Abstract : Generalized area spectral efficiency (GASE) was introduced as a metric to quantify the spectral utilization efficiency of wireless transmissions. Unlike other performance metrics, GASE takes into account the spatial property of wireless transmissions. In this paper, we extend the research on GASE from link level to network level. In particular, we consider a wireless ad-hoc network with Poisson distributed nodes. We take into account the cochannel interference and derive the generic closed-form moment generating function (MGF) expression of aggregate interference in such network. With the interference statistics, we calculate the ergodic capacity, affected area, and GASE of the network over Rayleigh fading channels. Furthermore, we analyze the effect of carrier sense multiple access with collision avoidance (CSMA/CA) mechanism on the GASE performance of such network. Finally, we propose a new cognitive paradigm that allows the secondary transmitters that are located outside the primary affected area to transmit. With this paradigm, we can achieve high ergodic capacity while effectively utilizing the space-spectrum resource of primary network. Besides, through mathematical analysis and numerical examples, we show that GASE provides a new perspective on transmission power selection and secondary network optimization.
Index terms : Aggregate interference, ergodic capacity, Poisson point process, Rayleigh fading.