实用的传感器网络拥塞控制方案不仅需要满足多项网络性能指标,而且必须控制开销很小,提出了一种满足上述要求的拥塞控制机制EasiCC(Easi Netcongestion control mechanism).在EasiCC中,数据流源节点将数据报文按比例划分到各优先等级中,各网络节点根据网络拥塞程度动态地、同步地调整报文过滤标准,结合报文过滤标准和报文优先级来调节网络流量,保证了无线信道带宽分配上的公平性;将网络准入控制和队列丢包手段相结合来调整网络流量,保证了网络综合性能指标.EasiCC控制开销很少,已在实际传感器网络平台中实现.模拟验证和实验测试结果显示,EasiCC能够公平地为各数据流分配发报速度和网络带宽,并且在报文传输成功率、传输延迟等性能指标上均有良好的表现。
It's necessary for wireless sensor network(WSN) applications to deal with network congestion, because the channel bandwidth of WSN is usually narrow. However, existing congestion control mechanisms don't run well on WSN platforms. A practical congestion control mechanism for WSN should not only have good network performance but also have little control cost, the proposed EasiCC (EasiNet congestion control mechanism) is such a congestion control mechanism. In EasiCC, the source nodes prorate data packets into several transporting priorities, all the network nodes adjust in-phase packet filtering threshold to adapt network congestion status. With the help of filtering threshold and packet priority, network bandwidth is fairly allotted among data streams. EasiCC uses a stepwise and exponential method to adjust network traffic, so as to reduce control messages between network nodes. EasiCC uses network access waiting time and packet queue overflowing to detect network congestion, uses network admittance and queue dropping simultaneously to ensure the integral network performance. EasiCC has been implemented in wireless sensor network test-bed and costs little on the network bandwidth, communication energy and node memory. Simulation and experimental results indicate that EasiCC can limit transmission delay effectively, reduce packet loss ratio remarkably, and provide bandwidth fairness between data streams at the same time.