Allocating resource on multiple user basis in OFDMA systems by using regenerative algorithm

Abstract

The main objective of our paper is to increase the throughput and fairness index and to decrease the power level. At most prioroty is to Provide very high data rate and coverage in forth co ming generation wireless networks. We propose regenerative algorithm in order to avoid work load balancing in memory. Joint al-location of resources using Auction-based algorithms, namely profiles of power at the source node as well as relay nodes and hence subcarrier assignment profile for multiple users amplify and forward orthogonal frequency division multiple access systems. It is based on either the marginal or relative marginal increase of data rate uing the subcarrier, each user is said to submit a bid and this acts as the base for the first algorithm which is based on sequential single item auction. The first bidding techniques is said to maximize the sum data rate While the second bidding strategy is said to maximize the index of fairness. In the dual cases, the user who submits the highest bid is assigned with the subcarrier. The algorithm is said to proceed in a sequential fashion while and until all subcarriers are left assigned. In order that reducing the synchronized interactions between the base-station and the users, here proposed is a one-shot auction algorithm, where each user submits bids for all subcarriers only once based on the Shapley value, a familiar cooperative-game theoretic concept. Depending on an estimation of the Shapley value, the user is said to evaluate each subcarrier. Based on the bids submitted and by using an iterative algorithm that enhances the fairness index, the subcarriers are assigned. To evaluate the performance of the proposed algorithms, hroughput and fairness indices are used. In order to reduce workload balance in the sensor memory, here proposed is a regenerative algorithm, where each user's data need to be stored in the memory. Numerical results are used to show the merits of each algorithm.