in order to project a reliable telecommunication network ,we have to measure the call lossrate of present traffic and to predzct the traffic of hajective network. In this paper we point out that afactor of real commun...in order to project a reliable telecommunication network ,we have to measure the call lossrate of present traffic and to predzct the traffic of hajective network. In this paper we point out that afactor of real communication needs for subscrihas should be added in studying the traffic source char-acterization. Generally, the predicted traffic does not equal to the real demends, Considered asubscriber's psychological facter for attempts in busy-hour, a mathematical medel of probable dertva-tive calls from each real communication demand is given,with which the calling-up probabity, andthe repeating attempt probability of original subscribers on line occupation are calculated and the rela-tionship between the traffic and the real cammunication need is predicted.展开更多
The network traffic prediction is important for service quality control in computer network.The performance of the traditional prediction method significantly degrades for the burst short-term flow.In view of the prob...The network traffic prediction is important for service quality control in computer network.The performance of the traditional prediction method significantly degrades for the burst short-term flow.In view of the problem,this paper proposes a double LSTMs structure,one of which acts as the main flow predictor,another as the detector of the time the burst flow starts at.The two LSTM units can exchange information about their internal states,and the predictor uses the detector’s information to improve the accuracy of the prediction.A training algorithm is developed specially to train the structure offline.To obtain the prediction online,a pulse series is used as a simulant of the burst event.A simulation experiment is designed to test performance of the predictor.The results of the experiment show that the prediction accuracy of the double LSTM structure is significantly improved,compared with the traditional single LSTM structure.展开更多
文摘in order to project a reliable telecommunication network ,we have to measure the call lossrate of present traffic and to predzct the traffic of hajective network. In this paper we point out that afactor of real communication needs for subscrihas should be added in studying the traffic source char-acterization. Generally, the predicted traffic does not equal to the real demends, Considered asubscriber's psychological facter for attempts in busy-hour, a mathematical medel of probable dertva-tive calls from each real communication demand is given,with which the calling-up probabity, andthe repeating attempt probability of original subscribers on line occupation are calculated and the rela-tionship between the traffic and the real cammunication need is predicted.
基金the research plan of State Grid Sichuan Electric Power Company,Chinathe research plan of the 10th Research Institute of China Electronics Technology Group Corporation(KTYT-XY-002).
文摘The network traffic prediction is important for service quality control in computer network.The performance of the traditional prediction method significantly degrades for the burst short-term flow.In view of the problem,this paper proposes a double LSTMs structure,one of which acts as the main flow predictor,another as the detector of the time the burst flow starts at.The two LSTM units can exchange information about their internal states,and the predictor uses the detector’s information to improve the accuracy of the prediction.A training algorithm is developed specially to train the structure offline.To obtain the prediction online,a pulse series is used as a simulant of the burst event.A simulation experiment is designed to test performance of the predictor.The results of the experiment show that the prediction accuracy of the double LSTM structure is significantly improved,compared with the traditional single LSTM structure.
