Design of Anfis based pacemaker controller having improved transient response and its FPGA implementation

Abstract

This paper presents the design and hardware implementation of an Adaptive Neuro-Fuzzy Inference System (Anfis) based controller for a cardiac pacemaker. Cardiac pacemakers are used for the patients who suffer from bradycardia: heart rate below 60 beats per minute, and pacemaker controllers are required to provide stability between the heart rate and the desired preset profile of the pacemaker. The proposed work aims to design a controller to improve the parameters of cardiovascular system response in terms of stability, steady-state error, transients, etc. The settling time is the major constraint in the design of a pacemaker controller; ideally, it is required to have no delay between the need of heartbeat and the stimulus delivery by the pacemaker. The combination of neural network with fuzzy logic makes the proposed controller more effective concerning ‘learning’ and speed. The membership function of fuzzy logic plays an important role in controller design; the optimal number of membership functions are selected, so as to avoid the controller design complexities. In the proposed scheme, we have designed a pacemaker controller with fifteen membership functions on MATLAB/Simulink. To validate the theoretical results of the proposed methodology, the proposed controller is implemented on the Artix 7 Field Programmable Gate Array (FPGA) board. The simulation results show that the control strategy of the proposed design is more suitable in terms of timing response and the hardware implementation shows that the proposed controller will consume less power. The settling time and the power consumption for the proposed pacemaker controller are found to be 1.3813sec and 8.04 mW respectively.