FreeCalypso > hg > gsm-net-reveng
changeset 100:a1868d31ce7f default tip
tfo-ut: add README
author | Mychaela Falconia <falcon@freecalypso.org> |
---|---|
date | Wed, 19 Mar 2025 01:39:57 +0000 |
parents | f8b501c8bfce |
children | |
files | tfo-ut/README |
diffstat | 1 files changed, 25 insertions(+), 0 deletions(-) [+] |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tfo-ut/README Wed Mar 19 01:39:57 2025 +0000 @@ -0,0 +1,25 @@ +Each of the 3 tfo-*.bin files in this directory is an extract from an MSC-side +E1 timeslot recording in a Nokia TCSM2 TFO session involving a cross-connect +between two TRAU channels. The extracts have been chosen to begin at the point +where the TRAU starts emitting TFO frames, thereby beginning with a series of +TFO frames that contain an embedded TFO_TRANS message. In each experiment, one +of the two cross-connected TRAU channels emitted two "plain" TFO frames (not +containing embedded TFO messages) in between the initial embedded TFO_TRANS and +the subsequent embedded TFO_REQ_L; this channel was chosen for the present +extracts. Each extract is thus 2560 PCM samples, containing 16 aligned TFO +frames: 5 carrying TFO_TRANS, 2 plain, 9 carrying TFO_REQ_L. + +In each of the three TCSM2 sessions from which these TFO frame extracts were +taken, the Ater UL input was fed with a stream of good DHFs (decoder homing +frames) for the respective codec, hence the TFO frame content is nothing but +these DHFs. In the case of FR and EFR captures, one of the 16 consecutive +frames also exhibits TAF=1. OTOH, the same does not hold for HR: the TRAU-8k +frame format used in TFO and in most Abis implementations carries TAF only for +BFIs and invalid SID, but not for good speech or valid SID frames. + +The tfo-*.hex files that are generated from tfo-*.bin by the Makefile in this +directory are intended to serve as unit tests for TFO frame extraction functions +that are currently being implemented in libosmotrau. The lack of frame content +diversity (nothing but DHF) is not seen as a problem: the objective here is to +test TFO frame extraction functions; the higher-level logic of TRAU frame +decoding has its own unit tests.