Study on active sleeping node detection method in sensor network based on multi-dimensional sliding window

Abstract

To overcome the problems of low coverage and detection accuracy in traditional detection methods, a multidimensional sliding window based active sleep node detection method for sensor networks is proposed. Firstly, we set up an active sleep node simulator and controller in the sensor network space to determine the active sleep range. Secondly, we design a multidimensional sliding window algorithm to determine anomalies in the transmission link by calculating the standard deviation of sensing information in the sliding window. Finally, the total length of data transmission is dimensionally transformed to achieve reliable detection of active sleep nodes. The experimental results show that the coverage rate of the detection results of this method is closer to 1, and its detection accuracy remains between 94.84% -97.32%, and the detection process time remains between 1.72 s-232 s. It has the advantages of strong reliability and high efficiency in applications.