Electrocardiography deals with the electrical activity of the heart. The condition of cardiac health is given by ECG and heart rate. A study of the nonlinear dynamics of electrocardiogram (ECG) signals for arrhythmia characterization is considered. The statistical analysis of the calculated features indicate that they differ significantly between normal heart rhythm and the different arrhythmia types and hence, can be rather useful in ECG arrhythmia detection. The discrimination of ECG signals using non-linear dynamic parameters is of crucial importance in the cardiac disease therapy and chaos control for arrhythmia defibrillation in the cardiac system. The four non-linear parameters considered for cardiac arrhythmia classification of the ECG signals are Spectral entropy, Poincaré plot geometry, Largest Lyapunov exponent and Detrended fluctuation analysis which are extracted from heart rate signals. The inclusion of Artificial Neural Networks (ANNs) in the complex investigating algorithms yield very interesting recognition and classification capabilities across a broad spectrum of biomedical problem domains. ANN classifier was used for the classification and an accuracy of 90.56% was achieved. Linguistic variables (fuzzy sets) are used to describe ECG features, and fuzzy conditional statements to represent the reasoning knowledge and rules. Good results have been achieved with this method and an overall accuracy of 93.13% is obtained.