The A3D-TC: An Adaptive Approach for Selecting Third Component Parameters to Generate Robust Turbo Codes

Abstract

In recent days, the introduction of third component in conventional turbo codes proved to be effective in improving the code performance. In such third component enabled turbo codes, the parameters such as permeability and permittivity rates seems to be very static and so such codes cannot perform under different noisy environments. In this paper, an adaptive third component turbo code (A3D-TC) is proposed to solve the aforesaid drawbacks. In A3D-TC, the third component parameters are made adaptive. This is accomplished by generating a GA-based knowledge source and feeding it to feed forward neural network. The network outputs third component parameters according to the noise and signal strengths so that bit error rate at decoding section can be minimized in an effective way. Extensive experimental results prove the performance of A3D-TC over the conventional 3D-TC under various noisy environments.

Optimization of Heuristic Algorithms for Improving Ber of Adaptive Turbo Codes

“…By having the special intelligence (SI), the error correction capability was improved in the proposed A3D-TC given in the paper [1]. This special intelligence decides the permeability and permittivity rates of the third component encoder.…”

“…But this increased the complexity and the parameters such as permeability and permittivity rates were constant and they could not perform well under noisy environments. This drawback was addressed in [1] by proposing A3D-TC. The bit error rate was minimized by generating parameters based on noise and signal strengths.…”