Statistical QoS Control of Network Coded Multipath Routing in Large Cognitive Machine-to-Machine Networks

Abstract

Machine-to-machine (M2M) communication enables many applications such as smart grid, vehicular safety, and health care among many others. To achieve ubiquitous data transportation among objects and the surrounding environment, deploying spectrum sharing M2M communications with existing wireless networks is a must. A general large-scale cognitive M2M network (CM2MN), adopting cognitive radio technology, consists of multiradio systems, the primary system (PS), and secondary system(s) with tremendous cooperative cognitive machines, under heterogeneous wireless architecture. For these CM2MNs, due to dynamic spectrum access (DSA) nature, there exists possibly unidirectional opportunistic wireless fading links and thus traditional flow control mechanisms at link level do not fit anymore. Furthermore, effective end-to-end quality-of-service (QoS) control is still required to provide a reliable transportation for such multihop CM2M communications. Facing the above challenges, we propose a novel statistical QoS control mechanism through cooperative relaying, realizing virtual multiple-input and multiple-output (MIMO) communications at session level . In particular, a probabilistic network coded routing algorithm and the statistical QoS guarantee are first proposed to coordinate and cooperate tremendous machines. Next, based on the proposed guarantee and routing algorithm, the statistical QoS control mechanism is designed to enable MIMO communications for the session traffic. Specifically, the diversity mode is used to deal with PS’s opportunistic nature and wireless fading, and the spatial multiplexing mode is employed to obtain the maximum end-to-end throughput. Simulation results confirm that under our control solution, the great improvements of end-to-end delay violation probability are obtained, thus practically facilitating network coded multipath routing in large CM2MNs.