Chapter 7: Digital Communications

TRUE/FALSE 1. Analog signals cannot be sent using digital techniques. ANS: F 2. Digitizing a signal can reduce distortion. ANS: T 3. Digitizing a signal can improve the signal-to-noise ratio. ANS: T 4. Morse code is an example of a binary digital transmission system. ANS: F 5. Digital signals can be modulated onto an analog carrier. ANS: T 6. Digitization removes noise and distortion from analog signals. ANS: F 7. A digital signal can be changed from a 1 to a 0 by noise. ANS: T 8. A regenerative repeater converts a degraded pulse into a new pulse. ANS: T 9. TDM is easy to implement with digital signals. ANS: T 10. Unlike analog, digital communications is not band-limited. ANS: F 11. The amount of digital data that can be sent is limited by the channel capacity (C). ANS: F 12. The digital data rate is limited by the bandwidth of the channel. ANS: T

13. The digital data rate is limited by the SNR of the channel. ANS: T 14. The digital data rate is limited by the number of levels transmitted. ANS: T 15. The Shannon-Hartley theorem describes how to obtain the Shannon limit for transmission. ANS: F 16. Digital data can be sent through a channel no matter how low the SNR is, as long as it is not zero. ANS: T 17. Analog signals must be sampled before they can be sent in digital form. ANS: T 18. Mathematically, it is not possible to completely reconstruct a band-limited signal from only samples of the signal. ANS: F 19. The Nyquist Rate is equal to half the highest frequency component of the analog signal. ANS: F 20. Natural sampling is also called "flat-top" sampling. ANS: F 21. Flat-top sampling requires a sample-and-hold circuit. ANS: T 22. Aliasing occurs when the sampling rate is too high. ANS: F 23. Foldover distortion occurs when the sampling rate is too low. ANS: T 24. Sampling is actually a form of modulation. ANS: T 25. The output of a sampler is a PDM signal. ANS: F

26. The most commonly used digital modulation scheme is PCM. ANS: T 27. In PCM, the number of levels is the same as the number of bits. ANS: F 28. Quantizing is converting a sample of an analog signal to a binary number. ANS: T 29. Quantizing always introduces some error. ANS: T 30. Quantizing always introduces some "noise". ANS: T 31. The significance of quantizing noise increases as the number of bits per sample increases. ANS: F 32. The dynamic range of a digital transmission system depends on the number of bits per sample. ANS: T 33. The bandwidth required by a digital transmission system depends on the number of bits per sample. ANS: T 34. Companding allows improved dynamic range for a given bandwidth. ANS: T 35. The companding system used in America is known as "A-Law" companding. ANS: F 36. Companding is basically a linear process. ANS: F 37. Companding can be done with analog circuitry. ANS: T 38. Digital companding is used by most modern telephone systems. ANS: T

39. With delta modulation, only one bit is transmitted per cycle. ANS: T 40. Delta modulation is particularly well suited to rapidly changing analog signals. ANS: F 41. Delta modulation is prone to "granular noise". ANS: T 42. Adaptive delta modulation reduces the occurrence of "slope-overload". ANS: T 43. A disadvantage of adaptive delta modulation is that it requires a higher bit rate than PCM. ANS: F 44. Digital data is put onto a cable using a line code. ANS: T 45. Unipolar line coding requires DC continuity. ANS: T 46. Bipolar RZ coding generates DC and low-frequency AC components. ANS: F 47. The Manchester line code provides strong timing information. ANS: T 48. The basic DS-1 signal consists of 12 voice channels. ANS: F 49. In DS-1, each analog voice channel is sampled 8000 times per second. ANS: T 50. In DS-1, the bit rate for each voice channel is 56 kbits per second. ANS: F 51. A T-1 frame contains 193 bits. ANS: T 52. T-1 uses AMI.

ANS: T 53. A T-1 cable can be twisted-pair copper wires. ANS: T 54. A T-1 line runs at 1.544 Mbits per second. ANS: T 55. In DS-1, bits are sometimes "stolen" from the voice channel to be used for signaling. ANS: T 56. "Lossy" compression involves transmitting all the data in the original signal, but with fewer bits. ANS: F 57. "Lossless" compression schemes look for redundancies in the data. ANS: T 58. "Run-Length" encoding is a type of lossless compression. ANS: T MULTIPLE CHOICE 1. The first digital code was the: a. ASCII code b. Baudot code ANS: C 2. In digital transmission, signal degradation can be removed using: a. an amplifier c. a regenerative repeater b. a filter d. all of the above ANS: C 3. TDM stands for: a. Time-Division Multiplexing b. Time-Domain Multiplexing ANS: A 4. Hartley's Law is: a. I = ktB b. C = 2B log2M ANS: A 5. The Shannon-Hartley theorem is: c. C = B log2(1 + S/N) d. SR = 2fmax c. Ten-Digital Manchester d. Ten Dual-Manchester c. Morse code d. none of the above

a. I = ktB b. C = 2B log2M ANS: B 6. The Shannon Limit is given by: a. I = ktB b. C = 2B log2M ANS: C 7. The Nyquist Rate can be expressed as: a. I = ktB b. C = 2B log2M ANS: D 8. Natural Sampling does not use: a. a sample-and-hold circuit b. true binary numbers ANS: A

c. C = B log2(1 + S/N) d. SR = 2fmax

c. C = B log2(1 + S/N) d. SR = 2fmax

c. C = B log2(1 + S/N) d. SR = 2fmax

c. a fixed sample rate d. an analog-to-digital converter

9. Which is true about aliasing and foldover distortion? a. They are two types of sampling error. b. You can have one or the other, but not both. c. Aliasing is a technique to prevent foldover distortion. d. They are the same thing. ANS: D 10. Foldover distortion is caused by: a. noise b. too many samples per second ANS: C 11. The immediate result of sampling is: a. a sample alias b. PAM ANS: B 12. Which of these is not a pulse-modulation technique: a. PDM c. PPM b. PWM d. PPS ANS: D 13. Quantizing noise (quantization noise): a. decreases as the sample rate increases b. decreases as the sample rate decreases c. decreases as the bits per sample increases d. decreases as the bits per sample decreases ANS: C c. PCM d. PDM c. too few samples per second d. all of the above

14. The dynamic range of a system is the ratio of: a. the strongest transmittable signal to the weakest discernible signal b. the maximum rate of conversion to the minimum rate of conversion c. the maximum bits per sample to the minimum bits per sample d. none of the above ANS: A 15. Companding is used to: a. compress the range of base-band frequencies b. reduce dynamic range at higher bit-rates c. preserve dynamic range while keeping bit-rate low d. maximize the useable bandwidth in digital transmission ANS: C 16. In North America, companding uses: a. the Logarithmic Law b. the A Law ANS: D 17. In Europe, companding uses: a. the Logarithmic Law b. the A Law ANS: B 18. Codec stands for: a. Coder-Decoder b. Coded-Carrier ANS: A 19. A typical codec in a telephone system sends and receives: a. 4-bit numbers c. 12-bit numbers b. 8-bit numbers d. 16-bit numbers ANS: B 20. Compared to PCM, delta modulation: a. transmits fewer bits per sample b. requires a much higher sampling rate ANS: D 21. In delta modulation, "granular noise" is produced when: a. the signal changes too rapidly c. the bit rate is too high b. the signal does not change d. the sample is too large ANS: B 22. Compared to PCM, adaptive delta modulation can transmit voice: a. with a lower bit rate but reduced quality c. only over shorter distances b. with a lower bit rate but the same quality d. only if the voice is band-limited c. can suffer slope overload d. all of the above c. Code-Compression d. none of the above c. the Law (alpha law) d. the Law (mu law) c. the Law (alpha law) d. the Law (mu law)

ANS: B 23. Which coding scheme requires DC continuity: a. AMI c. unipolar NRZ b. Manchester d. bipolar RZ ANS: C 24. Manchester coding: a. is a biphase code b. has a level transition in the middle of every bit period c. provides strong timing information d. all of the above ANS: D 25. The number of framing bits in DS-1 is: a. 1 b. 2 ANS: A 26. Framing bits in DS-1 are used to: a. detect errors b. carry signaling ANS: C 27. So-called "stolen" bits in DS-1 are used to: a. detect errors b. carry signaling ANS: B 28. The number of bits per sample in DS-1 is: a. 1 b. 2 ANS: D 29. The number of samples per second in DS-1 is: a. 8 k c. 64 k b. 56 k d. 1.544 106 ANS: A 30. The bit rate for each channel in DS-1 is: a. 1.544 Mb/s b. 64 kb/s ANS: B 31. In DS-1, bits are transmitted over a T-1 cable at: a. 1.544 MB/s c. 56 kb/s b. 64 kb/s d. 8 kb/s c. 56 kb/s d. 8 kb/s c. 4 d. 8 c. synchronize the transmitter and receiver d. all of the above c. synchronize the transmitter and receiver d. all of the above c. 4 d. 8

ANS: A 32. A T-1 cable uses: a. Manchester coding b. bipolar RZ AMI coding ANS: B 33. The number of frames in a superframe is: a. 6 b. 12 ANS: B 34. A typical T-1 line uses: a. twisted-pair wire b. coaxial cable ANS: A 35. "Signaling" is used to indicate: a. on-hook/off-hook condition b. busy signal ANS: D 36. A vocoder implements compression by: a. constructing a model of the transmission medium b. constructing a model of the human vocal system c. finding redundancies in the digitized data d. using lossless techniques ANS: B 37. Compared to standard PCM systems, the quality of the output of a vocoder is: a. much better c. about the same b. somewhat better d. not as good ANS: D COMPLETION 1. Digitizing a signal often results in ____________________ transmission quality. ANS: improved better 2. To send it over an analog channel, a digital signal must be ____________________ onto a carrier. ANS: modulated 3. To send it over a digital channel, an analog signal must first be ____________________. c. ringing d. all of the above c. fiber-optic cable d. microwave c. 24 d. 48 c. NRZ coding d. pulse-width coding

ANS: digitized 4. In analog channels, the signal-to-noise ratio of an analog signal gradually ____________________ as the length of the channel increases. ANS: decreases gets worse 5. The ____________________ value of a pulse is the only information it carries on a digital channel. ANS: binary 6. A ____________________ repeater is used to restore the shape of pulses on a digital cable. ANS: regenerative 7. There are techniques to detect and ____________________ some errors in digital transmission. ANS: correct 8. Converting an analog signal to digital form is another source of ____________________ in digital transmission systems. ANS: error noise 9. ____________________-division multiplexing is easily done in digital transmission. ANS: Time 10. All practical communications channels are band-____________________. ANS: limited 11. ____________________ Law gives the relationship between time, information capacity, and bandwidth. ANS: Hartley's 12. Ignoring noise, the _________________________ theorem gives the maximum rate of data transmission for a given bandwidth. ANS: Shannon-Hartley 13. The ____________________ limit gives the maximum rate of data transmission for a given bandwidth and a given signal-to-noise ratio. ANS: Shannon 14. ____________________ sampling is done without a sample-and-hold circuit.

ANS: Natural 15. The ____________________ Rate is the minimum sampling rate for converting analog signals to digital format. ANS: Nyquist 16. ____________________ distortion occurs when an analog signal is sampled at too slow a rate. ANS: Foldover 17. ____________________ means that higher frequency baseband signals from the transmitter "assume the identity" of low-frequency baseband signals at the receiver when sent digitally. ANS: Aliasing 18. The output of a sample-and-hold circuit is a pulse-____________________ modulated signal. ANS: amplitude 19. ____________________ modulation is the most commonly used digital modulation scheme. ANS: Pulse-code 20. ____________________ noise results from the process of converting an analog signal into digital format. ANS: Quantizing 21. ____________________ is used to preserve dynamic range using a reasonable bandwidth. ANS: Companding 22. In North America, compression is done using the ____________________-law equation. ANS: mu 23. In Europe, compression is done using the ____________________-law equation. ANS: A 24. A ____________________ is an IC that converts a voice signal to PCM and vice versa. ANS: codec 25. In a PCM system, the samples of the analog signal are first converted to ____________________ bits before being compressed to 8 bits. ANS: 12

26. The number of bits per sample transmitted in delta modulation is ____________________. ANS: 1 one 27. Delta modulation requires a ____________________ sampling rate than PCM for the same quality of reproduction. ANS: higher 28. ____________________ noise is produced by a delta modulator if the analog signal doesn't change. ANS: Granular 29. In delta modulation, ____________________ overload can occur if the analog signal changes too fast. ANS: slope 30. The ____________________ size varies in adaptive delta modulation. ANS: step 31. Adaptive delta modulation can transmit PCM-quality voice at about ____________________ the bit rate of PCM. ANS: half 32. Unipolar NRZ is not practical because most channels do not have ____________________ continuity. ANS: DC 33. In AMI, binary ones are represented by a voltage that alternates in ____________________. ANS: polarity 34. Long strings of ____________________ should be avoided in AMI. ANS: zeros 35. Manchester code has a level ____________________ in the center of each bit period. ANS: transition 36. Manchester coding provides ____________________ information regardless of the pattern of ones and zeros. ANS: timing 37. There are ____________________ channels in a DS-1 frame.

ANS: 24 38. DS-1 uses a ____________________ bit to synchronize the transmitter and receiver. ANS: framing 39. In DS-1, each channel is sampled ____________________ times per second. ANS: 8000 40. Data is carried over a T-1 line at a rate of ____________________ bits per second. ANS: 1.544 106 41. A group of 12 DS-1 frames is called a ____________________. ANS: superframe 42. From a group of twelve frames, signaling bits are "stolen" from every ____________________ frame. ANS: sixth 43. ____________________ compression transmits all the data in the original signal but uses fewer bits to do it. ANS: Lossless SHORT ANSWER 1. Use Hartley's Law to find how much time it would take to send 100,000 bits over a channel with a bandwidth of 2,000 hertz and a channel constant of k = 10. ANS: 5 seconds 2. Use the Shannon-Hartley theorem to find the bandwidth required to send 12,000 bits per second if the number of levels transmitted is 8. ANS: 2000 hertz 3. What is the Shannon Limit of a channel that has a bandwidth of 4000 hertz and a signal-to-noise ratio of 15? ANS: 16 kbps 4. What is the minimum required number of samples per second to digitize an analog signal with frequency components ranging from 300 hertz to 3300 hertz? ANS:

6600 samples/second 5. What is the approximate dynamic range, in dB, of a linear PCM system that uses 12 bits per sample? ANS: 74 dB 6. What is the approximate data rate for a system using 8 bits per sample and running at 8000 samples per second? ANS: 64 kbps 7. If bits were "stolen" from every DS-1 frame, what would the useable data-rate be for each channel in the frame? ANS: 56 kbps 8. Assuming maximum input and output voltages of 1 volt, what is the output voltage of a -law compressor if the input voltage is 0.388 volt? ANS: 0.833 volt