针对机会网络中由于节点移动、网络稀疏等各种原因通常导致网络拓扑动态变化大,消息源节点到汇聚节点之间往往不存在稳定的端到端的通信链路,提出了一种基于偏好顺序决策法(the technique for order preference by similarity to ideal so-lution,TOPSIS)的数据收集策略(data gathering based on the TOPSIS,DGT)。DGT策略根据节点的剩余能量、感知节点到汇聚节点的距离以及传感器节点的连通变化,采用TOPSIS评估选择下一跳中继节点。仿真实验表明,与现有的几种典型转发控制机制相比,DGT策略在保证较低传输延迟和较高传输成功率的基础上,通过减少节点间的转发次数,降低了网络传输开销。
With the rapid evolution of WSNs technology,it is very important to evaluate link quality quickly and accurately,so that the routing protocols can take relevant strategies in time to keep the entire network working steadily and efficiently. However,the issue of improving link quality assessment methods on physical layer is still open to research. To tackle this issue,a novel link quality assessment metric called S 3 LQA is proposed,which estimates the link quality of wireless sensor networks by CC2420 wireless radio frequency transceiver principles and free space propagation theory. The metric adopts both complete and incomplete packages to improve the evaluation performance effectively based on IEEE802. 15. 4 frame format and DSSS-O-QPSK mechanism. The experimental results show that the proposed method can improve energy cost and achieves better real-timing performance than traditional counting-based ( PRR) link quality assessment metric.
在分析现有机会网络转发控制策略的基础上,针对采用固定效用值阈值的机会网络转发控制,提出了一种基于节点能力状况的自适应转发控制策略(adaptive forwarding algorithm based on nodal capacity condition,AFNC)。该策略根据节点的能力状况计算阈值控制因子,自适应调整不同网络传输状况以及通信机会下的数据转发条件。仿真实验表明,与现有的几种典型转发控制相比,AFNC在保证较低传输延迟和较高传输成功率的基础上,通过减少节点间的转发次数,有效地降低了网络传输开销。
This paper proposes a chip correlation indicator (CCI)-based link quality estimation mechanism for wireless sensor networks under non-perceived packet loss. On the basis of analyzing all related factors, it can be concluded that signal-to-noise rate (SNR) is the main factor causing the non-perceived packet loss. In this paper, the relationship model between CCI and non-perceived packet loss rate (NPLR) is established from related models such as SNR versus packet success rate (PSR), CCI versus SNR and CCI-NPLR. Due to the large fluctuating range of the raw CCI, Kalman filter is introduced to do de-noising of the raw CCI. The cubic model and the least squares method are employed to fit the relationship between CCI and SNR. In the experiments, many groups of comparison have been conducted and the results show that the proposed mechanism can achieve more accurate measurement of the non-perceived packet loss than existing approaches. Moreover, it has the advantage of decreasing extra energy consumption caused by sending large number of probe packets.
