Signal strength imbalance due to imperfect power control and soft handoffs in direct-sequence code division multiple access (CDMA) cellular systems increases the mutual interference among users. This paper discusses the use of successive interference cancellation (SIC) for improving the reception of reverse link signals transmitted by cdma2000 mobile stations. A single-stage SIC algorithm has been implemented in a software defined radio (SDR) which employs general purpose digital signal processing (DSP) chips for interference cancellation and signal demodulation. Computer simulation of the DSP software and measured performance of the SDR have shown SIC to be an effective multiuser detector for the cdma2000 reverse link

This paper compares fixed-rate signaling (FRS) with variable-rate signaling (VRS) when used in conjunction with forward error-correction coding for meteor burst communications. VRS refers to the reduction of bandwidth to compensate for the signal decay, which occurs as a meteor trail diffuses. The performance is quantified by calculating the cutoff channel model which assumes that signal amplitude decays exponentially. It is that VRS achieves the limiting performance constant-amplitude channel if the signaling rate is adjusted to maintain a constant received energy per bit. It is also shown that the improvement of VRS over FRS is moderate to high. Small if the code rate is low and the decoder is provided with side information on the received signal level. Therefore, the use of low-rate coding and PRS is viable alternative to VRS for achieving good performance.>

The use of error correction coding for meteor burst communications is discussed. The cutoff rate is derived for a meteor burst channel model which assumes that the received signal amplitude decays exponentially. The message error rate of a coded system, obtained by means of computer simulation, is compared to that of an uncoded system. A tight upper bound on the performance of the coded system is presented. It is shown that coding can reduce the signal-to-noise ratio (SNR) required for reliable communications via meteor bursts by many decibels. The coding gain is larger at shorter communication ranges where meteor trails, decay faster. It is also shown that the additional improvement achieved by providing the decoder with side information on the instantaneous level of the received signal is small.>

The bit error probability performance of a differentiallycoherent phase-shift keyed (DPSK) modem with convolutional encoding and Viterbi decoding on time-varying fading channels is examined. We consider both the Rician and the lognormal channels. Bit error probability upper bounds on fully-interleaved (zero-memory) fading channels are derived and substantiated by computer simulation. It is shown that the resulting coded system performance is a relatively insensitive function of the choice of channel model provided that the channel parameters are related according to the correspondence developed as part of this paper. Finally, a comparison of DPSK with a number of other modulation strategies is provided.

The performance of short constraint length convolutional codes in conjunction with binary phase-shift keyed (BPSK) modulation and Viterbi maximum likelihood decoding on the classical Rician fading channel is examined in detail. Primary interest is in the bit error probability performance as a function ofE_{b}/N_{0}parameterized by the fading channel parameters. Fairly general upper bounds on bit error probability performance in the presence of fading are obtained and compared with simulation results in the two extremes of zero channel memory and infinite channel memory. The efficacy of simple block interleaving in combating the memory of the channel is thoroughly explored. Results include the effects of fading on tracking loop performance and the subsequent impact on overall coded system performance. The approach is analytical where possible; otherwise resort is made to digital computer simulation.