| Summary: | The central limit theorem (CLT)-based Gaussian approximation facilitates the probability analysis of cognitive radio (CR), but it incurs data rate loss due to the inaccurate configuration of the optimal sensing duration (OSD). To estimate the data rate loss, an approach to obtain a tight upper bound of the maximal achievable data rate over a Nakagami-m fading via-to-sense channel is proposed. Moreover, a direct formula is more desirable than an iteration scheme for a highly mobile CR that has to calculate the OSD more quickly. To meet this need, a strategy to achieve the OSD based on exponential interpolation is proposed. For Nakagami-Gamma shadowed fading via-to-sense channels, the proposed scheme is demonstrated to have advantages over the popularly utilized Hermite spline, such as: 1) a more tractable derivation of the probability density function (PDF) of the OSD and 2) a less complex calculation of the ergodicsensing capacity. These advantages are significantly favorable for a CR network to achieve low latency communication and swiftly perform quality of service management to adapt to the high mobility.
|
|---|
