Performance of Heuristic Algorithms in minimizing BER for Robust Turbo Codes

The third component introduced in conventional turbo codes in recent days is effective in improving the performance of the code. But under noisy environments the parameters such as permeability and permittivity rates are static and hence their adapatability to noisy environment is poor. To overcome the aforesaid drawback A3D-TC was proposed in our previous paper[29].In our present paper a comparison has been made between two heuristic algorithms viz.,Genetic algorithm and Simulated annealing algorithm. The parameters are made adaptive by generating a knowledge source using both Genetic algorithm and Simulating Annealing algorithm. The source output is fed to the feed forward neural network. The bit error rate at decoding section is tried to minimize in an effective way by the third component parameters generated from the neural network. Under various noisy environments, the experimental results compare the performance of A3D-TC using both Genetic algorithm and simulated annealing algorithm. Index terms A3D-TC, knowledge feeding, GA-based knowledge source, simulated annealing, adaptive, permeability rate, permittivity rate, and third component