Incentivizing Secondary Spectrum Trading: A Profit Perspective

Abstract

With the ever-increasingly connected mobile devices, demand for mobile broadband service is likely to outstrip spectrum capacity in the near-term. Without action to address this spectrum crisis, service quality is likely to suffer and prices are likely to rise. Fortunately, recent studies show that a large part of licensed spectrum remains under-utilized, which should allow concepts such as dynamic spectrum access/sharing, open access, and secondary spectrum market to alleviate the crisis. From the inception of the open access paradigm, it was clear that for it to work two issues must be adequately addressed: sensing and pricing. The first refers to the ability of a device to accurately detect channel opportunity and more generally to acquire information on the spectrum environment. The second refers to mechanisms that provide license holders with the right incentives so that they will willingly allow access by unlicensed devices. For the pricing issue, we formulate a contract design problem where a primary license holder wishes to profit from its excess spectrum capacity by selling it to potential secondary users. It needs to determine how to optimally price the excess spectrum so as to maximize its profit, knowing that this excess capacity is stochastic in nature, does not come with exclusive access. We adopt as a reference a traditional spectrum market where the buyer can purchase exclusive access with fixed/deterministic guarantees. When multiple primary holders exist, we develop a price competition model for the license holders selling on a secondary spectrum market. We introduce a regulator which can also be thought of as the sellers forming a coalition, whose role is to enable money transfer based on partial observations of the sellers' actions. We show that by proper design of the transfer mechanism, efficient equilibrium (profit-maximizing) can be achieved. On the sensing front, we propose a spectrum utilization model which uses stochastic differential equations (SDE) to model dynamic scattering and multipath fading channels, in particular, Rayleigh-distributed stationary channels. The SDE model can generate spectrum dynamics as a temporal process, and is shown to provide very good fit for real spectrum measurement data.