A joint abnormal event detection scheme based on compressed sensing for Internet of Things

In order to improve the efficiency of abnormal event detection for Internet of Things, this paper proposes an abnormal event detection scheme based on compressed sensing. Due to the low energy consumption target constraints, the cluster-head often has no ability to collect all the data of the terminal nodes. Therefore, a sparse model of terminal nodes is proposed, which is applied to the WSNs scenario. When the number of nodes to be collected is far beyond the ability to collect data, the sparse model is used to propose a distributed abnormal event detection scheme. Each cluster-head uses the compressed sensing method to collect the data. During the procedure, data collection rules are in accordance with the sparse parity check matrix of LDPC code. In the fusion center, the compressed sensing equations of these cluster-heads will be combined into a complete equation. Then the Message Passing algorithm based on Tanner Graph is used to reconstruct the abnormal event sensing results of the sensor nodes.