No abstract

In this paper, we consider the CSMA/CA multihop networks where the two end-nodes transmit their packets to each other and each intermediate node adopts network coding for delivering bidirectional flows. In addition, the neighbor nodes are randomly uniformly deployed with the Poisson Point Process. By varying the combination of the physical carrier-sensing range of the transmitter node and the target signal-to-interference ratio (SIR) set by the receiver node, we can control the interference level in the network and the degree of spatial reuse of a frequency band. The larger the carrier-sensing range is, the smaller the interference level, while the smaller the opportunity of getting a channel by a node. Similarly, the higher the target SIR value is, the more probable the retransmission (by the exponential random backoff) is, while the better the link quality on successful transmission is getting. Under this tradeoff context, we find the optimal combinations of these two factors that make the end-to-end throughput of the flow maximal for three different retransmission schemes.Index Terms-Carrier-sensing multiple access/collision avoidance (CSMA/CA), carrier-sensing range, cross-layer optimization, network coding (NC), Poisson point process (PPP), target signal-to-interference ratio (SIR), wireless multihop networks.

No abstract

From the results of the ray-tracing simulation of 28 GHz frequency band over downtown of Ottawa and New York University campus, we propose 3D large-scale channel models for urban city which are applicable to a mmWave version of 3D spatial channel model. Due to the change of geographical topology dimension from 2D to 3D, the height information of the transmitter should be reflected to the parameters of the channel model such as line-of-sight probability, Ricean K factor, path loss and shadow fading accordingly. In the simulation results, the line-of-sight probability depends on the height of the transmitter compared to the 2D model. The dual-slope path loss model is still appropriate in 3D channel model. We also propose the linear standard deviation function of the shadow fading, which shows a large standard deviation and an increasing trend with the distance.

No abstract