The Research Of Video Coding Based On SPIHT Over Wireless Multimedia Sensor Networks

Wireless Multimedia Sensor Networks (WMSNs) is a new type of sensor network, which is developed on traditional wireless sensor networks (WSNs), and have multimedia information-aware capabilities such as audio, video, images. Compared to traditional wireless sensor networks with only a simple function of environmental data collection, wireless multimedia sensor networks can achieve environmental monitoring with fine-grained and accurate information. It has very broad application prospects in the military, civil, commerce and other fields.For existing coding strategy in wireless multimedia sensor networks have high complexity of video coding strategy, and the transmission strategy does not take into account the importance of video content and other issues, this paper based on the SPIHT algorithm and multiple description coding techniques, carried on a in-depth study and a large number of experiments on multi-media video encoding and transmission robustness over wireless multimedia sensor networks, and achieved the following results:1. Based on the analysis of the SPIHT algorithm, and its advantages and disadvantages of the existing improved algorithm, this article proposed a new improved algorithm. In this paper, a new kind of tree construction algorithm is adopted, using the reordering technology of importance on the code words, and more important information in the bit stream priority encoded in the front-end, and achieved good results, improve PSNR more than 1~5dB, the lower the effective rate the more remarkable. At the same time, lay the foundation for the latter part of multiple description coding.2. For the wireless multimedia sensor networks have nature of resource constrained and energy constrained, this paper improved the existing 3D-SPIHT algorithm, using lifting wavelet integral transform and non-linked coding structure, improve coding speed more than 60%, greatly reducing the coding complexity, while reducing the resources and energy, to adapt to use in wireless multimedia sensor networks.3. On the basis of region of interest coding, proposed a stream content-based multiple description SPIHT coding method. The algorithm fully considered the importance of the zero-tree coding, distributed different bit rates to different zero-tree with different importance, through the importance of re-sorting, making redundant tree contains more important information to improve the reconstructed video quality.