Reconfiguration Decision Making Based on Ant Colony Optimization in Cognitive Radio Network
Wireless Personal Communications: An International Journal
| Hi-index | 0.00 |
Ant colony system (ACS) and a few improved ant colony optimization (ACO) methods for solving optimization problems with continuous domain were studied. Based on these work, a new ACO method for solving nonlinear geophysical multi-parameter inversion was presented. Several techniques are employed, such as decimal coding, mapping mechanism, roulette wheel selection of genetic algorithm (GA), renormalization, stochastic local search, etc. –namely after each ant moves along layers, update pheromone locally to the paths passed by the ants and globally to the paths passed by optimal ants. Then the probability of the path to be chosen by later ants depends on the amount of pheromone. The more pheromone the path has, the higher probability to be chosen. The ants will find the shortest path through such a positive feedback manner that is the satisfactory solution of the problem. Mainly we apply it to seismic wavelet extraction, wave impedance inversion and gravity anomaly inversion of several geologic bodies. The numerical results exhibit its feasibility and efficiency. We also compare the performances of ACO with simulated annealing (SA) and GA. The experimental results show the ACO method has the attributes of higher speed of convergence and better effect of inversion.