Application of Collective Synchronization to Sensor Data Processing

Abstract

This paper is concerned with a new approach for sensor data processing for sensor networks in which multiple sensors are supposed to monitor overlapping part of the environment. The proposed method is based on collective synchronization of nonlinear coupled virtual oscillators whose natural frequencies are set to sensor data. Under appropriate conditions, the oscillators are locked to a common frequency. This convergent behavior is interpreted as data processing for bleaching and classification of noisy sensor data.