Resource Optimization For Massive MIMO Systems With Energy Harvesting

Massive multiple-input multiple-output(Massive MIMO)can sufficiently enhance system capacity and spectrum efficiency.With the improvement of spectrum efficiency,the overall amount of energy consumption and carbon emission are remarkable.Therefore,the contradiction between resource shortage and sustainable development has become increasingly prominent.Green communication is imperative.As one of the enabling technologies for green communication,energy harvesting(EH)attracts considerable interests from both industry and academia.This dissertation deals with resource optimization for massive MIMO systems based on hybrid energy harvesting and simultaneous transmission of wireless information and power(SWIPT).The main achievements and results of this dissertation are listed as follows:(1)For the power optimization problem of massive MIMO system based on hybrid energy harvesting,a single-cell,multi-user system model is established in this paper.Specifically,a finite-capacity energy harvester is employed at the base station in conjunction with fixed power grid for system energy supply.Based on the above model,this paper formulates throughput maximization problem,taking into account the circuit energy consumption and the transmission power.Two optimization algorithms are proposed that: a)When complete channel state information(CSI)as well as energy harvesting process are available,the throughput maximization problem is formulated as a typical convex optimization problem.Lagrange dual decomposition technique can be used to derive the optimal power variables.Simulation results indicate that the proposed offline scheme not only presents the best performance in comparison with the other schemes,but also minimizes the consumption of non-renewable energy.b)However,the complete CSI and energy harvesting process can not be acquired in practice.Due to its offline ideal characteristics,this paper heuristically presents an online scheme that merely utilizes the known CSI and the statistics of the energy harvesting process.At the same time,the idea of dynamic programming is adopted.Simulation results indicate that the heuristic online scheme attains a performance pretty close to the offline one,but it is much easier to implement.(2)For the resource allocation algorithm design for massive MIMO based on SWIPT,this paper theoretically derives the approximate expressions of harvested energy and downlink achievable rate in massive MIMO system with imperfect CSI.An optimization algorithm is also proposed to minimizie the transmission power of base station.The algorithm jointly optimizes the power-spliting ratio,the transmission power and power consumption of pilots.Under the insurance of the minimum transmission rate and the harvested energy,it can provide a sustainable and predictable power supply.Simulation results are provided to demonstrate the accuracy of the approximate expressions and the effectiveness of our proposal.