Short code with prepared long and short sequence for OFMD IEE802.11n

IEEE_802_11_OFDM_Preamble.py

#prepare OFDM

import numpy as np
from matplotlib import pyplot as plt 
import scipy
import scipy.signal

fs = 20e6

class OFDM:
    pass

ofdm = OFDM()
ofdm.K = 64                    # Number of OFDM subcarriers
ofdm.Kon = 52                   # Number of switched-on subcarriers
ofdm.CP = 16                     # Number of samples in the CP
ofdm.ofdmSymbolsPerFrame = 30     # N, number of payload symbols in each frame
ofdm.L = ofdm.K//2   
ofdm.first_subcarrier=6
ofdm.Middle_zero=33

def IEEE802_11_short_sequence(): #without 0 in middle - generation is dealing with that
    return np.sqrt(13/6)*np.array([0,0,1+1j,0,0,0,-1-1j,0,0,0,1+1j,0,0,0,-1-1j,
                                   0,0,0,-1-1j,0,0,0,1+1j,0,0,0,0,
                                   0,0,0,-1-1j,0,0,0,-1-1j,0,0,0,1+1j,0,0,0,
                                   1+1j,0,0,0,1+1j,0,0,0,1+1j,0,0])


def IEEE802_11_long_sequence(): #without 0 in middle - generation is dealing with that
    return np.array([1, 1, -1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 1, 1, -1, -1, 1, 1, -1, 1, -1, 1, 1, 1, 1, 0,
                     1, -1, -1, 1, 1, -1, 1, -1, 1, -1, -1, -1, -1, -1, 1, 1, -1, -1, 1, -1, 1, -1, 1, 1, 1, 1])

def random_qam(ofdm):
    qam = np.array([1+1j, 1-1j, -1+1j, -1-1j]) / np.sqrt(2)
    return np.random.choice(qam, size=(ofdm.Kon), replace=True)


def ofdm_modulate(ofdm, qam,n=None,cp=True,zero_included=False):
    fd_data = np.zeros(ofdm.K, dtype=complex)
    off = (ofdm.K - ofdm.Kon)//2
    print("off",off)
    if False==zero_included:
        assert (len(qam) == ofdm.Kon)
        fd_data[off:(off+int(len(qam)/2))] = qam[:int(len(qam)/2)]  # modulate in the center of the frequency
        fd_data[(off+1+int(len(qam)/2)):off+1+2*int(len(qam)/2)] = qam[int(len(qam)/2):]
    else:
        fd_data[off:off+len(qam)] = qam  # modulate in the center of the frequency
       
    fd_data = np.fft.fftshift(fd_data)
    if n!=None:
        symbol = np.fft.ifft(fd_data,n) * np.sqrt(n)
    else:
        symbol = np.fft.ifft(fd_data) * np.sqrt(ofdm.K)
    if True == cp:
        print("time:",len(symbol)*1/fs + len(symbol[-ofdm.CP:])*1/fs)
        return np.hstack([symbol[-ofdm.CP:], symbol])
    else:
        print("time:",len(symbol)*1/fs)
        return symbol
    
def IEEE802_11_ofdm_preamble(ofdm):
    
    samples_short = samples_long = None
    samples_short=64
    samples_long=64
    print("0.8e-6/fs",samples_short*1/fs,samples_long*1/fs)
    short=[]
    short = ofdm_modulate(ofdm, IEEE802_11_short_sequence(),samples_short,cp=False,zero_included=True)
    for n in range(1):
        short = np.hstack([short,ofdm_modulate(ofdm,IEEE802_11_short_sequence(),samples_short,cp=False,zero_included=True)])
    short = np.hstack([short[:int(len(short)/4)],short])
    long=[]
    long = ofdm_modulate(ofdm, IEEE802_11_long_sequence(),samples_long,cp=False,zero_included=True)
    for n in range(1):
        long = np.hstack([long, ofdm_modulate(ofdm,IEEE802_11_long_sequence(),samples_long,cp=False,zero_included=True)])
    

    long=np.hstack([long[-ofdm.CP*2:],long]);
    print(len(long)*1/fs,len(short)*1/fs)
    preamble=np.hstack([short,long]);
    return preamble
    
# preamble = short
preamble=IEEE802_11_ofdm_preamble(ofdm);
print("Preamble time: ",len(preamble)*1/fs)
plt.figure(figsize=(20,5))
plt.subplot(121)
t = np.arange(0,len(preamble)/fs,1/fs);
plt.plot(abs(preamble))

plt.subplot(122)
f = np.linspace(-ofdm.K/2, ofdm.K/2, 4*len(preamble[170:]), endpoint=False)

plt.plot(f,abs(np.fft.fftshift(np.fft.fft(preamble[170:], 4*len(preamble[170:]))/np.sqrt(len(preamble)))))
plt.show()