Adaptive equalization for co-channel interference in a multipath fading environment

The paper concerns the feasibility and achievable performance of adaptive filtering in an interference-limited multipath fading environment as encountered in indoor wireless communications. In a typical cellular radio application, the performance-limiting impairment is interference due to synchronous data streams from other co-channel and adjacent channel users (CCI and ACI). The receiver under consideration employs an adaptive fractionally spaced decision feedback equalizer (DFE) Which exploits the correlation of the cyclostationary interference to achieve superior performance relative to the worst case when the interference is stationary noise. The paper presents ideal calculations which confirm that significant DFE performance gains are potentially achievable by explicitly accounting for the cyclostationary CCI. Two adaptive DFE strategies are considered. One approach is to adapt the DFE directly using iterative algorithms such as least mean square (LMS) or recursive least squares (RLS). Another approach is to compute the minimum mean square error DFE using an RLS channel impulse response (CIR) estimate and a sample estimate of the CCI autocorrelation obtained from the CIR estimation error during training. The best approach for adaptive equalization, in terms of adaptation speed and system performance, is to employ an RLS DFE which does not explicitly estimate the CIR or the CCI autocorrelation.<
>