Coding-spreading tradeoff in CDMA systems

Abstract

In this thesis we investigate the use of low rate codes primarily to provide the total bandwidth expansion required for a CDMA system. Comparing different combinations of coding and spreading with a traditional DS-CDMA, as defined in the IS-95 standard, allows the criteria to be defined for the best coding-spreading tradeoff in CDMA systems. The analysis of the codingspreading tradeoff is divided into two parts. The first part is dedicated to the study of the deterministic components of the problem. This includes the different factors with non-random behavior that the system's designer can determine. The processing gain, the code characteristics and the number of users are well-defined variables that can determine the overall performance and can consequently affect the tradeoff. The second part of the study is dedicated to analyzing different combinations of coding and spreading with no ideal channel estimation and interference reduction techniques. Small-scale fading channel conditions are emulated through Nakagami-m distribution. Large-scale path loss was incorporated through the extended Hata model while Lognormal shadowing considered the fluctuations on the received power at points with the same distance to the transmitter. We assessed the performance of different combinations of coding and spreading considered in two cases: a worst-case scenario in which the mobile user was located at the corner of a hexagon cell in a seven-cell cluster and a more realistic scenario in which the user could be physically located anywhere in the cell, following a uniform probability distribution function. Furthermore, we investigated the improvement in performance generated by interference reduction techniques, such as sectoring and power control.