388 PRINCIPLES OF DIGITAL DATA TRANSMISSION REFERENCES 1. A. Lender, “Duobinary Technique for High Speed Data Transmission,” /EEE Trans. Commun. Electron. vol. CE-82, pp. 214-218, May 1963 2, A, Lender, “Corelative Level Coding for Binary-Data Transmission,” IEEE Spectrum, vol. 3, o. 2, pp. 104-115, Feb. 1966. 3. P. Bylanski and D. G. W, Ingram, Digital Transmission Systems, Peter Peregrinus Lid., Hertshire, England, 1976. 4. H. Nyquist, “Certain Topics in Telegraph Transmission Theory.” AIEE Trans., vol. 47, p. 817, April 1928. 5. E.D. Sunde, Communication Systems Engineering Technology, Wiley, New York, 1969. 6. R.W. Lucky and H. R. Rudin, “Generalized Automatic Equalization for Communication Channels,” IEEE Int. Commun. Conf, vol. 2, 1966. 7. W. F. Trench, “An Algorithm for the Inversion of Finite Toeplitz Matrices,” J. SIAM, vol. 12, pp. 515-522, Sept. 1964. 8. A. Lender, Chapter 7, in Digital Communications: Microwave Applications, K. Feber, Ed, Prentice- PROBLEMS 724 12-2 723 124 Hall, Englewood Ciiffs, NJ, 1981. Consider a full dth rectangular pulse shape PO) =MG/Ts) (a) Find PSDs for the polar, on-off, and bipolar signaling. (b) Sketch roughly the PSDs and find their bandwidths. For each case, compare the bandwidth to the case where p() isa half-width rectangular pulse. (a) A random binary data sequence H1OL00101- is tansmitted by using a Manchester (pit. phase) line code with the pulse p) shown in Fig. 7.7a, Sketch the waveform (0). (©) Derive Sy(/), the PSD of « Manchester (split-phase) signal in pat (a) assuming 1 and 0 ‘equally likey. Roughly sketch this PSD and find is bandwidth If the pulse shape is ; no-=n(qin) use differential code (see Fig. 7.18) to derive the PSD for a binary signal, Determine the PSD SO, ‘The duobinary line coding proposed by Lender is also ternary ike bipolar, but it requires only half the bandwidth of bipolar, In practice, duobinary coding is indirectly realized by using a special pulse shape as discussed in Sec, 7.3 (see Fig. 7-18). In this code, a0 is transmitted by no pulse, and a1 is transmitted by a pulse p(t) or —p() using th following rule. A 1 is encoded by the same pulse as that used for the previous 1 if there are an even number of Os between them. It is encoded by a pulse of opposite polarity if there are an odd number of Os between them. A number 0 is considered to be an even number. Like bipolar, this code also has a single error detection capability, because correct reception implies that between successive pulses of the same polarity, an even number of @s must occur, and between successive pulses of opposite polarity, an odd number of Os must occur.134 132 133 134 Figure P.7.3-4 Problems 389 (a) Assuming half-width rectangular pulse, sketch the duobinary signal y(t) fr the random binary sequence 1110001101001010. (b) Determine Ro, Ry, and Rp for this code. Assume (or you may show if you like) that Ry, = 0 for all > 2. Find and sketch the PSD for this line code (assuming half-width pulse). Show ‘that its bandwidth is Ry /2 Hi, half that of bipolar. Data at a rate of 6 kbit/s isto be transmitted over a leased line of bandwidth 4 kHz by using [Nyquist eriterion pulses. Determine the maximum value ofthe roll-off factor r that can be used. Ina certain telemetry system, there are eight analog measurements, each of bandwidth 2 kHz. ‘Samples ofthese signals are time-division- multiplexed, quantized, and binary-coded. The error in sample amplitudes cannot be greater than 1% of the peak amplitude, (2) Determine L, the number of quantization levels. (b) Find the transmission bandwidth Br if Nyquist criterion pulses with roll-off factor r = are used, The sampling rate must be atleast 25% above the Nyquist rat, ‘Alleased telephone line of bandwidth 3 KHz is used to transmit binary data. Calculate the data rate (in bits per second) that can be transmitted if we use: (a) Polar signal with rectangular half-width pulses. (b) Polar signal with rectangular full-width pulses. () Polar signal using Nyquist criterion pulses of r = 0.25. (@) Bipolar signal with rectangular half-width pulses. (©) Bipolar signal with rectangular full-width pulses. ‘The Fourier transform P(f) of the basic pulse p(é) used in a certain binary communication system is shown in Fig. P7.3-4 (a) Fromthe shape of P(f) explain at what pulse rate this pulse would satisfy Nyquist’s criterion, (b) Find p(e) and verify that this pulse does (or does not) satisfy the Nyquist’s criterion, (© If the pulse does satisfy the Nyquist criterion, what is the transmission rate (in bits per second) and what is the roll-off factor? PU) 135 136 pulse p() whose spectrum P(f) is shown in Fig. P7.3-5 satisfies Nyquist’s criterion. Iff ‘MHZ and fy = 1.2 MHz, determine the maximum rate at which binary data can be transmitted by this pulse using Nyquist’s criterion. What isthe roll-off factor? [Binary data ata rate of | Mbit is to be transmitted by using Nyquist criterion pulses with P(/) shown in Fig. P7.3-5, The frequencies f and fy of the spectrum are adjustable. The channel availabe forthe transmission of this data has bandwidth of 700 kHz. Determine fy and fy and the roll-off factor.390 PRINCIPLES OF DIGITAL DATA TRANSMISSION Figure P.7.3-5 + rn 1 os : i 73-7 738 739 73-10 734 14d SME > Show thatthe inverse Fourier transform of P(f) in Eq, (7.39) is indeed second erterion pulse ‘PCO given in Eq. (7.38). Hint: Use Eq. (3.32) to find the inverse transform of P(f) in Eq, (7.39) and express sine (x) in the form sin x/x. ‘Show thatthe inverse Fourier transform of P(f) (the raised cosine pulse spectrum in Eq. (7.35) is the pulse p(t) given in Eq (7.36). Hint: Use Eq, (332) to find the inverse transform of P(f) in Eq. (7.39) and express sine (x) in the form sin x/s. ‘Show that there exists one (and only one) pulse p(t) of bandwidth Ry/2 Hi that satisfies the ctiterion of second criterion pulse [Eg. (7.37)}. Show that this pulse is given by sin (xR) (0) = {sine (eRyt) + sine fe Rye ~ 75)}} = —SA TR PCO) = (sin Gr Ryo) + sine in RS C= To = Es and its Fourier transform is P(F) given in Eq, (7.39). “Hint: For a pulse of bandwidth R,/2, the Nyguist interval is 1/R = Tp and the conditions (7.37) give the Nyquist sample values at r = =, Use the interpolation formula (Eq (6.10)] with B = Rg/2, Ts = Tp toconsuct (0). In determining P(/), recognize that (1++e"F"1T) = eM inMls e-infTy, {In a binary data transmission using duobinary pulses, sample values were read as follows: 120 ~2 ~200 ~202002000 -2 (a) Explain if there is any error in detection. (b) If there is no detection error, determine the received bit sequence. Ina binary data transmission using duobinary pulses, sample values ofthe received pulses were read as follows: 12000 ~200 ~20200 ~20220 ~2 (a) Explain if there is any error. (b) Can you guess the correct transmitted digit sequence? There is more than one possible ‘correct sequence, Give as many correct sequences as possible, assuming that more than one ‘detection eror is extremely unlikely. In Example 7.2, when the sequence $ = 101010100000111 was applied to the input of the scrambler in Fig. 720s, the output T was found to be 101110001101001. Verify that when this