Comments on Partial Parallel Interference Cancellation for CDMA

In this letter we comment on partial parallel interference cancellation as discussed in a recent paper. The work in [1] showed that by multiplying symbol estimates by a factor less than unity in early stages of cancellation, the performance of parallel cancellation can be improved relative to full ("brute force") cancellation. The authors of [1] attribute the performance improvement to weighting symbol estimates according to reliability. We show that when soft correlator outputs are used for symbol or channel estimates, the improvement form partial cancellation is largely due to a reduction int he decision statistic bias in addition to symbol estimate reliability.