[Top][All Lists]

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

Re: [Discuss-gnuradio] Find and look to center frequency and phase of ba

From: Marcus Müller
Subject: Re: [Discuss-gnuradio] Find and look to center frequency and phase of bandwidth with help of stored original version of that
Date: Tue, 22 Aug 2017 10:29:58 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.2.1

Dear Mohammad,

Basically, what you want to do has been a bit "hyped" under the term of
"full duplex" in the last years. Essentially, all you need to do to
still be able to receive while you transmit is cancelling out the
effects of your known transmission on your reception.

For systems with large dynamic range, there's usually no way around
doing this at least partially in analog hardware – that is why you'll
find /circulators/ in many radar architectures. For more benign ratios
of transmit to receive power, you can do that simply by "simulating"
what the TX->RX crosstalk would be in software, and subtracting that
from the receive signal. I think that is what you're trying to do with
your LMS algorithm?

Sadly, we don't know what's wrong with that algorithm. Maybe if you
could describe in more detail what you're doing, mathematically, we can
help you. I'll take a well-labeled hand drawing of a spectrum over an
unlabeled signal plot any day, so don't be shy to share :)

You also might want to read the proceedings of this year's GRCON w.r.t.
the DySPAN 2017 challenge; there's a full-duplex system in real world
usage :)

Best regards,


On 21.08.2017 14:52, mohammad nejati wrote:
> Dear Marcus
> There are one 5 Mhz modulated bandwidth centered and transmited on
> 1ghz . in the cable i add 1 Mhz modulated bandwidth centered in the
> same occupied bandwidth for example 1ghz or 1.001 ghz or ..
> i want do some thing like this in satellites link :
> http://www.newtec.eu/frontend/files/leaflet/bwc0900-bandwidth-canceller.pdf
> Thank you .
> On Sun, Aug 20, 2017 at 8:41 PM, mohammad nejati <address@hidden> wrote:
>> Hi ,
>> It is difficult to explain what i need but i want to try it .
>> I have a single modulated carrier with 5 Mhz bandwidth i send it throw
>> cable and receive delayed version of that with a good SNR like 45dB .
>> I store i/q sample in last stage of transmitter ,
>> In cable i add one 1 Mhz carrier with lower power into exactly 5 mhz
>> equipped bandwidth of carrier (1Mhz carrier is hidden in 5 Mhz carrier
>> due to lower signal power) .
>> what i want to do is in receiver subtract transited signal from
>> received signal in a proper delay and phase and recover 1 Mhz carrier
>> .
>> I use LMS algorithm for adaption and subtraction and that work only
>> when received i/q sample have exactly same center frequency with
>> stored transmitted i/q sample which in reality with SDR board and zero
>> if architecture it is almost impossible . i don't want find center
>> frequency of received carrier with method like costas loop , because
>> modulated signal may have difference modulation like 64 QAM , 16 QAM ,
>> 32APSK , BPSK ... i want find and lock received center frequency and
>> phase with help of that i stored a correct version of that before
>> transmit .
>> i attached an image of subtracted signal and recovered smaller carrier
>> (which is in an ideal condition without frequency drift).
>> Thank you ,
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio

reply via email to

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