Modelling is an activity that fits within the family of methodologies in what is often referred to as the scientific method of enquiry. This method requires the statement of a research question. Evidence is then gathered through a series of activities that explore the research question. Finally, a conclusion is reached. Alternatively, the evidence may permit a hypothesis to be put forward for further investigation.

So how does modelling help? Consider a question on spectrum coexistence: To what extent can application specific license exempt devices satisfactorily share with licensed WiMAX networks in the urban environment? This can be answered by outlining the representative networks in a modelling tool, developing scenarios for the operation of the various devices and determining the degree of interference between the devices and networks. From this an answer to the research question can be reasoned.

Consider another question: How many sites are needed for national coverage given two different spectrum block sizes and two different network deployment strategies? This can be answered by auto-planning a series of networks. Curves are drawn for coverage versus site count for a variety of traffic loading. This is repeated for different block sizes, resulting in a series of curves. The site count is asymptotic to 98% coverage and the number of sites needed for various percentages can be given.

Modelling is a powerful way of revealing evidence to support, or refute, research questions. What’s your question?