Service-Oriented Hybrid-Database-Assisted Spectrum Trading: A Blueprint for Future Licensed Spectrum Sharing

Licensed spectrum sharing (LSS) is an emerging scheme to improve spectrum efficiency by authorizing certain secondary spectrum licensees (SSLs) to share spectra with the primary spectrum licensees (PSLs). Spectrum trading is an effective way to implement LSS by offering incentives for PSLs to open their spectra. Although many research efforts have been devoted to such an economic-based scheme, most of them mainly focus on the trading part from the economic perspective, and the service provisioning related issues when implementing the trading-based LSS (TLSS) are generally ignored. In this article, we study the spectrum trading system from the service-oriented perspective considering all three aspects on SSLs, spectrum market, and PSLs, to promote the implementation of TLSS for the future. First, we introduce a cognitive capacity harvesting network architecture for SSLs to enable them to use the purchased PSLs' spectra for service provisioning. Then, considering the overhead on spectrum trading and sharing-based spectrum usage, we develop a hybrid-database-assisted model to facilitate TLSS, where information storage and trading process are carried locally, and the trading contextual information is clarified as well. Finally, we propose a service-oriented two-tier trading scheme, where two potential trading methods are discussed, namely, ex-post trading and ex-ante trading, which are especially suitable for trading the spectra with short-term and long-term availability, respectively.