Multiple random access and multiuser detection in wireless networks

This dissertation is devoted to the design and analysis of multiuser interference avoidance and suppression schemes in wireless communication networks. We focus on two contention based medium access schemes: random multiple access and DS-CDMA.; The thesis has four major parts. In the first part, the stability region of slotted ALOHA with finite number of queues is upper and lower bounded by using a new ranking technique. This technique can be used to intelligently set up the useful dominant systems. In the second part, we consider the capture effect on throughput and energy efficiency of random multiple access. Number of power levels, transmission power and packet length are optimized to increase the throughput of the multiple-power-level random access system under the system average power constraint. It is proved that under a mild condition on decodability threshold, a single-power-level system achieves the maximum achievable throughput. Analysis of multiuser detection in DS-CDMA system is performed in the third and fourth part. We show new graphic properties of indecomposable error sequences in the third part. Those properties are used to upper bound the error probability of the maximum likelihood multiuser detection. In the fourth part, we analyze the energy efficiency of the power controlled system using multiuser detection. It is shown that the use of nonlinear detection has fundamental impact on power control schemes of the system.