Research On Wireless Network Scheduling And Capacity Problem Based On Fading Model

In recent years, with the rapid development of wireless communication technology and sensor technology, wireless networks have been widely used in more and more applications.While the wireless nodes typically interfere with each other in the actual communications, how to reduce the interference between wireless nodes and then obtain a more size of concurrent communication links is one of the most important problems to be considered during the designing process of wireless networks. To some sense, this problem determines directly the throughput capacity, lifetime, connectivity, routing and so on. Therefore, how to design effective algorithms to achieve more concurrent transmissions or parts of them in a short time is necessary.Wireless link scheduling is to say, construct a sub-set with some properties(e.g., low interference, concurrent transmission, success) associated with n communication links. Most existing results about this problem have always been under the assumption of graph-based model or Signal to Interference plus Noise Ratio(SINR) model. While the SINR model represents a significant improvement over previous approaches, it still uses a limited view of signal propagation. The main assumption is that receiving power is deterministic after a certain distance.In contrast, in reality signal propagation is by no means deterministic. Rayleigh fading model not only takes cumulative interference into account, but also considers the impact of environment on signal propagation. Thus, we study link scheduling in wireless networks with Rayleigh fading interference.The probability that communication links transmit successfully is one of essential indicators to evaluate link scheduling algorithms, which is associated with the size of scheduling set. If the number of ongoing links is too few, the network throughput will be low as a result. On the contrary, the cumulative interference between links is excessive, which can result in several links failing to transmit, the successful probability of link transmission will be lower. In spite of we study wireless link scheduling with Rayleigh fading model instead of SINR model, we still ensure that a similar or higher success probability can be achieved. Actually, we show that a more size of scheduling set can be obtained.In this paper, we first present concurrent neighbor set based link scheduling algorithm(CNS_LS). In fact, we may ignore the fact that the part of scheduling set is constructed, if algorithms continue to select links far in distance into scheduling set, we do not consider that these links may result in more bad interference for remaining unselected links which are more far in distance. Therefore, one possible result is that links, which are further in distance, do notsatisfy SINR constraint. Finally, based on CSMA/CA scheme and neighbor set, we propose a distributed algorithm IDSA, it can schedule all links in ??log n? time slots and give an??1/ ???? approximation factor. In the third section, we propose transmission reliability and security based link scheduling algorithms with the existence of eavesdroppers, namely SLS(Secrecy based link scheduling) and ISLS(Improved-SLS), respectively. In fact, with the location of passive eavesdroppers unknown, we analyze the size of scheduling set, transmission outage probability and secrecy outage probability from the view of theoretical results and simulation. Furthermore, we design distributed link scheduling algorithm in this model to reduce secrecy outage probability as much as possible. By application of algorithms TSDLS and RSOP,compared with existing algorithms, a secrecy outage probability which is less than 0.1 is obtained.