Achievable Capacity Under the Interference Temperature Model

The interference temperature model was proposed by the FCC in 2003 as a way to dynamically manage and allocate spectrum resources. It would allow unlicensed radios to sense their current RF environment and transmit in licensed bands, provided their transmission does not raise the interference temperature for that frequency band over the interference temperature limit. It never received much interest because nobody was sure exactly how to use it or how if it would work. This research focuses on a mathematical analysis of the interference temperature model in an effort to examine the relationships between the capacity achieved by the unlicensed network and the interference caused to the licensed network. We develop a model for the RF environment and determine probability distributions governing interference temperature as a function of various elements in the model. We then determine bounds on the amount of interference caused by implementing such a system. We examine model environments for a wireless WAN and a wireless LAN, each coexisting with a licensed carrier. For each, we quantify both the impact on the licensed signal and also the capacity achieved by our underlay network. By substituting numeric values for RF environments in which the interference temperature model might be applied, we show that achievable capacity is very small, while the impact the licensee can be very large. Based on this, we propose alternate usages for interference temperature and ways to boost capacity.