Mobile Intelligent Computing in Internet of Things: An Optimized Data Gathering Method Based on Compressive Sensing

Abstract

In order to alleviate the impacts of the rapid network energy exhaustion and the unreliable links on the data gathering in the Internet of Things (IoT), mobile intelligent computing based on compressive sensing date gathering (MIC-CSDG) algorithm is proposed in this paper, which could improve the data reconstruction accuracy. We conduct research from the following four links. First, this method employs mobile intelligent computing to derive the multi-hop function among sensor nodes, which is further utilized to determine the proportional relationship for the sensor nodes. Second, based on the sparse matrix, an observation matrix is designed with low correlation to mitigate the influences of the data packet loss on the entire IoT system and improve the data reconstruction accuracy for the sink node. Then, the acknowledge mechanism for the data forwarding strategy is employed to improve the reliability of the data transmission among clusters. Therefore, reliable data handover is accomplished for the multi-path routing data among different nodes. The results which are about the simulation shows that the loss rate of the packet is equal to 40%, the data reconstruction error of the MIC-CSDG algorithm still remains lower than 5%. Compared with other existing algorithms, the data forwarding time is reduced by 16.36%, while the average network energy consumption is reduced by 23.59%. Therefore, the validity and efficiency of the proposed method are verified.