Resource Allocation for Multiple-User AF-OFDMA Systems Using the Auction Framework

We develop auction-based algorithms for joint allocation of resources, i.e., power profiles at the source and relay nodes and subcarrier assignment profile for multiple-user amplify-and-forward (AF) orthogonal frequency-division multiple-access (OFDMA) systems. The first algorithm is based on sequential single-item auction, where each user submits a bid based on either the marginal increase or the relative marginal increase in the data rate using the subcarrier. The first bidding strategy maximizes the sum data rate, whereas the second bidding strategy maximizes the fairness index. In both cases, the subcarrier is assigned to the user who submits the highest bid. The algorithm proceeds in a sequential fashion until all subcarriers are assigned. To reduce the synchronized interactions between the base station and the users, we propose a one-shot auction algorithm, where each user submits bids for all subcarriers at once based on the Shapley value, a well-known cooperative-game theoretic concept. The user evaluates each subcarrier based on an estimation of the Shapley value. The subcarriers are then assigned based on the submitted bids using an iterative algorithm that maximizes the fairness index. The throughput and fairness indices are used to evaluate the performance of the proposed algorithms. Numerical results are used to show the merits of each algorithm.