Decision-feedback interference suppression in CDMA systems: a ML-based semiblind approach
Signal Processing - Special section: Security of data hiding technologies
| Hi-index | 0.07 |
Decision feedback sequence estimation (DFSE), which is a reduced-complexity alternative to maximum likelihood sequence estimation (MLSE), can be used effectively for equalization of intersymbol interference (ISI) as well as for multiuser detection. The algorithm performs very well for whitened (minimum-phase) channels. For nonminimum-phase channels, however, the algorithm is not very effective. Moreover, DFSE requires a noise-whitening filter, which may not be feasible to compute for time-varying channels such as a multiuser direct-sequence code division multiple access (DS-CDMA) channel. Noise-whitening is also cumbersome for applications that involve bidirectional equalization such as the global system for mobile communication (GSM) system. In such conditions, it is desirable to use the Ungerboeck (1974) formulation for sequence estimation, which operates directly on the discrete-time unwhitened statistic obtained from conventional matched filtering. Unfortunately, DFSE based on matched filter statistics is severely limited by untreated interference components. We identify the anticausal interference components, using an error probability analysis. This leads us to a modified unwhitened decision feedback sequence estimator (MUDFSE) in which the components are canceled, using tentative decisions. We obtain approximate error probability bounds for the proposed algorithm. Performance results indicate that the modified algorithm, used on unwhitened channels with relatively small channel correlations, provides similar performance/complexity tradeoffs as the DFSE used on the corresponding whitened minimum-phase channels. The algorithm is especially attractive for multiuser detection for asynchronous DS-CDMA channels with long spreading codes, where it can achieve near-MLSE performance with exponentially lower complexity