Interference Modelling
and Cell Design in Microcellular Networks

PhD Thesis by Brendan C. Jones
Electronics Department
Macquarie University NSW 2109


Microcellular technologies are being developed to provide wireless communications to very large numbers of people at a much higher user density than is possible with conventional macrocell systems. Ubiquitous deployment of a high quality wireless communication system will require a methodology that facilitates rapid, low cost, and accurate system design. As the number of deployed microcells increases, site-by-site engineering may become too time consuming and costly.

The fundamental challenge in assessing microcell system performance is to model the interference of many users transmitting in a service area. This thesis develops a new microcell interference model to characterise microcell system performance under high density conditions. This model enables interference and cell radius distributions in microcell systems to be computed, and cell coverage quality to be related to basic system design parameters including cell spacing, user load, mobile terminal distribution, and channel spill characteristics.

The microcell interference model is based upon a parameter called the `Interference to Noise Ratio' or INR. It is shown that the INR and maximum possible cell radius for a mobile terminal are related by a simple function. The INR enables cell radius distributions to be calculated, providing a new method of analysing microcell coverage performance.

Through theory and simulation, it is shown that microcell systems are inherently more interference-limited than macrocell systems. Microcell systems consequently exhibit a larger cell radius variation than macrocell systems, and this variation is extremely sensitive to the location of the closest interferers. Terminal mobility is implicitly a part of the new microcell interference model. This enables characterisation of these interference effects based upon the statistical distribution of interferers.

Mathematical analysis is performed, under simplifying assumptions, to obtain closed form expressions for the INR and cell radius density and distribution functions as a function of three terminal distribution models, arbitrarily specified channel spills, and single and dual-slope propagation models. Monte Carlo simulations that do not make the simplifying assumptions of the analyses are performed to test the theory. The results indicate that fair to good agreement is obtained under a range of conditions. Avenues for improving the accuracy of the theory, and the applicability of the theory to a wider range of conditions are explored.

One avenue for improved interference analysis lies in better prediction of the statistics of the closest approach of an interferer. A novel analysis is presented which derives the channel reuse ratio distributions in DCA microcell systems for both cochannel interferers and adjacent channel interferers of arbitrary strength. It is shown that the closed form expressions obtained give the fundamental lower limit to channel reuse under the assumed conditions. Monte Carlo simulations are used to test the theory, and the results show very good accuracy in predicting cochannel reuse ratio distributions, and excellent accuracy in predicting adjacent channel reuse ratio distributions.

This analysis also proves that microcell systems using DCA exhibit fundamentally closer channel reuse than macrocell systems using FCA. Moreover, the closer frequency reuse is one of the fundamental reasons for the difference observed between the interference distributions, and ultimately performance, of FCA and DCA systems.

This research was undertaken as a step towards developing a methodology to generate microcell base station deployment locations using only gross design parameters of the microcell system and propagation environment. A sample microcell network design is presented to illustrate some of the concepts that would be involved in a microcell design methodology.

Thesis submitted 24 June 1996. Degree Awarded 16 May 1997.

Publication list

Web Site © 2013 email Brendan Jones