discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] 3G on USRP


From: Chris Kilgour
Subject: Re: [Discuss-gnuradio] 3G on USRP
Date: Mon, 04 Jul 2005 22:35:58 -0700
User-agent: Mozilla Thunderbird 1.0.2-0.fdr.1.2 (X11/20050514)

tss ooty wrote:

Dear GNU Radio Team & Members,

I am looking for using USRP as my development kit for implementation of 3G standards 1xEvDO and 1xRTT.

I would like to know if there is any work already done in developing the CDMA2000 PHY on the USRP kit. It would even be very helpful if someone points out any feasibility of making cdma2000 protocol work on USRP. Is there any feature or limitation on the USRP board that makes it suitable or unsuitable for 3G standards 1xEVSO and 1xRTT?
IMHO, your biggest challenges stem from needing custom logic, generally because CDMA2000 is a synchronous system with fairly tight timing requirements. For example, one requirement will be power-control which requires the reverse channel to react to a recovered power-control bit from the forward channel (and hence alter the reverse power by up to 1dB) within 500us, every 1250us. Given the buffering latencies at each stage of hardware, USB, and processing time in the host, you'll probably have to accomplish this power control in the FPGA. In order to extract power-control bits in logic you'll need a quadrature despreader and (masked) long-code generator there too. The despreader will have to be syncronized to the network according to a pilot searcher, and if you hope to support soft-handoff then it will basically have to be a whole rake receiver done in logic. Continuing on that line, soon you realize that most of the whole PHY design requires custom logic. If you can find a way to reduce the latencies to an acceptable level, then it may be possible to avoid the custom logic. An interesting starting point might be a forward-only "monitor" implementation, which of course could be done purely in software since there would be no tightly-timed reverse transmissions to make.
Please help me out in this regard since I am finding it very difficult to evaluate the same and wondering why there is already not existing an implementation in the OpenSDR project...

 TIA a lot for all your help.

regards,
./tss





reply via email to

[Prev in Thread] Current Thread [Next in Thread]