iCover - Indoor Coverage Simulation Tool

Abstract

In the age of the Communications, when everybody needs to stay connected, to their mobile phone operator, or internet, the lack of coverage in indoor environments is one of the main worries. People needs to stay always connected, does not care if it is a shopping mall, an underground station or an airport. In the last years many mobile phone operators and wireless networks installers, started to research the phenomena of the propagation of signals in indoor systems attending to the needs of the population. When designing a communications system, one of the critical points is modelling the channel of radio communication. Once we have an appropriate model of propagation we can predict the behaviour of the signals when propagating in our environment. The propagation models allow us to predict the power (And thus the loss) of a signal received at a certain distance of the transmitter, what allows us to plan wireless networks with more accuracy. The principles of radio transmission in indoor environments are the same than in outdoors environments, reflection, refraction, dispersion, but the conditions vary a lot from one environment to another. The lack of line of sight between transmitter and the point of measurement is a disadvantage that makes harder to determine the bandwidth of the channels and the quality of the transmission, because of the presence of obstacles like walls. We have to take in account the distribution of the floors, the height of the walls, the materials they are made of and even the furniture. The actual system for modelling indoor propagation is by mean of measurements taken in the area where we want to know the coverage. With those measures make a simulation and determine the coverage. This system involves having to send workers to the place of study, what means a waste of money, human resources, and time. This led them to the development and study of various propagation models that could be implemented in simulation tools that could predict the coverage without needing to take those in-the-field measurements. Many indoor propagation models have been developed, both empirical and deterministic, each one with its own advantages and disadvantages. Some work better in some environments and circumstances than others. There are many propagation models and we will discuss them in the chapter 1 of this document, basically we can divide them in two groups, the deterministic ones and the empirical ones. The deterministic ones are base on Maxwell equations and the empirical ones are based on experimental measurements. There are many simulation tools based in those propagation models, but there is another model to take in account. In 1998 three researchers, Kwok-Wai Cheung, Jonathan H-M Sau and R.D Murch developed a new indoor propagation model [1]. This model is an empirical model that provides accuracy similar to the ray-tracing techniques, like the uniform theory of diffraction (UTD), requiring lower computation times. Because of that, it was selected as base for the tool developed in this project. This model will be explained deeply in chapter 2. The objective of this project was to develop a simulation tool based on that empirical model keeping the low computation times and accuracy of the model, offering a user friendly interface with commercial purposes. The platform chosen for developing this tool has been Matlab® from MathworksTM. The reason for selecting this option was because the programming language used in Matlab® is very similar to C, is compatible with other programming languages, it provides many graphic tools and has its own mathematical libraries. The result has been a tool able to predict the looses of a wireless system in an indoor environment, taking in account the number of floors, the angle of incidence of the signal with the floors and walls, the material of the walls, the looses due to the distance between transmitter and point of measurement and the influence of the diffraction in the corners of the walls. How this tool works will be explained deeply in chapter 3. After comparing our tool with other ray-tracing based tools, and some experimental measurements, the accuracy of the empirical model has been proved to be pretty good and with low computation times. The comparison between both tools will be explained deeply in chapter 4. So, is this the best propagation model? Is this tool finished or needs improvement? In chapter 5 you will find all the conclusions obtained from this project and what I think are the next work lines to take in account.