This program simulates a basic communication system using Matlab, and it plots BER curves in order to compare the performance of several codification algorithms. It includes the following components:
main.m is the main file.
Voz.wav is the audio file that contains the message. The program plots the input signal.
fundamental_frequency: the fundamental frequency of the message in the audio file.
x: vector that contains the message.
This module plots the first input signal with the levels of quantization and the quantized signal.
level: the number of levels for quantization.
option_quantization: quantization processes. Available options:
- 1 - Uniform
- 2 - Mu-Law
- 4 - A-Law
xq: quantized message.
quantization_error: quantization error.
This simulation codifies the message using the following methods:
- Hamming (7,4)
- Convolutional codes: soft decision
- Convolutional codes: hard decision
This module modulates the message according the codification, and it plots the constellation for the selected modulation.
option_modulation: modulation processes. Available options:
- 1 - BPSK
- 2 - QPSK
- 3 - BPSK and QPSK
Message modulated according to the selected modulation:
BPSK variables: bitsm1 (no codification), bitsm2 (Hamming), bitsm3 (Convolutionl)
QPSK variables: bitsmqpsk1 (no codification), bitsmqpsk2 (Hamming), bitsmqpsk3 (Convolutional)
This module uses a loop in order to simulate an AWGN channel with several Eb/N0 values. Eb/N0 is the energy per bit to noise power spectral density ratio. Value between 1 and 6, where 6 is for the least noisy channel.
Demodulation and decodification are performed in this loop. The program plots the BER curves of several codification algorithms.
Probability of error gives the average rate of occurrence of decoding errors. Pe error for the codification algorithms according to the selected modulation:
BPSK variables: errorpe_bpsk_nocod (no codification), errorpe_bpsk_hamming (Hamming), errorpe_bpsk_hard (Convolutionl: Hard decision), errorpe_bpsk_soft (Convolutional: Soft decision)
QPSK variables: errorpe_qpsk_nocod (no codification), errorpe_qpsk_hamming (Hamming), errorpe_qpsk_hard (Convolutional: Hard decision)
This program was developed during the communication course "Comunicación y codificación digital" at Universidad de las Fuerzas Armadas ESPE.