Two-stage nondata-aided adaptive linear receivers for DS/CDMA

Closed-form steady-state performance analysis of the signal-to-interference plus noise ratio (SINR) at the output of well-known adaptive implementatio ns of the linear minimum mean-square error (MSE) receiver for direct-sequen ce code-division multiple access show that nondata-aided (NDA) schemes may suffer from a considerable performance degradation with respect to their da ta-aided counterparts. Motivated by this fact, we propose a new two-stage N DA scheme where symbol-by-symbol predecisions at the output of a first adap tive stage are used to train a second stage. We derive closed-form steady-s tate performance analysis for both the two-stage and classical decision-dir ected schemes, taking into account detection errors in decision-directed ad aptation. Our analysis shows that the SINR of the two-stage algorithm is cl ose to optimal over a large range of values, while the SINR of the decision -directed scheme is far from optimal when the optimal SINR is small, Finall y, we consider the case of time-varying fading channels. We derive modified recursive Least square and least mean square adaptation schemes by conside ring SINR maximization rather than MSE minimization (that is useless under the assumption of zero-mean random channels), The resulting two-stage recei ver shows good tracking properties in heavy near-far conditions (at least f or moderate normalized Doppler bandwidth), while the decision-directed rece iver may easily loose tracking after deep fades.