A Fuzzy Message Priority Arbitration Approach for Sensor Networks

This paper presents a novel fuzzy multinode communication priority protocol for intelligent network sensors. Each node is part of an intelligent sensor network. The node n
_N
basically consists of three logic blocks: a digital interface, application, and network communication. The network communication is across a serial bus system defined as controller area network (CAN). The fuzzy arbitration system was designed with simulation software assuming two variables. The first variable corresponds to the message size or size of the information to be transmitted and the second variable is the rate of change of the sensor variable. A dynamic auto-programming priority system was developed with a range of 0 to 15 priorities. The system is implemented with nine fuzzy rules. The best defuzzification approach was obtained from an experiment using five different methods. In order to evaluate the performance of the proposed method it is compared against a static approach. The FMPAS approach can work like a dynamic priority arbitration system and it is a new alternative to avoid the static disadvantages of other paradigms. Besides, the proposed methodology establishes the fundamentals to create reconfigurable communication in intelligent network sensors.