Font Size: a A A

Interleaved serial concatenated convolutional codes with direct-sequence code division multiple access multicarrier modulation in fading channels

Posted on:2002-01-15Degree:Ph.DType:Dissertation
University:University of Colorado at Colorado SpringsCandidate:Petzold, Mark CarlFull Text:PDF
GTID:1468390011492390Subject:Engineering
Abstract/Summary:
This dissertation evaluates the combination of interleaved serial concatenated convolutional codes and multicarrier direct sequence spread modulation. This combination of forward error correcting code and modulation is powerful and versatile. This combination is evaluated in a Doppler spread channel for both differential phase shift keying (DPSK) and binary phase shift keying (BPSK) with perfect channel state information. Theoretical bounds are developed and both theoretical and simulation results are given for several variations of interleaver length, constituent codes, number of carriers, and various fading conditions.; Changing the number of carriers was shown to have a slight impact on the performance, mostly for the slow fading case (fdT = 0.002), where the fade duration was longer. Since the error correcting code used was a rate ¼, as the number of carriers was increased from four to 16, more symbols were sent, effectively increasing the data rate. Increasing the length of the interleaver was shown to have the most impact, especially for slow fading (fdT = 0.002). This is due to the fade duration as compared to the median symbol separation provided by the pseudo-random interleaver. It is also shown that the number of iterations of the decoder can be altered to exchange between hardware complexity and performance. It is shown that four iterations of the decoder provide most of the performance available from the decoder, while minimizing complexity. Increasing the constraint length of the constituent codes also improved performance, at the cost of increased complexity.; The theoretical bounds developed were shown to be effective at higher signal-to-noise ratios, and worked well for DPSK, but showed some weakness due to the required matrix inversions for BPSK. However, for fast fading ( fdT < 0.04) the bounds were easily calculated and showed good performance.
Keywords/Search Tags:Fading, Codes, Modulation, Performance
Related items