The Application Of Queuing Theory To Alleviate Traffic Congestion In GSM

Traffic congestion is an important factor which effects the GSM network run quality. By the traffic congestion reason, the MS can not connect to the network in the user's call processes, this will mar the consumer's mood, influence their faithfulness for the network, so some one will abandon the network, the profit of the GSM network operators will be damaged. Therefore, it is very important for every mobile telecommunication corporation that some methods must be executed in order to alleviate the network traffic congestion.There are a lot of habitual methods to alleviate the traffic congestion, such as building new 1800M BTSs, hardware expansion, cell splitting, controlling the coverage ranges of the BTS by adjusting their antenna's angle, amending the radio interface parameters for the traffic, performing the half-rate function, etc. These methods have some disadvantages in alleviating traffic congestion in the countryside. In fact, there is another way which is often neglected by many people to solve the traffic congestion in the user's call processes, that is, we can use Queuing Theory. This paper is just about how to solve the GSM traffic congestion problem by the use of Queuing Theory.The main idea of this paper is, on the base of the GSM theory and criterion, in order to benefit the people's sense about the network, we can introduce Queuing Theory into the MS calling process and establish M/M/n mathematic queuing model, so that we can discuss the method of alleviating GSM traffic congestion. Whereafter, OPNET simulation software is used to simulate the selected actual cell with traffic congestion from one country, with the given intensity of the user's call, trying to prolong the network's processing time, the simulation result shows that the number of the call applications which stay in the system is increased, this means that it is possible to increase the user's number which system can hold by prolonging the system's processing time to a certain value. Thereby, call back rate may be reduced to a low level, and the system capacity is highly improved. At last, in the actual GSM network several cells are selected to unlock the queuing function, the presented outcome shows the same modality as the OPNET simulation results. All of these prove that it is effective to solve the problems of the GSM traffic congestion via using Queuing Theory. The conclusion of this paper is valuable for those engineers whose tasks is to optimize the performance of their GSM network.