CODEWITHSUNDEEP
pythonpython-3.xnumpyscipy
Uzair
Uzair

Reputation: 231

Getting Value Error in Python function when calling signal.firwin

I am using Least Square filter function in python which is firwin and it is in signal library and when i am calling the function it is raising Value Error.

My Code Snippet:

Fs = 100

epochs = n/Fs

nyquist = Fs/2

lower_filter_bound = 7;

upper_filter_bound = 13;

transition_width = 0.7;

filter_order = np.round((Fs/lower_filter_bound))

idealresponse = [ 0, 0, 1, 1, 0, 0 ];

filterName = 'Band pass filtered';

F = [0, (((1-transition_width)*lower_filter_bound)/nyquist), 
(lower_filter_bound/nyquist),(upper_filter_bound/nyquist),(((1+transition_width)*upper_filter_bound)/nyquist), nyquist/nyquist];

filterweights = sig.firwin(filter_order, F, idealresponse)

Error:

Traceback (most recent call last):
  File "File.py", line 34, in <module>
    filterweights = sig.firwin(filter_order, F, idealresponse)

raise ValueError("Invalid cutoff frequency: frequencies must be "
ValueError: Invalid cutoff frequency: frequencies must be greater than 0 and less than nyq.

Upvotes: 0

Views: 2511

Answers (1)

Warren Weckesser
Warren Weckesser

Reputation: 114811

The least squares FIR filter design function in scipy is scipy.signal.firls (not scipy.signal.firwin).

firls requires an odd number of taps, so you'll have to ensure that filter_order is odd.

If firwin is actually the function that you meant to use, then take another look at the docstring. In particular:

  • firwin does not take an argument for the ideal response. It is only given the band edges in the cutoff argument.
  • The description of the cutoff argument specifically says this argument must not contain 0 and the Nyquist frequency.

  • You appear to be creating a bandpass filter. There is an example of this in the docstring:

    Band-pass:

>>> f1, f2 = 0.1, 0.2
>>> signal.firwin(numtaps, [f1, f2], pass_zero=False)
array([ 0.06301614,  0.88770441,  0.06301614])

  • The first argument of firwin must be an integer, not a float.

Here's how you implement your filter using firwin:

lower = lower_filter_bound/nyquist
upper = upper_filter_bound/nyquist
filterweights = sig.firwin(int(filter_order), [lower, upper], pass_zero=False)

If you need more flexibility in the design of your FIR filter, take a look at scipy.signal.firwin2.

Upvotes: 1

Related Questions