Two-hop delay analysis in cognitive machine-to-machine networks with poisson point process model

End-to-end delay is an important metric for network performance. This paper studies the end-to-end delay in a two-hop Cognitive Machine-to-Machine (CM2M) network, where a pair of primary users (PUs) coexist with a secondary (SR) network with many two-hop line M2M communication links. The occupancy of the licensed channel by PUs is modelled as a continuous-time two-state (busy and idle) Markov process. Secondary (SR) network is composed of many two-hop line routes, each of which consists of a source node, a relay node and a destination. The locations of the source nodes are assumed to be the realizations of independent Poisson Point Process in each time slot. Secondary users (SUs) deploy a combination of TDMA and ALOHA medium access control (MAC) protocol. We first quantify the packet success probability for single-hop when the channel is idle for SUs, where the capture of a packet is determined by signal-to-interference ratio (SIR) experienced by the receiver. We then evaluate the delays of each hop (1st hop and 2nd hop delay) and the two-hop delay. Numerical results show that a better end-to-end performance could be obtained through proper network parameters settings. Simulation results verify the correctness of theoretical analysis.