一种基于干扰感知概率预测的无线异构Mesh网络带宽统计联合优化算法

Joint Optimization of Statistical Bandwidth Based on Interference-Aware Probability Prediction for Wireless Heterogeneous Mesh Networks

文中提出了一种基于IEEE 802.11的无线多媒体异构网络可用带宽评估算法,该算法通过研究分组碰撞概率和退避时间的相互影响计算发送端与接收端空闲时间的重叠概率,同时通过区分节点的繁忙状态和载波侦听状态来提高重叠概率估计的准确性。网络节点根据感知的信道剩余可用带宽,综合考虑流内和流间干扰,计算路径可行的发送速率和路径代价函数,以此选择最佳分组转发路径。NS2仿真结果表明,文中提出的带宽优化算法,与现有算法相比能够更加准确的感知链路的可用带宽,提高网络吞吐量,避免网络拥塞,为多媒体业务流的接入提供更好的QoS保障。

A new available-bandwidth prediction algorithm for IEEE 802.11 wireless multimedia networks is proposed. The proposed algorithm can calculate the overlap probability of the idle channel time be- tween the sender and the receiver using the relation of collision probability and back-off time, and im- prove the accuracy of the overlap probability through the discrimination of the channel state and carrier sensing. According to the residual available bandwidth at each node, the algorithm can predict the feasi- ble sending rate of the path and the path cost function considering inter-flow and intra-flow interferences to select the optimal forwarding path among all feasible paths. Simulation results show that the proposed algorithm can accurately calculate the channel bandwidth and improve the network throughput with quality of service （QoS） of multimedia flows.