Joint Power and Channel Allocation for Cognitive Radios

Spectrum scarcity is the major challenge facing all parties working in the telecommunications industry. As recently shown, the main problem of spectrum unavailability is due to spectrum inefficient utilization rather than spectrum scarcity. Inspired by this argument, a paradigm shift in spectrum management is achieved by the aid of smart radios called cognitive radios that can opportunistically access the available spectrum. In this context, we address the problems of power control and channel assignment jointly. Namely, given a set of available channels which is determined dynamically, we devise a near-optimal yet simple algorithm with linear complexity targeting capacity maximization of a cognitive radio network while jointly optimizing power and channel allocation among users (cognitive radios) while respecting total power constraints per individual users. In an extended version, we also consider a conservative design where signal-to-interference-plus-noise ratio (SINR) constraints at the primary users have are imposed. Simulated experiments illustrate the near optimal performance of the devised algorithm compared to the exhaustive search approach. Moreover, the limits of the system capacity are quantified given both ideal (no SINR constraints) and conservative (with SINR constraints) designs.