| At a time where there is an ever growing move towards wireless communications, it is important to provide the literature with a rigorous, accurate performance analysis tool for DS CDMA systems since most third generation systems are based on this multiple access scheme. In this light, this thesis is concerned with providing an analysis scheme to calculate the error probabilities of DS CDMA systems in additive white Gaussian noise (AWGN) and multipath fading channels with diversity combining when correlator receivers are used to coherently receive the signal. Unlike the popular approximations used for performance analysis, the scheme developed here is valid for systems at a wide variety of operating points.;Developing an accurate analysis scheme for DS CDMA is important for two reasons. First, it provides a better understanding of such systems and second, such an analysis scheme could be used to find just how accurate the popular approximations that are currently in use to estimate the system performance are and whether they have any limitations.;The error probability expressions derived throughout the thesis are based on a Fourier series approximation of the error function, Q( x). Once the error probability expressions are derived, we evaluate the performance of biphase and quadriphase spread, coherent DS CDMA systems in AWGN and multipath fading channels. Synchronous, quasi-synchronous and asynchronous systems are considered. For multipath fading channels, we investigate the effects of diversity combining on the system performance. We consider selection diversity, equal gain combining and maximal ratio combining. We investigate the effects of nonzero synchronization errors on the system capacity and conclude that the such errors cause significant capacity reductions in the system. We observe that in all cases considered, the synchronization errors effect the system performance by potentially reducing the energy of the desired signal component of the received signal and by introducing self interference.;We also calculate error probabilities for various DS CDMA systems using the Standard Gaussian Approximation (SGA), Improved Gaussian Approximation (IGA) and the Fourier series based expression we derived in this thesis. The scenarios are selected such that conditions for the validity of the Central Limit Theorem are questionable. We investigate a scarcely populated system, a system with dominant interferers, a multi-service system with uneven number of subscribers to the offered services and system in a frequency non-selective Rayleigh fading channel. We observe that the SGA in all of these cases yields optimistic results, especially when the number of users active in the system is small. The IGA on the other hand, gives accurate results for a scarcely populated system, a system with a dominant interferer and a multi-service system. However, in the dominant interferer and multi-service cases, if the strength of the dominant interferer is significantly large, or if the signaling rates between the different services are large, the IGA does not work at all. A modification can be made to overcome this problem at the expense of a reduced accuracy. For systems in fading channels neither of the Gaussian approximations seems to converge to the true error probability. |
