Research On Multi-Core Parallel Computing In Video Service

With the development of network technology,streaming media technology is continuously developing,streaming media technology widely used in network television,video conferencing systems,video surveillance systems,and other places.Now people are in high demands of streaming media technology,how can we improve the real-time video system, On the one hand is to enhance network bandwidth,on the other hand is to raise its compression rate.In order to improve the efficiency of compression, MPEG standards are put forward.MPEG is an abbreviation of Moving Picture Expert Group,under the leadership of the International Organization for Standardization(ISO/IEC), after nearly 10 years of fruitful work for the multimedia computer system,sports image compression technology standardization and practical use,have made tremendous contributions.In 1998 the formal launch of MPEG-4 standard, which is mainly used in video telephony(Video Phone), video e-mail(Video E-mail)and the Electronics News (Electronic News).MPEG-4 using a very narrow bandwidth, through reconstruction frame data compression and transmission technology to enable the data to minimize the best image quality.At present,many chip manufacturers at home and abroad are committed to the development of multi-core technology.Multicore chips can reduce the frequency of this system and reduce the power consumption at the same time, increase the speed of computing chips.In order to improve the speed of compression,in conjunction with the current rapid development of multi-core technology for the com- pression process to be improved.First,this paper presentes some basic concepts and algorithms in MPEG-4 standard,including video services system,the overall system architecture,MPEG video compression algorithms and motion projections and so on.Then,through the study on the MPEG-4 standard compression algorithm,we find that there are parallel in many compression algorithm.Parallel is mainly reflected in the decoding of the discrete cosine transform coding,as well as inter-frame motion estimation.This paper proposed two parallel algorithms for intra-and inter-frame coding,and realizing algorithm and simulation on the dual-core UP-OMAP5910 platform.The experimental results show that, after the using of parallel algorithm,compression speed is greatly enhanced.Finally,in order to further improve the compression rate,we have introduced a multi-core load balancing technology, after a brief introduction about the traditional load balancing technology,we propose a new model of load balancing. In this Model,the cores are divided into three level, The First Level core,the second level core and the third level core.All three levels core do different work.Among all the second level cores,we proposed distribute load balancing.And Between the second core and its third cores, we proposed Centralized Load Balancing.The model is used on the PPDDME algorithm.Experimental data indicate that, the combination of load balancing technology and parallel compression algorithm improves the efficiency of the system video services to a great extent.