Dynamic Multiple Description Wavelet Based Image Coding Using Enhanced Particle Swarm Optimization

摘要

In this paper, dynamic multiple description wavelet based image coding using enhanced particle swarm optimization (EPSO) is proposed for effective multimedia transmission in the dynamically changing network topologies such as mobile ad-hoc networks (MANETs). In wireless networks, multimedia transmission has become a challenging problem because of changing conditions over time such as packet loss rate (PLR), capacity/bandwidth, network and link conditions. The proposed method is used to simultaneously create optimal multiple description coding (MDC) or sub-images based on the dynamically changing capacity/bandwidth of wireless joint/disjoint communication channels/paths. To perform it, PSO is enhanced by adding a position mutation scheme as a generation feature to employ the proposed dynamic cost function to find the optimal solution within in a search space. To create MDs efficiently, three different wavelet transformation methods which are Discrete Wavelet Transform (DWT), Dual Tree-Complex Wavelet Transform (DT-CWT), and Stationary Wavelet Transform (SWT) are investigated. In this manner, a wavelet transform method is first applied to the given input image to create the sub-bands or set of coefficients as a search space, and then EPSO is employed to find optimal results in the search space to generate MDs. The proposed approach produces good quality of descriptions for effective multimedia transmission over dynamically changing and unreliable communication networks, particularly in wireless and ad-hoc networks (MANETs). The results of the proposed method are compared with the state-of-the-art techniques and simulation results demonstrate that using EPSO with DT-CWT provides a significant improvement of optimal and balanced MDC generation with good quality at different bit-rates.