IPTV Quality Monitoring Strategy Based On Decoding Cache Information

In recent20years, with the popularity of the Internet and the development of high speed computer networks people’s lives are facing great changes. And the IPTV business will undoubtedly become one of the most attentions in the Internet business field. However, as a real-time streaming media transmission services, stable and high speed broadband networks are needed to sustain the services and the network performance is directly related to the merits of the quality of service. So the QoS indexes which are used to assess ordinary network applications such as packet loss, delay cannot completely reflect the performance of IPTV business.In the IPTV business field user experience of the video quality always draws most attention of the service providers. So the QoE (Quality of Experience) is introduced to the assessment of streaming media’s quality. Since basing on the users’view, QoE can effectively reflect the quality of the video and provide reliable information to the service providers. When the concept QoE was introduced, at first people measured it by subjective scores. It means service providers should invite the users to score the video to get its QoE assessment, but this method requires a large number of people involved and cannot get the real-time feedback. So recently the objective assessing method is accepted as a main way to get QoE assessment. This method requires mounting a probe in the network of IPTV, and the probe will collect all kinds of objective indexes to help assess the QoE. However QoE is subjective item, this method requires complex algorithm to transform these objective ones into QoE.In this paper we propose a strategy which can help the service providers get real-time QoE information from set-top box of IPTV. We design and develop a module which can be loaded on operating system of the set-top box to monitor the status of the decode buffer. It feedbacks the media lost rate information and underflow information of buffer while the media player is running. As getting these buffer indexes streaming media mid-ware is responsible for putting them into analyzing module to do QoE assessing.Experimental results show that this strategy can help the service providers to get effectively real-time QoE information. And this way can be ported to different hardware platforms.