Rayleigh Fading
in [Matlab]
|
Prev: Regarding Image Processing in Matlab..Urgent...please help
Next: how to get a count data for each group
From: kiran on 14 Jan 2010 22:20 Hi frds., I am designing one CDMA system. In that i have to use rayleigh fading channel. When I referred to some other previous channel design they are working for normal txdata (randomly generated data no data processing on Tx'ed data). http://www.mathworks.com/matlabcentral/fileexchange/21058-ber-for-bpsk-in-rayleigh-channel But in my program I am doing encoding(using convolution encoder) the tx'ed data bits and spreading (PN sequence) and both are perferctly working. When I am used preivous rayleigh fading channel its not working for coding or spreading or for both. Could any body give me any reference or any matlab code that help me. I need simple rayleigh fading channel (not like doppler shift no delay) just for now single path rayleigh fading I need. Thank you in advance.
From: Ishwinder Singh on 3 Feb 2010 01:25 Hi Kiran, I am facing exactly the same problem as you mentioned. My encoding-decoding (Convolutional encoder-Viterbi decoder) set works fine with AWGN but the moment i subject it to Rayleigh fading, it doesn't show any gain with respect to the uncoded data. For Rayleigh fading, i am using Rayleighchan() function in the Matlab's Communications blockset. Request Mathworks for any leads in this regard. Regards Ishwinder "kiran " <kirankumar501(a)gmail.com> wrote in message <hiomt7$1b3$1(a)fred.mathworks.com>... > Hi frds., > > I am designing one CDMA system. In that i have to use rayleigh fading channel. When I referred to some other previous channel design they are working for normal txdata (randomly generated data no data processing on Tx'ed data). > > http://www.mathworks.com/matlabcentral/fileexchange/21058-ber-for-bpsk-in-rayleigh-channel > > But in my program I am doing encoding(using convolution encoder) the tx'ed data bits and spreading (PN sequence) and both are perferctly working. > When I am used preivous rayleigh fading channel its not working for coding or spreading or for both. > Could any body give me any reference or any matlab code that help me. > I need simple rayleigh fading channel (not like doppler shift no delay) just for now single path rayleigh fading I need. > > Thank you in advance.
From: Idin Motedayen-Aval on 3 Feb 2010 15:29 Ishwinder Singh wrote: > Hi Kiran, > > I am facing exactly the same problem as you mentioned. My encoding-decoding > (Convolutional encoder-Viterbi decoder) set works fine with AWGN but the > moment i subject it to Rayleigh fading, it doesn't show any gain with > respect to the uncoded data. > > For Rayleigh fading, i am using Rayleighchan() function in the Matlab's > Communications blockset. > > Request Mathworks for any leads in this regard. > > Regards > Ishwinder Ishwinder & Kiran, My guess is that you are not correcting the phase rotation introduced by the Rayleigh channel. The file referred to by Kiran does this on line 34 (yHat = y./h;). Since the channel state (h) is assumed to be perfectly known to the receiver, we can simply do the above operation. You can see my models also: http://www.mathworks.com/matlabcentral/fileexchange/22316-communication-systems-reference-curves (Check out the Rayleigh Fading models) These also assume perfect channel state information (CSI) at the receiver. Mind you, both files also assume flat fading. Multipath fading (frequency selective) will be more complicated to deal with and correct. If the channel is unknown at the receiver, you will need to do channel estimation, etc. (i.e. more advanced equalization). There are demos in the Communications Blockset that show equalization. HTH, Idin > "kiran " <kirankumar501(a)gmail.com> wrote in message > <hiomt7$1b3$1(a)fred.mathworks.com>... >> Hi frds., >> >> I am designing one CDMA system. In that i have to use rayleigh fading >> channel. When I referred to some other previous channel design they >> are working for normal txdata (randomly generated data no data >> processing on Tx'ed data). >> >> http://www.mathworks.com/matlabcentral/fileexchange/21058-ber-for-bpsk-in-rayleigh-channel >> >> >> But in my program I am doing encoding(using convolution encoder) the >> tx'ed data bits and spreading (PN sequence) and both are perferctly >> working. >> When I am used preivous rayleigh fading channel its not working for >> coding or spreading or for both. >> Could any body give me any reference or any matlab code that help me. >> I need simple rayleigh fading channel (not like doppler shift no >> delay) just for now single path rayleigh fading I need. >> >> Thank you in advance. -- Idin Motedayen-Aval The MathWorks, Inc. zq=[4 2 5 -15 -1 -3 24 -57 45 -12 19 -12 15 -8 3 -7 8 -69 53 12 -2]; char(filter(1,[1,-1],[105 zq])), clear zq
From: angelo on 14 Feb 2010 07:05
On Feb 3, 9:29 pm, Idin Motedayen-Aval <run.signat...(a)MATLAB.for.email> wrote: > Ishwinder Singh wrote: > > Hi Kiran, > > > I am facing exactly the same problem as you mentioned. My encoding-decoding > > (Convolutional encoder-Viterbi decoder) set works fine with AWGN but the > > moment i subject it to Rayleigh fading, it doesn't show any gain with > > respect to the uncoded data. > > > For Rayleigh fading, i am using Rayleighchan() function in the Matlab's > > Communications blockset. > > > Request Mathworks for any leads in this regard. > > > Regards > > Ishwinder > > Ishwinder & Kiran, > My guess is that you are not correcting the phase rotation introduced by > the Rayleighchannel. The file referred to by Kiran does this on line > 34 (yHat = y./h;). Since thechannelstate (h) is assumed to be > perfectly known to the receiver, we can simply do the above operation. > You can see my models also:http://www.mathworks.com/matlabcentral/fileexchange/22316-communicati... > (Check out the Rayleigh Fading models) > These also assume perfectchannelstate information (CSI) at the > receiver. Mind you, both files also assume flat fading. Multipath > fading (frequency selective) will be more complicated to deal with and > correct. > > If thechannelis unknown at the receiver, you will need to dochannelestimation, etc. (i.e. more advanced equalization). There are demos in > the Communications Blockset that show equalization. > > HTH, > Idin > > > > > > > "kiran " <kirankumar...(a)gmail.com> wrote in message > > <hiomt7$1b...(a)fred.mathworks.com>... > >> Hi frds., > > >> I am designing one CDMA system. In that i have to use rayleigh fading > >>channel. When I referred to some other previouschanneldesign they > >> are working for normal txdata (randomly generated data no data > >> processing on Tx'ed data). > > >>http://www.mathworks.com/matlabcentral/fileexchange/21058-ber-for-bps.... > > >> But in my program I am doing encoding(using convolution encoder) the > >> tx'ed data bits and spreading (PN sequence) and both are perferctly > >> working. > >> When I am used preivous rayleigh fadingchannelits not working for > >> coding or spreading or for both. > >> Could any body give me any reference or any matlab code that help me. > >> I need simple rayleigh fadingchannel(not like doppler shift no > >> delay) just for now single path rayleigh fading I need. > > >> Thank you in advance. > > -- > Idin Motedayen-Aval > The MathWorks, Inc. > zq=[4 2 5 -15 -1 -3 24 -57 45 -12 19 -12 15 -8 3 -7 8 -69 53 12 -2]; > char(filter(1,[1,-1],[105 zq])), clear zq hi i am doing the same and i have same problem! i am doing my master thesis on LTE.i am trying to implement mmse and svd estimation but i have some probleme this is my mmse code,could some how help me?it doesn't work. i have my pdp and its length is L. SNR = 1/(noiseVar); beta=1; power_delay_profile=[power_delay_profile zeros(1,fftlen-(L))]; power_delay_profile=fft(power_delay_profile); power_delay_profile=diag(power_delay_profile(carriers)); % Calculate frequency correlation matrix R_hh =power_delay_profile; R_hy = R_hh + (((beta/SNR)* eye(length(carriers)))); % Calculate LMMSE estimate h_LMMSE = (R_hh*inv(R_hy)) * h_LS; please help! this is the paper i refer http://www.sps.ele.tue.nl/members/C.K.Ho/paper/Robust%20MMSE%20channe... |