A Stochastic Geometry Approach to Delay Analysis for Mega LEO Satellite Networks

Compared to terrestrial networks, satellite-toground communications face severe propagation delay challenges, which limits the number of retransmissions. In this paper, we focus on the local delay in mega downlink low earth orbit satellite networks, which is defined as the mean times required for a message successfully transmitted. Modeling the locations of the satellites as a binomial point process and considering directional transmission, we derive the conditional success probability under a given network topology. On this basis, we give the analytical expression for the local delay. To highlight the influence of interference on the local delay, we also analyze the case without considering the aggregated interference (i.e., the noise-limited case). Numerical results verify our analysis and illustrate the effect of key system parameters, such as the directional antennas size, the satellites’ number and altitude on the local delay.