in this thesis, orthogonal frequency division multiple access (ofdma) system is considered. assuming perfect knowledge of instantaneous channel conditions for all users, we propose resource allocation algorithms to minimize the total transmission power subject to the constraints of the requirements of each user’s data rate and bit error rate (ber) which is referred to as margin adaptive (ma) problem, or to maximize the overall spectral efficiency while simultaneously satisfying the requirements of each user’s d ata rate, ber and base station (bs)’s transmission power constraint which is referred to as rate adaptive (ra) problem. by converting the above problems into linear integer programming problems, a branch-and-bound method based optimal algorithm and a fast suboptimal algorithm are proposed. the proposed branch-and- bound method based optimal algorithm offers the same optimal performance as full-search algorithm with remarkably reduced computational complexity. the proposed suboptimal algorithm , which is based on the formulation with constraints considered and greedy approach, can be used to solve both ma and ra optimization problems by satisfying the constraints one by one without any bit loading or transmission power distribution assumptions. compared with other suboptimal methods, the performance of this suboptimal algorithm is close to the optimal one with even lower computational complexity. index terms-adaptive modulation, frequency selective fading channel, multi-access communication, multiuser channel, channel capacity, orthogonal frequency division multiple access (ofdma), power control, resource management.