A Mixed Integer Programming Approach for 5G Spectrum Auction Strategy Optimization

In most countries, mobile network operators compete through auctions in order to acquire spectrum, necessary resource to deliver sustainable services to customers in the forth-coming decades. These auctions can present complex mechanism designs which make them hard to analyze, and therefore to participate in. In this paper we propose a new approach based on mixed integer linear programming to address the problem of finding an optimal strategy for a bidder, considering a full information auction. This approach enables to simulate realistic instances of auctions, and to quantify the performances of the optimal strategies found against the Straightforward Bidding strategy, a widely studied participation strategy for multi-round auctions.