SAT: A new application mapping method for power optimization in 2D — NoC

Application mapping is one of the challenging and most imperative research areas in Networks on Chip (NoC). Mapping of NoC is an NP-hard problem and varieties of heuristics methods have been adopted to solve it so far. Heuristic search methods are generally modeled to fit a specific problem rather than range of applications. Based on this study, we propose improvised cluster based mapping with meta-heuristic search algorithm called simulated annealing with tabu search (SAT) to analyze and optimize the power consumption of NoC based systems. The performance and efficiency of proposed methodology has been verified with the experiments conducted on various types of benchmarks in NoC. Experimental results show a promising improvement in power consumption by 5.5 % for PIP & MWD benchmarks, 4.67 % for VOPD and 4.17 % for MPEG-4 benchmark compared to the best known related work.