A Spatial Correlation Based Energy Transmission in Wireless Sensor Networks

Modern pace in hardware technology such as low cost CMOS cameras and microphones has prompted the evolution of Wireless Multimedia Sensor Networks. This network of sensors is equipped with multimedia devices such as microphones and cameras to capture multimedia data such as images, videos and audios streams along with scalar data from the environment. High data rate and directional sensing range of video sensors demands a novel solution in order to reduce the amount of data and workload on individual node. So, in this reference, this paper proposed a rare approach by utilizing spatial correlation that exists in the visual information captured by captured by the video sensors having overlapped field of views. By estimating the overlapped region in the video frames, a binary mask has been calculated that has been resent to the video sensors in order to receive only desired portion of consecutive video frames by mutual cooperation of different video sensors. Energy analysis based on the number of bits needed to be transmitted from each video sensor node has also been studied. Experimental results shows that the proposed approach is capable of transmitting videos with low computational cost and without much affecting the video quality and that has been shown by calculating the PSNR values. And also proposed approach is compared with video stitching method. Further simulations shows that proposed method can increase network lifetime by using less energy levels and is an energy efficient approach.