Energy Analysis of Compressed Sensing in Wireless Multimedia Sensor Network

___________________________________________________________________________________________________ Abstract— A Wireless Sensor Network (WSN) consists of several sensor nodes deployed in inaccessible areas for monitoring temperature, pressure, vibration, sound, motion etc. A WSN is used for variety of applications such as military, civil, industrial automation, medical, home automation, fleet monitoring, habitat monitoring, preventing theft etc. The availability of inexpensive hardware such as CMOS cameras and microphones has led to the development of Wireless Multimedia Sensor Networks (WMSN) which is used for image and video applications. In case of video applications the captured data will be too large if transmitted as such so it has to be compressed before transmission. Compression in traditional video encoding makes use of motion estimation and motion compensation techniques which requires intensive operations that lead to significant energy consumption and also the storage required is high. This drawback can be addressed by Compressed sensing, an emerging technique that directly obtains the desired samples, thereby reducing the energy consumption, storage capacity and bandwidth used in the network. It is used for reconstructing a signal from the M<<N measurements obtained from sparse or compressible signals, where N is the number of samples required for Nyquist sampling. Compressed sensing can overcome the drawbacks of traditional video encoders by simultaneously sensing and compressing the data at low complexity. The original signal can be recovered from measurements using basis pursuit and greedy algorithms. The objective of this paper is to implement a video compressed sensing framework using Gaussian measurement matrix and reconstruct it using Orthogonal Matching Pursuit algorithm and further transmission energy is analysed for the video compressed sensing framework .