Re: [Discuss-gnuradio] Successive Interference Cancellation using GR

I was wondering whats the limit for set_history() of gr::block. 

For SIC, my intuitive feeling is that I have to buffer the samples, a lot of samples. If I take the simple case where PDU lengths are very less. 1 byte for 802.11 and 1 byte for 802.15.4, 

I will end in 560 samples for 802.11 and 898 X 5 ~ 4000 samples for 802.15.4 (multiplied by 5 for upsampling the 4MHz signal of 802.15.4 to 20 MHz)

So the set_history should save at least 4000 previous samples. 

I am not sure how it will affect the case where both 802.11 and 802.15.4 being decoded. 

On Thu, Nov 2, 2017 at 9:42 PM, Sumit Kumar <sumit.kumar@alumni.iiit.ac.in> wrote:

I want to implement a system where I will receive overlapped wifi and zigbee on the same channel using same RF front end. Then in the base band I wish to do interference cancellation. 

I understand that most of the time wifi is stronger than zigbee. 

So if the preamble of wifi is not damaged, I will be able to get the channel estimates of wifi. 

Hopefully if the wifi frame has not suffered much due to interference of zigbee(which is very much possible), I can decode and regenerate the wifi signal using my channel estimates. 

Then I will subtract this regenerated wifi signal from the mixed signal (wifi + zigbee).

The residue should give me a relative clean signal from which decoding of zigbee might be possible.

Here is graphic for reference: 

http://bit.ly/2gZ4Nkk

y1 is a received signal which is x1 + x2. + noise. 

First we demodulate x2 (probably because it is stronger), then re modulate it. Then subtract it from y1 and then decode x1. 

Regards

Sumit 

On Thu, Nov 2, 2017 at 9:02 PM, Marcus Müller <marcus.mueller@ettus.com> wrote:

Hi Sumit,

I wrote that but can't remember the context (it seems there was some).

So, maybe, let's start from scratch. What is it that you want to implement as a whole that includes Successive Interference Cancellation? What is the application you're building?

Best regards,
Marcus

On 02.11.2017 20:56, Sumit Kumar wrote:

Hi, 

I was looking for some previous work and I found this link

https://lists.gnu.org/archive/html/discuss-gnuradio/2016-01/msg00304.html

where Marcus replied about the difficulty to realize such module in GR. 

I am sorry but I could't understand it properly. 

Can someone elaborate it :) 

I plan to perform SIC on a 802.15.4 waveform which is corrupted by 802.11 and vice versa. 

Regards

Sumit 

_______________________________________________  Discuss-gnuradio mailing list  Discuss-gnuradio@gnu.org  https://lists.gnu.org/mailman/listinfo/discuss-gnuradio  

_______________________________________________
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio