Digital Communication: UE17EC352: VI Sem ECE, PESU: Jan-May 2020 Experiment-5: Symbol Error Rate For QPSK 02-03-2020
Digital Communication: UE17EC352: VI Sem ECE, PESU: Jan-May 2020 Experiment-5: Symbol Error Rate For QPSK 02-03-2020
Digital Communication: UE17EC352: VI Sem ECE, PESU: Jan-May 2020 Experiment-5: Symbol Error Rate For QPSK 02-03-2020
In this program, we wish to plot the SER for QPSK. Similar to the
BPSK case, here too we consider the signal space model directly. The two
dimensional signal space is modeled as the complex plane.
Steps:
1. Generate the N = 107 length sequence s(n) of random QPSK
√ symbols:
1 + j, 1 − j, −1 + j and −1 − j (Here Eb = 1 and E = 2)
2. Set the Eb /N0 range (in dB) as -4:2:12.
3. For each value of Eb /N0 , perform the following steps:
(a) For the given value of Eb /N0 , find the noise variance.
(For example, if Eb /N0 = x dB, then N0 = 10−x/10 since Eb = 1.
The noise variance is N0 /2 )
(b) Generate the complex zero mean white gaussian noise sequence
w(n) of length N for that variance. (Use randn function). Here
w(n) = wr (n) + jwi (n), where both wr (n) and wi (n) have zero
mean and variance N0 /2.
(c) Find x(n) = s(n) + w(n)
(d) Recover the input s(n) from x(n). To do this, find the quadrant
in which each value of x(n) falls by using the two BPSK decisions
real(x) > 0 and imag(x) > 0.
(e) Find the number of errors, and the symbol error rate (no. of
errors/N ).
(f) Find the probability of error using the expression derived in class.
4. Plot the symbol error rate (on log scale) vs Eb /N0 (in dB). Use the
semilogy function.
5. Plot the theoretical SER (the probability of error expression) and see
if the two curves match.
6. Also plot the ”scatter plot” (use the scatterplot function in MATLAB)
of the received signal x(n) for different Eb /N0 values.