Title :
Solving NoC mapping problem with improved particle swarm algorithm
Author :
Zhengxue Li ; Yang Liu ; Mingsong Cheng
Author_Institution :
Sch. of Math. Sci., Dalian Univ. of Technol., Dalian, China
Abstract :
Network on chip (NoC) mapping problem belongs to quadratic assignment problem, and quadratic assignment problem is a classical combinatorial optimization problem which has been proved to be a NP-complete problem. In this paper, the widely used 2D-mesh as a NoC topology is studied, and the important factors influencing the performance of the system-power consumption and delay are modeled. Then the approximate optimal solution is found using the improved discrete particle swarm optimization algorithm. Simulation experiments show that very good optimal results are obtained.
Keywords :
approximation theory; combinatorial mathematics; computational complexity; delays; network-on-chip; particle swarm optimisation; power aware computing; quadratic programming; 2D-mesh; NP-complete problem; NoC mapping problem; approximate optimal solution; combinatorial optimization problem; delay; discrete particle swarm optimization algorithm; network on chip mapping problem; quadratic assignment problem; system-power consumption; Equations; Optimization; Quality of service;
Conference_Titel :
Advanced Computational Intelligence (ICACI), 2013 Sixth International Conference on
Conference_Location :
Hangzhou
Print_ISBN :
978-1-4673-6341-9
DOI :
10.1109/ICACI.2013.6748466