Fuzzy Logic Based Multidimensional Link Quality Estimation for Multi-Hop Wireless Sensor Networks

Abstract

In the applications of multi-hop wireless sensor networks, the effective evaluation of wireless link quality which can improve the network reliability is a crucial problem in the design of network protocols. The lossy and dynamic nature of the radio channel, however, makes link quality estimation a great challenge. First, we use data-driven modeling and statistical inference methods to model the burstiness of wireless link packet loss, which reflects the reliability of packet retransmission in the data link layer. Then, from a multidimensional view of characterizing the wireless link quality, we present the fuzzy logic based link quality indicator (FLI), which is a comprehensive reflection of single hop reliability of packet delivery, link volatility, and packet loss burst. Finally, we implement a wireless link quality estimator based on FLI in the collection tree protocol (CTP). Through the experimental comparison with the original link quality estimator in the CTP, the network performance with the new link quality estimator guarantees high end to end delivery reliability, while the average routing depth of the nodes and the dynamic changes of network topology are reduced.