Spatial Interference Cancelation for Mobile Ad Hoc Networks: Perfect CSI

TL;DR

This paper analyzes how perfect channel state information and multiple antennas enable spatial interference cancelation, significantly boosting the capacity of mobile ad hoc networks by applying stochastic geometry and zero-forcing beamforming.

Contribution

It provides a mathematical analysis of network capacity with spatial interference cancelation using perfect CSI and multiple antennas, deriving scaling laws and demonstrating capacity gains.

Findings

Transmission capacity scales as a power law with outage probability.

Adding extra antennas greatly increases capacity, often by an order of magnitude.

Spatial interference cancelation significantly improves network throughput.

Abstract

Interference between nodes directly limits the capacity of mobile ad hoc networks. This paper focuses on spatial interference cancelation with perfect channel state information (CSI), and analyzes the corresponding network capacity. Specifically, by using multiple antennas, zero-forcing beamforming is applied at each receiver for canceling the strongest interferers. Given spatial interference cancelation, the network transmission capacity is analyzed in this paper, which is defined as the maximum transmitting node density under constraints on outage and the signal-to-interference-noise ratio. Assuming the Poisson distribution for the locations of network nodes and spatially i.i.d. Rayleigh fading channels, mathematical tools from stochastic geometry are applied for deriving scaling laws for transmission capacity. Specifically, for small target outage probability, transmission capacity is proved to increase following a power law, where the exponent is the inverse of the size of antenna array or larger depending on the pass loss exponent. As shown by simulations, spatial interference cancelation increases transmission capacity by an order of magnitude or more even if only one extra antenna is added to each node.