencies and reports any LTE cells that were identified. A cell is considered identified if the MIB can be decoded and passes the CRC check. LTE-Tracker is a program that continuously searchers for LTE cells on a particular frequency and then tracks, in realtime, all found cells. With the addition of a GPS receiver, this program can be used to obtain basic cellular coverage maps. The author had only tested it on Ubuntu 12.04 but with some frustrating work replacing cmake files and compiling dependencies I made it work on 10.04. Scanner is very useful to get your dongles frequency offset reliably and Tracker is very pretty. Remember to let your rtlsdr dongle warm up to equilibrium temperature before checking frequency error. Kalibrate. or kal, can scan for GSM base stations in a given frequency band and can use those GSM base stations to calculate the local oscillator frequency offset. The code was written by Joshua Lackey and made rtlsdr accessible by stevem. There is also a windows build made by Hoernchen. Let your rtlsdr dongle warm up to equilibrium temperature before running the test. When youre using this to find your frequency error its important to use the - e option to specify intial error. 270k of bandwidth is used for GSM reception and if the error of the dongle is too large the FCCH-peak is outside the range. I compiled some install process and example usage notes . Simple FM (Stereo) Receiver simplefmrcv also by patchvonbraun is the best sounding and tuning commercial FM software in my opinion. He released a major update to his gnuradio creation at the end of October. my DongleLogger : I wrote these scripts do automatic generation of 1D spectrograms, per frequency time series plots of total power, and 2D spectral maps over arbitrary frequency ranges using multiple dongles at once. There is almost no DSP done and it is very simple but the wideband spectrograms and time series can be informative and fun regardless. It uses gnuplot for graphics generation. Obsolete. Use rtlpower instead . A simple, GRC-based tool for small-scale radio astronomy, providing both Total Power and Spectral modes. It has a graphical stripchart display, and a standard FFT display. It also records both total-power and spectral data using an external C program that records the data along with timestamps based on the Local Mean Sidereal Time. This is another incredible tool by patchvonbraun. It does all the heavy lifting of integration over time and signal processing to get an accurate measurement of absolute power over a range. With it he has managed to pick out the transit of the milky way at the neutral hydrogen frequency using rtlsdr sticks and a pair of yagi antenna. The log file format is text and fairly easy to parse with gnuplot but it comes with processsimpletpdat for cutting it into the bits you want and making total power or spectral component graphs. Itll make a directory called simpleradata in your home by default. Dont forget to set the --devid to rtl otherwise gnuradio wont find the gr-osmosdr source and itll substitute a gaussian noise source. Ear to Ear Oak made this wideband total power scanner that generates 1D spectrum plots over any tunable ranges with arbitrary integration times. It can update a matplotlib python plot GUI in real time and has the ability to output cvs values as well as an internal format. Its very useful for finding whats broadcasting in your area quickly. Using its csv output and gnuplot I visualized a scan from 54-1100 MHz . If you want to use the data in gnuplot you have to sort it and make sure the header is commented out. You can comment out the header manually but I instead prefixed a hash to the log writing behavior at line 786, Pager stuff Thomas Sailers multimon. Linux Radio Transmission Decoder which I use to (try to) decode pager transmissions around 930Mhz. And more recently Dekar s multimonNG. a fork with improved error correction, more supported modes, and nixosxwindows support. Dekar also supplied a GRC receiver for pagers to decode pager transmissions in real-time using fifos. zarya has made rtlflex. py. a gnuradio based flex decoder for pagers. It can be used to decode dutch p2000 messages, for example. This fills a gap in multimon-ng pager support. DongleLogger: my pyrtlsdr lib based spectrogram and signal strength log and plotter Obsolete. Use rtlpower instead. Automatic generation of and html gallery creation of wideband spectrograms using multiple rtlsdr dongles to divide up the spectrum. It also produces narrow band total charts, and other visualizations. (not live): erewhon. superkuhgnuradiolive - click the spectrograms for time series plot These scripts cause the rtlsdr dongle to jump from frequency to frequency as fast as they can and take very rough total power measurement. This data is stored in human readable logs and later turned into wideband spectrograms by calling gnuplot. In order to further increase coverage of any given spectrum range multiple instances of the script can be run at once in the same directory adding to the same logs. Their combined output will be represented in the spectrogram. I dont know much python but the python wrapper for librtlsdr pyrtlsdr was a bit easier to work with than gnu radio when I wanted to do simple things without a need for precision or accuracy . Actualy receivers with processing could be made with it too, but not by me. This is the gist of what it does, The pyrtlsdr library can be downloaded by, I have used the test. py matplotlib graphical spectrogram generator that came with pyrtlsdr as a seed from which to conglomerate my own program for spectrum observation and logging. Since I am not very good with python I had to pull a lot of the logic out into a perl script. So everything is modular. As of now the python script generates the spectrogram pngs and records signal strength (and metadata) in frequency named logs. It is passed lots of arguments. These arguments can be made however you want, but I wrote a perl script to automate it along with a few other useful things. It can generate a simple html gallery of the most recent full spectral map and spectrograms with each linked to the log of past signal levels. Or it can additionally generate gnuplot time series pngs (example ) and link those intead of the raw logs. It also calls LTE Cell Scanner and parses out the frequency offset for passing to graphfreq. py for correction. I no longer have it running because of the processor usage spikes which interrupt daily tasks. In the past Id have rsync updating the public mirror with a big pipe every Modifying pyrltsdr As it is pyrtlsdr does not have the getset functions for frequency correction even if I sent the PPM correct from the perl script. Since the hooks () were already in librtlsdr. py (line 60-66) but just not pythonized in rtlsdr. py they were easy to add to the library. These changes are required to use frequency correction and make the int variable errppm available. I have probably shown that I dont know anything about python with this description. I forked roger-s pyrtlsdr on github and added them there for review or use, githubsuperkuhpyrtlsdrcommitffba3611cf0071dee7e1efec5c1a582e1e344c61. I apologize for cluttering up the pyrtlsdr namespace with such trivial changes but Im new to this and github doesnt allow for private repositories. What you should be using instead. rtlpower was recently (2013-08-20) released by keenerd. It does most of what my scripts do, except much better, faster, and easier. I highly recommend you try it first. RTLSDR Scanner by Ear To Ear Oak is awesome for generating 1D wideband spectrograms. Enoch Zembecowicz made a polished and useful sdr logging tool Panteltjes rtlsdrscan is another tool like RTLSDR Scanner for 1D total power scans. It is a good reference for using librtlsdr with C. If you are serious about measuring total power over one 2.5 Mhz range then simplera. or simple radio astronomy, is best. fast version: see below donglelogger-faster. tar. gz - all needed files including pyrtlsdr radioscanfaster. pl - pyrltsdr using script frequency setting and incrimenting, sampling, and logging. graphfreqsfaster5.py - option passing, log parsing, plot making, frequency corrections wrapper, html image gallery generation graphfreqsgnuplot. py - legacy functions slow version: graphfreqs. py - pyrltsdr using script sampling, and logging radioscan. pl - manages graphfreqs, parses logs, makes plots, gets frequency corrections, generates gallery The faster version Speed ups, Inline C usb reset, and avoiding dongle reinitialization. (less options) cli switchesoptions These two scripts do fast scans within python from x to y frequency. Enabled it with - fast and make sure to set start and stop frequency with - f1 and - f2. Do not use - flist with this option. This is an example output spectral map (a spectrogram with a silly name). This example output above shows the overloading effects of using a wideband discone that picks up off-band noise. Each column is made up of small squares colored by intensity of the signal. Since the scripts start at the low frequency and sweep to high there is a small time delay between the bottom and top (see it more clearly zoomed in ). And this is represented as the slant of the row. Sometimes strong signals will swamp out others resulting in discontinuities displaying as small dark vertical bands. Or fast (-fast) scan a smaller range with smaller range (-f1,-f2: 24-80Mhz), with smaller samplerate (-r: 250 Khz) at smaller intervals (-s: 400Khz steps) with a gain of 30. Only output a large spectrogram of all frequencies to the directory specified with - d2 as spectral-map. png. This example does not use frequency offset correct (-c) for even faster speeds. Combining multiple rtlsdr devices for greater speed By splitting up the spectrum into multiple smaller slices and giving them to multiple dongles the time required for one scan pass can be greatly improved. The above spectrogram is made with 2 dongles, one for the lower half and one for the upper. It is from ryannathans who also contributed the code for for specifying device ID . This is as simple as running the script twice but giving each instance a different - dev argument to specify device ID. You can run as many rtlsdr devices with my scripts as you wish (up to the USB and CPU limits). If they are using the same directory (-d2) their log data will be combined automagically for better coverage. Outlier signals skewing your color map scale Sometimes I get corrupt samples that show a signal level of 60dB. These skew the scale of the output spectrograms. If I notice that they have occurred during a long run Ill use grep to find them and remove them manually. I replace the signal level with the level of the previous non-corrupt sample. In the future Ill build this kind of outlier removal in to the scripts, or sanity check before writing them. All the incremental improvements in speed Ive made above are okay but not very easy to maintain with multiple script types (bashperlpython). Im slowly putting together an Inline C based perl wrapper for exposing librtlsdrs functions within a perl script to write this as a standalone in perl. This is slow work because Ive never done anything like it before. rtlsdrperl - what if there were a perl wrapper for librtlsdr Well, there never will be. But heres some example code anyway. Older version graphfreqs. py You have to have the modified pyrtlsdr with the getset functions for frequency correction. LTE Cell Scanner should also be installed so the CellSearch binary is available. Then download the two scripts above and put them in the same directory. For large bandwidths sampled this feature, ppm error correction, has an unnoticably small effect but I wanted to add it anyway. To call the spectrogramlog generator by itself for 431.2 Mhz at 2.4MSs with a gain of 30 and frequency correction of 58 PPM use it like, Ive disabled the matplotlib (python) per frequency spectrogram plots for frequencies over 1 Ghz because theres not much going on up there. Also, the x-axis ticks and labels become inaccurate for some reason. Logs and format The signal strength logs, named by frequency (e. g. 53200000.log), use unix time and are comma seperated with newlines after each entry. In order of columns it is: unix time. relative signal level. gain in dB, PPM correction. It also generates a log file with all frequencies for use with gnuplot, all. log. This file has unixtime first, then frequency, then gain and ppm error. radioscan. pl The radioscan. pl script is used to automate calling graphfreqs in arbitrary steps. To generate plots and signal strength for 52 Mhz to 1108 Mhz with a gain of 30, sample rate of 2.4MSs, and an interval between center frequencies of 1.2 Mhz, call it like, cli switchesoptions Because I can use the default directories I keep it running like the below, but anyone else should make sure to set - d2. TunerUSB freeze solution with unplugging edit: as of Jan 5th 2013, librtlsdr has added soft reset functionality Since graphfreqs. pys initializing and calling of rtl-sdr happens so frequently there are sometimes freezes. To fix these the USB device has to be reset. In the past I would accomplish this by un and re-plugging the cord manually. But that meant lots of downtime when I was away or sleeping. So, Ive added in a small C program to the perl script using Inline::C that exposes a function, resetusb(). It is used if the eval loop around the graphfreqs call takes more than 10 seconds. This means you need Inline::C to run this script . To look at the original C version with a good explanation of how to use it click here . Page Sections My rtlsdr receiver wgnuradio implementation of the 11 GHz VSRT solar interferometer As far as I understand it, the VSRT design is a subset of intensity interferometer that uses the frequency error between multiple 11 GHz satellite TV low noise downconverter block (LNBF) clocks to create a beat frequency in the total power integrated. I am basically copying the MIT Haystack Very Small Radio Telescope (VSRT) but replacing the discrete component integrator and USB video input device with an rtlsdr dongle. The idea is to spend as little on hardware as possible. With modern LNBF the error between same model parts is about 30 ppm which results in beat frequencies of 100 KHz at the 10 GHz of the mixers. With this kind of front-end there are no nulls but the fringe modulation can still be read out as variations in count of histogram bins that contain the beat frequency (in the total power fft). This intensity measurement proxy traces out the the envelope of the fringes and varies as a sinc function of distance between antenna. Knowing this and the distance can give you high angular diameter and position measurements of very bright radio sources. Historical and other context. For a detailed mathematical explanation of VSRT see MIT Haystacks VSRT Introduction . There is also a thread on the Society for Amateur Radio Astronomers list discussing the VSRT design. The more general concept of intensity interferometry, where you correlate total power instead of frequency, was originally developed by Hanbury-Brown amp Twiss. Roger Jennison was around too. The Early Years of Radio Astronomy: Reflections Fifty Years after Janskys Discovery by W T Sullivan (2005) is an excellent source about Hanbury Brown and Twisss side of it. The chapter The Invention and Early Devlopment of The Intensity Interferometer (pdf) is fascinating. Also see The Development of Michelson and Intensity Long Baseline Interferometry (pdf). It covers not only the technical concepts but also historical context, detailed hands-on implementations, and other personal anectdotes. And check out Jennisons book Radio Astronomy (1966)) as he invented the process of phase closure which uses a third antenna signal combined mathematically to recover some of the missing phase information. Arranged in a triangle of projected baselines the phase errors cause equal but opposite phase shifts in ajoining baselines, canceling out in the closure phase. The MIT Haystack groups managed to resolve individual sunspots groups moving across the solar disk using with the technique with the VSRTs. An interferometer is an instrument that combines two signals (normally from two detectors) in a manner that the signals interfere to produce a resultant signal. The resultant signal is usually the vector sum of the two signals, but in some cases it is the product or some other mix. The traditional interferometer, usually studied and analyzed in physics courses, combines the two signals in a way that both amplitude and phase information are used. By varying the positions of the two detectors, it is possible to synthesize an effective aperture that is equivalent to the separation of the detectors and to reconstruct the impinging wavefront, thus providing significant information about the extent and structure of the signal source. The traditional phase-sensitive interferometer requires retention of the signal phase at each detector the phase-sensitive interferometry technique will not be discussed in detail here. A special case of the interferometer is the intensity interferometer, which performs an intensity correlation of signals from the two detectors. Although in the intensity interferometer the phase information from the two antennas is discarded, the correlation of the two signals remains useful. Aperture synthesis is not practical, but some important source characteristics may be determined. I think the VSRT is a special case of intensity interferometer where you dont try to align samples by time after recording. Instead you just look for the baseline distance sinc pattern in total power at the beat frequency of the unsynchronized clocks. Implementation so far. So far Ive only done it with manual pointing screwed to a board. The interferometry correlation is done with a satellite tv market stripline power combiner at the intermediate frequency (IF, 950-1950 MHz) and then an rtlsdr dongle is used to measure the total power of a 2.4 MHz bandwidth of the intermediate frequency range. I use a gnuradio-companion flowgraph to take the total power and then do a fourier transform of the total power. In this fourier transform the fringes show up as a modulation of the count in the FFT bins which correspond to the difference in frequency between the two downconverters. In my case this is about In the Haystack VSRT memos a line drop amplifier, or two, are sometimes put behind the respective LNBF IF coax outputs or the power combiner. With the rtlsdr dongle and relative short (lt10m) baselines of RG6 this isnt required. The GUI allows for setting the exact 2.4 MHz bandwidth of the IF range to sample and the total power FFT bin bandpass to where and what the LNBF beat frequency is. The file name is autogenerated to the format, The time embedded in the filename is later used by a perl script, vsrtlogtimeplot. pl, which converts and metadata tags the binary records to gnuplot useable text csv format for making PNG plots. total power modes (tp-modes. grc) vsrtlogtimeplot. pl Who else helped I consulted with patchvonbraun a lot for the softwaregnuradio side. He gave me an example of how to use the WX GUI Stripchart and I would not have guessed I needed to square the values from the beat frequency bins after the first squaring for taking total power. He made a generic simulator for dual free running clocks LNBF intensity interferometers. You dont even need to have an rtlsdr device to run it only an up to date install of gnuradio. It is an easy way to understand how to do interferometry without a distributed clock signal. With this setup on a 1 meter baseline and a intermediate tuning frequency of 1.6 GHz IF (10700 MHz(1600 MHz950 MHz) 11350 MHz) the main beamwidth would be about 70(c11GHz)1m), or 1.9 degrees. This does not resolve the solar disk ( 0.5 deg) during drift scans. I have been told that the magnitude goes down in a SINC pattern as you widen the baseline and approach resolving the source but I will not resolve the sun initially. In the VSRT Memos Development of a solar imaging array of Very Small Radio Telescopes a computationally complex way to resolve individual action regions is done with a 3rd dish providing phase closure in the array on a slanted north-south baseline in addition to the existing east-west baseline. I try to point my dishes so that the Earth is passing the sun through the beam at 12:09pm (noon) each day. To aid in pointing a cross of reflective aluminum tape is applied center of the dish. This creates a cross of light on the LNBF feed when it is in the dish focal plane and the dish is pointed at the sun. The picture below is from later in the day, the one of the left shows the sun drifting out of the beam as it sets. I made my LNBF holders out of small pieces of wood compression fit in the dish arm. There are grooves for the RG6 coax to fit ground out with a rotary tool. The PVC collars have slots cut in the back with screws going into the wood to set the angle. The screenshot shows a short run near sunset on an otherwise cloudy day. The discontinuities are me running outside and manually re-pointing the dishes. But it does highlight how the beat frequency of the 2 LNBF varies as they warm up when turned on. It starts down at 90 KHz but within 10 minutes it rises to 115 KHz. After it reaches equilibrium the variation is -1 KHz. I could change the existing 80-120 KHz bandpass to a 110-120 KHz bandpass and have better sensitivity. But that bandwidth is something that has to be found empirically with each LNBF pair and set manually within the GUI for now. patchvonbraun said it was feasible to identify the frequency bins with the most counts and that there was an example within the simplara code, You could even have a little helper function, based on a vector probe, that finds your bin range, and tunes the filter appropriately. The below close up of indoor testing showing how everything is connected on the rtlsdr side showing the power injector, e4k based rtlsdr (wrapped in aluminum tape), and the stripline based satellite power combiner for correlation. The two rg6 quadshield coaxial lines going from the power combiner to the ku band LNBF are as close to the same length as I could trim them. I use a 1 amp 18v power supply and coaxial power injector to supply power to the LNB and any amplifiers. This voltage controls linear polarization (horiztonalvertical) and it can be changed by putting a few 1 amp 1N4007 in series with the power line to drop the voltage. Accessory scripts. tp-modes. grc produces binary logs that are pretty simple. The count of the LNBF beat frequency bins in the bandpass are saved as floats represented as 4 pairs of hexadecimal. When the integration time is set to the default 1 second then one 4 byte data point is written to the log every 0.5 seconds. I highly recommend not changing this for now. There is no metadata or padding. Heres a screenshot of a run using the utility bless, In order to convert the binary logs of 4 byte records into something gnuplot can parse I use a simple perl script, Now I have the filename which gives the time the gnuradio-companion grc file started running. This is not the time I hit the record button and started logging. The offset is a second or two. Ignoring that, it is possible to use the start time encoded in the log file name to figure out when a particular measurement was taken. To do that I have to know the interval between entries saved to the binary log. To know what time a log record corresponds to, take the time from the filename and then add 0.5 seconds the index of the 4 byte entry in the binary log. This should be possible to write into the until loop so it outputs time instead of just index i. The below example is a hacky version of my log parser that does just this. Heres an example output . Now I just have to make up a good gnuplot format and integrate the calls into the perl script. Computer controlled pointing, mechanical and software differences Manually repositioning the dishes swamps out the signal of interest as the target leaves the beamwidth. For any decent measurements I need computer controlled pointing. This means the Haystack idea of two coupled Diseqc 1.2 compatible motor positioners mounted one on the other. In their design both dishes are mounted on a single PVC tube hooked to one of the positioners with a metal extension. My satellite dish mounts cant rotate like theirs so Ill have to modify this design a bit. They use a serial relay to push the buttons on a physical Diseqc 1.2 motor controller remote. That seemed a bit convoluted to me. I bought a SkyStar2 DVB-S pci card and under linux send raw Diseqc commands out by calling xdipo which accesses the linux DVB interface. It has both a GUI and cli interface. Unfortunately xdipo cannot send through Diseqc switches. I had to add manual motor commands to tune-s2 which did support switches but not manual motor commands. This version which supports manual stepping mode is available at githubsuperkuhtune-s2-stepping. Another alternative Diseqc motor controller I didnt persue would be using a 192 KHz USB soundcard and the DiSEqC Audio Generator software from Juras-Projects. The documentation for the hardware side of the audio generator is 404 now, but Juras responded to an email of mine with the schematics attached . Since the bent motor shafts that came will my motors looked really difficult to drill through I thought Id use straight hex holed shafts to make everything mechanically simpler. I found reidsupplyskuHHS-18 and ordered a couple. Unfortunately my measurements of the dish motor shaft flat-to-flate size were off. The Reid hex holed shaft hole is just a tiny bit too large. This was easily fixed by wrapping a couple turns of masking tape around the shaft to increase the diameter. This is often how fishing rod handles are made. I also encountered this construction technique on Jarrod Kinseys CO2 laser pages . The hardest part of all this is drilling an 8mm hole precisely normal to the curved outside surface of the hex hole shaft. The first step is to flatten the area with a hand file. This took me about 10 minutes. I had previously ordered and received two carbide drill bits, one small to sub-drill the intial hole and then one 8mm for the final hole. A drill press and small vice are quired to actually drill the holes. And even then its really tricky. My first two attempts resulted in holes not quite normal to the surface of the hex flat. I could only use roll or taper pins to secure the shaft. Luckily I bought 2x shafts just in case. I also had to drill 4 additional 8mm holes in the 2x satellite dish motor mounts to make holes for level mounting instead of at a tilt. The VRST guys got lucky with their sat motor mounts having a long slot. The diameter of easily available PVC is slightly to small for the dish mounting clamp. This is remedied like the motor shafts by wrapping wide masking tape to size and optionally epoxy coatingsanding it. The dish motors used in the VSRT project were Stab HH90. These have come down in cost since the VSRT memos were written and are still widely available. In order to control these motors a system to send DISEqC 1.2 commands is needed. The first option would be to faithfully replicate the VSRT implementation. They do it in a rather roundabout way but at least it is tested and known to work with their software. Unfortunately the specific hardware used has become rare, is mostly shipped from overseas, or is expensive. My chosen method of HH90 motor control is a single DVB-S card under linux with DVB API 5.x w my modified tune-s2 and optionally xdipo. This can be combined with a DiSEqC switch to scale to control of multiple motors relatively cheaply. I do sun alt-az position calculation by using a small pysolar python script. I have not yet completed the scripts to turn alt-az positions of the sun at my location into motor step commands. Hopefully I can use some of the USAL fuctions in tune-s2 for that. Both require PVC pipe, tools like drills, 8mm drill bits and smaller sub-drill bit, hand saws, files, and potentially a welder (though liberal J-B Weld would probably work). Diseqc switches problems and solutions. It turns out that xdipo alone cannot deal with motors behind Diseqc switches. This means it can only control one Diseqc motor at once. Controlling two would require 2x Skystar 2 pci cards. Luckily there are other options. CrazyCats tune-s2 supports Diseqc switches and addressing. It normally only provides for motor commands using the USAL system which isnt too helpful. But I was able to modify the code to support manual motor position commands while retaining the switch support. xdipo could still be used in theory by calling tune-s2 to set the Diseqc switch to the appropriate portmotor and then calling xdipo as normal. But it is easier to just use the modified tune-s2 for everything. This gutted version of tune-s2 for manual motor commands is available at: The functions I added are basically just look up arrays with Diseqc bus commands for different steps in the clockwise or counter-clockwise directions. In Diseqc the packets have 4 sections. Check out the Diseqc Bus Functional Specification (pdf) for a better explanation with more detail. The first, Framing byte represent if the command is from the receiver or diseqc device and wether it needs a reply. For my table these are all just EO which means its a packet from the receiver with no response required. Most commands are EO but it goes up to E7. The second, Address specifies which types of Diseqc devices should listen (ex: LNB, switch, motor, polarizer). For motors this is 32 The third is Command. This is a huge list of values of which only 68 and 69 are relevant. They are Drive Motor East and Drive Motor West respectively. The Command byte is only relevant to their specific devices specified via the Address byte. The remaining bytes of the packet are Data and how theyre interpreted depends on the Command bytes specifying a specific type of command. For motor movement there are three options. 00 makes the motor turn until a Diseqc stop command is sent. The second mode is positive values for the bytes, 01 to 7F. They represent an amount of time to turn the motor. Or by specifying negative byte values 80 to FF the motor is rotated a number of steps. This last is best and detailed in the Positioner Application Note (pdf) with an excerpt below, The number of steps to make is given by the additional count needed to make the parameter byte reach zero (or overflow to zero if the byte is considered as unsigned). Thus the byte FF (hexadecimal) requests only one step, FE two steps, and for example F9 requests 7 steps. With my motors each step corresponds to about 0.1 deg. Using this information I made up a table of Diseqc packets for each rotation direction. For addressing specific ports of the Diseqc switch tune-s2s normal functions are used. They are called before the motor position commands are sent. Usage of the modified tune-s2 is pretty simple. The only differences are two new cli switches and not needing to give it tuning parameters. They each take any value from 0 to 10 like, This would cause the satellite dish motor on port 1 of the Diseqc switch to step 1 position counter-clockwise. To send the same command of stepping 1 position counter-clockwise to the other motor, The stepping argument values 0 through 10 are mapped on a fairly arbitrary set of actual steps. This results from just doing array index look ups in the above packet tables, Calculating solar position and using that to decide how many steps to step per axis Figuring out where the sun is in the sky in terms of an alt-az format is made simple by pysolar. Figuring out how to turn that position into sequences of steps on the motors is much, much harder. These values are relative to the pysolar reference frame which is given by their diagram, My setup is pointed directly south. So for this example time that means I need to calculate the number of steps required to turn the (top) altitude motor 41.5 up from level and the (bottom) azimuth motor 35.8 degrees to the left (east). Decoding Pager Data with multimon andor gnu radio receivers The hardest part of this is figuring out what kind of pager system you have. I spent a long time trying to decode the local FLEX pager system with decoders that did not support it. Written by Thomas Sailer, HB9JNXAE4WA, multimon (multimon. tar. bz2 ) supports decoding a large number of pager modulations. FLEX is not one of them. Scroll down for FLEX . On June 29th 2012 dekar told me about his updated fork of multimon, multimonNG. with better error correction and more modulations supported. As of right this instant those on 64bit linux should just use the existing makefile and not qmake or qt-creator to compile it. For the windows users (or anyone wanting more info) theres a precompiled version and blog post. Make sure to disable all the demodulators you dont need. I think especially ZVEI is quite spammy. This and this is what pocsag sounds like if youre wondering. When I originally started playing and wrote this there were only a couple options for rtlsdr receivers to use with the multimon decoder. I used patchvonbrauns multimode to save. wavs and dekars pager example GRC I modified for OsmoSDR sources linked below for raw, real time decoding. Lately (as of late 201213) a large number of receivers have been released that dont depend on GNU Radio. rtlfm is one and theres an example usage below. real time decoding rtlfm real time decoding wdekars pagerfifo Dekar s multimonNG. a fork with improved error correction, more supported modes, and nixosxwindows support. In the screenshots below the signal is not pocsag. I thought it might be zwei but now Im not so sure its even pager data. Test samples of pocsag that Dekar links on his blog decode just fine. pagerfifoweb. grc In order to decode the pager data in real time you should use a first-in first-out file (fifo). Dekars pagerfifo is designed to do that but youll need to set the correct file paths for the File Sink yourself. In the copy downloadable here the File Sinks path is set to tmppagerfifo. raw. You should be able to run it without editing once youve made that fifo. Make sure to start multimon reading the fifo before you begin GRC and execute the receiver. In my personal copy of dekars pagerfifo the file and audio sinks are enabled while the waterfall, wav, and other sinks are disabled. To enable the disabled (grey) block select them and press e (d to disable). The audio sink is set to pulseaudio (pulse). FLEX Pagers Unfortunately it turned out my local pagers were all using FLEX. and so not supported by any of the above software. But the procedures might still be useful for someone. Decoding FLEX can be done with the software PDW, but it is windows only. In GNU Radio there is additionally gr-pager. which is supposed to support flex, but many implementation scripts for it are GNU Radio 3.6.5 or older and getting stuff to work with 3.7 requires namespace changes. mothrans flexhackrf is one of these. Since the rtlsdr receivers can but shouldnt do 3.125 MSs, like flexhackrf of uhdflux, what they use natively for the bandwidth, and so decimation, and pretty much everything else have to be re-written too. Ive attempted to start this and you can see a copy here . A couple days after I wrote the above paragraph zarya came on rtlsdr on freenode and mentioned his rtlsdr supprting FLEX decoder written months before. It is easy to use and works great This script runs at a 250 KSs sample rate and decodes 12.5 KHz channel only. Internally it uses gnuradios optfir to generate low pass taps that wide to use witih a frequency xlating FIR filter. It then passes that to gr-pagers flexdemod. later . argilo (Clayton Smith) has also put together an osmosdr source based gr. pager flex decoder for his GNU Radio tutorial series. The below output is heavily censored and edited to avoid disclosing or reproducing sensitive information but it gives you an idea of the type of messages. I tried for over a year before successfully decoding local pager signals. Now that I have I think it is a bad idea. There is far much too much private information in cleartext. I dont plan to try again. (old) gqrx install notes When I wrote this up the original version by csete didnt support the hardware yet but mathis, phirsch, Hoernchen, and perhaps others Ive missed from rtlsdr on freenode had added librtlsdr support to gqrx their repos are still listed by commented out. These days csete has added in rtlsdr support so you can use his original repository. Use with Ubuntu 10.04 and distros with old Qt lt 4.7 You will almost certainly not get this error. But, someone might, so Im leaving it here to be indexed. If youre like me and run an older distribution then your Qt libraries will be out of date and lack a function required for generating the name of the files to be saved when recording. Initially I thought it was a qtcreator thing so I tried to get more information by doing it manually, qmake make g - c - pipe - O2 - Iusrlocalincludegnuradio - Iusrlocalinclude - Iusrlocalinclude - Iusrlocalincludegnuradio - DREENTRANT - DREENTRANT - Iusrincludelibusb-1.0 - Wall - W - DREENTRANT - DQTNODEBUG - DQTNODEBUGOUTPUT - DVERSION0.0 - DQTNODEBUG - DQTGUILIB - DQTCORELIB - DQTSHARED - Iusrshareqt4mkspecslinux-g - I. - Iusrincludeqt4QtCore - Iusrincludeqt4QtGui - Iusrincludeqt4 - I. - I. - o dockaudio. o qtguidockaudio. cpp qtguidockaudio. cpp: In member function void DockAudio::onaudioRecButtonclicked(bool): qtguidockaudio. cpp:100: error: currentDateTimeUtc is not a member of QDateTime make: dockaudio. o Error 1 To get it to compile on these systems youll have to do the below. (edit: This little change is now added into phirschs .) Ubuntu 10.04 has old Qt libs and gqrx uses a function call not in them. So, while I was waiting for Qt 4.74 to compile I decided to try a hack. I removed that function call with a static string of text. edit I later found comparable functions for Qt 4.6 and older. If you are using qtcreator like the docs suggest you can double click on the error and go to the line. If not, it was in. Sourcesqtguidockaudio. cpp replace, With something like this. And itll compile and run correctly on my specific machine. Compiling LTE Cell Scanner and LTE Tracker on Ubuntu 10.04 Before starting make sure to have a fortran compiler, FFTW, BLAS, and LAPACK libraries installed from the repositories. If youre using 12.04 just follow the instructions on the github page and everything is trivial. For 10.04 (lucid) users the the initial hurdle is cmake. LTE Cell Scanner requires cmake 2.8.8 and Ubuntu 10.04 only has 2.8 the finding of BLAS and LAPACK libraries will fail like, Until you open CMakeList. txt and change the version number on first line to 2.8.0. After fixing that the BLAS and LAPACK issues come in, You can see my installation notes before I figured it out. To fix it I searched for people complaining of similar problems on other projects and then replaced my system files with theirs, FindBLAS. cmake . LAPACK will also fail this way. I used this arbitray cmake file, code. googlepqmcpacksourcebrowsetrunkCMakeFindLapack. cmaker5383. And this is a local backup in case that disappears. After fixing the cmake issues compile and install the latest IT (ITPP 4.2). Make sure to completely remove the old ITPP 4.0.7 libraries from the Ubuntu repository. When LTE Cell scanner compiles you can go back and restore the. bak cmake files. The rate of scan is about 0.1 Mhz per 10 seconds. Both positive and negative frequency offsets happen, but rarely in the same dongle. LTE Tracker I havent used as much yet (recently released ) but it is included in the github repository cloned initially and should be compiled as well if you did the above. Check out the authors site for videos of its use since an ascii paste of the ncurses like interface wouldnt tell you much. Tapi. the start looks like this, Slightly altered GNU Radio Companion flowcharts The GUI stuff in Gnu Radio was rather an afterthought. Nobody really expected that youd use it to build actual applications, but rather just use it as a way of making test jigs for your signal flows. This section is my notes on how I made basic examples work, and how I edited those examples in very simple and often broken ways. Also, since gqrx, multimode, and other intergrated receivers came out I dont see any need to update these as things change. Most of this is very old. While there are links to the originals in the summaries, these descriptions are of the versions modified by me usually just sample rate and GUI stuff. While the sample rate or tuner width I set may be some large number, itll become obvious what the limits of each other as you scan about and see the signal folding or mirroring. Using sample rates above 2.4 MSs with rtlsdr is not recommended. It does create aliases all over. If youre using GNU Radio 3.7 dont even bother trying with any. grc files hosted here. FM: patchvonbrauns simple (stereo) fm receiver - harder setup, best reception, best sound, 2.048 MSs, -600Khz fine tune lindis fm receiver easy setup, good reception, good sound, 3.2 MSs, -600Khz fine tune superkuhs offset fm receiver - easy setup, okay reception, okay sound, 2.8 MSs, -900Khz tune, -50Khz fine. 2h20s beginner fm (mono) receiver - easy to understand, easy setup, okay sound, 2.8 MSs, no fine tune SSB: OZ9AECs SSB Receiver - SSB rx and record to disk, seperate playback script. 1 MSs, -1k fine tune. If it comes with a python file, try that first before generating one from the GRC file. When tuning, make sure to hit enter again if it doesnt work the first time or tunes to the wrong frequency. Always hit autoscale to start, and for FFT displays try using the average settings. I have set all audio sinks to pulse (pulseaudio) instead of say, hw:0,0 (ALSA). You might have to change that. To get a list of hardwareuse aplay - l. Thatll show the various cards and devices. Use the format, hw:X, Y where hw:CARDX, DEVY. Some flowcharts have variables for it, others put it directly in the Audio Sink element. If you hear something interesting you can try comparing it to indentified samples from kb9ukddigital or hfradio. org. ukhtmldigitalmodes. html. or the windows program, Signals Analyzer. Check radioreference or wireless2.fcc. govUlsAppUlsSearchsearchAdvanced. jsp to see whats in the USA area at a given frequency. Multiple Dongles There are two ways to specify the use of multiple dongles. The first, correct, way is to set the Num Channels in the OsmoSDR Source block to 2 and then specify the device IDs in Device Arguments like, rtl0 rtl1. Each specified device is seperate with a space from the previous one. The not so correct but still working way is to use multiple OsmoSDR Source blocks with Num Channels set to 1 and each with its respective Device Arguments field set to rtl0 or rtl1, or so on. The OsmoSDR Source block has extensive help files at the bottom of its properties if you scroll down. To enable a block, select it and press e. To disable a block select it and press d. When disabled blocks will appear darker gray. If you open a. grc file and it looks like there are blocks missing (red error highlights and no connections between them) then it is likely the name of the block changed during some GNU Radio update. If your install is more than a month or two old this often happens. Update GNU Radio. Its easier to type in 1e6 than 1000000 so use scientific notation when you can in variable fields. If you double click on an element in a flowchart it usually includes a helpful Documentation: of most the variables to be set at the bottom. The GUI element grid position is a set of two pairs of numbers: y, x,a, b where the first pair y, x, is position (y row, x column) and a, b is the span of the box. If you enter a Grid Position and it overlaps with another element itll turn red and report the error and where the origin is of the element it overlaps with. The tab effect is done with notebooks. For RTL2832 Source the minimum sample rate is 800KSs, its gr-baz() and generally not updated. Use OsmoSDR source. Its under OsmoSDR, not Sources on the right panel). It has a 1MSs minimum sample rate. Its not recommended to use sample rates above 2.4M. In older versions of gr-osmosdr and rtl-sdr I think automagic gain control (AGC) was on all the time so you didnt have to set the gain explicitly in the source in GRC. New versions require that and also require setting the chan 0. freq to something. The dongles seem to have noise at their 0Hz center frequency so the best performance is from selecting a band 100-200Khz offset from the center (depending on signal type). patchvonbrauns simplefmrcv is a great example of that. patchvonbrauns simplefmrcv (this summary is outdated) The best sounding software Ive found for listening to FM is patchvonbrauns Simple FM (Stereo) Receiver. I dont think it is very simple it includes many advanced FM specific features like extraction of the 19k (pilot) tone next to some commercial FM broadcasts. It used to do RDS, I hear, and older versions checked into CGRAN still have it, but it is removed for simplicitly in this version. lindis FM receiver Original: lindi. iki. filindignuradiortl2832-cfile-lindi-fm. grc. this was an example posted to rtlsdr by lindi. It used a file source which was decoded to wav and saved to disk. Seen in the screenshot above. Modified: superkuhrtl2832-cfile-lindi-fmedit. grc. had a frontend GUI and an increased sample rate. Right now the rate of the audio files saved out is. not very useful. But it sounds fine. Seen below. 3.2 MSs field of view, tune -900Khz 2h20s simple fm receiver 2.8 MSs field of view, no fine tuning. 2h20 made available. with thorough explaination, a bare bones FM (mono) receiver to learn how to use GNU Radio. This was the first one I managed to get to work. Because the original h202s uses the RTL2832 Source and not the OsmoSDR Source you might experience tuner crashes if you scan too quickly. Make sure to unreplug in the dongle after these. Its best just to enter the frequency as a number. Be aware this section of this page was written many months ago when rtl-sdr was different and I had little idea of what I was doing. xzero has since manually added signal seeking to this example . My edit of 2h20s simple receiver does not add much, but I did replace the RTL2832 source with an OsmoSDR source to avoid tuner crashes. I also increased the sample rate to 2.8MSs (to see more spectrum) and then increased the decimation in the filter from 4 to 8 to compensate so everything still decodessounds right. I also remove the superfluous throttle block. my offset tuning recording example 2.8 MSs field of view, -900Khz tuning, -50Khz fine. This takes parts from a bunch of the other example receivers and repurposes them in presumably incorrect but seemingly working ways. It is a basic example of how to offset the tuner 200khz away from the center to avoid the noise there. I started with 2h20s simple tuners GUI framework and removed almost all of the content. I copied, with inaccurate trial and error changes of sample rate and filter offset, sections of the offset tuning and other advanced bits from simplefmrcv and wfmrx. grc. The tuner is tuned 200Khz. The freqxlating filter is tuned 200Khz. The the bandpass filter is specified in a variable, The net result is that the frequency of interest comes out of the tuner 200Khz below DC, and the freqxlater lifts it up by 200Khz, and then its bandpassed. I also blindly copied the RF power display, a toggle for saving the audio files out to disk, and a -900khz tuning slider from other receivers. I added a second fine tune -50Khz. This is done by setting the frequency of the Xlating FIR filter to, where freqoffset is the frequency offset from center (200Khz in this case), fine is the ID of a wx gui slider for regular tuning, and finer is the same for fine tuning. In order for the frequency display to show the proper value it was correspondingly set to a variable ID curfreq, I also made the current frequency display a editable text field so you can tune, copy, and paste. There are good examples of how notebook positioning works and includes simple scripting examples for the file field. This flowchart is simple enough to learn from but includes many elements pulled out of the very complex simplefmrcv from patchvonbraun. Without his explanation of the offset process I wouldnt have figured it out. All blocks are layed out by type and GUI elements in the same order as they appear when run. This should help you figure out Grid and Notebook positioning. The sound is only okay. I think the signal is being clipped off at the edges a little bit. I am not sure if it is required to install patchvonbrauns simplefmrcv to use this, I do use some of his custom filter stuff. Use the Waterfall for scanning through channels. Once located, look at the offset from 0 on the bottom display. Use that to set the tuning (and fine) slider and wiggle it till you get the signal crossing the band in the Second Filter top display. Switch to FFT view and look at the bottom First Filter display, use tuning and fine tuning to center the peak on the First Filter display. Or the other way around. Its personal preference. Ignore the noise you see at higher frequencies (900Mhz) at 0.2Mhz baseband. Although sometimes it gets folded in depending on tuning. SSB Receiver and data Recorder Created by Alexandru Csete OZ9AEC the notes say, Simple SSB receiver prototype. This comes from the GNU Radio GRC examples repository over at githubcsetegnuradio-grc-examplestreemasterreceiver I changed the way it saves samples for the sister decoder program by adding automatic generation of file names and an onoff tickbox toggle for recording. You might want to change the default directory by editing the variable prefix. The key was in the File Sink file field. I also changed the GUI so it was easier to find signals. Use the saved. bin files with ssbrxplay to hear. Jumping around in frequency is a lot smoother when reading from disk instead of the dongle. How to use rtlsdr and hackrf on a fresh odroid-u3 with ubuntu Related Pages Page Sections Type, sayYour message here. after the end of any URL on my site and hit enter to leave a comment. You can view them here. An example would be, superkuhrtlsdr. htmlsayThis is a comment.