IEEE Transactions on Cognitive Communications and Networking | 2019

Improving Liquidity in Secondary Spectrum Markets: Virtualizing Spectrum for Fungibility

 
 
 

Abstract


Pricing mechanisms in the form of auctions have been the main method for spectrum assignment in the United States for over 20 years. The spectrum auctions carried out by the federal communications commission constitute a primary market for spectrum and have been affected by lack of flexibility which has resulted in inefficiencies in spectrum assignment, especially in environments where spectrum is considered scarce. In recent years, we have observed significant efforts to increase efficiency in spectrum assignment and use. Among those efforts is the design and adoption of secondary markets. Secondary markets have the potential to address inefficiencies arising in primary markets over time or those that occur through features of auction mechanisms by enabling spectrum to be assigned to users who value it the most. Furthermore, liquid secondary markets have enabled the explicit management of risk in other markets, such as agriculture and commodities, through futures and options trading. In this paper, we advance the study of liquidity in secondary markets that was begun in our previous work. We explore: 1) the reasons that may have hindered the emergence of liquid secondary markets for radio spectrum and 2) what we might change to promote secondary markets. With these objectives in mind, we study various configurations for the design of secondary markets, which account for the physical constraints inherent to electromagnetic spectrum. In addition, we study technical alternatives that would permit us to develop an appropriate, tradeable, and spectrum-related commodity. The results of our analysis show that lack of fungibility has an adverse impact on secondary market liquidity. To address this outcome, we propose virtualization of spectrum resources into fungible chunks and show that this improves market liquidity by yielding viable market outcomes in all the scenarios we tested.

Volume 5
Pages 252-266
DOI 10.1109/TCCN.2019.2901787
Language English
Journal IEEE Transactions on Cognitive Communications and Networking

Full Text