Researches And Applications On Load Based Resource Management Strategies In Wireless Networks

In this study,we apply an existing load coupling model to investigate the radio resource management problem.Based on this model,the computation of cell loads can be characterized as the solution of fixed-point problems.Previous studies have analyzed the existence of the fixed point of load coupling mappings by constructing both linear lower and upper bounds.One of the main contributions of this thesis is to extend these approaches to investigate the feasibility of power coupling mappings,which have been derived from the load coupling model and are used to compute the transmit power for given cell load patterns.Although the load coupling model has been widely used in many applications,little attention has been paid to the validation of the model itself.Toward this end,we use the network simulator ns-3 to evaluate its performance in load estimation.The model outputs are shown to lead to a strong linear relationship with the numerical results obtained from simulation,which indicates the possibility of highly accurate predictions of cell load levels using this model.Taking link adaption schemes in real networks into consideration,we impose an additional constraint on the maximum achievable data rate in the model.Significant performance enhancement in estimation accuracy is observed with use of this improved model in a scenario where many users are located close to base stations.As an exemplary application of the load concept,we propose a mathematical framework for downlink heterogeneous networks,where cell load factors are used to measure the average resource consumption.By leveraging tools from stochastic geometry,we compute the average cell load levels across tiers via fixed-point equations that are dependent upon network parameters and users' data demands.Based on the proposed model,the effect of the association bias on achieving balanced network loads is investigated in two-tier networks.The optimum association bias factor,although not expressed in closed form,can be found efficiently through the standard bisection search method.As another outcome of our study,we consider joint power and time allocation for wireless powered communication networks,where the transmission time plays a similar role as load factors in OFDMA networks.We propose a novel system model that allows users to first harvest energy from the power station in downlink,and then concurrently send data information to their access points in uplink.For fixed users' data demands,we proceed an iterative algorithm to compute the transmit power and time of each user by using the properties of standard interference mappings.We further investigate the minimum throughput maximization problem,which aims at achieving high data rates but also ensures fairness among users.