Access Methods For AdHoc Networking Based On Backoff Mechanism

The accelerating use of smartphones and the development of an exciting class of mobile Internet devices(MIDs) are creating an explosion of data transfer across wireless networks. At the same time, wireless ad-hoc network gets more and more attention. The rapid development of wireless ad-hoc network put forward higher requirements of the performance of the network. The design of MAC protocol adapting to wireless ad-hoc network faced with great challenge. The MAC protocol with a good performance is bound to promote the rapid development of wireless ad-hoc network, which is beneficial for army and people’s life. So the research of MAC protocol in wireless ad-hoc network has important practical significance.Under the specific background, the basic theory of MAC protocol in wireless ad-hoc network was introduced in this paper firstly. The characteristics, classification and performance metrics of the MAC protocol was studied. At the same time, the principle of IEEE802.11 was introduced in detail, with pointing out important part that influences the performance of the MAC protocol largely: binary index retreat algorithm. This paper introduced the improved algorithm of retreat so far, leading to the resignation of BEB.After theoretical study, we used 802.11 DCF model in NS2 under different scenarios for the simulation analysis. The results of simulation show the network nodes have different channel occupancy rate, which was called unfairness. Next, through the simulation analysis of MILD comparing with the simulation results of BEB under the same scenario, we can see advantages and disadvantages of MILD. Then a Markov model was established for 802.11 DCF, best packet transmission probability to reach the maximum saturation throughput was figured out through the mathematical deduction.Finally, we put forward with Probability-Selection-based Back-off. Then by modifying the code underlying protocol of 802.11 protocol in NS2, SPB was set up based on 802.11 model in NS2. Comparing the result of simulation of SPB with BEB and MILD, we can found that SPB has a good performance in fairness and adaptability.