Next generation cellular networks are expected to support multimedia applications and wide user mobility anytime and everywhere. As a result, several challenges arise from the network's perspective; for example how to support the increasing demand for wireless access while guaranteeing the requested quality of service, QoS, using a limited set of radio resources/channels. Reducing the cell size is proposed as a means of increasing the system capacity. However, it complicates the resource management tasks during handoff due to the time restriction. In such situations, the handoff rate is dramatically high and ensuring the QoS is more difficult due to the high variability in the user's mobility. These factors stimulate the need for more efficient and computationally tractable algorithms to be implemented in real time. Efficient resource management during handoff is a key element for the success of the potential cellular mobile networks. In this dissertation, we address the problem of channel management during handoff. We propose a new channel allocation scheme for improving the quality of service at the network access level. The proposed algorithm prioritizes handoff call requests over new call requests. The goal is to reduce the handoff failures while still making efficient use of the network resources. The performance measure is formed as a function of new call and handoff call blocking probabilities. The problem is formulated as a semi-Markov decision process. A simulation-based learning algorithm is then developed to approximate the optimal control policy online using the generated samples from direct interactions with the network. First we adopt a model-free learning scheme and subsequently we introduced a class of learning schemes that are based on an approximate model that is estimated simultaneously while learning a control policy. The estimated model is used to direct the search for an optimum policy. Extensive simulations are provided to assess the effectiveness of the proposed algorithms under a variety of traffic conditions. Comparisons with some well-known allocation policies, such as complete sharing and guard-channel policies, are also presented. Simulation results show that for the traffic conditions considered in this dissertation, the proposed schemes, while more broadly applicable, have a comparable performance to the optimal guard channel approach.
展开▼
