In [1]-[5], we have developed a mathematical theory of spectrum jails to enable light-handed spectrum regulation. The major result of this work is that trust can be guaranteed at certification time regardless of the technical solution for finding spectrum holes. Then, performance can improve as technology gets better. This paper is the policy-oriented case for why a new model of trust is needed in spectrum and why spectrum jails are a good starting point for implementing this new trust model. Whereas earlier work developed the mathematical theory assuming a stand-alone enforcement system, this paper shows that spectrum jails can actually be implemented through databases. The database, instead of directing exactly where and when the secondary can transmit, will control the opportunity for the secondary to find spectrum holes. The secondary can then use whatever technical means it desires to recover spectrum holes, understanding that it will no longer be allowed to do so if it causes interference. Interestingly, our results show that the database should gate access to all bands, including unlicensed and exclusively licensed bands. Normally, all devices have access to the ISM bands, and those certified to have access to the licensed bands have permission to use these bands. However, to actually deter harmful interference, the databases must have the power to disable access to these other bands as well. The result is a system that operates as a spectrum tribunal, except run by robots at a fast time scale and a very low cost. It can also adapt its infrastructure to the local situation, and easily evolve as technology changes.
展开▼
