Research On Throughput Capacity And Resource Optimization Of Multi-hop Wireless Networks

Multihop wireless networks are infrastructure-less and self-orgonized wireless networks featuring multi-hop transmission, which includes ad hoc networks, wireless sensor networks and wireless mesh networks. As a new type of wireless networks, a multi-hop wireless network inherits the traditional problems of wireless and mobile communications, such as bandwidth optimization, power control, and transmission quality enhancement. In addition, distinctive features of multihop wireless networks make them very different from traditional infrastructure-based wireless networks in many aspects, including network organization, architecture and protocol design etc.Former researches on multihop networks are mainly focused on routing protocols, MAC protocols and internetworking, and many achievements have been obtained.However, there are still some challenging issues in this area. The first one is network capacity, including the fundamental capacity bounds, the relationship between capacity and delay and the maximum feasible throughput considering overhead of actual protocols. The second one is the technique to reduce interference and improve the utilization of network recourses. The third, for energy-constrained ad hoc networks and sensor networks, how to save energy consumption of netwok nodes is also an essential task. In addition, many advanced physical-layer technologies are emerging in multi-hop networks. Therefore, how to optimize the performance of these technologies becomes indispensable.The work in this dissertation is mainly devoted to the analysis of network capacity and the optimization of resource utilization. The main achievements are as follows:1) The queueing delay caused by multihop transmission and the relationship between throughput capacity and average delay in a finite sized network are analyzed. The network model and traffic model are proposed, and the everage end-to-end delay of the traffic is derived based on ideal transmission control and routing mechanism. The delay-limited capacity in the form of analytic expressions is obtained.2) To improve the transmission efficiency of broadcast operation in multihop wireless networks, the idea of space-covered broadcasting for global optimization is proposed. The optimal distribution of forwarding nodes that can cover and doubly cover the network is presented, and the optimal forwarding efficiency is analyzed.3) A 1-hop neighbor information based broadcast method called Triple Forwarding Broadcast (TFB) is proposed. In TFB, the total number of retransmissions is quite small since each node only needs to select 2-3 forwarding nodes, and high deliverability can be achieved for the set of forwarding nodes provide nearly double coverage to the network. Therefore, TFB can greatly improve the transmission efficiency.4) A broadcast method that does not need any neighbor information called Space-Covered Broadcast (SCB) is also proposed. In SCB, the distribution of forwarding nodes is optimized to achieve space coverage, and the overhead in terms of bandwidth, storage and computation are reduced. Furthermore, SCB is more practical for it adaptive to the physical channel conditions and can avoid transmission errors.5) We mathematically formulate the channel allocation, interface assignment and multicast routing as a joint linear optimization problem. The constraints of available channels and interfaces and interference between channels are unitedly considered and the objective of multicast routing is to minimize the total number of channels occupied by each node. Thus the capability of concurrent transmissions at different links is increased. Since finding the optimal solutions are not trivial for large-scale networks, we also present a heuristic method to achieve higher aggregated throughput and lower end-to-end delay.The thesis are divided into three parts.The first part (chapter 2) investigates the delay-constrained capacity of a finite-sized nultihop wireless network. The second part (chapter 3 and chapter 4) considers how to improve the transmission efficiency of broadcast operation in multihop wireless networks. In this part, we first analyze the minimum forwarding nodes needed to cover the network where nodes are ideally distributed, and the idea of space-covered broadcasting is proposed. We find the optimal distribution of forwarding nodes that can doubly cover the network, and the optimal forwarding efficiency is derived. On this basis, we propose the 1-hop neighbor information based broadcast method TFB and the broadcast method SCB that does not need any neighbor information. The third part (chapter 5) investigates the joint optimization problem of multicast routing and channel allocation in multi-channel wireless mesh networks. Finally, a summary of the dissertation is presented: it reviews the main results of the work and presents some open issues in this area.