Resource Allocation Technology For OFDMA Wireless Networks With Imperfect Channel State Information

With the rapid increase in the requirements for high data-rate communications,the network resources,such as spectrum and energy,have been consumed in an unprecedented way.Due to the abilities of achieving high spectral efficiency(SE)and eliminating inter symbol interference in frequency selective fading channel,orthogonal frequency division multiple access(OFDMA)is envisioned to become the major access scheme for the next generation wireless communications.In OFDMA wireless networks,it is necessary to allocate the network resource adaptively for improving energy efficiency(EE)and SE further.Generally speaking,when channel state information(CSI)is available,the sophisticated channel-aware resource allocation schemes could achieve the higher EE and SE compared to the conventional channel-blind resource allocation schemes.In practical wireless communication systems,CSI is obtained by the means of channel estimation with the received pilot symbols.However,the obtained CSI never be perfect in practice networks because of the estimation error and limited estimation cost.Treating CSI as perfect,when this is not the case,can deteriorate system performance significantly.Therefore,this dissertation devotes to investigate the resource allocation for OFDMA wireless communication systems with the consideration of resource consumption for acquiring CSI and performance degradation as a result of imperfect CSI.The contents and contributions of this dissertation are outlined as follows.In chapter 2,the subcarrier and power allocation is investigated to maximize EE for point-to-point data transmission in multi-user OFDMA networks,whilst considering imperfect CSI,channel estimation cost and selfishness of users.The resource optimization problem is formulated to handle inter competition(the resource competition among different users)and intra competition(the resource assignment between channel estimation and data transmission for each individual user).We employ the non-cooperative game with pricing approach to solve the optimization problem.By analyzing optimal relationship between channel estimation cost and total occupied resource for each user,the two-level resource competition is simplified to the problem including only inter competition,which is modeled as a non-cooperative subcarrier and power allocation game with subcarrier pricing.In this game,the utility function is defined as the number of received data bits per joule of energy,which is associated with the estimate error and cost,the data transmit power and the number of data-subcarriers.We prove the existence,uniqueness and Pareto efficiency of Nash equi-librium(NE)for the proposed resource allocation game.Furthermore,a distributed resource allocation algorithm is designed to achieve the NE.In chapter 3,the tradeoff between EE and SE is studied in downlink OFDMA systems,whilst considering the channel estimation cost and the corresponding effect of imperfect CSI on SE and EE.The problem is formulated as a multi-objective optimization to determine the optimal pilot transmission power,data transmission power and subcarrier assignment,and then transformed into a single-objective optimization problem,which is a non-convex mixed-combination non-linear programming(MCNLP)and NP-hard.To address it,we propose an efficient algorithm by adopting alternating optimization and convex optimization methods in lower power region as well as approximate conversion and branch-and-bound methods in high power region.Simulation results analyze and validate the performance of EE-SE tradeoff.In chapter 4,with various practical considerations including imperfect CSI,stochastic packet arrivals,time-varying channels and queue stability requirements of all users,the joint power and subcarrier allocation is investigated to minimize power consumption for the downlink OFDMA systems.To this end,this problem is formulated as the stochastic optimization model to minimize the time-averaged power consumption,whilst keeping all queues at the base-station(BS)stable.The data transmission rate is defined as a function of the transmit power,assigned subcarrier and estimation error.With the aid of Lyapunov optimization and dual decomposition methods,we propose a dynamic resource allocation algorithm DPSAI.The analytical bounds for the time-averaged power consumption and queue length achieved by our proposed algorithm is determined which depend on the channel estimation error.Moreover,the theoretical analysis and simulation results show that the proposed algorithm reduces the energy consumption at the expense of queue backlog(i.e.,achieves a energy-queue tradeoff),and quantitatively strikes the energy-queue tradeoff by simply tuning an introduced control parameter V.In chapter 5,based on the quantitative relationship between channel estimation error and cost,we further investigate the resource optimization between channel estimation and data transmission,whilst considering time-varying channel,stochastic packet arrivals,and queue stability demands.This problem is formulated as a stochastic optimization model,which jointly optimizes pilot power,data power,and data subcarriers allocation,to maximize the time-averaged data transmission rate subject to the queue stability,maximum transmit power and subcarrier allocation constraints.Since data transmission rate is non-concave function with respect to decision variables,the formulated optimization model falls into thescope of non-convex MCNLP stochastic optimization problem,which is NP-hard.To tackle this problem,we proposed dynamic resource allocation algorithm referred as DRA-JRAP2.Furthermore,the lower bound of time-averaged data transmission rate and the upper bound of time-averaging queue length achieved by DAR-JRAP2 are also derived.