Coefficients are 'NaN' while using dsp.LMSFilter

Hi, I am new to this field. Please help me in using LMS functions.

I have two sets of data.

data1 - signal+noise (data samples acquired from arduino)

data2 - noise (samples acquired from arduino)

I want to use LMS algorithm to remove the artifacts in data1. I have replaced the random data from LMS link example with my data.

But it returns 'NaN' for all [y,e,w] coefficients.

Why is that ? Is the input value range too high ? Are there any limitations to use this function?

I use MATLAB R2013a version.

signal = data1(:); % 5120x1 matrix ; minValue=0 ; maxValue=600
noise = data2(:); % 5120x1 matrix ; minValue=0 ; maxValue=900
L=length(signal);

nfilt = fir1(10,0.018); % lowpass filter
coeffs = nfilt.' -0.1;
fnoise = filter(nfilt,1,noise); 

d = signal + fnoise;
mu = 0.1;          % Set the step size for algorithm updating.

lms = dsp.LMSFilter(11,'Method','LMS','StepSize',mu,'InitialConditions',coeffs)
[y,e,w] = step(lms,noise,d);

figure()
plot(0:L-1,d(1:L,:),0:L-1,signal(1:L,:),0:L-1,e(1:L,:));
title('Noise Cancellation by the Sign-Data Algorithm');
legend('added noise Data','Actual signal','Result of Noise Cancellation','Location','NorthEast');