Study On Link Quality Monitoring Mechanism For Event-Driven Wireless Sensor Network

According to data transmission mode, the wireless sensor network (WSN) can fall into the following category -- event-driven WSN, periodical sampling WSN and inquiry-based WSN. The project comes from National Nature Science Foundation. The event-driven WSN have the characteristics of multi-phase, paroxysm, and large number of data with high redundancy. The thesis studies on link quality monitoring mechanism, so as to obtain the link quality timely, to find out unsteady links and to warn early some abnormal scenarios such as link fault. It can ensure information aware send to sink node timely and accurately.This thesis introduces the classification of Link Quality Monitoring Mechanism for Wireless Sensor Network. Based on comparative analysis of current research of link quality monitoring mechanism, it presents a new mechanism named Link Quality Monitoring Mechanism for Event-driven Wireless Sensor Network (ED-LQMM). ED-LQMM defines two network statuses—regular status and event status. On regular status, there is lower data flow rate, and requirement for real time is not strict. ED-LQMM adopts passive exploration methods to get link quality and it employs Dynamic Sample & Table Check (DSTC) algorithm to determine the quantity of sampling Probe Packet in different PRR (Packet Reception Rate) period. Furthermore, it takes Exponentially Weighted Moving Average (EWMA) to deal with the monitoring result of link quality. On event status, there is a higher data flow rate, and it is sensitive to the fluctuation of packet loss. ED-LQMM employs active probe method to get link quality, which needn't extra probe packets. PPR is counted with times of data retransmission. The exception of link will be handled with EWM.The thesis addresses the design implementation of ED-LQMM using nesC language on TinyOS. The energy-consumption and accuracy of ED-LQMM is tested. Consecutive packet lost and random packet lost of event message is analyzed. Compared to general link quality monitoring mechanism for WSN, the results show that ED-LQMM remarkably decreases the energy-consumption caused by numerous of Probe Packet, and it prolongs the life of the network. The reliability of transportation is improved.