Research Of Image Distributed Processing In WMSNs

WMSNs is on the basis of WSNs with introducing audio, video, image,sound, large-capacity information.It's widely used in environmental monitoring, battlefield surveillance, traffic monitoring,ect.Besides self-organizing,unattended it also combines perception of media-rich multimedia technology. On the one hand,it is traditional self-organizing,multi-hop routing, resource constraints and other common characteristics.On the other hand,the energy distribution,QoS requirments,sensor model and also has a significant personal characteristics.As energy of nodes, processing power,storage capacity of WMSNs isvery limited,efficient use of energy is the primary objective of WMSNs.The compression and transmission of image is critical.Compression,filtering,and data fusion of information processing technology are widely used in WSNs. Taking account of the limited energy of single node,processing power,we can use the processing power of multiple nodes, with "in network computing" we can complete the compression of multimedia information and upload the small amount of data with sufficient information to the next target step by step.Because of the special requirements,based on the improvement of existing algorithms,a distributed image compression algorithm is proposed.First because the Walsh-Hadamard transform is simple,efficient and enengy focused,so it can be applied to the image compression algorithm.lt is very suitable for computer implementation.Compared to DCT transfom and wavelet transform,it requires less storage space and improves the speed of operation.The simulation results show that the algorithm with Walsh-Hadamard transform is more simple and rapid than DCT-based algorithm.Secondly taking account of the characteristics of wireless channel in WSNs,we propose an adaptive image compression mechanism with channel estimation.Bacause wireless channel is signal-attenuation,large delay,high bit err rate and burst data transmission,multimedia information processing needs proper Qos,such as delay tolerance,data bandwidth,ect.Based on accurate channel state estimation,we can use adaptive quantization table, ensure that the quality of image meet the needs and get higher compression ratio.Simulation results show that the algorithm in intensive WMSNs can prolong the network life cycle.Finally according to the previous improvement we proposed the adaptive distributed image compression algorithm with Walsh-Hadamard transform.In the transform stage, the Walsh-hadamard tranfrom is used to reduce the node processing overhead. In the quantitative phase,we use the channel estimation mechanism to quantilize adaptively and make higher compression ratio.The simulation results show that the proposed algorithm with limited resources and deployment of intensive WMSNs nodes,it prolongs the network lifetime significantly.