Abstract: an adaptive sub-carrier allocation algorithm based on proportional fairness was investigated in the downlink of multi-user OFDM system, which considered allocation of sub-carriers to each user to maximize the sum of user data rate, subject to constraints on total power. Initialization of the allocation was done by assigning the sub-carrier with the highest channel gain to each user independently according to the relative proportion of data rate and the channel state. Conflict problem happened when two or more users chose one same sub-carrier. The priority of user to choose the sub-carrier was determined according to the average channel gain. Two conflict resolution algorithms were studied where the lager average channel gain was, the higher one user’s priority was or vice versa. Numerical results and comparisons with various conventional adaptive schemes in Rayleigh fading environment are provided to demonstrate the potential of our proposed techniques. It is shown that the proposed sub-carrier allocation algorithm can achieve good tradeoff between fairness and spectrum efficiency as well as provide almost the same capacity gain as the quasi-optimal method. At the same time, our scheme can meet the requirement of real time systems with low complexity and less computation.