Frequency response analysis of respiratory sounds and comparative study for windowing techniques

Respiratory diseases are common and can be potentially deadly. These diseases can be diagnosed by using Stethescope. This auscultation process is most commonly used by the doctors. Auscultation provides important information to diagnose abnormalities and disorders in the respiratory system to identify respiratory diseases. The measurement of the different subjects data and Permanent record of the ralated data is difficult to maintain by auscultation using a stethescope. The use of stethoscope for diagnosing abnormalities has many limitations which overcome with the use of digital signal processing technique. In this paper, the piezoelectric transducer linked with the mircophone, so as to pick up the respiratory sound signal and record it. Then through the audio input the further processing of the Signal was done in MATLAB. Bandpass IIR filter is used for the Signal processing such as respiratory sound signal filtering and to remove noise. Design technique of bandpass filter is proposed using various type of window function Hanning, Hamming, Blackman and Kaiser window. Fast forier transform is plotted to observe the frequency distribution o f a signal and which is useful for the computation of Power spectral density graph. Power spectral density graph is plotted to obeserve the frequency at maximum peak value of the waveform and the maximum frequency is calculated. Our objective is to determine that which window is showing better response with bandpass filter for respiratory sound analysis. The results obtained using the proposed hardware then calculated and compared to the prerecorded lung sounds that were obtained online.