WiFi干扰环境下传感器节点新型信道接入方式研究

The applicant’s Young Scientists Fund studies the medium access problem for the newly randomly deployed sensor nodes, so as to establish effective communication links between the initial nodes to implement the self-organize duty. Now, the project is going to be completed. The main achievements include: 5 papers published by international authoritative journals (as the first author for 3 of them), 1 paper published by international journal (as the first author), and 5 papers published by famous international conferences (as the first author or the first authors are the applicant’s students). The key techniques proposed in the project are applied in the applicant’s demos accepted by SenSys 2009 and 2010.. Based on the research work of the project above, the applicant considers the phenomena that sensor nodes in practice are usually seriously interfered by WiFi in the same frequency band, and proposes to further studies the media access techniques for sensor nodes to withstand the interference of WiFi. Moreover, the applicant will theoretically optimize the link throughput between sensor nodes under the WiFi environment. The applicant gets out of the restriction of the traditional opportunistic access idea and can achieve significant progress. The novel idea can be summarized as follows. Based on the collision recovery idea instead of the traditional collision avoidance idea, the proposed project will utilize the busy channel duration with some channel business metric predicted instead of only the idle channel duration in tradition, and propose novel wireless channel access techniques with adaptively changing the coding gain and the size of packets, which would be much promising to remarkably improve the link throughput between sensor nodes in WiFi environment.

申请人承担的青年基金项目，通过研究传感器节点在初始随机部署环境下信道接入问题，实现节点间初始有效通信链路的建立及自组网任务。课题现接近尾声，取得的成绩包括：发表国际权威期刊论文5篇(第一作者3篇)、国际期刊论文1篇(第一作者)，国际著名会议论文5篇(第一作者或第一作者为申请人学生)。相关成果在申请人SenSys’09及''10的demos上得到应用。. 基于该课题，申请人拟针对实际传感器节点通信链路被日益增多的同频段WiFi严重干扰现象，继续研究传感器节点抗WiFi干扰的信道接入技术，并在理论上优化链路吞吐量。申请人跳出前人基于机会式接入的干扰避免思想，有望取得突破性进展。其思路可概括为：利用信道短时相关预测的“干扰度”——而非传统仅考虑信道完全空闲时段，基于干扰恢复思想——而非传统干扰避免思想，提出自适应编码增益和包长优化的新型无线信道接入方式，有望显著提升节点链路吞吐量。

随着WiFi应用和室内无线传感器网络应用的广泛普及，WiFi对低功耗的室内无线传感器网络通信的干扰问题愈来愈严重。一直以来，无线传感器网络协议设计被动的应对环境中的干扰，导致在WiFi干扰稍严重情况下，无线传感器网络通信性能急剧恶化。本项目针对占据环境干扰主导地位的WiFi信号进行主动监测分析，让传感器节点合理适应WiFi信道当前特征，可使得WiFi环境下传感器节点通信问题得到显著改善，因而具有重要的应用价值。.项目主要研究内容有：WiFi信道占闲模型的建立；最优自适应编码策略的计算；物理层参数自适应优化；以及完整抗WiFi干扰的信道接入协议设计。.项目重要结果如下：1）WiFi信号的达到呈现帧群的分布；帧群到达时间、以及帧群大小，在100ms的窗口内，较好符合Pareto模型，且不同窗口内分布参数相互独立。2）汉明（12,8）、汉明（7,4）和BCH（15,5）、BCH（15,1）是四种适宜在传感器节点上运行、应对不同WiFi干扰度的编码方案。3）可以在不影响传感器节点自适应通信情况下保持对WiFi实时建模的时效性。4）自适应物理层参数（包长和速率）可以实现WiFi负荷20Mbps下传感器节点不低于10kbps的吞吐量。.项目科学意义在于：1）回答了WiFi信道是否可建模、以及在多大时间尺度上可建模问题；2）回答了能否在自适应通信过程中进行实时信道建模问题。这两个科学问题的回答，有力支持了WiFi干扰下传感器节点最优自适应信道接入方式的设计和实现，并具有重要的实际应用价值，将极大改善无处不在的WiFi信号对广泛应用的无线传感器网络的传输性能。