Cognitive radio networks enable the sharing of the wireless spectrum with licensed users, but impose challenges due to the fluctuating nature of the spectrum as well as the diverse quality-of-service (QoS) requirements. To address these challenges, a spectrum-aware COGnitive radio NETwork (COGNET) based on orthogonal frequency division multiplexing (OFDM) is proposed.
In COGNET, five major research directions are investigated: (1) First, a frequency agile, wideband analog/RF front-end technology is developed, which is robust to non-linearity (RF/analog IC design). (2) To detect the presence of licensed users, a fast detection scheme is developed by exploiting the spatial diversity inherent to multi-user networks (spectrum sensing), (3) To select the best channel, novel decision methods are proposed that consider channel information, licensed user activities, and application requirements (spectrum decision). (4) A dual-mode spectrum sharing framework is proposed, which enables access to existing networks as well as coordination between cognitive radio users (spectrum sharing). (5) A spectrum mobility management framework is proposed to achieve efficient operations as cognitive users switch between the channels when a licensed user is detected (spectrum mobility). Based on these research topics, a testbed is developed to demonstrate the protocols developed on cognitive radio transceivers.
Although COGNET was originally proposed for both infrastructure and ad hoc networks, due to the budget cut, research topics on ad hoc networks ("cooperative spectrum sensing for COGNET ad-hoc access" and "spectrum-adaptive route recovery for COGNET ad hoc access") will not be investigated.
