The heart sound is processed and displayed in the form of waveform called as phonocardiogram (PCG) signal. The PCG signal can be obtained from electronic stethoscope or from phonocardiograph is used as primary tool for diagnosis of heart valvular diseases. There are two heart sounds: S1 known as" Lub" occurs due to the opening and closure of Atreioventricular valve and S2 is known as" dub" sound occur due to opening and closure of semilunar valve. The PCG signal acquired from electronic stethoscope or from phonocardiograph contains environmental noise along with the heart sound. It is a challenge to separate heart sound from noise. LMS are Adaptive finite impulse response filter, whose coefficients or weights change over time to adapt to the statistics of a signal. LMS has low computational complexity and it minimize the mean square value of the error signal. In this work the random noise is correlated with PCG signal. The noisy PCG signal is processed for noise cancellation using Least Mean Square (LMS) adaptive algorithm in MATLAB. The PCG signal processed by LMS algorithm is measured based on the parameters namely Minimum Mean Square Error (MMSE), computational complexity and peak signal to noise ratio (PSNR).