The Research On Traffic Forecasting Model In Mobile Communication

Long-term and stable operation of mobile network depends on timely and effective network planning and optimization, the network operators'initiatively changes successfully address changes in user behavior and trends. Therefore, forecasting mobile communication traffic based on traffic statistics data and other business information on trends, then determining the capacity of the project. All the area set the corresponding traffic model to calculate effective capacity of the equipment for determining the design capacity to meet the type and amount of equipment. It is very necessary for the mobile operators. The main contents:1. Based on the speciality of traffic load in the paper, an ultra-short-term traffic load forecasting model based on weighted markov chain was proposed. The model needs less samples, operates fast, forecasts precisely, and is able to verify. Besides, this model applies the Fuzzy Fisher Criterion's clustering algorithm,which is based on the limitation of itself. The proposed method not only avoids the dependence of the validity of clustering on the space distribution of samples, but also raises the clustering efficiency. Through analyzing the experiment results, the algorithm can be used in mobile traffic load forecasting with rapid speed and high precision.2.Based on the speciality of traffic load in the paper, a medium-term traffic load forecasting model based on microcanonical annealing-support vector machines (SVM) was proposed. Appropriate parameters are very crucial to SVM forecasting ability, the optimal parameters selection was achieved by random walks of demons in the state space of microcanonical annealing (MA) algorithm. The verification on the model with real traffic data shows that, this algorithm will offer better results with higher probability to hit the global optimum than Simulated Annealing (SA) algorithm, objective function value also decreased faster, and have high precision.