论文部分内容阅读
针对广电网络中用户视频点播延迟时间不能保证、服务质量不稳定的状况,提出了一种新的视频服务器服务质量(QoS)控制方法,基于BP神经网络比例积分微分控制,以响应的延迟率与设定值之间的差值和变化率作为网络的学习依据,通过动态调整BP神经网络学习的方法,依据网络学习的实际情况,快速地调整服务请求的响应时间,保证视频服务器的服务质量。仿真实验表明,BP神经网络能够很好地控制请求的出错率和响应时间,为广电用户的视频点播服务提供更加稳定的服务质量。
Aiming at the situation that the user’s video on demand delay time can not be guaranteed and the quality of service is unstable in the radio and television network, a new quality of service (QoS) control method for video server is proposed. Based on proportional-integral-derivative control of BP neural network, Set the value of the difference and the rate of change as a learning network basis, through the dynamic adjustment of BP neural network learning method, according to the actual situation of network learning, quickly adjust the response time of service requests to ensure that the video server’s service quality. Simulation results show that BP neural network can well control the request error rate and response time, and provide more stable quality of service for video-on-demand service of broadcasting and TV users.