Cognitive spectrum access in femtocell networks exploiting nearest interferer information

In this paper, we consider the sensing-based dynamic spectrum access problem for femtocells in the two-tier network. We propose to exploit the information of the nearest macrocell base stations (MBSs) activity (i.e., on or off activity on the target frequency band) so that each femtocell base station (FBS) can make its access decision differently depending on the activity of the nearest MBS and the detected interference plus noise energy level. We then develop an analytical model to evaluate the performance of the proposed sensing-based access scheme in terms of success probability and total throughput by employing the stochastic geometry technique. We also provide an efficient and low-complexity algorithm to determine the sensing time and access thresholds for throughput maximization. Numerical studies are then conducted to validate the analysis and demonstrate that our proposed spectrum access with the information of nearest MBS activity achieves better total throughput performance than other scheme with no sensing or sensing without nearest MBS activity information.