In this dissertation, we solve the Resource Allocation (RA) problem of a Multiple Input Single Output (MISO)--Orthogonal Frequency Division Multiplexing Access (OFDMA) system supporting minimum rates. This problem can be modelled as a non-linear Mixed Integer Program (NLMIP). We are interested in various kinds of methods to solve this problem.;First, our focus is on an off-line method that gives near-optimal solutions that serve as benchmark for more practical methods. For this purpose, we propose a method based on the optimal solution of the dual problem. We obtain a dual function and solve the dual problem through subgradient iterations. Then, we find upper and lower bounds for the optimal solution and verify that the duality gap is small for the system configurations studied. Therefore, the dual optimal serves as a reference for any feasible solution produced by the heuristic methods. The dual formulation also gives a better insight into the problem, as it shows us the relation between the problem's feasibility and the minimum rate requirements.;To obtain more practical methods, we propose two heuristics that have very low computational complexity and give performances not far from the optimal. We compare their performance against other methods proposed in the literature and find that they give a somewhat lower performance, but support a wider range of minimum rates while reducing the computational complexity of the algorithm by several orders of magnitude. The gap between the objective achieved by the heuristics and the upper bound given by the dual optimal is not large. For example, in our experiments this gap is 10.7% averaging over all performed numerical evaluations for all system configurations. The increase in the range of the supported minimum rates when compared with the method reported in the literature is 14.6% on average. This increase is achieved by considering the rate constraint dual variables in the user power allocation stage. The proposed heuristics select a set of users for each subcarrier, but contrary to other reported methods used to solve the throughput maximization problem, they consider the set of real-time (RT) users to ensure that their minimum rate requirements are met. Then, they solve a power allocation problem for fix subcarrier assignment, which is a convex problem that is simpler to solve. We use efficient algorithms for the subcarrier assignment and power allocation stages to solve the problem much quicker.;Finally, we adapt the algorithms to solve the RA problem in a single cell using LTE (Long Term Evolution)--Advanced technology. The methods examined in this dissertation are part of the new set of algorithms needed to support the high rate applications and spectral efficiency required in the wireless access of upcoming 4G networks.
展开▼
