1. 1 Topic 4: Physical Layer - Chapter 8: Data Communication Fundamentals Business Data Communications, 4e

2. 2 Outline  Characteristics of Electromagnetic Signals  Data, Signal, and Transmission  Analog Transmission of Digital Data  Digital Transmission of Analog Data  Digital Transmission of Digital Data

3. 3 Electromagnetic Signals  Function of time Analog (varies smoothly over time) Digital (constant level over time, followed by a change to another level)  Function of frequency (more important) Spectrum (range of frequencies) Bandwidth (width of the spectrum)

4. 4 Periodic Signal Characteristics S(t) = A sin(2  ft +  ) Amplitude (A): signal value, measured in volts Frequency (f): repetition rate, cycles per second or Hertz Period (T): amount of time it takes for one repetition, T=1/f Phase (  ): relative position in time, measured in degrees

5. 5 Bandwidth  Width of the spectrum of frequencies that can be transmitted if spectrum=300 to 3400Hz, bandwidth=3100Hz  Greater bandwidth leads to greater costs  Limited bandwidth leads to distortion

6. 6 Bandwidth on a Voice Circuit  Human hearing ranges from about 20 Hz to about 14,000 Hz (some up to 20,000 Hz). Human voice ranges from 20 Hz to about 14,000 Hz.  The bandwidth of a voice grade telephone circuit is 0 to 4000 Hz or 4000 Hz (4 KHz).  Guardbands prevent data transmissions from interfering with other transmission when these circuits are multiplexed using FDM.

7. 7 Bandwidth on a Voice Circuit

8. 8  It is important to note that the limit on bandwidth is imposed by the equipment used in the telephone network.  The actual capacity of bandwidth of the wires in the local loop depends on what exact type of wires were installed, and the number of miles in the local loop.  Actual bandwidth in North America varies from 300 KHz to 1 MHz depending on distance.

9. 9 Data  Analog data Voice Images  Digital data Text Digitized voice or images

10. 10 time (sec) amplitude (volts) 1 cycle frequency (hertz) = cycles per second phase difference Analog Signaling  represented by sine waves

11. 11 Phase Frequency: 1 Period/Sec = 1 Hertz   Phase

12. 12 Three Components of Data Communication  Data Analog: Continuous value data (sound, light, temperature) Digital: Discrete value (text, integers, symbols)  Signal Analog: Continuously varying electromagnetic wave Digital: Series of voltage pulses (square wave)  Transmission Analog: Works the same for analog or digital signals Digital: Used only with digital signals

13. 13 Data Transmissions  Analog Transmission of Analog Data Telephone networks (PSTN)  Digital Transmission of Digital Data A computer system  Analog Transmission of Digital Data Uses Modulation/Demodulation (Modem)  Digital Transmission of Analog Data Uses Coder/Decoder (CODEC)

14. 14 Digital Coding  Character: A symbol that has a common, constant meaning.  Characters in data communications, as in computer systems, are represented by groups of bits [1’s and 0’s].  The group of bits representing the set of characters in the “alphabet” of any given system are called a coding scheme, or simply a code.

15. 15 Digital Coding  A byte consists of 8 bits that is treated as a unit or character. (Some Asian languages use 2 bytes for each of their characters, such as Chinese.)  (The length of a computer word could be 1, 2, 4 bytes.)  There are two predominant coding schemes in use today:  United States of America Standard Code for Information Interchange (USASCII or ASCII)ASCII  Extended Binary Coded Decimal Interchange Code (EBCDIC)EBCDIC

16. 16 Advantages of Digital Transmission  The signal is exact  Signals can be checked for errors  Noise/interference are easily filtered out  A variety of services can be offered over one line  Higher bandwidth is possible with data compression

17. 17 Why Use Analog Transmission?  Already in place  Significantly less expensive  Lower attenuation rates  Fully sufficient for transmission of voice signals

18. 18 Analog Encoding of Digital Data  Data encoding and decoding technique to represent data using the properties of analog waves  Modulation: the conversion of digital signals to analog form  Demodulation: the conversion of analog data signals back to digital form

19. 19 Methods of Modulation  Amplitude modulation (AM) or amplitude shift keying (ASK)  Frequency modulation (FM) or frequency shift keying (FSK)  Phase modulation or phase shift keying (PSK)  Differential Phase Shift Keying (DPSK)

20. 20 Amplitude Shift Keying (ASK)  In radio transmission, known as amplitude modulation (AM)  The amplitude (or height) of the sine wave varies to transmit the ones and zeros  Major disadvantage is that telephone lines are very susceptible to variations in transmission quality that can affect amplitude

21. 21 Amplitude Modulation and ASK

22. 22 Frequency Shift Keying (FSK)  In radio transmission, known as frequency modulation (FM)  Frequency of the carrier wave varies in accordance with the signal to be sent  Signal transmitted at constant amplitude  More resistant to noise than ASK  Less attractive because it requires more analog bandwidth than ASK

23. 23 Frequency Modulation and FSK

24. 24 Phase Modulation and PSK

25. 25 Phase Shift Keying (PSK)  Also known as phase modulation (PM)  Frequency and amplitude of the carrier signal are kept constant  The carrier signal is shifted in phase according to the input data stream  Each phase can have a constant value, or value can be based on whether or not phase changes (differential keying)

26. 26 011 Differential Phase Shift Keying (DPSK) 0

27. 27 Sending Multiple Bits Simultaneously

28. 28 Sending Multiple Bits Simultaneously 3  /2  11  10  /2  01 0 00

29. 29 Sending Multiple Bits Simultaneously In practice, the maximum number of bits that can be sent with any one of these techniques is about five bits. The solution is to combine modulation techniques. One popular technique is quadrature amplitude modulation (QAM) involves splitting the signal into eight different phases, and two different amplitude for a total of 16 different possible values.

30. 30 Sending Multiple Bits Simultaneously Trellis coded modulation (TCM) is an enhancement of QAM that combines phase modulation and amplitude modulation. It can transmits different numbers of bits on each symbol (6-10 bits per symbol). The problem with high speed modulation techniques such as TCM is that they are more sensitive to imperfections in the communications circuit.

31. 31 Example  Use a drawing to show how the bit pattern 11100100 would be sent using a combination of 1-bit Amplitude Modulation and 1-bit Phase Modulation (1AM+1PM).

32. 32 Modem  An acronym for modulator-demodulator  Uses a constant-frequency signal known as a carrier signal  Converts a series of binary voltage pulses into an analog signal by modulating the carrier signal  The receiving modem translates the analog signal back into digital data

33. 33 Modem Standards  V.22 1200-2400 baud/bps (FM)  V.32 and V.32bis full duplex at 9600 bps (2400 baud at QAM) bis uses TCM to achieve 14,400 bps.  V.34 for phone networks using digital transmission beyond the local loop. 59 combinations of symbol rate and modulation technique symbol rates 3429 baud. Its bit rate is up to 28,800 bps (TCM-8.4)  V.34+ up to 33.6 kbps with TCM-9.8

34. 34 Modem Standards (Cont’d)  V.42bis data compression modems, accomplished by run length encoding, code book compression, Huffman encoding and adaptive Huffman encodingHuffman encoding MNP5 - uses Huffman encoding to attain 1.3:1 to 2:1 compression. it uses Lempel-Ziv encoding and attains 3.5:1 to 4:1.Lempel-Ziv V.42bis compression can be added to almost any modem standard effectively tripling the data rate.

35. 35 Voice Grade Modems

36. 36 Data Compression  How fast if using V.42bis V.32 - 57.6kbps V.34 - 115.2 kbps V.34+ - 133.4 kbps V.90 ?

37. 37 Data Compression There are two drawbacks to the use of data compression:  Compressing already compressed data provides little gain.  Data rates over 100 Kbps place considerable pressure on the traditional microcomputer serial port controller that controls the communications between the serial port and the modem.

38. 38 Analog Channel Capacity: BPS vs. Baud  Baud=# of signal changes per second. ITU-T now recommends the term baud rate be replaced by the term symbol rate.  BPS=bits per second  In early modems only, baud=BPS. The bit rate and the symbol rate (or baud rate) are the same only when one bit is sent on each symbol.  Each signal change can represent more than one bit, through complex modulation of amplitude, frequency, and/or phase  Increases information-carrying capacity of a channel without increasing bandwidth  Increased combinations also leads to increased likelihood of errors

39. 39 Digital Transmission of Analog Data  Codec = Coder/Decoder  Converts analog signals into a digital form and converts it back to analog signals  Where do we find codecs? Sound cards Scanners Voice mail Video capture/conferencing

40. 40 Codec vs. Modem  Codec is for coding analog data into digital form and decoding it back. The digital data coded by Codec are samples of analog waves.  Modem is for modulating digital data into analog form and demodulating it back. The analog symbols carry digital data.

41. 41 Digital Encoding of Analog Data  Primarily used in retransmission devices  The sampling theorem: If a signal is sampled at regular intervals of time and at a rate higher than twice the significant signal frequency, the samples contain all the information of the original signal.  Pulse-code modulation (PCM) 8000 samples/sec sufficient for 4000hz

42. 42 Pulse Code Modulation (PCM) Analog voice data must be translated into a series of binary digits before they can be transmitted. With Pulse Code Modulation (PCM), the amplitude of the sound wave is sampled at regular intervals and translated into a binary number. The difference between the original analog signal and the translated digital signal is called quantizing error.

43. 43 PCM

44. 44 PCM

45. 45 PCM

46. 46 PCM PCM uses a sampling rate of 8000 samples per second. Each sample is an 8 bit sample resulting in a digital rate of 64,000 bps (8 x 8000).

47. 47 Converting Samples to Bits  Quantizing  Similar concept to pixelization  Breaks wave into pieces, assigns a value in a particular range  8-bit range allows for 256 possible sample levels  More bits means greater detail, fewer bits means less detail

48. 48 Analog/Digital Modems (56k Modems)  The basic idea behind 56K modems (V.90) is simple. 56K modems take the basic concepts of PCM and turn them backwards. They are designed to recognize an 8-bit digital signal 8000 times per second.V.90  It is impractical to use all 256 discrete codes, because the corresponding DAC output voltage levels near zero are just too closely spaced to accurately represent data on a noisy loop. Therefore, the V.90 encoder uses various subsets of the 256 codes that eliminate DAC output signals most susceptible to noise. For example, the most robust 128 levels are used for 56 Kbps, 92 levels to send 52 Kbps, and so on. Using fewer levels provides more robust operation, but at a lower data rate.

49. 49 Downstream vs. Upstream

50. 50 Downstream vs. Upstream

51. 51 Analog/Digital Modems (56k Modems) Noise is a critical issue. Recent tests found 56K modems to connect at less than 40 Kbps 18% of the time, 40-50 Kbps 80% of the time, and 50+ Kbps only 2 % of the time. It is easier to control noise in the channel transmitting from the server to the client than in the opposite direction. Because the current 56K technology is based on the PCM standard, it cannot be used on services that do not use this standard.

52. 52 Digital Encoding of Digital Data  Most common, easiest method is different voltage levels for the two binary digits  Typically, negative=1 and positive=0  Known as NRZ-L, or nonreturn-to-zero level, because signal never returns to zero, and the voltage during a bit transmission is level

53. 53 Differential NRZ  Differential version is NRZI (NRZ, invert on ones)  Change=1, no change=0  Advantage of differential encoding is that it is more reliable to detect a change in polarity than it is to accurately detect a specific level  Used for low speed (64Kbps) ISDN

54. 54 Problems With NRZ  Difficult to determine where one bit ends and the next begins  In NRZ-L, long strings of ones and zeroes would appear as constant voltage pulses  Timing is critical, because any drift results in lack of synchronization and incorrect bit values being transmitted

55. 55 Biphase Alternatives to NRZ  E.g. Manchester coding and Differential Manchester coding  Require at least one transition per bit time, and may even have two  Modulation rate is greater, so bandwidth requirements are higher  Advantages Synchronization due to predictable transitions Error detection based on absence of a transition

56. 56 Manchester Code  Transition in the middle of each bit period  Transition provides clocking and data  Low-to-high=1, high-to-low=0  Used in Ethernet

57. 57 Differential Manchester  Midbit transition is only for clocking  Transition at beginning of bit period=0  Transition absent at beginning=1  Has added advantage of differential encoding  Used in token-ring

58. 58 Digital Encoding Illustration

59. 59 Transmission Timing - Asynchronous vs. Synchronous  Sampling timing – How to make the clocks in a transmitter and a receiver consistent?  Asynchronous transmission – sending shorter bit streams and timing is maintained for each small data block.  Synchronous transmission – To prevent timing draft between transmitter and receiver, their clocks are synchronized. For digital signal, this can be accomplished with Manchester encoding or differential Manchester encoding.

60. 60 Digital Interfaces  The point at which one device connects to another  Standards define what signals are sent, and how  Some standards also define physical connector to be used

61. 61 Generic Communications Interface Illustration

62. 62 DTE and DCE

63. 63 RS-232C (EIA 232C)  EIA’s “Recommended Standard” (RS)  Specifies mechanical, electrical, functional, and procedural aspects of the interface  Used for connections between DTEs and voice-grade modems, and many other applications

64. 64 *EIA-232-D  new version of RS-232-C adopted in 1987  improvements in grounding shield, test and loop-back signals  the prevalence of RS-232-C in use made it difficult for EIA-232-D to enter into the marketplace

65. 65 *RS-449  EIA standard improving on capabilities of RS-232-C  provides for 37-pin connection, cable lengths up to 200 feet, and data rates up to 2 million bps  covers functional/procedural portions of R-232-C electrical/mechanical specs covered by RS- 422 & RS-423

66. 66 *Functional Specifications  Specifies the role of the individual circuits  Data circuits in both directions allow full-duplex communication  Timing signals allow for synchronous transmission (although asynchronous transmission is more common)

67. 67 *Procedural Specifications  Multiple procedures are specified  Simple example: exchange of asynchronous data on private line Provides means of attachment between computer and modem Specifies method of transmitting asynchronous data between devices Specifies method of cooperation for exchange of data between devices

68. 68 *Mechanical Specifications  25-pin connector with a specific arrangement of leads  DTE devices usually have male DB25 connectors while DCE devices have female  In practice, fewer than 25 wires are generally used in applications

69. 69 DB-25 Female DB-25 Male *RS-232 DB-25 Connectors

70. 70 *RS-232 DB-25 Pinouts

71. 71 *RS-232 DB-9 Connectors  Limited RS-232

72. 72 *RS-422 DIN-8  Found on Macs DIN-8 MaleDIN-8 Female

73. 73 *Electrical Specifications  Specifies signaling between DTE and DCE  Uses NRZ-L encoding Voltage < -3V = binary 1 Voltage > +3V = binary 0  Rated for <20Kbps and <15M greater distances and rates are theoretically possible, but not necessarily wise

74. 74 *RS-232 Signals (Asynch) Odd Parity Even Parity No Parity