Chaotic fractal walk trainer for sonar data set classification using multi-layer perceptron neural network and its hardware implementation

First, this study proposes the use of the newly developed Stochastic Fractal Search (SFS) algorithm for training MLP NNs to design the evolutionary classifier. Evolutionary classifiers, often experience problems of slow convergence speed, trapping in local minima, and non-real-time classification. This paper also use four chaotic maps to improve the performance of the SFS. This modified version of SFS has been called Chaotic Fractal Walk Trainer (CFWT). To assess the performance of the proposed classifiers, these networks will be evaluated using the two benchmark datasets and a high-dimensional practical sonar dataset. For endorsement, the results are compared to four popular meta-heuristics trainers. The results show that new classifiers suggest better performance than the other benchmark algorithms, in terms of entrapment in local minima, classification accuracy, and convergence speed. This paper also implements the designed classifier on the Filed Programmable Field Array (FPGA) substrate for testing the real-time processing ability of the proposed method. The results of the real application prove that the designed classifiers are applicable to high-dimension challenging problems with unknown search spaces.