ProTOMAC: Proactive Transmit Opportunity Detection at the MAC Layer for Cognitive Radios

摘要

Traditional Cognitive Radio Networks aim to utilize the radio spectrum white space. We present a radically different approach which exploits the excess Signal-to-Noise ratios at which the primary network nodes operate. This paper introduces ProTOMAC (Proactive Transmit Opportunity Detection at MAC Layer), the first spectrum access paradigm that leverages the gray space present at the MAC layer via a novel conceptual extension of the Interference Temperature (IT) approach. The ProTOMAC architecture is based on detecting changes in the primary network packet statistics. Specifically, we develop novel techniques based on iteratively updated information theoretic divergence measures along with rapid and accurate kernel density estimators. These enable the secondary network to operate subject to an interference constraint that ensures a given primary network Quality Of Service (QOS). ProTOMAC presents a generalized scheme for coexistence of dissimilar Packet Based Networks. We provide an efficient implementation of ProTOMAC using commercial off-the-shelf equipment. The coexistence efficacy of ProTOMAC is validated on an IEEE 802.11g WLAN via establishing a 500 kbps secondary link with a range of 3 meters with an early interference detection time of 240-600 ms. For the first time, identification and exploitation of hitherto unutilized transmission opportunities in Packet Based Networks has been made possible.