Executive Summary : | Fractional order has been used for the design of various analog filters due to the advantage of larger frequency range and flexibility in shaping the frequency response. Researchers have proven that the performance of the FOF have shown improvement in comparison with integer order filters in the design of bandpass filters and the study concluded that the fractional order filters in general have versatile responses. On the other hand, Butterworth filters have been extensively used for analog fractional order filter design with different design approaches. A comparison with integer order filters for different frequency ranges, FOF have shown improved performance and satisfies frequency domain specifications. In biomedical engineering, Fractional Order Filters (FOF) are very useful for signal conditioning. The existing literature shows that more work is being done in the design of fractional order calculus based filters in biomedical signal processing for removal of noise and enhance signal information. Fractional-order digital filters have developed to provide an alternative solution to higher-order integer-order filters, with increased design flexibility and better performance. This mathematical tool is also applicable for modelling complex biological systems. Tuning of fractional-order filter coefficients to meet the required performance indices such as integral square error, integral absolute error and integral time absolute error are gaining interest among the scientists working in the area of FOF. This is an unconstrained, high-dimensional and multimodal optimization problem, which is solved using several heuristic algorithms such complex optimization problems. They are classified into two broad categories, stochastic and deterministic. In deterministic methods, the given initial feasible guess proceeds towards the final solution based on certain deterministic guidelines. In stochastic methods, a search is made for the better solution in a probabilistic manner. Among deterministic methods, the simplex method of Nelder and Mead is the most popular one used for tuning the fractional order. Stochastic algorithms such as genetic algorithm (GA), particle swarm optimization (PSO), artificial bee colony (ABC), adaptive GA, improved electromagnetic-like algorithm with GA (IEMGA) and differential harmony search algorithm have been used to tune fractional-order controller’s coefficients. The above literature motivates us to investigate and suggest suitable fractional order filters for biomedical applications. The outcome of the project is to get the original / reconstructed bio signal using the FOF and hence it is useful for medical practitioners / researchers across India for identification and disease diagnosis for the early stages diseases. |