This Small Business Innovation Research (SBIR) Phase I project approach to address the next challenge in wireless networking: to increase overall network throughput rates well in excess of a Gbps utilizing broadband links. This is an order of magnitude increase relative to today's systems. Achieving such high rates requires the nodes to incorporate learning and interference mitigation techniques, an optimum co-design of the MAC and PHY layers, and extensive experimentation. The key contribution of this work will be to identify a highly agile cognitive radio platform that can meet the demands of Gbps+ networking. Such a radio platform must (a) accurately sniff, identify, and characterize both friendly (inter network) or foreign (intra network) interference; (b) be quickly reconfigurable to morph itself into the optimal radio for a given set of environmental and user specified parameters; (c) learn interference conditions and leverage the knowledge towards improved network and MAC layer protocols. Development of such a platform requires a departure from traditional approaches and hinges on successful integration of state of the art signal processing and communication algorithms with the SDR framework. Additionally, comprehensive RF/Baseband co-design and optimization is needed, plus efficient learning and cataloguing techniques that are closely coupled to and under direct control of the network layer.
Although home data networking has pretty much cut the umbilical cord that tethered all computers and peripherals, that umbilical cord has not been fully severed in the enterprise due to the higher demand on aggregate network throughput. Moreover, the desire to broadcast video inside the home is also being hampered by the capabilities and robustness of today's wireless LAN solutions. This work addresses these issues by leapfrogging next generation WLAN activities and aggressively pursuing a high throughput highly agile physical layer co-designed with an efficient MAC. The work will help demonstrate the limits of WLAN capabilities and will help drive current and future standards activities in this domain. Moreover, it will help demonstrate the tremendous throughput improvements that are achievable when each node actively senses, learns, and ultimately adapts to its channel and environmental conditions.
