emcware
commented
1 year ago I took a stab at this with some help from gpt4. import numpy as np import matplotlib.pyplot as plt
""" This example uses random QAM data and does not include channel coding or any of the intricacies involved in a real-world communication system. It is a simple simulation meant to illustrate the basic OFDM signal structure. """
def generate_ofdm_symbol(num_subcarriers, cp_length, data_symbols): """ :param num_subcarriers: :param cp_length: :param data_symbols: :return: """
Perform IFFT to move from frequency domain to time domain
ifft_result = np.fft.ifft(data_symbols, n=num_subcarriers)
# Add cyclic prefix by taking the last cp_length samples and prepending them to the symbol
cyclic_prefix = ifft_result[-cp_length:]
ofdm_symbol_with_cp = np.concatenate([cyclic_prefix, ifft_result])
return ofdm_symbol_with_cpParameters
num_subcarriers = 64 # Number of OFDM subcarriers cp_length = 16 # Length of the cyclic prefix num_symbols = 10 # Number of OFDM symbols to generate
Generate random QAM data symbols for illustration purposes
data_symbols = np.random.choice([1 + 1j, -1 + 1j, -1 - 1j, 1 - 1j], size=(num_symbols, num_subcarriers))
Generate the OFDM waveform
ofdm_waveform = np.array([])
for symbol in range(num_symbols): ofdm_symbol = generate_ofdm_symbol(num_subcarriers, cp_length, data_symbols[symbol]) ofdm_waveform = np.concatenate([ofdm_waveform, ofdm_symbol])
Plot the IQ constellation of the OFDM waveform
plt.figure(figsize=(5, 5)) plt.scatter(ofdm_waveform.real, ofdm_waveform.imag) plt.title('OFDM IQ Constellation') plt.xlabel('In-Phase') plt.ylabel('Quadrature') plt.grid(True) plt.show()
OFDM waveforms are difficult to generate, and there's a significant barrier to entry given the complexity of the modulation type. Let's make it easier to build and analyze OFDM waveforms.
Multi-carrier OFDM is on the table here, but that might be a stretch goal.
Request is to add OFDM waveform type to wfmBuilder.py