Multiple event detection in wireless sensor networks using compressed sensing

Event Detection is one of the main applications of wireless sensor networks (WSN). However, due to the noisy sensed data of sensors and the wireless channel noise, it's difficult to guarantee the accuracy of detection, especially in multiple event detection. In this paper, we proposed a multiple event detection scheme using compressed sensing (CS). By analogy with CS problem, the efficient recovery algorithms of CS can be used to reconstruct the source signal that contains multiple simultaneous events. Moreover, the events may not change much, so the source signals at two adjacent time instants have high redundancy. This temporal correlation is also utilized in our scheme to improve the detection accuracy. In the proposed scheme, not only the position but also the value of an event can be achieved. Three algorithms of CS are used in our scheme to show the advantages on detection probability over the traditional decentralized detection methods using Bayesian.