Protecting Primary Users in Cognitive Radio Networks: Peak or Average Interference Power Constraint?

This paper considers spectrum sharing between a cognitive radio (CR) and a primary radio (PR) where the CR protects the PR transmission by regulating the resultant interference power level at the PR receiver to be below some predefined threshold. The interference-power constraint at the PR receiver is usually one of the following two types: average interference power (AIP) constraint that regulates the average power level over different fading states and peak interference power (PIP) constraint that limits the peak power level at each fading state. From CR's perspective, AIP constraint is more favorable than PIP constraint because of its more flexibility for dynamic power allocations. On the contrary, from the perspective of protecting the PR, the more restrictive PIP constraint appears at a first glance to be a better option. Some surprisingly, this paper shows that in terms of various forms of capacity limits achievable for the PR fading channel, namely, ergodic and outage capacities, AIP constraint is also superior over PIP constraint. This result is based upon an interesting interference diversity phenomenon, i.e., variable interference power levels at the PR receiver in the AIP case are more advantageous over constant ones in the PIP case for minimizing the resulted PR capacity losses. Therefore, AIP constraint leads to larger fading channel capacities over PIP constraint for both CR and PR transmissions.