Increasing users’ tendency to high data rate applications motivated researchers in various fields of communications to intensively look for inherent abilities of the communication systems which are useful to increase the system efficiency. Multi-user diversity is the interesting and unique characteristic of the multi-user wireless systems, such as cellular networks. In such systems, at the same time that some users experience deep fading, there may be some other users whose channel condition is good. As a consequence, at any time in a large network (a network with many users), with high probability, there is a user whose channel is near its peak. By utilizing this concept, the total throughput of a multi-user system is increasable. There is a simple channel assignment policy that fully exploits the existing multi-user diversity. At each time, the policy assigns the channel to the user who has the best channel condition at that time. This policy is called opportunistic channel assignment. Opportunistic channel assignment is unfair in the sense that the user with a poor channel condition has little chance to be served even in a long time interval. Since fairness is an important requirement of many multi-user scenarios, it is desirable to find channel assignment policies which make proper compromise between utilizing multi-user diversity and fairness. Adding fairness constraint to multi-user diversity utilization, makes the channel assignment problem complicated. In this thesis, we study the channel assignment problem in multi-channel wireless systems. We try to find solutions to the channel assignment problem with the objective of utilizing multi-user diversity under fairness constraint. Toward this goal, different definitions of fairness are considered and under each definition, channel assignment problem is expressed as an optimization problem. We propose two channel assignment policies for multi-channel wireless systems and prove the optimality of them. Each policy corresponds to a specific fairness definition and by optimality of a policy we mean the highest throughput among all policies which provide corresponding fairness. We adapt our policies to practical applications and demonstrate their efficiencies through simulations. Moreover, we investigate the trade off between utilizing multi-user diversity and fairness through some appropriate simulation results. Key Words: Wireless Communications Systems, Multi-User Diversity, Channel Assignment, Fairness.
