A Message-Passing Approach to Self-Organizing Internet-of-Things Based Public Safety Networks

IEEE Access

卷:6

Illsoo Sohn¹  Sang Hyun Lee² 

[1] Department of Computer Science and Engineering, Seoul National University of Science and Technology, Seoul, South Korea;

[2] School of Electrical Engineering, Korea University, Seoul, South Korea;

关键词: Distributed algorithm;    Internet-of-Things;    message-passing algorithm;    public safety network;    spanning tree with hop limit;   

DOI  :  10.1109/ACCESS.2018.2882250

来源: DOAJ

【 摘 要 】

This paper develops a distributed self-organizing strategy for an Internet-of-Things (IoT)based public safety network (IPSN). Recent advances in wireless broadband and multimedia services have evolved PSNs extensively. Third generation partnership project long term evolution (3GPP-LTE) becomes a basic platform for deploying public safety networks around the world. There have been extensive studies on LTE-based PSNs satisfying mission-critical requirements. However, most studies focus on investigating PSNs with network infrastructure, and there is little progress on infrastructure-less PSNs, where base stations and network coordinators become destroyed or impaired. This paper focuses on infrastructure-less PSNs, where battery-powered individual IoT devices cooperate to construct the network without any central coordination. It is aimed at maximizing the network survival time of the IPSN while satisfying mission-critical requirements. A highly nonlinear nature of the network construction problem with several constraints renders the optimization task very challenging. In addition, no coordinator exists in IPSNs and all nodes are subject to the identification of distributed strategy to achieve the goal. To this end, a state-of-the-art message-passing framework is introduced to develop a novel distributed algorithm. The major benefit originates from the controllability of the limit on wireless link hops to meet the data reliability and transmission latency required for mission-critical IPSNs. We also establish the proof on the optimality. The proposed technique converges rapidly and keeps the computation load per IoT device low, which makes it attractive for practical implementation. Simulation results verify that the proposed approach outperforms various existing approaches considerably and consistently.

【 授权许可】

Unknown