File Exchange

image thumbnail

SINR-based k-coverage probability in cellular networks

version 1.4 (12.5 KB) by

Calculating the k-coverage probability with Poisson process-based cellular network models.

16 Downloads

Updated

View License

The scripts calculate the k-coverage probability (based on SINR* values) in a single-tier cellular network using a method based on a homogeneous Poisson process model. More details are found in the (submitted) work [1], which presents the model that these scripts are based on.

The script funProbCov.m uses a inclusion-exclusion-like formula and two types of integral to calculate the k-coverage probability in a network with log-normal shadowing (though the shadowing distribution can be somewhat arbitrary [1]) and without fading.

The simpler integral I_n uses a quadrature method or a simple analytic formula (for the zero-noise or “interference limited” case). The more complex high-dimensional integral J_n uses quadrature methods for low dimensions and quasi-random (Sobol) integration for higher (n>2) dimensions.

The script funProbCovFade.m calculates 1-coverage probability for a network with Rayleigh fading (exponentially distributed with unit mean) and log-normal shadowing. Numerical integration of hypergeometric function 2F1 is used when the model has noise. A close-form solution with 2F1 is used in the no noise case.

Simulation scripts are also included for comparison purposes. All network base stations are sampled on a disk region. The disk region needs to be large enough to reduce "edge effects", which become more prominent when fading is included.

Running the file TestSimVsInt.m is a good place to start. The fading-incorporated model is demonstrated in TestSimVsIntFade.m. Default parameters are based on a Walfisch-Ikegami model for a urban environment. The (submitted) associated work [1] has more details on the equations and the model.

*SINR = signal-to-interference-and-noise-ratio

[1] H.P Keeler, B. Błaszczyszyn and M. Karray, 'SINR-based k-coverage probability in cellular networks with arbitrary shadowing', 2013. Online at http://arxiv.org/abs/1301.6491

Comments and Ratings (3)

This file gives a great insight on how the multi-tier network works and also it counts as a good tutorial on network simulation.
I am wondering if the Weighted Voronoi Teselian diagram was included.
Thank you

meng liu

meng liu

Updates

1.4

The funJn.m has been updated. It uses a fact, discovered in later work*, that eta values sum to one.

*Blaszczyszyn and Keeler, Studying the SINR process of the typical user in Poisson networks by using its factorial moment measures, submitted 2014

1.3

Fixed up some typos and improved some of the code slightly.

1.2

Uses a different (slightly better) numerical integration method for funIn.m.

1.1

Owner.

MATLAB Release
MATLAB 7.9 (R2009b)

Download apps, toolboxes, and other File Exchange content using Add-On Explorer in MATLAB.

» Watch video