Re: [Discuss-gnuradio] gr-channel: adding ETSI/3GPP channel model taps

Hi,

I'm working on something similar. I'm also trying to map channel parameters to input parameters for the gr-channel blocks and validate the results.

I wondered whether there were any news on that and whether the python code is available somewhere.

Best,

Bastian

On 30 Dec 2015, at 02:33, Marcus Müller <marcus.mueller@ettus.com> wrote:

I must admit I did that, but feel unsure about how many sines I'd need to use to simulate spread.
The result I got with 8 and standard doppler spread don't look overly healthy, and osmocore/gr-gsm has a hard time understanding noise-free synthetic bcch bursts after going through the fading model.
Any advice on that?

Best regards,
Marcus

Am 29. Dezember 2015 22:16:53 MEZ, schrieb Johnathan Corgan <johnathan@corganlabs.com>:

On Tue, Dec 29, 2015 at 11:50 AM, Marcus Müller <marcus.mueller@ettus.com> wrote:

 

ETSI TS 145 005 V[1] specifies the relevant GSM channel models.

I do have a bit of python code that converts those to 10MS/s sampled
FIRs. Should I just add something to gr-channels python code that gives
you FIR taps for different of these models?

​It would be useful to see how well the Annex C tables can be mapped to the Frequency Selective Fading Model block parameters and do some resulting simulations with these.  The block simulates the time-varying effects of doppler and Rician/Rayleigh fading given a power delay profile and other relevant parameters.​