Research On Energy-efficient Resource Allocation Based On Energy Harvesting In Cognitive M2M Networks

The ultimate goal of the Internet of Things(IoT)is to create an interconnected global network that enables things to exchange information and collaborate with each other,anytime,anywhere,anyway,thereby supporting new applications such as smart cities,intelligent transportation,and smart grids.Machine to Machine(M2M)communication is the key technology to successfully realize IoT,which enables devices to communicate with each other independently without any human intervention.Therefore,the traffic demand for different services of M2M is increasing.However,massive deployment of M2M devices still faces many challenges,such as limited battery capacity of devices and lack of spectrum resources.At present,cognitive M2M communication based on energy harvesting(EH-CM2M)can effectively solve the above problems by enabling M2M transmitters(M2M-TXs)to harvest energy from surrounding radio frequency signals and to reuse resource blocks(RBs)of cellular users(CUs)in an opportunistic manner.However,complex interference environments and strict Quality of Service(QoS)requirements pose new challenges to resource allocation optimization.Therefore,it is necessary to design an effective resource allocation optimization scheme to maximize the energy efficiency of M2M-TXs in EH-CM2M networks.This paper considers the energy efficiency maximization problem of M2M-TXs in EH-CM2M communication networks.A two-stage three-dimensional matching algorithm is proposed by jointly optimizing power control,time allocation,channel selection and peer discovery.In the first stage,M2M-TXs,M2M receivers(M2M-RXs)and RBs are temporarily matched,and then the power control and time allocation sub-optimization problem is solved by combining alternating optimization(AO)algorithm,nonlinear fractional programming algorithm,and linear programming algorithm to establish the preference lists of M2M-TXs.In the second stage,the pricing-based matching algorithm is used to solve the channel selection and peer discovery sub-optimization problem based on the preference list obtained in the first stage.Finally,the simulation results show that the energy efficiency of the proposed algorithm can be significantly improved comparing with the other two heuristic algorithms.