2.1 Introduction

As noted in Chapter 1, code-division multiple access (CDMA) implemented with direct-sequence spread-spectrum (DS-SS) modulation continues to gain popularity as a multiple-access technology for personal, cellular, and satellite communication services. Also as noted in Chapter 1, multiuser detection techniques can substantially increase the capacity of CDMA systems, and a significant body of research has addressed various such schemes. Considerable recent attention has been focused on the problem of adaptive multiuser detection [184, 185]. For example, methods for adapting the linear decorrelating detector that require the transmission of training sequences during adaptation have been proposed in [73, 334, 335]. An alternative linear detector, the linear minimum mean-square-error (MMSE) detector, however, can be adapted either through the use of training sequences [4, 307, 325, 403] or in the blind mode (i.e., with prior knowledge of only the signature waveform and timing of the user of interest) [183, 549]. Timing-free blind multiuser detection has been considered recently in [62], while noncooperative multiuser detection, in which none of the signature waveforms is known, is treated in [596]. Blind adaptation schemes are especially attractive for the downlinks of CDMA systems, since in a dynamic environment it is very difficult for a mobile user to obtain accurate information about other active users in the channel, such as their signature waveforms, and frequent use of training sequences wastes channel bandwidth. There are primarily two approaches to blind multiuser detection: the direct matrix inversion (DMI) approach and the subspace approach. In this chapter we present batch algorithms and adaptive algorithms under both approaches. For the sake of exposition, we first treat simple synchronous single- path CDMA channels and present the principal techniques for blind multiuser detection. We then generalize these methods to the more general asynchronous CDMA channels with multipath effects. The rest of this chapter is organized as follows . In Section 2.2 we introduce the synchronous CDMA signal model and linear multiuser detectors. In Sections 2.3 and 2.4 we discuss the direct approach and the subspace approach to blind multiuser detection, respectively. In Section 2.5 we present analytical performance assessments of direct and subspace multiuser detectors. In Section 2.6 we discuss various subspace tracking algorithms for adaptive implementations of subspace blind multiuser detectors. In Section 2.7 we treat blind multiuser detection in general asynchronous CDMA systems with multipath channels. Finally, in Section 2.8 we present the mathematical derivations and proofs for some results in this chapter.

The following is a list of the algorithms appearing in this chapter.

• Algorithm 2.1 : DMI blind linear MMSE detector ”synchronous CDMA

• Algorithm 2.2 : LMS blind linear MMSE detector ”synchronous CDMA

• Algorithm 2.3 : RLS blind linear MMSE detector ”synchronous CDMA

• Algorithm 2.4 : QR-RLS blind linear MMSE detector ”synchronous CDMA

• Algorithm 2.5 : Subspace blind linear detector ”synchronous CDMA

• Algorithm 2.6 : Blind adaptive linear MMSE detector based on subspace tracking ”synchronous CDMA

• Algorithm 2.7 : Subspace blind linear multiuser detector ”multipath CDMA

• Algorithm 2.8 : Adaptive blind linear multiuser detector based on subspace tracking ”multipath CDMA

• Algorithm 2.9 : Blind linear MMSE detector in multipath CDMA with correlated noise ”SVD-based method

• Algorithm 2.10 : Blind linear MMSE detector in multipath CDMA with correlated noise ”CCD-based method