How can I measure maximum frequency component present in my recorded EEG signal so that I can use right filter ahead?

I am working on a single channel EEG dataset but not able to apply hamming window based Low Pass Filter to it for denoising. I am using DenoiseF = filter (Hd,m); {Hd= window function, m= dataset} but it says "??? Undefined function or method 'filter' for input arguments of type 'cell'." Also I need to know maximum frequency component in my signal so that I can decide whether I need to filter it or it is already in usable range. Someone please help. May be I am not using right codes in Matlab.

Answer 1

[Star Strider]

See the documentation on the fir1 (link) function. It uses a Hamming window by default.

Use the filtfilt function to do the actual filtering.

[JAGRITI SAINI]

Thank You Sir.. It worked Now.. But please can you help me to know what should be my filter order for better results? I am trying this formula for calculation of filter order:

if true
>> f1=0.2;
>> f2=60;
>> fs=250;
>> m = (0.3*f1)/(fs/2);
>> M=round(8/m);
>> N = M-1;
>> b = fir1(N,0.5,'low');
  end

However, I am not interested to define f1, as I need a low pass filter with cut frequency range 60 Hz or 62 Hz. my data set length is= 4097 Here I am getting filter order N = 16666.. That seems too high.. Is it right? I was thinking to use low filter order like somewhere between 20 to 80 for reduced computational complexity. What should be right procedure to choose filter order here?

[Star Strider]

My pleasure.

The easiest way to calculate the filter order is to use another window function, and then let the window function design your filter using your desired parameters.

Example (bandpass filter):

Fs = 96E+3;                                                     % Sampling Frequency (Hz)
fcuts = [850  891  1122  1163];                                 % Frequency Vector (Hz)
mags =   [0 1 0];                                               % Magnitude (Defines Passbands & Stopbands)
devs = [0.05  0.01  0.05];                                      % Allowable Deviations
[n,Wn,beta,ftype] = kaiserord(fcuts,mags,devs,Fs);
n = n + rem(n,2);
hh = fir1(n,Wn,ftype,kaiser(n+1,beta),'scale');

figure(1)
freqz(hh, 1, 2^16, Fs)
set(subplot(2,1,1), 'XLim', [0 1500])                           % Set Frequency Axis To Show 0-500 Hz
set(subplot(2,1,2), 'XLim', [0 1500])                           % Set Frequency Axis To Show 0-500 Hz

See the documentation on kaiserord for details.

Other options exist, and an IIR filter that approximates the steep rolloff of the FIR filter is a Chebyshev Type II design. You can use the designfilt function, or design it yourself.

Example (different bandpass filter):

Fsp = 1000;                                     % Sampling Frequency
Fn = Fsp/2;                                     % Nyquist Frequency
Wp = [1.0 49]/Fn;                               % Passband Frequencies (Normalised)
Ws = [0.5 50]/Fn;                               % Stopband Frequencies (Normalised)

Rp=10;                                          % Passband Ripple (dB)
Rs=30;                                          % Stopband Ripple (dB)
[n,Ws] = cheb2ord(Wp,Ws,Rp,Rs);                 % Calculate Filter Order
[z,p,k] = cheby2(n,Rs,Ws);                      % Design Filter
[sos,g] = zp2sos(z,p,k);                        % Convert To Second-Order-Section For Stability

figure(1)
freqz(sos, 2^16, Fsp)
set(subplot(2,1,1), 'XLim',[0 100])             % ‘Zoom’ X-Axis
set(subplot(2,1,2), 'XLim',[0 100])             % ‘Zoom’ X-Axis

IIR filters have the advantage if being shorter than FIR filters, important with shorter data vectors. For your application, I would use the Chebyshev Type II IIR filter.