Digital Communication Systems Lecture3 Prof Dr Habibullah Jamal

Digital Communication Systems Lecture-3, Prof. Dr. Habibullah Jamal Under Graduate, Spring 2008 1

2

3

4

Baseband Demodulation/Detection n In case of baseband signaling, the received signal sis already in pulse-like form. Why is then is demodulator required? n Arriving baseband pulses are not in the form of ideal pulse shapes, each one occupying its own symbol interval. n The channel (as well as any filtering at the transmitter) causes intersymbol interference (ISI). n Channel noise is another reason that may cause bit error is channel noise. 5

Effect of Channel Figure 1. 16 (a) Ideal pulse. (b) Magnitude spectrum of the ideal pulse. 6

Figure 1. 17 Three examples of filtering an ideal pulse. (a) Example 1: Good-fidelity output. (b) Example 2: Good-recognition output. (c) Example 3: Poor-recognition output. 7

Effect of Noise 8

3. 1. 2 Demodulation and Detection Figure 3. 1: Two basic steps in the demodulation/detection of digital signals n. The digital receiver performs two basic functions: q Demodulation, to recover a waveform to be sampled at t = n. T. q Detection, decision-making process of selecting possible digital symbol 9

3. 2 Detection of Binary Signal in Gaussian Noise n For any binary channel, the transmitted signal over a symbol interval (0, T) is: n The received signal r(t) degraded by noise n(t) and possibly degraded by the impulse response of the channel hc(t), is (3. 1) Where n(t) is assumed to be zero mean AWGN process n For ideal distortionless channel where hc(t) is an impulse function and convolution with hc(t) produces no degradation, r(t) can be represented as: (3. 2) 10

3. 2 Detection of Binary Signal in Gaussian Noise n The recovery of signal at the receiver consist of two parts q Filter n Reduces the effect of noise (as well as Tx induced ISI) n The output of the filter is sampled at t=T. This reduces the received signal to a single variable z(T) called the test statistics q Detector (or decision circuit) n Compares the z(T) to some threshold level 0 , i. e. , where H 1 and H 2 are the two possible binary hypothesis 11

Receiver Functionality The recovery of signal at the receiver consist of two parts: 1. Waveform-to-sample transformation (Blue Block) n Demodulator followed by a sampler n At the end of each symbol duration T, predetection point yields a sample z(T), called test statistic (3. 3) Where ai(T) is the desired signal component, and no(T) is the noise component 2. Detection of symbol n Assume that input noise is a Gaussian random process and receiving filter is linear (3. 4) 12

n Then output is another Gaussian random process Where 0 2 is the noise variance n The ratio of instantaneous signal power to average noise power , (S/N)T, at a time t=T, out of the sampler is: (3. 45) n Need to achieve maximum (S/N)T 13

3. 2. 2 The Matched Filter n n n Objective: To maximizes (S/N)T Expressing signal ai(t) at filter output in terms of filter transfer function H(f) (Inverse Fourier transform of the product H(f)S(f)). (3. 46) where S(f) is the Fourier transform of input signal s(t) Output noise power can be expressed as: (3. 47) n Expressing (S/N)T as: (3. 48) 14

n Now according to Schwarz’s Inequality: (3. 49) n n Equality holds if f 1(x) = k f*2(x) where k is arbitrary constant and * indicates complex conjugate Associate H(f) with f 1(x) and S(f) ej 2 f. T with f 2(x) to get: (3. 50) n Substitute in eq-3. 48 to yield: (3. 51) 15

n Or and energy E of the input signal s(t): n Thus (S/N)T depends on input signal energy and power spectral density of noise and NOT on the particular shape of the waveform n Equality for function H 0(f) such that: holds for optimum filter transfer (3. 54) (3. 55) n For real valued s(t): (3. 56) 16

n n n The impulse response of a filter producing maximum output signal-to -noise ratio is the mirror image of message signal s(t), delayed by symbol time duration T. The filter designed is called a MATCHED FILTER Defined as: a linear filter designed to provide the maximum signal-to-noise power ratio at its output for a given transmitted symbol waveform 17

3. 2. 3 Correlation realization of Matched filter n A filter that is matched to the waveform s(t), has an impulse response q h(t) is a delayed version of the mirror image of the original signal waveform Signal Waveform Mirror image of signal waveform Impulse response of matched filter Figure 3. 7 18

n n n This is a causal system q Recall that a system is causal if before an excitation is applied at time t = T, the response is zero for - < t < T The signal waveform at the output of the matched filter is (3. 57) Substituting h(t) to yield: (3. 58) n When t=T, (3. 59) 19

n The function of the correlator and matched filter are the same n Compare (a) and (b) q From (a) 20

n From (b) But n At the sampling instant t = T, we have n This is the same result obtained in (a) n Hence 21

Detection n Matched filter reduces the received signal to a single variable z(T), after which the detection of symbol is carried out The concept of maximum likelihood detector is based on Statistical Decision Theory It allows us to q formulate the decision rule that operates on the data q optimize the detection criterion 22

Probabilities Review n n n P[s 1], P[s 2] a priori probabilities q These probabilities are known before transmission P[z] q probability of the received sample p(z|s 1), p(z|s 2) q conditional pdf of received signal z, conditioned on the class si P[s 1|z], P[s 2|z] a posteriori probabilities q After examining the sample, we make a refinement of our previous knowledge P[s 1|s 2], P[s 2|s 1] q wrong decision (error) P[s 1|s 1], P[s 2|s 2] q correct decision 23

How to Choose threshold? n Maximum Likelihood Ratio test and Maximum a posteriori (MAP) criterion: If n else n Problem is that a posteriori probabilities are not known. Solution: Use Bay’s theorem: n n 24

n MAP criterion: n When the two signals, s 1(t) and s 2(t), are equally likely, i. e. , P(s 2) = P(s 1) = 0. 5, then the decision rule becomes n This is known as maximum likelihood ratio test because we are selecting the hypothesis that corresponds to the signal with the maximum likelihood. n In terms of the Bayes criterion, it implies that the cost of both types of error is the same 25

n Substituting the pdfs 26

n Hence: n Taking the log of both sides will give 27

n Hence where z is the minimum error criterion and 0 is optimum threshold n For antipodal signal, s 1(t) = - s 2 (t) a 1 = - a 2 28

n. This means that if received signal was positive, s 1 (t) was sent, else s 2(t) was sent 29

Detection of Binary Signal in Gaussian Noise The output of the filtered sampled at T is a Gaussian random process 30

Matched Filter and Correlation n n The impulse response of a filter producing maximum output signal-to-noise ratio is the mirror image of message signal s(t), delayed by symbol time duration T. The filter designed is called a MATCHED FILTER and is given by: 31

Bay’s Decision Criterion and Maximum Likelihood Detector n Hence where z is the minimum error criterion and 0 is optimum threshold n For antipodal signal, s 1(t) = - s 2 (t) a 1 = - a 2 32

Probability of Error n Error will occur if q s 1 is sent s 2 is received q n s 2 is sent s 1 is received The total probability of error is the sum of the errors 33

n If signals are equally probable n Numerically, PB is the area under the tail of either of the conditional distributions p(z|s 1) or p(z|s 2) and is given by: 34

n n The above equation cannot be evaluated in closed form (Q-function) Hence, 35

Table for computing of Q-Functions 36

A vector View of Signals and Noise n N-dimensional orthonormal space characterized by N linearly independent basis function {ψj(t)}, where: n From a geometric point of view, each ψj(t) is mutually perpendicular to each of the other {ψj(t)} for j not equal to k. 37

n Representation of any set of M energy signals { si(t) } as a linear combinations of N orthogonal basis functions where N M. where: 38

n Therefore we can represent set of M energy signals {si(t) } as: n Waveform energy: Representing (M=3) signals, with (N=2) orthonormal basis functions 39

Question 1: Why use orthormal functions? n In many situations N is much smaller than M. Requiring few matched filters at the receiver. n Easy to calculate Euclidean distances n Compact representation for both baseband passband systems. Question 2: How to calculate orthormal functions? n Gram-Schmidt orthogonalization procedure. 40

n Examples 41

n Examples (continued) 42

Generalized One Dimensional Signals n One Dimensional Signal Constellation 43

n Binary Baseband Orthogonal Signals q Binary Antipodal Signals q Binary orthogonal Signals 44

Constellation Diagram n n n Is a method of representing the symbol states of modulated bandpass signals in terms of their amplitude and phase In other words, it is a geometric representation of signals There are three types of binary signals: q Antipodal n Two signals are said to be antipodal if one signal is the negative of the other n The signal have equal energy with signal point on the real line q ON-OFF n Are one dimensional signals either ON or OFF with signaling points falling n on the real line 45

q n With OOK, there are just 2 symbol states to map onto the constellation space n a(t) = 0 (no carrier amplitude, giving a point at the origin) n a(t) = A cos wct (giving a point on the positive horizontal axis at a distance A from the origin) Orthogonal q Requires a two dimensional geometric representation since there are two linearly independent functions s 1(t) and s 0(t) 46

q Typically, the horizontal axis is taken as a reference for symbols that are Inphase with the carrier cos wct, and the vertical axis represents the Quadrature carrier component, sin wct Error Probability of Binary Signals n Recall: n Where we have replaced a 2 by a 0. 47

n To minimize PB, we need to maximize: or n We have n Therefore, 48

n The probability of bit error is given by: 49

The probability of bit error for antipodal signals: n n n The probability of bit error for orthogonal signals: The probability of bit error for unipolar signals: 50

n n Bipolar signals require a factor of 2 increase in energy compared to orthogonal signals Since 10 log 102 = 3 d. B, we say that bipolar signaling offers a 3 d. B better performance than orthogonal 51

Comparing BER Performance n For the same received signal to noise ratio, antipodal provides lower bit error rate than orthogonal 52

Relation Between SNR (S/N) and Eb/N 0 n In analog communication the figure of merit used is the average signal power to average noise power ration or SNR. n In the previous few slides we have used the term Eb/N 0 in the bit error calculations. How are the two related? n Eb can be written as STb and N 0 is N/W. So we have: n Thus Eb/N 0 can be thought of as normalized SNR. n Makes more sense when we have multi-level signaling. n Reading: Page 117 and 118. 53