Traffic forecasting is an important task which is required by overload warning and capacity planning for mobile networks. Based on analysis of real data collected by China Mobile Communications Corporation (CMCC) Heilongjiang Co. Ltd, this paper proposes to use the multiplicative seasonal ARIMA models for mobile communication traffic forecasting. Experiments and test results show that the whole solution presented in this paper is feasible and effective to fulfill the requirements in traffic forecasting application for mobile networks.

Traffic forecasting for mobile networks with multiplicative seasonal ARIMA models

Accurate traffic dimensioning plays an important role in telecommunications network planning and is particularly important for the performance analysis of mobile and wireless networks. The main goal of cellular operators now a day with the increasing number of mobile subscribers is to keep the subscribers satisfied with the delivered quality of service ( QoS ) . Traffic analysis & optimization is thus paramount importance for network planning and design. They are useful in areas such as network architecture comparisons, network resource allocations and performance evaluations. In order to achieve the best performance, service providers have to monitor and optimize their network continuously .A Operation Support Subsystem (OSS) with an online database is responsible for collection of data on live networks. In this paper, we have summarized the measurements taken on an operative GSM-900 & GSM-1800 network to evaluate and optimize traffic. Here we have established statistically the following facts (i) The peak hour in Orissa varies in four different MSC areas.(ii) Week day traffic is different from weekend traffic and the traffic intensity of weekday traffic is more than that of weekend traffic.(iii) Traffic channel availability is well within limits which implies that the network is well planned.(iv) SDCCH congestion is beyond threshold limits which implies operator is losing big revenue due to this. (v) In some cases the traffic carried exceeds the maximum full rate traffic resulting utilization of half rate TCH & thus reduction in speech quality.

Traffic Analysis and Optimization of GSM Network

Global System for Mobile Communications (GSM) is currently one of the most widely and most demanding telecommunication applications in the world. Researchers have reported that radiation from base stations may be dangerous to public health and that some human diseases are related to RF fleld exposure. Considering that in the past almost all the EM radiation assessments were focused on the Maximum transmission power of base station, and no statistical analyses have been performed on transmitted power's variation with the tra-c. An accurate method for predicting electromagnetic (EM) radiation from GSM base stations is proposed in this paper. It is based on the Poisson distribution of GSM-transmitted signals to calculate GSM transmitted power at difierent time periods. The theoretical calculation data flts well with the measurement data. Measurement results conflrm that electromagnetic radiation varies with changing tra-c and power density at difierent times with varying tra-c strength is more accurate than implementing only maximum power (20W) calculation. In some occasion, maximal power density is about 6"W/cm 2 for 15m in rush hours, but minimum power density is only 0.03"W/cm 2 for 15m in idle hours.

Accurate Method to Estimate EM Radiation from a GSM Base Station

With the evolution of various new types of application services for mobile devices, cellular operators have started providing multiple subscription plans to the mobile users. The plan subscribed determines the Quality of Service (QoS) to be provided to the user, and operators distinguish users as priority users and non-priority users, accordingly. To ensure better QoS to the priority users, necessary resources must be reserved at the base station. This demands analyzing the network traffic to predict future traffic pattern. This paper pioneers the use of machine learning methods to forecast mobile user traffic pattern for providing better QoS to the priority users. We analyze two different supervised learning methods, Naive Bayes Classifier and Logistic Regression, used for prediction of probable times when a priority user would be active. The prediction results are applied to the user scheduling strategies for efficient bandwidth management, thus improving system capacity and reduce blocking. Simulations on multiple real-life datasets validate the model and predict the mobile user pattern with very high accuracy, along with significant reduction in priority user service blocking ratio and improvement in their capacity.

A learning based mobile user traffic characterization for efficient resource management in cellular networks

As mobile networks are growing fast in size and complexity, radio network resource planning needs to be optimized to cope with both increased traffic demand and spectrum scarcity. Although many studies have been reported in the area of mobile network planning in terms of traffic distribution analysis and mobility modeling, relatively few studies have been achieved regarding network planning based on real-world data and measurements. A realistic model has been developed in order to evaluate the performance of a real GSM network providing a voice service in an urban area. In this framework, our contribution addresses an important issue: how to monitor the performance measurements collected in the operations and maintenance centre (OMC) to propose a complete study including a realistic mobility model and a qualitative and quantitative traffic characterization that constitute a generic planning model which can be customized by any operator for mobile network deployment.

Real mechanisms for mobile networks modeling and engineering

The core task of cellular system planning is to design an optimal radio network which provides the largest amount of traffic for a given number of channels at a specified level of quality of service. During this process, these objectives are achieved by an accurate traffic characterization and a precise analysis of mobile users' behavior in terms of mobility and traffic. The aim of this paper is to review traffic engineering issues and to develop analysis techniques to extract the relevant parameters from measurements realized on an operating GSM network. The results can be used to model the traffic in cellular networks and apply it in the planning process.

Traffic characterization for mobile networks

This paper presents a realistic characterization of the channel holding time in an operating GSM network covering an urban area. Our approach is entirely statistical: the data set to be fitted has been provided by the OMC (operations and maintenance center). The Kolmogorov Smirnov goodness of fit test is used in order to test the suitability of the exhibited distribution. From this statistical analysis, it can be concluded that the negative exponential distribution is a good approximation for describing the channel holding time.

H. Khedher

Papers

Traffic characterization for mobile networks

Channel holding time characterization in real GSM network