This is a modified version of osmo-tetra (original README contents are a
few lines below).

This version is patched to work with telive - a program for live listening
of tetra traffic and recording it, displaying network data etc. 

The most important changes are:
- support parsing of SDS messages (text messages similar to SMS), including some location protocols
- support parsing of D-SETUP, D-RELEASE, D-NWRK-BROADCAST
- sending of data and voice traffic via UDP (for consumption with telive or some other program)
- incorporate float_to_bits into tetra-rx
- implement a crude AFC into tetra-rx and float_to_bits (enables the signal to be less precisely tuned)

For more info about osmo-tetra-sq5bpf and telive:

Main docs (read this first!): https://github.com/sq5bpf/telive/raw/master/telive_doc.pdf 
Telive repository: https://github.com/sq5bpf/telive
Modified osmo-tetra: https://github.com/sq5bpf/osmo-tetra-sq5bpf
Script to automatically download/compile/install the acelp codec: https://github.com/sq5bpf/install-tetra-codec
Forum thread: http://forums.radioreference.com/voice-control-channel-decoding-software/302347-tetra-decoding.html
Tutorial: http://www.rtl-sdr.com/rtl-sdr-tutorial-listening-tetra-radio-channels/

What follows is the original content of the README file:

TETRA MAC/PHY layer experimentation code
(C) 2010-2011 by Harald Welte <[email protected]>
======================================================================

This code aims to implement the sending and receiving part of the
TETRA MAC/PHY layer.

If you read the ETSI EN 300 392-2 (TETRA V+D Air Interface), you will
find this code implementing the parts between the MAC-blocks (called
type-1 bits) and the bits that go to the DQPSK-modulator (type-5 bits).

It is most useful to look at Figure 8.5, 8.6, 9.3 and 19.12 in conjunction
with this program.

You will need libosmocore (http://bb.osmocom.org/trac/wiki/libosmocore) to link.


== Demodulator ==

src/demod/python/cpsk.py
	* contains a gnuradio based pi4/DQPSK demodulator, courtesy of KA1RBI
src/demod/python/tetra-demod.py
	* call demodulator on a 'cfile' containing complex baseband samples
src/demod/python/usrp1-tetra_demod.py
	* use demodulator in realtime with a USRP1 SDR
src/demod/python/usrp2-tetra_demod.py
	* use demodulator in realtime with a USRP2 SDR

The output of the demodulator is a file containing one float value for each symbol,
containing the phase shift (in units of pi/4) relative to the previous symbol.

You can use the "float_to_bits" program to convert the float values to unpacked
bits, i.e. 1-bit-per-byte


== PHY/MAC layer ==

=== library code ===

Specifically, it implements:
lower_mac/crc_simple.[ch]
	* CRC16-CCITT (currently defunct/broken as we need it for
	  non-octet-aligned bitfields)
lower_mac/tetra_conv_enc.[ch]
	* 16-state Rate-Compatible Punctured Convolutional (RCPC) coder
lower_mac/tetra_interleave.[ch]
	* Block interleaving (over a single block only)
lower_mac/tetra_rm3014.[ch]
	* (30, 14) Reed-Muller code for the ACCH (broadcast block of
	  each downlink burst)
lower_mac/tetra_scramb.[ch]
	* Scrambling
lower_mac/viterbi*.[ch]
	* Convolutional decoder for signalling and voice channels
phy/tetra_burst.[ch]
	* Routines to encode continuous normal and sync bursts
phy/tetra_burst_sync.[ch]


=== Receiver Program ===

The main receiver program 'tetra-rx' expects an input file containing a
stream of unpacked bits, i.e. 1-bit-per-byte.


=== Transmitter Program ===

The main program conv_enc_test.c generates a single continuous downlinc sync
burst (SB), contining:
	* a SYNC-PDU as block 1
	* a ACCESS-ASSIGN PDU as broadcast block
	* a SYSINFO-PDU as block 2

Scrambling is set to 0 (no scrambling) for all elements of the burst.

It does not actually modulate and/or transmit yet.


== Quick example ==

	# assuming you have generated a file samples.cfile at a sample rate of
	# 195.312kHz (100MHz/512 == USRP2 at decimation 512)
	src/demod/python/tetra-demod.py -i /tmp/samples.cfile -o /tmp/out.float -s 195312 -c 0
	src/float_to_bits /tmp/out.float /tmp/out.bits
	src/tetra-rx /tmp/out.bits