• Stars
    star
    137
  • Rank 266,121 (Top 6 %)
  • Language
    C++
  • License
    GNU General Publi...
  • Created over 5 years ago
  • Updated 4 months ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Software decoder for FM/AM broadcast radio with AirSpy R2 / Mini, Airspy HF+, and RTL-SDR

airspy-fmradion

  • Version 20230529-0
  • For MacOS (both Intel and Apple Silicon) and Linux

Contributing

See CONTRIBUTING.md for the details.

Known issues and changes

Please read CHANGES.md before using the software.

What is airspy-fmradion?

  • airspy-fmradion is a software-defined radio receiver for FM and AM broadcast radio, and also NBFM/DSB/USB/LSB/CW/WSPR utility communications, specifically designed for Airspy R2, Airspy Mini, Airspy HF+, and RTL-SDR.

What does airspy-fmradion provide?

  • Mono or stereo decoding of FM broadcasting stations
  • Mono decoding of AM stations
  • Decoding NBFM/DSB/USB/LSB/CW/WSPR stations
  • Buffered real-time playback to soundcard or dumping to file
  • Command-line interface (only)

Usage

# Portaudio output
airspy-fmradion -t airspy -q \
    -c freq=88100000,srate=10000000,lgain=2,mgain=0,vgain=10 \
    -b 1.0 -P -

# 16-bit signed integer WAV output (pipe is not supported)
airspy-fmradion -t airspyhf -q \
    -c freq=88100000,srate=768000 \
    -b 1.0 -W output_s16_le.wav

# 32-bit float WAV output (pipe is not supported)
airspy-fmradion -m am -t airspyhf -q \
    -c freq=666000 \
    -b 0.5 -G output_f32_le.wav

airspy-fmradion requirements

For the latest version, see https://github.com/jj1bdx/airspy-fmradion

Recommended utilities

Branches and tags

  • Official releases are tagged
  • main is the "production" branch with the most stable release (often ahead of the latest release though)
  • dev is the development branch that contains current developments that will be eventually released in the main branch
  • Other branches are experimental (and presumably abandoned)
  • The master branch is deprecated and to be deleted. Use main branch.

Prerequisites

Airspy HF+ firmware

Use the latest version of Airspy HF+ firmware, available at Airspy HF+ Dual Port and Airspy HF+ Discovery Web pages.

airspy-fmradion sets the default sampling rates to 384kHz for FM broadcast, and 192kHz for the other modes. Old Airspy HF+ firmwares do not support the lower sampling rate other than 768kHz.

Required libraries

Note: the main (formerly master) branch of libvolk is now required from v0.8.1.

If you install from source in your own installation path, you have to specify the include path and library path. For example if you installed it in /opt/install/libairspy you have to add -DAIRSPY_INCLUDE_DIR=/opt/install/libairspy/include -DAIRSPYHF_INCLUDE_DIR=/opt/install/libairspyhf/include to the cmake options.

Debian/Ubuntu Linux

  • sudo apt-get install cmake pkg-config libusb-1.0-0-dev libasound2-dev libairspy-dev libairspyhf-dev librtlsdr-dev libsndfile1-dev portaudio19-dev

macOS

brew tap pothosware/homebrew-pothos
brew tap dholm/homebrew-sdr #other sdr apps
brew update
brew install portaudio
brew install libsndfile
brew install rtl-sdr
brew install airspy --HEAD
brew install airspyhf --HEAD
brew install volk

Install the latest libvolk

Install libvolk as described in INSTALL-latest-libvolk.md.

Dependency installation details

libvolk

libairspyhf

Use the latest HEAD version.

libairspy

Note: this is applicable for both macOS and Linux.

Install and use the latest libairspy --HEAD version for:

  • Working airspy_open_devices(), required by airspy_open_sn(). See this commit for the details.
  • Proper transfer block size. if_blocksize for Airspy HF+ is reduced from 16384 to 2048, following this commit.

git submodules

r8brain-free-src and readerwriterqueue are the submodules of this repository. Download the submodule repositories by the following git procedure:

  • git submodule init
  • git submodule update

Installing

To install airspy-fmradion, download and unpack the source code (as well as the r8brain-free-src submodule) and go to the top level directory. Then do like this:

  • mkdir build
  • cd build
  • cmake ..

CMake tries to find librtlsdr. If this fails, you need to specify the location of the library in one the following ways:

cmake .. \
  -DAIRSPY_INCLUDE_DIR=/path/airspy/include \
  -DAIRSPY_LIBRARY_PATH=/path/airspy/lib/libairspy.a
  -DAIRSPYHF_INCLUDE_DIR=/path/airspyhf/include \
  -DAIRSPYHF_LIBRARY_PATH=/path/airspyhf/lib/libairspyhf.a \
  -DRTLSDR_INCLUDE_DIR=/path/rtlsdr/include \
  -DRTLSDR_LIBRARY_PATH=/path/rtlsdr/lib/librtlsdr.a

PKG_CONFIG_PATH=/path/to/airspy/lib/pkgconfig cmake ..

For using static analyzers such as OCLint and Clangd, run the following commands:

cd build
ln -s `pwd`/compile_commands.json ..
cmake -DCMAKE_EXPORT_COMPILE_COMMANDS=ON ..

Compile and install

  • make -j4 (for machines with 4 CPUs)
  • make install

Copying binary to another directory

On M1 Mac, using cp causes a trouble. Use the following command to properly install the command to a local directory:

install -o user -m 0700 -c -s build/airspy-fmradion $(HOME)/bin

Basic command options

  • -m devtype is modulation type, one of fm, nbfm, am, dsb, usb, lsb, cw, wspr (default fm)
  • -t devtype is mandatory and must be airspy for Airspy R2 / Airspy Mini, airspyhf for Airspy HF+, rtlsdr for RTL-SDR, and filesource for the File Source driver.
  • -q Quiet mode.
  • -c config Comma separated list of configuration options as key=value pairs or just key for switches. Depends on device type (see next paragraph).
  • -d devidx Device index, 'list' to show device list (default 0)
  • -M Disable stereo decoding
  • -R filename Write audio data as raw S16_LE samples. Use filename - to write to stdout
  • -F filename Write audio data as raw FLOAT_LE samples. Use filename - to write to stdout
  • -W filename Write audio data as RF64/WAV S16_LE samples. Use filename - to write to stdout (pipe is not supported)
  • -G filename Write audio data as RF64/WAV FLOAT_LE samples. Use filename - to write to stdout (pipe is not supported)
  • -P device_num Play audio via PortAudio device index number. Use string - to specify the default PortAudio device
  • -T filename Write pulse-per-second timestamps. Use filename '-' to write to stdout
  • -b seconds (ignored, remained for a compatibility reason)
  • -X Shift pilot phase (for Quadrature Multipath Monitor) (-X is ignored under mono mode (-M))
  • -U Set deemphasis to 75 microseconds (default: 50)
  • -f Set Filter type
    • for FM: wide and default: none, medium: +-156kHz, narrow: +-121kHz
    • for AM: wide: +-9kHz, default: +-6kHz, medium: +-4.5kHz, narrow: +-3kHz
    • for NBFM: wide: +-20kHz, default: +-10kHz, medium: +-8kHz, narrow: +-6.25kHz
  • -l dB Enable IF squelch, set the level to minus given value of dB
  • -E stages Enable multipath filter for FM (For stable reception only: turn off if reception becomes unstable)
  • -r ppm Set IF offset in ppm (range: +-1000000ppm) (Note: this option affects output pitch and timing: use for the output timing compensation only!
  • -A (For FM only) Experimental 10Hz-step IF AFC

Major changes

Timestamp file format

  • For FM: pps_index sample_index unix_time if_level
  • For the other modes: block unix_time if_level
  • if_level is in dB

Rate compensation for adjusting audio device playback speed offset

  • Background: some audio devices shows non-negligible offset of playback speed, which causes eventual audio output buffer overflow and significant delay in long-term playback.
  • How to fix: compensating the playback speed offset gives more accurate playback timing, by sacrificing output audio pitch accuracy. A proper compensation will eliminate the cause of increasing output buffer length, by sending less data (lower sampling rate) to the conversion process.
  • You can specify the compensation rate by ppm using -r option.
  • How to estimate the rate offset: when elapsed playback time is Tp [seconds] and output buffer length (buf= in the debug output) increases during the time is Ts [seconds], the compensation rate is (Ts/Tp) * 1000000 [ppm].
  • For example, if the output buffer length increases for 1 second after playing back for 7 hours (25200 seconds), the offset rate is 1/25200 * 1000000 ~= 39.68ppm.
  • +- 100ppm offset is not uncommon among the consumer-grade audio devices.
  • +- 100ppm pitch change may not be recongizable by human.

Caveats for the rate compensation

  • Do not use this feature if the per-sample accuracy is essential.
  • Do not use this feature for non-realtime output (for example, to files).
  • Output audio pitch increases as the offset increases.
  • Too much compensation will cause output underflow.
  • This feature causes fractional (non-integer) resampling by IfResampler class, which causes more CPU usage.

Smaller latency

  • v0.9.2 uses smaller latency algorithms for all modulation types and filters. The output frequency characteristics may be different from the previous versions.

Audio gain adjustment

  • Since v0.4.2, output maximum level is back at -6dB (0.5) (adjust_gain() is reintroduced) again, as in pre-v0.2.7
  • During v0.2.7 to v0.4.1, output level was at unity (adjust_gain() is removed)
  • Before v0.2.7, output maximum level is at -6dB (0.5)

Audio and IF downsampling is performed by r8brain-free-src

  • Output of the stereo decoder is downsampled to 48kHz
  • 19kHz cut LPF implemented for post-processing resampler output
  • Audio sample rate is fixed to 48000Hz
  • r8b::CDSPResampler24 is used for IF resampling

Phase discriminator uses GNU Radio fast_atan2f()

  • From v0.7.8-pre0, GNU Radio fast_atan2f() which has ~20-bit accuracy, is used for PhaseDiscriminator class and the 19kHz pilot PLL.
  • The past fastatan2() used in v0.6.10 and before was removed due to low accuracy (of ~10 bits)
  • Changing from the past atan2() to fast_atan2f() showed no noticeable difference of the THD+N (0.218%) and THD (0.018%). (Measured from JOBK-FM NHK Osaka FM 88.1MHz hourly time tone 880Hz, using airwaves after the multipath canceler filter of -E36)
  • The past fastatan2() allowed +-0.005 radian max error
  • libm atan2() allows only approx. 0.5 ULP as the max error for macOS 10.14.5, measured by using the code from "Error analysis of system mathematical functions " by Gaston H. Gonnet (1 ULP for macOS 64bit double = 2^(-53) = approx. 10^(15.95))

FM multipath filter

  • A Normalized LMS-based multipath filter can be enabled after IF AGC
  • IF sample stages can be defined by -E options
  • Reference amplitude level: 1.0
  • For Mac mini 2018 with 3.2 GHz Intel Core i7, 288 stages consume 99% of one CPU core
  • This filter is not effective when the IF bandwidth is narrow (192kHz)
  • The multipath filter starts after discarding the first 100 blocks. This change is to avoid the initial instability of Airspy R2.
  • Note: this filter recalculates the coefficients for every four (4) samples, to reduce the processing load.

Multipath filter configuration

  • v0.7.3 and later: -E36 for 108 previous and 36 after stages (ratio 3:1). The multipath filter order: (4 * stages) + 1
  • For reference only: v0.7.3-pre1 and before: -E72 for 72 previous and 72 after stages (ratio 1:1), summary: set the -E parameter to 1/2 of the previous value
  • Rule of thumb: -E36 is sufficient for a stable strong singal (albeit with considerable multipath distortion).

FM L-R signal boosted for the stereo separation improvement

  • Teruhiko Hayashi suggested boosting L-R signal by 1.017 for a better stereo separation. Implemented since v0.7.6-pre3.
  • DiscriminatorEqualizer removed since v1.7.6-pre3 (needs more precise compensation, presumably with an FIR filter.

FM deemphasis error prevention

  • Teruhiko Hayashi suggested applying deemphasis before the sampling rate conversion, at the demodulator rate, higher than the audio output rate. Implemented since v0.7.6.

No-goals

  • CIC filters for the IF 1st stage (unable to explore parallelism, too complex to compensate)
  • Using lock-free threads (boost::lockfree::spsc_queue didn't make things faster, and consumed x2 CPU power)

Filter design documentation

General characteristics

  • If resampling converter affects the passband characteristics

For FM

  • FM Filter coefficients are listed under doc/filter-design

For NBFM

  • Deviation: normal +-8kHz, for wide +-17kHz
  • Output audio LPF: flat up to 4kHz
  • NBFM Filter coefficients are listed under doc/filter-design
  • default filter width: +-10kHz
  • Narrower filters by -f options: middle +-8kHz, narrow +-6.25kHz
  • Wider filters by -f options: wide +-20kHz (with wider deviation of +-17kHz)
  • Audio gain reduced by -3dB to prevent output clipping

For AM and DSB

  • AM Filter coefficients are listed under doc/filter-design
  • default filter width: +-6kHz
  • Narrower filters by -f options: middle +-4.5kHz, narrow +-3kHz
  • Wider filters by -f options: wide +-9kHz

For SSB (USB/LSB)

  • For USB: shift down 1.5kHz -> LPF -> shift up 1.5kHz
  • For USB: shift up 1.5kHz -> LPF -> shift down 1.5kHz
  • Rate conversion of 48kHz to 12kHz and vice versa for the input and output of LPF
  • 12kHz sampling rate LPF: flat till +-1200Hz, -3dB +-1320Hz, -10dB +-1370Hz, -58.64dB at +-1465Hz

For CW

  • Zeroed-in pitch: 500Hz
  • Filter width: +- 100Hz flat, +-250Hz for cutoff
  • Uses downsampling to 12kHz for applying a steep filter

For WSPR

  • Set USB TX carrier frequency for reception (as shown in WSPRnet)
  • Pitch: 1500Hz
  • Filter width: +- 100Hz flat, +-250Hz for cutoff
  • Uses downsampling to 12kHz for applying a steep filter

AGC algorithms

Simple AGC with Tisserand-Berviller Algorithm

  • Reference: Etienne Tisserand, Yves Berviller. Design and implementation of a new digital automatic gain control. Electronics Letters, IET, 2016, 52 (22), pp.1847 - 1849. ff10.1049/el.2016.1398ff. ffhal-01397371f
  • Implementation reference: https://github.com/sile/dagc/
  • Implemented for IF AGC since 20220808-0
  • Implemented for AF AGC since 20220808-3

Airspy R2 / Mini modification from ngsoftfm-jj1bdx

Feature changes

  • Finetuner is removed (Not really needed for +-1ppm or less offset)

Airspy R2 / Mini configuration options

  • freq=<int> Desired tune frequency in Hz. Valid range from 1M to 1.8G. (default 100M: 100000000)
  • srate=<int> Device sample rate. list lists valid values and exits. (default 10000000). Valid values depend on the Airspy firmware. Airspy firmware and library must support dynamic sample rate query.
  • lgain=<x> LNA gain in dB. Valid values are: 0, 1, 2, 3, 4, 5, 6, 7, 8 ,9 ,10, 11 12, 13, 14, list. list lists valid values and exits. (default 8)
  • mgain=<x> Mixer gain in dB. Valid values are: 0, 1, 2, 3, 4, 5, 6, 7, 8 ,9 ,10, 11 12, 13, 14, 15, list. list lists valid values and exits. (default 8)
  • vgain=<x> VGA gain in dB. Valid values are: 0, 1, 2, 3, 4, 5, 6, 7, 8 ,9 ,10, 11 12, 13, 14, 15, list. list lists valid values and exits. (default 0)
  • antbias Turn on the antenna bias for remote LNA (default off)
  • lagc Turn on the LNA AGC (default off)
  • magc Turn on the mixer AGC (default off)

Airspy HF+ modification from airspy-fmradion v0.2.7

Sample rates and IF modes

  • This version supports 768kHz zero-IF mode and 384/256/192kHz low-IF modes

Conversion process for 768kHz zero-IF mode

  • LPFIQ is single-stage
  • IF center frequency is down Fs/4 than the station frequency, i.e: when the station is 76.5MHz, the tuned frequency is 76.308MHz
  • Airspy HF+ allows only 660kHz alias-free BW, so the maximum alias-free BW for IF is (660/2)kHz - 192kHz = 138kHz

Conversion process for 384/256/192kHz low-IF mode

  • IF center frequency is not shifted

Airspy HF configuration options

  • freq=<int> Desired tune frequency in Hz. Valid range from 0 to 31M, and from 60M to 240M. (default 100M: 100000000)
  • srate=<int> Device sample rate. list lists valid values and exits. (default 384000). Valid values depend on the Airspy HF firmware. Airspy HF firmware and library must support dynamic sample rate query.
  • hf_att=<int> HF attenuation level and AGC control.
    • 0: enable AGC, no attenuation
    • 1 - 8: disable AGC, apply attenuation of value * 6dB

RTL-SDR

Sample rate

  • Valid sample rates are from 1000000 to 1250000 [Hz]
  • The default value is 1200000Hz

RTL-SDR configuration options

  • freq=<int> Desired tune frequency in Hz. Accepted range from 10M to 2.2G. (default 100M: 100000000)
  • gain=<x> (default auto)
    • auto Selects gain automatically
    • list Lists available gains and exit
    • <float> gain in dB. Possible gains in dB are: 0.0, 0.9, 1.4, 2.7, 3.7, 7.7, 8.7, 12.5, 14.4, 15.7, 16.6, 19.7, 20.7, 22.9, 25.4, 28.0, 29.7, 32.8, 33.8 , 36.4, 37.2, 38.6, 40.2, 42.1, 43.4, 43.9, 44.5, 48.0, 49.6
  • srate=<int> Device sample rate. valid values in the [900001, 3200000] range. (default 1152000)
  • blklen=<int> Device block length in bytes (default RTL-SDR default i.e. 64k)
  • agc Activates device AGC (default off)
  • antbias Turn on the antenna bias for remote LNA (default off)

File Source driver

  • Reads an IQ signal format file and decode the output.
  • This device is still experimental.

File Source configuration options

  • freq=<int> Frequency of radio station in Hz.
  • srate=<int> IF sample rate in Hz.
  • filename=<string> Source file name. Supported encodings: FLOAT, S24_LE, S16_LE
  • zero_offset Set if the source file is in zero offset, which requires Fs/4 IF shifting.
  • blklen=<int> Set block length in samples.
  • raw Set if the file is raw binary.
  • format=<string> Set the file format for the raw binary file. Supported formats: U8_LE, S8_LE, S16_LE, S24_LE, FLOAT

Authors and contributors

  • Joris van Rantwijk, primary author of SoftFM
  • Edouard Griffiths, F4EXB, primary author of NGSoftFM (no longer involving in maintaining NGSoftFM)
  • Kenji Rikitake, JJ1BDX, maintainer/developer/experimenter
  • András Retzler, HA7ILM, for the former AF/IF AGC code in csdr
  • Youssef Touil, Airspy Founder, aka Twitter @lambdaprog, for the intriguing exchange of Airspy product design details and the technical support
  • Iowa Hills Software, for their FIR and IIR filter design tools
  • Brian Beezley, K6STI, for his comprehensive Web site of FM broadcasting reception expertise and the idea of Quadrature Multipath Monitor
  • Ryuji Suzuki, for reviewing the FM multipath filter coefficients and suggesting putting more weight on picking up more previous samples from the reference point than the samples after
  • Teruhiko Hayashi, JA2SVZ, the creator of FM FPGA Tuner popular in Japan, for reviewing the measurement results of FM broadcast reception of airspy-fmradion, and various constructive suggestions
  • Takehiro Sekine, for suggesting using GNU Radio's VOLK for faster calculation, and implementing Filesource device driver
  • Takeru Ohta, for his Rust implementation of Tisserand-Berviller AGC algorithm
  • Cameron Desrochers, for his readerwriterqueue implementation of a single-producer-single-consumer lock-free queue for C++

License

  • As a whole package: GPLv3 (and later). See LICENSE.
  • csdr AGC code: BSD license.
  • Some source code files are stating GPL "v2 and later" license.

Repository history

More Repositories

1

sfmt-erlang

sfmt-erlang: SIMD-oriented Fast Mersenne Twister (SFMT) for Erlang
Erlang
70
star
2

sshrpc

Erlang SSH RPC module (experimental)
Erlang
39
star
3

tinymt-erlang

This software is no longer maintained. For archive/reference use only. -- Tiny Mersenne Twister (TinyMT) for Erlang
Erlang
26
star
4

erltrek

An Erlang Star Trek game
Erlang
16
star
5

arduino-atmega16u2

Arduino ATmega16U2 USB chip firmware for DFU and USB Serial
C
11
star
6

macos-audio-streaming-tips

macOS to Linux audio streaming tips including PyAudio, PortAudio, gstreamer, ffmpeg
Python
11
star
7

dbskkd-cdb

SKK dictionary server based on cdb
C
11
star
8

avrhwrng

Hardware Random Number Generator with Arduino
C
10
star
9

emprng

Erlang/OTP rand module prototype repository until OTP 20 (for reference use only) - multiple PRNG suite
Erlang
9
star
10

rtl_power-fm-multipath

FM broadcasting multipath distortion estimation by D/U ratio measurement with rtl_power
Python
9
star
11

bsdtrek

BSD Star Trek game
C
8
star
12

exs64

Xorshift64star for Erlang
Erlang
7
star
13

WWV

WWV/WWVH emulator. Full format including 100 Hz timecode. Speech by local OS's synthesizer; OSX is best (originally forked from ka9q/WWV, but the original is gone)
C
7
star
14

nsplit

Erlang mapping function library splitting a job to n-process concurrent execution
Erlang
7
star
15

v6hex

IPv6 hex address library for Erlang
Erlang
6
star
16

neug

An unofficial clone of FST NeuG
C
6
star
17

ent

ENT - A Pseudorandom Number Sequence Test Program
C
6
star
18

freebsd-dev-trng

An entropy injection device driver for FreeBSD
C
6
star
19

wspr

WSPR for Lubuntu/Ubuntu
Python
6
star
20

xorshiftplus-c

Xorshift*/Xorshift+ public domain C code archive
C
6
star
21

ads-b-analysis

dump1090 data analysis scripts in Python and Geographiclib
Python
5
star
22

wspr-cui

WSPR command-line tools
C
5
star
23

chopstx

STM32F0 / STM32F103 thread library chopstx: unofficial private clone
C
5
star
24

sd-erlang-public

Software Design誌連載「Erlangで学ぶ並行プログラミング」のサポートページ
Erlang
5
star
25

plug_static_ls

Serving directory Index for Plug/Phoenix Static Assets
Elixir
5
star
26

bearfort

Project Bearfort repository for Erlang Factory SF Bay 2016
C
4
star
27

exsplus128

Xorshift128plus for Erlang
Erlang
4
star
28

xorshiftplus

Archive of Sebastiano Vigna's xorshift+ and xorshift*
C
4
star
29

infnoise-windows

Infinity Noise TRNG Program for Windows and Visual Studio
C
4
star
30

exs1024

Xorshift1024star for Erlang
Erlang
4
star
31

apresse

APRS software for Erlang and Elixir (for Code BEAM STO 2018)
JavaScript
4
star
32

sfmt-extstate

SIMD-oriented Fast Mersenne Twister code, rewritten to handle internal state table explicitly
C
4
star
33

dump1090-airspy-experimental

Dump1090 is a Mode S decoder for Software Defined Radio (SDR) devices including RTL-SDR and Airspy.
C
4
star
34

TinyMT

Tiny Mersenne Twister original code with TinyMT-Jump
C++
4
star
35

soxr-code

Sox Resampling Library (forked)
C
4
star
36

tinymtdc-longbatch

TinyMT pre-calculated parameter list. See https://zenodo.org/record/5834954 for an alternative archive.
Shell
4
star
37

openbox-clearlooks-solarized

Solarized-colored Clearlooks Theme for Openbox
3
star
38

FourQlib

unofficial clone of Microsoft FourQlib
C
3
star
39

dot-kerlrc

Configuration files for kerl
Shell
3
star
40

erlang-freebsd-port

Tentative private FreeBSD Erlang Port for 18.2
Makefile
3
star
41

sudoku-erlang

Sudoku solver in Erlang
Erlang
3
star
42

infnoise-freebsd

Infinity Noise TRNG Program for FreeBSD
C
3
star
43

avrdice

Arduino Uno R3 C++ code showing random dice number sequence
C++
3
star
44

plug_url_redirector

URL redirector for Plug/Phoenix
Elixir
2
star
45

luatrek

Star Trek in Lua
Lua
2
star
46

cmoy-bdx

Chu Moy headphone amplifier schematics and etc.
2
star
47

sphinxconjp2014-examples

LaTeX documentation example for SphinxCon JP 2014
Python
2
star
48

bash-3.2-osx-fix

OS X bash-3.2 fix (with import-functions patch)
C
2
star
49

fraucheky

GNU GPL container for embedded devices
C
2
star
50

cowboy_beam_status

An Erlang Cowboy example program
Erlang
2
star
51

freebsd-gnu-libiconv-hack

This port is for FreeBSD 10 only. Not applicable for FreeBSD 11. -- GNU libiconv LD_PRELOAD hack for solving possible i18n/m17n problems in FreeBSD 10 and later
Shell
2
star
52

tew

This archive is no longer maintained - Tokyo Erlang Workshop public repository for Kenji Rikitake
1
star
53

dj1yfk-dxcc

DJ1YFK's DXCC Lookup Utility (forked)
Perl
1
star
54

tqsllib-freebsd-port

FreeBSD port for tqsllib (now WITHOUT GUI tqsl/tqslcert)
C++
1
star
55

indexme

links files into an HTML index for viewing and easy overview
Perl
1
star
56

kichijoji.pm-19

Kichijoji.pm 19 presentation, 2-AUG-2019
XS
1
star
57

efsfbay2015-slides

Erlang Factory SF Bay 2015 slides
1
star
58

infnoise-linux

Infinity Noise TRNG Program for Linux
C
1
star
59

sphinx-latex-table-test

Sphinx LaTeX multirow and multicolumn table creation bug example
TeX
1
star
60

emacs-gift-generic-mode

Emacs Moodle GIFT file format Generic Mode
Emacs Lisp
1
star
61

oueees-201606-public

Slides and figures of my talks on computer networks and information security, at Osaka University, {14, 21, 28} -JUN-2016
1
star
62

euc2016-erlang-prng

Erlang User Conference 2016 presentation source for "Fifteen Ways to Leave Your Random Module"
1
star
63

oueees-201706-public

Slides and figures of my talks on computer networks and information security, at Osaka University, {13, 20, 27} -JUN-2017
1
star
64

avr_md5

AVR/Arduino ATmega168/328 (Arduino Uno compatible) MD5 optimized assembly implementation
C
1
star
65

ipsj-tokai-20121029-public

Slides and figures of my talk on Erlang/OTP and Information Security, at Nagoya University, 29-OCT-2012
1
star
66

erlang-elixir-color-logo

Erlang and Elixir Color Logo. Licensed CC BY-NC-SA
1
star
67

adif_io

ADIF parsing library for Python (forked from Gitlab)
Python
1
star
68

oueees-202106-public

Slides and figures of my talks on computer networks and information security, at Osaka University, June 2021
1
star
69

poudriere_elixir_web

Web server for viewing Poudriere results
Elixir
1
star
70

sd_ep06_eqc

A QuickCheck-CI example for sd-erlang episode 6
Erlang
1
star
71

goadiftools

Go tools for amateur radio ADIF files
Go
1
star
72

gonnet-math-accuracy

Error analysis of system mathematical functions, by Gaston H. Gonnet, ETH, Informatik
C
1
star
73

hamsci-202012-freqdata

HamSCI December 2020 Eclipse Festival of Frequency Measurement result at the radio station JJ1BDX
R
1
star
74

gossn

Solar activity data retriever
Go
1
star
75

writing-otp-modules

Repository for "Writing A New Erlang/OTP Module for Beginners" (Erlang and Elixir Factory SF Bay Area 2017)
HTML
1
star
76

ctydat-entitycode

Conversion tables between CTY.DAT Main Prefixes and ARRL DXCC Entity Codes
Raku
1
star