Log-normal shadowing meets SINR: A numerical study of Capacity in Wireless Networks

The capacity of wireless multi-hop networks has been studied extensively in recent years. Most existing work tackles the problem from an asymptotic perspective and assumes a simplified physical layer model, as e.g., the protocol interference model or the path loss radio propagation. Real life network planning, provisioning and deployment can only be done with more precise statements about capacity in finite networks. With this in mind, we adopt a numerical approach based on Monte-Carlo methods to study capacity under various interference and radio propagation models, including the physical interference model and log-normal shadowing radio propagation. Our results indicate that, depending on the interference model, capacity may experience a three phase transition related to the connectivity of the network. We further show that throughput capacity increases in the presence of randomized radio propagation, even above the critical node density. Our analysis of the numerical data illustrates that log-normal shadowing creates more interference, but decreases the total amount of transmissions to be scheduled.