Resource Allocation for Energy Harvesting Communications

Edición de la obra Resource Allocation for Energy Harvesting Communications

With the rapid development of energy harvesting technologies, a new paradigm of wireless communications that employs energy harvesting transmitters has become a reality. The renewable energy source enables the flexible deployment of the transmitters and prolongs their lifetimes. To make the best use of the harvested energy, many challenging research issues arise from the new paradigm of communications. In particular, optimal resource (energy, bandwidth, etc.) allocation is key to the design of an efficient wireless system powered by renewable energy sources. In this thesis, we focus on several resource allocation problems for energy harvesting communications, including the energy allocation for a single energy harvesting transmitter, and the joint energy and spectral resource allocation for energy harvesting networks. More specifically, the resource allocation problems discussed in this thesis are summarized as follows. We solve the problem of designing an affordable optimal energy allocation strategy for the system of energy harvesting active networked tags (EnHANTs), that is adapted to the identification request and the energy harvesting dynamic. We formulate a Markov decision process (MDP) problem to optimize the overall system performance which takes into consideration of both the system activity-time and the communication reliability. To solve the problem, both a static exhaustive search method and a modified policy iteration algorithm are employed to obtain the optimal energy allocation policy. We develop an energy allocation algorithm to maximize the achievable rate for an access-controlled energy harvesting transmitter based on causal observations of the channel fading states. We formulate the stochastic optimization problem as a Markov decision process (MDP) with continuous states and define an approximate value function based on a piecewise linear fit in terms of the battery state. We show that with the approximate value function, the update in each iteration consists of a group of convex problems with a continuous parameter and we derive the optimal solution to these convex problems in closed-form. Specifically, the computational complexity of the proposed algorithm is significantly lower than that of the standard discrete MDP method. We propose an efficient iterative algorithm to obtain the optimal energy-bandwidth allocation for multiple flat-fading point-to-point channels, maximizing the weighted sum-rate given the predictions of the energy and channel state. For the special case that each transmitter only communicates with one receiver and the objective is to maximize the total throughput, we develop efficient algorithms for optimally solving the subproblems involved in the iterative algorithm. Moreover, a heuristic algorithm is also proposed for energy-bandwidth allocation based on the causal energy and channel observations. We consider the energy-bandwidth allocation problem in multiple orthogonal and non-orthogonal flat-fading broadcast channels to maximize the weighted sum-rate given the predictions of energy and channel states. To efficiently obtain the optimal allocation, we extend the iterative algorithm originally proposed for multiple flat-fading point-to-point channels and further develop the optimal algorithms to solve the corresponding subproblems. For the orthogonal broadcast channel, the proportionally-fair (PF) throughput maximization problem is formulated and we derive the equivalence conditions such that the optimal solution can be obtained by solving a weighted throughput maximization problem. The algorithm to obtain the proper weights is also proposed. We consider the energy-subchannel allocation problem for energy harvesting networks in frequency-selective fading channels. We first assume that the harvested energy and subchannel gains can be predicted and propose an algorithm to efficiently obtain the energy-subchannel allocations for all links over the scheduling period based on controlled water-filling. The proposed algorithm is shown to be asymptotically optimal when the bandwidth of the subchannel goes to zero. A causal algorithm is also proposed based on the Q-learning technique that makes use of the statistics of the energy harvesting and channel fading processes.