The traffic of highly compressed video data in communication networks is increasing. Besides, the development of wireless networks, especially in the access network layer and the increase in video transmission over error-prune links have been challenged, since the loss of small piece of information makes it hard to retrieve the video data. There are various ways to deal with these channel errors. Error concealment is highly effective way to deal with the errors in multimedia environment. These methods, especially in real-time applications that cannot retransmit the lost packets is more significant.Error concealment techniques attempt to improve the visual quality for the audience. In this thesis, a novel particle-filter-based error concealment is developed in order to correct the noisy motion vectors which are derived from two other error concealment techniques. In order to model these motion vectors, a Gaussian Mixture based model is developed and several assumptions are considered. H.264/AVC standard is widely used as a major coding system. Especially after the introduction of internet TV, international attention is attracted to this codec. The proposed methods in this thesis are tested and compared with existing algorithms and techniques in H.264/AVC platform. According to the results, the proposed methods can deal with non-uniform frames better than other ones. Besides, the blockiness is suppressed highly in the proposed methods. The effectiveness of the proposed algorithms have been tested on various frames at different motion speeds and uniformity. Final results of the performance evaluation is based on peak signal-to-noise ratio and compuational complexity is measured in the second unit. These results show that the proposed methods are able to enhance the video quality up to 4 dB. Keywords : Error concealment, video H.264/AVC coding, particle filter, Boundary Matching Algorithm, Gaussian Mixture distribution
