Signal Processing Algorithms for CDMA-Based Wireless Communications

Wireless communication systems rely on a multiple-access technique, i.e., a mechanism to divide the common transmission medium among dierent users. Code-division multiple-access (CDMA) is a multiple-access technique that has received considerable attention in recent years. In a CDMA system, each user spreads his information-bearing signal into a wideband signal, using specic code information. All users then transmit their wideband signal within the same frequency and time channel. This thesis deals with the development of receivers for various CDMA systems. Digital signal processing plays a central role in this development. In recent literature, so-called multi-user receivers have become very popular. These receivers take into account the full structure of the multi-user interfer- ence (MUI i.e., the interference originating from the other users. However, they have a rather high computational complexity. In the rst part of this the- sis, we therefore design new alternatives to multi-user receivers, yielding a lower computational complexity and a comparable performance. These alternatives are based on the concept of block spreading and have the remarkable property to completely remove the MUI, without using any channel information. In the second part of this thesis, we will look at CDMA systems with multiple receive antennas (spatial oversampling). A popular receiver for such a system rst applies some kind of space-time processing to transform the received wide- band signals into an estimate of the wideband signal transmitted by the desired user and then reverses the spreading operation corresponding to this user. This receiver is known as the space-time RAKE receiver. We show how the desired user’s code information can be used to design space-time RAKE receivers in a blind fashion, i.e., without relying on training sequences. Block processing as well as adaptive processing methods are considered.