System Throughput Performances Analyses with Stochastic Geometry in Unlicensed Spectrum Systems

With the increasing shortage of spectrum resources, the use of unlicensed spectrum is garnering more attention. This paper establishes a theoretical framework based on stochastic geometry to analyze the throughput performances of a cellular network with unlicensed spectrum. Considering that both cellular and WiFi users adopt a listen-before-talk (LBT) spectrum sharing scheme, we model it as a hard core point process (HCPP). Our analysis focuses on the combined effect of cellular user and WiFi user density on the throughput performances, and obtains their integral solution. The results show that the use of unlicensed spectrum can increase the throughput of cellular users when the densities of the two types of users are considered, which are of practical value to holistic designs and deployments of License Assisted Access (LAA) systems.