The objective of the project is to accurately predict the performance of the Smart Metering Implementation Programme (SMIP) service on constricted bandwidth systems, such as 2G only. It is necessary to perform some form of simulation of the likely SMIP events success rates when being initiated alongside load from other services such as legacy Voice and other non-SMIP background or interactive Data services.
Event based simulation allows for individual types of traffic to be generated by separate random processes, and as each is competing for the same set of cell radio resources then the probability of session success can be estimated based on the simulated load. The impact of varying levels of voice traffic and other Non-SMIP traffic on the success probability of SMIP procedures can then be estimated.
The model will need to consider how the combination of different services i.e. Voice, non-SMIP data and SMIP service all interact with each other. The model will need to simulate the birth and death of voice calls, non-SMIP data sessions and SMIP sessions. Each session, when initiated, will need to be allocated radio resource as per the description below. On completion of each session, the resource will be released and made available for another random birth event. Each of the 3 services will have independent random arrival processes. Statistics on the success rates for each service is the key output of the simulation.