Code-Aided Interference Suppression In DS-CDMA Communication Systems

One of the chief advantages of the spread-spectrum (SS) communications system is that the wideband signals can share a channel with narrowband communication signals, without one unduly interfering with the other. The SS signals have inherent antijamming capability and intrinsic resistance to any type of narrow-band interference (NBI). Even so, substantial performance gains can be achieved through the use of active NBI suppression prior to dispreading and demodulating. Earlier approaches sought to enhance the natural antijam capabilities of single-user SS systems. More recently, emphasis has shifted to multiple-access communication systems, particularly direct-sequence code division multiple access (DS-CDMA) systems, which play a leading role in current and future wireless networks.This thesis investigates the code-aided interference suppression techniques in DS-CDMA communication system. The code-aided methods exploit the spreading codes through the application of multiuser detection (MUD) techniques, designed to suppress the MAI, to NBI cancellation. The basic idea is to split a digital NBI into a set of nonoverlapping virtual users so that the overlay of an SS signal on digital NBI can be regarded as a CDMA system, and MUD tools can be applied. MUD techniques can suppress a much broader class of NBIs such as autoregressive processes and sinusoidal tones. The linear code-aided techniques, based on linear MMSE detector or Zero-Forcing detector are introduced in DS-CDMA communication system and their performance against both MAI and NBI is analyzed. The statistical results and numeral results show that the code-aided methods outperform all of the previous linear or nonlinear methods of NBI suppression, while simultaneously suppressing the MAI. The maximum-likelihood code-aided technique, which makes use of essentially all that is known about the useful signals and interference, is also introduced in this thesis.The future wireless communications systems must be able to accommodate the multimedia traffic such as voice, data and video, which have the different data rates and requirements of quality of service. This makes it imperative to develop effective multirate CDMA multiuser detectors. Reliably detecting each user requires the joint suppression of both MAI and NBI. This paper concentrates on the joint suppression of NBI and MAI by the application of the MUD techniques in variable spreading length (VSL) DS-CDMA system. The basic idea of MUD detector designed for multirate communication systems relies on decomposing each user into several virtual users to convert the original problem into a single rate multiuser problem and the code-aided techniques in DS-CDMA communication system can be applied. The optimum detector, MMSE detector, and Zero-Forcing detector for VSL DS-CDMA systems in the presence of narrowband interference are proposed, by exploiting some prior information of the received signals. Their performance against combined MAI and NBI is analyzed and the statistical results and numeral results are given. Also, code-aided blind rate detection for VSL DS-CDMA systems is considered under the circumstances that the receiver has no knowledge of the actual data-rate of the signal of interest.