Scribd
Upload
29 views3 pages

Raised Cosine Filter

The document describes the implementation of a raised cosine filter in both frequency and time domains using Python's numpy and matplotlib libraries. It includes the definition of the filter function, parameter initialization, and the generation of plots for various roll-off factors. Additionally, it demonstrates how to visualize the filter's response in both domains.

Uploaded by

Bonnie Mundia
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
29 views3 pages

Raised Cosine Filter

The document describes the implementation of a raised cosine filter in both frequency and time domains using Python's numpy and matplotlib libraries. It includes the definition of the filter function, parameter initialization, and the generation of plots for various roll-off factors. Additionally, it demonstrates how to visualize the filter's response in both domains.

Uploaded by

Bonnie Mundia
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

24/11/2024, 21:49 RaisedCosineFilter

In [ ]:

Importing numpy and matplotlib.plt libraries


In [1]: import numpy as np
import matplotlib.pyplot as plt

Defining the function of a raised cosine


In [2]: def raised_cosine(f, T, r):
"""
Raised cosine filter in frequency domain.

Parameters:
f: Frequency array (Hz)
T: Symbol period (s)
r: Roll-off factor (0 <= r <= 1)

Returns:
H: Filter response (magnitude) in frequency domain
"""
B = (1 + r) / (2 * T) # Bandwidth
H = np.zeros_like(f)

for i, freq in enumerate(f):


abs_f = abs(freq)
if abs_f <= (1 - r) / (2 * T):
H[i] = 1
elif (1 - r) / (2 * T) < abs_f <= (1 + r) / (2 * T):
H[i] = 0.5 * (1 + np.cos(np.pi * T / r * (abs_f - (1 - r) / (2 * T))
else:
H[i] = 0
return H

Intiating the parameters of the raised cosine function


In [3]: # Parameters
T = 1 # Symbol period (s)
fs = 10 * T # Sampling frequency (Hz)
f = np.linspace(-2 / T, 2 / T, 1000) # Frequency range

# Roll-off factors
roll_off_factors = [0, 0.2, 0.5, 1]

Raised cosine spectrum in frequency domain


In [4]: # Frequency domain plots
plt.figure(figsize=(12, 6))
for r in roll_off_factors:
H = raised_cosine(f, T, r)
plt.plot(f, H, label=f"r = {r}")

plt.title("Raised Cosine Spectrum (Frequency Domain)")

file:///C:/Users/bonfa/Documents/Machine Learning/RaisedCosineFilter.html 1/3


24/11/2024, 21:49 RaisedCosineFilter

plt.xlabel("Frequency (Hz)")
plt.ylabel("Magnitude")
plt.grid()
plt.legend()
plt.show()

Raised cosine impulse response in time domain


In [5]: # Time domain impulse responses
time = np.linspace(-5 * T, 5 * T, 1000) # Time range
plt.figure(figsize=(12, 6))
for r in roll_off_factors:
H_f = raised_cosine(f, T, r)
h_t = np.fft.ifftshift(np.fft.ifft(np.fft.fftshift(H_f))) # IFFT to get imp
plt.plot(time, np.abs(h_t), label=f"r = {r}")

plt.title("Raised Cosine Impulse Response (Time Domain)")


plt.xlabel("Time (s)")
plt.ylabel("Amplitude")
plt.grid()
plt.legend()
plt.show()

file:///C:/Users/bonfa/Documents/Machine Learning/RaisedCosineFilter.html 2/3


24/11/2024, 21:49 RaisedCosineFilter

file:///C:/Users/bonfa/Documents/Machine Learning/RaisedCosineFilter.html 3/3

You might also like