Abstract
Spectrum auction has been considered a promising solution to release the under-utilized spectrum from primary spectrum licensee to potential secondary users. In this paper, we introduce an auction-based resource allocation scheme between a spectrum broker and base stations (BS)s that belong to a wireless service provider (WSP). Each BS submits the proportion of resources it wishes to lease from each auctioned spectrum band based on its true demand and the auctioneer's offered price. The spectrum broker leases the resource blocks (RB)s of its under-utilized spectrum bands to the WSP base stations using an iterative resource allocation algorithm. The proposed algorithm is based on a fair spectrum auction. The spectrum broker who manages the under-utilized spectrum resources plays the role of an auctioneer. First, we establish utility functions for the auctioneer and each participating BS. Then we maximize these utilities in iterations in order to achieve optimal price and allocated resources. Furthermore, we introduce a cheat-prevent auction mechanism based on ascending-bid auction to prevent greedy BSs from cheating activities to obtain lower prices for the auctioned resources. Finally we present simulation results on the performance of the proposed resource leasing mechanism.