A Novel Recovery Technique For Lost Internet Video Frames Simulation and Evaluation OMAR S. ESSA Computer Science Dep., Faculty of Science, Menoufia University, EGYPT NIKOS E MASTORAKIS WSEAS European Office Agiou Ioannou Theologou 17-13, 15773, Zografou, Athens, GREECE http://www.wseas.org/mastorakis Abstract: Internet was designed to support traditional and multimedia applications. Multimedia applications suffer from unacceptable delay, jitter and data loss. Data loss has the largest impact on quality. When the retransmission technique is used to recover lost multimedia packets, more obstacles, like the delay time resulted from lost packets retransmission operation, network traffics overloading, and buffer management, are faced. All above demonstrated problems are due to the real-time communication features. In this paper, we extend our lost audio recovery technique to recover the video frames. Finally, NS2 is used to build a simulated environment for testing our new video technique as regards the MPEG transmission. Key-Words: - Multimedia Loss, Internetworking, Video Frames, TCP/IP, Multimedia Protocols. 1 Introduction Data loss in multimedia streams transmission can impact the continuity in the display. Data loss can occur involuntarily from network congestion or system buffer overflows, or voluntarily in order to avoid congestion at a client, server or network router. Too much data loss can result in unacceptable media quality. To compensate for data loss, much work has been done to find effective data loss recovery techniques. There are two categories of data loss recovery techniques: sender driven and receiver-based. Each of them has its own strength and weakness. These techniques have proven to be fairly effective (but not sufficient) for audio stream data loss, but have yet to be applied to video. In this paper, we extend our audio technique to recover the lost video frames [1], [2]. This paper is proceeding as follows; in section 2, our video technique is introduced. The detailed simulation is demonstrated in section 3. 2 Our Technique The basic idea of our video recovery technique is to classify the video objects depending on there importance. We can define the importance of the object as its ability of providing the users with a helpful meaning that aid them in video understanding. To reach our target, we should analyze the video streams. This analysis provides us with the sensitive parts for the human eyes, and the data that can be understood by the video context. The obstacles that may be faced in our technique are; how we can determine the features of the video objects, how these objects are classified, and what are the parameters that controlled in the object importance (object prioritization). We will answer these questions through our technique description In the following subsections, video contents analysis, video objects priority, and Real-Time Transport Protocol (RTP) header manipulation, are demonstrated. Finally, simulated experiments and results comments are showed. 2.1 Video contents analysis To analyze and describe the video streams, we should answer two questions stated as follows: 1) How the video streams are transmitted after taken by a camera to the perception? 2) What are the video contents? 2.1.1 How the video streams are transmitted? Video streams pass through four layers until viewed by the perception. The first layer is called a sensor layer. This layer takes video shots for the target place using a video camera. These video shots are passed to the second layer that is called a video- processing layer as a signal structure. The video- processing layer comprises a source media coding, alarms, and video database handling. Consequently, the video streams are transmitted within the network Proceedings of the 5th WSEAS Int. Conf. on System Science and Simulation in Engineering, Tenerife, Canary Islands, Spain, December 16-18, 2006 47