Chirp spread spectrum demodulator (LoRa-like): the PCM audio data is written to the decoder and the payload of the received frames with a correct CRC is written to the output. The input can be 8, 16 or 32 bit signed samples with any number of channels (only CSSConfig::channel is evaluated). More...

#include <CSSDecoder.h>

Inheritance diagram for tinymodem::CSSDecoder:

[legend]

Collaboration diagram for tinymodem::CSSDecoder:

[legend]

Classes
struct	Peak
	Peak of a spectrum. More...

Public Member Functions
	CSSDecoder ()=default
	CSSDecoder (Print &out)
CSSConfig	defaultConfig ()
	Provides a default configuration: css7 (SF 7, 1000 Hz).
void	setAudioInfo (audio_tools::AudioInfo info) override
	Defines the format of the input audio data.
audio_tools::AudioInfo	audioInfo () override
bool	begin (CSSConfig config)
	Starts the processing with the indicated configuration.
bool	begin () override
void	end () override
	Processes the remaining samples and reports the statistics.
void	flush () override
	Processes the remaining samples (e.g. at the end of the input).
size_t	write (const uint8_t *data, size_t len) override
	Provide PCM data to the decoder.
bool	isCarrier () override
	true while a frame is received
bool	isFinished () override
	true if the decoding was stopped (rx_one)
size_t	frameCount ()
	Number of received frames with a correct CRC.
size_t	errorCount ()
	Number of frames which could not be decoded.
float	frequencyOffset ()
	Frequency offset of the last frame in Hz.
CSSConfig &	config ()
	The configuration which was provided in begin().
	operator bool () override
Public Member Functions inherited from tinymodem::ModemDecoder
void	setStatusOutput (Print &out)
	Defines the output for the ### CARRIER / ### NOCARRIER messages.
void	setQuiet (bool quiet)
	Suppresses the status messages.
const char *	mime () override

Protected Types
enum	State { SEARCH , SYNC , HEADER , PAYLOAD }

Protected Member Functions
void	resetState ()
float	noise ()
	uniform random value in the range -0.5 to 0.5
void	addSample (float x)
	Mixes a sample to the baseband and resamples it to the bandwidth.
void	addBaseband (float re, float im)
Peak	spectrum (bool up, float offset)
float	signedBin (float pos)
	signed value of a bin (-n/2 to n/2)
void	processWindow ()
void	search ()
	Preamble: the same bin in consecutive windows.
void	sync ()
float	estimateDrift ()
	Change of the preamble peaks per symbol (least squares fit).
void	data ()
	Header and payload symbols.
void	endOfCarrier (bool ok, int corrected=0)
Protected Member Functions inherited from tinymodem::ModemDecoder
void	status (const char *fmt,...)

Protected Attributes
CSSConfig	cfg
bool	is_active = false
bool	is_finished = false
PcmInput	input
uint32_t	noise_seed = 1
int	n = 128
int	m = 256
SimpleFFT	fft
SampleRateConverter	converter
float	mix_re = 1
float	mix_im = 0
float	mix_step_re = 1
float	mix_step_im = 0
uint32_t	mix_count = 0
audio_tools::Vector< float >	ring_re
audio_tools::Vector< float >	ring_im
audio_tools::Vector< float >	fft_re
audio_tools::Vector< float >	fft_im
audio_tools::Vector< float >	power
audio_tools::Vector< float >	chirp_re
audio_tools::Vector< float >	chirp_im
int64_t	count = 0
int64_t	next_start = 0
State	state = SEARCH
int	hits = 0
int	candidate = 0
int	sync_windows = 0
float	last_up = 0
float	ups [MAX_UPS]
int	nups = 0
float	drift = 0
float	rate_factor = 1
float	sync_offset = 0
float	freq_offset = 0
audio_tools::Vector< uint16_t >	symbols
int	nsymbols = 0
int	needed = 0
int	frame_len = 0
float	snr_total = 0
size_t	frames = 0
size_t	errors = 0
Protected Attributes inherited from tinymodem::ModemDecoder
Print *	p_status = nullptr
bool	is_quiet = false

Static Protected Attributes
static constexpr int	RING = 4
static constexpr int	OVERSAMPLING = 2
static constexpr int	PREAMBLE_HITS = 3
static constexpr float	TRACKING_GAIN = 0.3f
	tracking of the remaining offset per symbol
static constexpr float	DRIFT_GAIN = 0.02f
	tracking of the remaining drift per symbol
static constexpr float	MIN_DRIFT = 0.08f
	min drift in bins per symbol for the correction of the sample rate
static constexpr int	MAX_UPS = 16
	max preamble windows for the drift estimation

Detailed Description

Chirp spread spectrum demodulator (LoRa-like): the PCM audio data is written to the decoder and the payload of the received frames with a correct CRC is written to the output. The input can be 8, 16 or 32 bit signed samples with any number of channels (only CSSConfig::channel is evaluated).

The signal is mixed to the complex baseband and resampled to the bandwidth, so that a symbol has N = 2^SF samples. A symbol is multiplied with the conjugate chirp (dechirped): the result is a tone, whose FFT bin is the symbol value. The preamble is detected by the same bin in consecutive windows; the bins of the up and down chirps provide the time and frequency offset. During the frame the timing follows a different clock of the sender.

The receiver must use the same ecc_bytes as the transmitter.

Copyright: GPLv3

Member Enumeration Documentation

◆ State

enum tinymodem::CSSDecoder::State

protected

Enumerator
SEARCH
SYNC
HEADER
PAYLOAD

Constructor & Destructor Documentation

◆ CSSDecoder() [1/2]

tinymodem::CSSDecoder::CSSDecoder ( )

default

◆ CSSDecoder() [2/2]

tinymodem::CSSDecoder::CSSDecoder ( Print & out )

inline

Member Function Documentation

◆ addBaseband()

void tinymodem::CSSDecoder::addBaseband	(	float	re,
		float	im )

inlineprotected

◆ addSample()

void tinymodem::CSSDecoder::addSample ( float x )

inlineprotected

Mixes a sample to the baseband and resamples it to the bandwidth.

◆ audioInfo()

audio_tools::AudioInfo tinymodem::CSSDecoder::audioInfo ( )

inlineoverride

◆ begin() [1/2]

bool tinymodem::CSSDecoder::begin ( )

inlineoverride

◆ begin() [2/2]

bool tinymodem::CSSDecoder::begin ( CSSConfig config )

inline

Starts the processing with the indicated configuration.

◆ config()

CSSConfig & tinymodem::CSSDecoder::config ( )

inline

The configuration which was provided in begin().

◆ data()

void tinymodem::CSSDecoder::data ( )

inlineprotected

Header and payload symbols.

◆ defaultConfig()

CSSConfig tinymodem::CSSDecoder::defaultConfig ( )

inline

Provides a default configuration: css7 (SF 7, 1000 Hz).

◆ end()

void tinymodem::CSSDecoder::end ( )

inlineoverride

Processes the remaining samples and reports the statistics.

◆ endOfCarrier()

void tinymodem::CSSDecoder::endOfCarrier	(	bool	ok,
		int	corrected = 0 )

inlineprotected

◆ errorCount()

size_t tinymodem::CSSDecoder::errorCount ( )

inline

Number of frames which could not be decoded.

◆ estimateDrift()

float tinymodem::CSSDecoder::estimateDrift ( )

inlineprotected

Change of the preamble peaks per symbol (least squares fit).

◆ flush()

void tinymodem::CSSDecoder::flush ( )

inlineoverridevirtual

Processes the remaining samples (e.g. at the end of the input).

Implements tinymodem::ModemDecoder.

◆ frameCount()

size_t tinymodem::CSSDecoder::frameCount ( )

inline

Number of received frames with a correct CRC.

◆ frequencyOffset()

float tinymodem::CSSDecoder::frequencyOffset ( )

inline

Frequency offset of the last frame in Hz.

◆ isCarrier()

bool tinymodem::CSSDecoder::isCarrier ( )

inlineoverridevirtual

true while a frame is received

Implements tinymodem::ModemDecoder.

◆ isFinished()

bool tinymodem::CSSDecoder::isFinished ( )

inlineoverridevirtual

true if the decoding was stopped (rx_one)

Implements tinymodem::ModemDecoder.

◆ noise()

float tinymodem::CSSDecoder::noise ( )

inlineprotected

uniform random value in the range -0.5 to 0.5

◆ operator bool()

tinymodem::CSSDecoder::operator bool ( )

inlineoverride

◆ processWindow()

void tinymodem::CSSDecoder::processWindow ( )

inlineprotected

◆ resetState()

void tinymodem::CSSDecoder::resetState ( )

inlineprotected

◆ search()

void tinymodem::CSSDecoder::search ( )

inlineprotected

Preamble: the same bin in consecutive windows.

◆ setAudioInfo()

void tinymodem::CSSDecoder::setAudioInfo ( audio_tools::AudioInfo info )

inlineoverride

Defines the format of the input audio data.

◆ signedBin()

float tinymodem::CSSDecoder::signedBin ( float pos )

inlineprotected

signed value of a bin (-n/2 to n/2)

◆ spectrum()

Peak tinymodem::CSSDecoder::spectrum	(	bool	up,
		float	offset )

inlineprotected

Dechirped spectrum of the window at next_start: up or down chirp, shifted by the frequency offset (in bins). The power of both parts of the chirp is added, which is robust against any time offset.

◆ sync()

void tinymodem::CSSDecoder::sync ( )

inlineprotected

Waits for the down chirps: up peak = tau + offset, down peak = -tau + offset

◆ write()

size_t tinymodem::CSSDecoder::write	(	const uint8_t *	data,
		size_t	len )

inlineoverride

Provide PCM data to the decoder.

Member Data Documentation

◆ candidate

int tinymodem::CSSDecoder::candidate = 0

protected

◆ cfg

CSSConfig tinymodem::CSSDecoder::cfg

protected

◆ chirp_im

audio_tools::Vector<float> tinymodem::CSSDecoder::chirp_im

protected

◆ chirp_re

audio_tools::Vector<float> tinymodem::CSSDecoder::chirp_re

protected

◆ converter

SampleRateConverter tinymodem::CSSDecoder::converter

protected

◆ count

int64_t tinymodem::CSSDecoder::count = 0

protected

◆ drift

float tinymodem::CSSDecoder::drift = 0

protected

◆ DRIFT_GAIN

float tinymodem::CSSDecoder::DRIFT_GAIN = 0.02f

staticconstexprprotected

tracking of the remaining drift per symbol

◆ errors

size_t tinymodem::CSSDecoder::errors = 0

protected

◆ fft

SimpleFFT tinymodem::CSSDecoder::fft

protected

◆ fft_im

audio_tools::Vector<float> tinymodem::CSSDecoder::fft_im

protected

◆ fft_re

audio_tools::Vector<float> tinymodem::CSSDecoder::fft_re

protected

◆ frame_len

int tinymodem::CSSDecoder::frame_len = 0

protected

◆ frames

size_t tinymodem::CSSDecoder::frames = 0

protected

◆ freq_offset

float tinymodem::CSSDecoder::freq_offset = 0

protected

◆ hits

int tinymodem::CSSDecoder::hits = 0

protected

◆ input

PcmInput tinymodem::CSSDecoder::input

protected

◆ is_active

bool tinymodem::CSSDecoder::is_active = false

protected

◆ is_finished

bool tinymodem::CSSDecoder::is_finished = false

protected

◆ last_up

float tinymodem::CSSDecoder::last_up = 0

protected

◆ m

int tinymodem::CSSDecoder::m = 256

protected

◆ MAX_UPS

int tinymodem::CSSDecoder::MAX_UPS = 16

staticconstexprprotected

max preamble windows for the drift estimation

◆ MIN_DRIFT

float tinymodem::CSSDecoder::MIN_DRIFT = 0.08f

staticconstexprprotected

min drift in bins per symbol for the correction of the sample rate

◆ mix_count

uint32_t tinymodem::CSSDecoder::mix_count = 0

protected

◆ mix_im

float tinymodem::CSSDecoder::mix_im = 0

protected

◆ mix_re

float tinymodem::CSSDecoder::mix_re = 1

protected

◆ mix_step_im

float tinymodem::CSSDecoder::mix_step_im = 0

protected

◆ mix_step_re

float tinymodem::CSSDecoder::mix_step_re = 1

protected

◆ n

int tinymodem::CSSDecoder::n = 128

protected

◆ needed

int tinymodem::CSSDecoder::needed = 0

protected

◆ next_start

int64_t tinymodem::CSSDecoder::next_start = 0

protected

◆ noise_seed

uint32_t tinymodem::CSSDecoder::noise_seed = 1

protected

◆ nsymbols

int tinymodem::CSSDecoder::nsymbols = 0

protected

◆ nups

int tinymodem::CSSDecoder::nups = 0

protected

◆ OVERSAMPLING

int tinymodem::CSSDecoder::OVERSAMPLING = 2

staticconstexprprotected

baseband samples per chip: the two parts of a chirp (before and after the wrap around) are separate tones, which makes the detection robust against a fractional time offset

◆ power

audio_tools::Vector<float> tinymodem::CSSDecoder::power

protected

◆ PREAMBLE_HITS

int tinymodem::CSSDecoder::PREAMBLE_HITS = 3

staticconstexprprotected

◆ rate_factor

float tinymodem::CSSDecoder::rate_factor = 1

protected

◆ RING

int tinymodem::CSSDecoder::RING = 4

staticconstexprprotected

◆ ring_im

audio_tools::Vector<float> tinymodem::CSSDecoder::ring_im

protected

◆ ring_re

audio_tools::Vector<float> tinymodem::CSSDecoder::ring_re

protected

◆ snr_total

float tinymodem::CSSDecoder::snr_total = 0

protected

◆ state

State tinymodem::CSSDecoder::state = SEARCH

protected

◆ symbols

audio_tools::Vector<uint16_t> tinymodem::CSSDecoder::symbols

protected

◆ sync_offset

float tinymodem::CSSDecoder::sync_offset = 0

protected

◆ sync_windows

int tinymodem::CSSDecoder::sync_windows = 0

protected

◆ TRACKING_GAIN

float tinymodem::CSSDecoder::TRACKING_GAIN = 0.3f

staticconstexprprotected

tracking of the remaining offset per symbol

◆ ups

float tinymodem::CSSDecoder::ups[MAX_UPS]

protected

The documentation for this class was generated from the following file:

src/modem/CSSDecoder.h

Classes

Public Member Functions

Protected Types

Protected Member Functions

Protected Attributes

Static Protected Attributes

Detailed Description

Member Enumeration Documentation

◆ State

Constructor & Destructor Documentation

◆ CSSDecoder() [1/2]

◆ CSSDecoder() [2/2]

Member Function Documentation

◆ addBaseband()

◆ addSample()

◆ audioInfo()

◆ begin() [1/2]

◆ begin() [2/2]

◆ config()

◆ data()

◆ defaultConfig()

◆ end()

◆ endOfCarrier()

◆ errorCount()

◆ estimateDrift()

◆ flush()

◆ frameCount()

◆ frequencyOffset()

◆ isCarrier()

◆ isFinished()

◆ noise()

◆ operator bool()

◆ processWindow()

◆ resetState()

◆ search()

◆ setAudioInfo()

◆ signedBin()

◆ spectrum()

◆ sync()

◆ write()

Member Data Documentation

◆ candidate

◆ cfg

◆ chirp_im

◆ chirp_re

◆ converter

◆ count

◆ drift

◆ DRIFT_GAIN

◆ errors

◆ fft

◆ fft_im

◆ fft_re

◆ frame_len

◆ frames

◆ freq_offset

◆ hits

◆ input

◆ is_active

◆ is_finished

◆ last_up

◆ m

◆ MAX_UPS

◆ MIN_DRIFT

◆ mix_count

◆ mix_im

◆ mix_re

◆ mix_step_im

◆ mix_step_re

◆ n

◆ needed

◆ next_start

◆ noise_seed

◆ nsymbols

◆ nups

◆ OVERSAMPLING

◆ power

◆ PREAMBLE_HITS

◆ rate_factor

◆ RING