Distributed video coding for underwater acoustic channels

The underwater study and communication have gained attention due to the military, commercial and biological purpose. The video monitoring of the underwater activity is carried out by video processing node (VPN) which form a sensor network. The VPN must be power efficient and less complex. The video coding standard such as MPEG and H.26X have a computationally more complex encoder for their working and are not suitable for many to one application like VPN. For applications such as underwater VPN, low power surveillance networks and wireless video cameras, the encoder needs to be simpler. Distributed Video Coding (DVC) is a technique which can be used for these applications. The underwater acoustic medium is affected by parameters such as absorption, attenuation and ambient noise parameters. These can be mitigated via modelling and using appropriate modulation technique. The primary objective of the current research was to design and develop DVC technique based on Slepian-Wolf and Wyner-Ziv theorems, which is resilience to transmission errors and has a high compression efficiency. The main concern of the design was to shift the motion estimation part from encoder end to the decoder through which low complexity is obtained at encoder. The channel characterization including study of absorption and ambient noise parameters are carried out with the help of empirical formulae. The simulation results show that the performance of the DVC Codec developed is close to the performance of conventional codec which provides good reconstruction with a high PSNR with up-to 70% level of compression. The architecture conforms to be stable for variation in size of GOP. The ARPS provides better estimation by taking less computational steps for calculating global minimum. The complete system developed assures 10% increase in bit error rate (BER) performance. However, the developed system needs to be validated for the real-time scenario.