Effective And Cooperative Resource Management For Green Internet Of Things

With the flourishing development of new IoT applications,such as smart grid and smart transportation,network terminal deployment is gradually becoming dense and heterogeneous,which makes the network energy consumption increasing day by day,the spectrum resources are increasingly scarce,and the terminal computing capacity is exploding.The development of IoT faces three major challenges:energy limitation and inconvenience for terminal provision,network spectrum resource shortage,and terminal computation capacity insufficiency.Facing the new situation and new challenges,it is important to introduce clean and environment-friendly energy to provide sustainable energy provision for network terminals and reduce environmental pollution and resource waste caused by traditional energy sources to achieve the goal of "carbon peaking and carbon neutral".Subject to the random and intermittent nature of renewable energy arrivals,it is difficult to directly apply the traditional research on resource management for green energy-powered IoTs.To fully utilize green energy and enhance network energy efficiency,the paper conducts research on effective and cooperative resource management for green IoTs from three perspectives:green energy deployment and scheduling,large-scale terminal access management,and multi-access edge computing allocation.Based on optimization theory,the research investigates distributed deployment and scheduling for green energy,multi-device access mechanism with green energy provisioning,and green computing offloading decision and its resource allocation strategy under multiple application scenarios.By jointly optimizing network energy,communication,computing,and storage resources,the paper gets rid of the dependence on traditional energy sources and improves network energy efficiency.The main work and innovations of this paper are as follows:(1)To solve the energy shortage and inconvenience problem for IoT terminals,this paper distributes the deployment of green energy sources to networks for providing continuous energy provision from the perspective of"open source",getting rid of the dependence on large power grids and mobile charging devices.Considering the random and intermittent solar energy arrival,the paper investigates the cooperative Simultaneous Wireless Information and Power Transfer(SWIPT)mechanism in full-duplex mode with solar energy provisioning and proposes the cooperative SWIPT energy scheduling strategy by jointly optimizing the distributed deployment of energy sources and the cooperative SWIPT energy scheduling of solar-powered cluster heads in fullduplex mode.The simulation verifies that the proposed strategy can fully consume solar energy,extend the network lifetime,and maximize energy efficiency.(2)To address the dual scarcity of energy and spectrum resources caused by large-scale terminal access to the network,this paper introduces nonorthogonal multiple access(NOMA)to improve the network spectrum efficiency based on the Green IoT in Chapter 2.Considering the impact of large-scale terminal access on the complexity of serial interference cancellation(SIC)and network energy consumption at the NOMA receiver,the hybrid TDMA-NOMA multiple access mechanism for solar-powered networks is studied.Considering the causal constraints of solar energy arrival,network quality of service,and stable decoding at the receiver,the paper proposes an energy-efficiency prioritized optimization algorithm,which jointly optimizes NOMA clustering,solar energy broadcast power control,and time allocation to maximize system energy efficiency.Simulation verifies that the proposed algorithm can support large-scale terminal access while improving the network solar energy utilization.(3)For the high energy consumption and high computing capacity demand for IoT computing-intensive applications,this paper integrates MEC and solar harvesting technology to build a long-term computing task offloading and resource management model for green IoT.According to the difference in service attributes,the paper classifies it into rigid task offloading,soft task offloading,and mixed task offloading,which realizes adaptive offloading to specified optimization targets under dynamic energy constraints by jointly optimizing multidimensional heterogeneous resources,such as network energy,communication,computation,and storage.Facing the rigid offloading scenario with indivisible tasks,this paper considers the day and night volatility of solar energy arrival and the network demand all-time monitoring and proposes a joint optimization strategy for terminal computation frequency,energy broadcast power control,and time slot allocation with the goal of network time-frame computation capacity balancing.For the soft offloading scenario with separable tasks,this paper focuses on optimizing the network’s long-term energy efficiency under the constraints of solar energy arrivals and the imperfect channel state information.Taking into account the energy and data queue stability,the paper proposes a joint online optimization algorithm based on Lyapunov optimization for soft task offloading decisions and resource allocation to maximize the network’s long-term energy efficiency by optimizing the terminal computing frequency,solar energy broadcast power,and time allocation.Facing the task heterogeneous terminal hybrid offloading scenario,this paper focuses on optimizing the network long-term computation capacity problem under the coupling of different offloading modes of network heterogeneous terminals and proposes a joint online optimization algorithm based on Lyapunov optimization for mixed task offloading decision and resource allocation considering non-perfect channel states and delay constraints.