The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> System Simulators >> Simulink Upconversion Problem
https://designers-guide.org/forum/YaBB.pl?num=1307460921

Message started by Paodoc on Jun 7^th, 2011, 8:35am

Title: Simulink Upconversion Problem
Post by Paodoc on Jun 7^th, 2011, 8:35am

Hi. First of all, sorry for my english Very Happy

I'm trying to simulate a 64-QAM modulation with LDPC coding and raised cosine transmit filter.
The source block has a sample time of Ts=4.873148e-10, seed = 25 and outputs frames of dimension 32400 bits. Then the LDPC block codes with a coderate 1/2 and outputs frames of 64800 bits. The LDPC uses the parity check matrix of the DVBS2 standard.
Thereafter we have the rectangular 64-QAM baseband modulator (M=6) with a gray constellation ordering and minumum distance between symbols = 2. Then we have the Raised Cosine Transmit Filter which has a group delay of 4, upsampling factor N=8 and rolloff = 0.2.
After that we have the AWGN channel with EbNo=1000 (so high to see the delay between tx and rx signals by comparing the 2 signals in the workspace. To do that I choose in the source block the probability of a zero to be 0.01 so that the sequences are almost only ones and are easy to compare to find the dealy for the BER block).
Than we have the Raised Cosine Receive Filter with the same parameters of the transmit filter. The QAM demodulator uses Hard Decision and outputs a sequence of zeros and ones. Because of a bug, the LDPC decoder only works with bipolar signals, so I introduced a BPSK modulator follow by a Re() block to transform the signals in -1 and +1. In this way the LDPC decoder works good and I can use the Error Rate Calcolation block to calculate the BER. In this block the delay is 48, given by the product of 6*8 which are M=6 and the cosine filter upsampling factor N=8.

At this point I have a nice working simulation with a very low BER.

THE PROBLEM

Now I need to upconvert the signal at 100 GHz.
I've tried to use the Upconversion block used in the demo simulation "Passband Modulation". You can find it under the simulink Help -> Library browser Help -> communication blockset -> Demos -> modulation

Using this block I set in the DSP the sample time as Tc = Ts/86400 and the samples per frame = 86400. In this way we have the product of two signals with the same frame time and frame dimension.
I can't understand why there is the presence in the Upconversion block of the Re() and in the downconversion of the complex conjugate.
Anyway I've tried to use this block but my signals is just upconverted to 1.8 GHz and the BER becomes 0.5 Sad
I can't understand why is this happening. Any help will be very appreciated.

I've uploaded the setting file (to run before the MDL file) and the .mdl file here:

http://www.megaupload.com/?d=SXH8J3QT
http://www.megaupload.com/?d=WY4VWGYF

ty for you help