MTI 7602 PCM Modulation and Demodulation

Page 1 of 1 MTI 7602 PCM Modulation and Demodulation Contents Aims of the Exercise Learning about the functioning principle of the pulse-code modulation (quantisation, coding, time division multiplex operation) Presentation of the advantages of the two commonly used companding processes (following A Law and µ Law) Overview of Exercises Pulse-Code Modulation Exercise 1: Quantisation Quantisation and quantisation errors without companding Presentation of the data word for transmission of sinusoidal and direct voltage Exercise 2: Companding process Presentation of the quantisation with companding following A Law (13 segmented characteristic) and following µ Law (15 segmented characteristic) Presentation of the data word for transmission of direct voltage using the two companding processes Presentation of the companding by direct comparison of input and output voltage Direct presentation of the companding characteristics Distortions in the case of a small output signal Exercise 3: Time division multiplex operation Principle of the time division multiplex operations for PCM transmission routes Presentation of the time division multiplex principle for the transmission of two direct voltages via a PCM transmission link Presentation of the time division multiplex principle for the transmission of a direct voltage and sinusoidal alternating voltage via a PCM transmission link

Page 1 of 3 MTI 7602 PCM Modulation and Demodulation Introduction Pulse Code Modulation, PCM The multiple usage with time division multiplex is also vulnerable to amplitude interference which is typical of PAM signals. However, this interference can be considerably reduced by digitising the PAM signal in discrete numerical values and transmitting these signals in the form of a serial, digital signal. The digital information is obtained by dividing the PAM information signal into defined stages of amplitude - the so-called "quantising intervals" - and then assigning a binary codeword to each stage. The more bits that this codeword contains, the higher is the resolution of the amplitude stages. For example, in a CD player, the resolution is 16 bits (= 65563 stages); in this exercise, the ISDN standard of 8 bits ( = 256 stages) is used. Thus, a PCM modulator consists initially, of the PAM modulator circuit described above followed by the stages responsible for quantising and coding. Since the quantising and coding processes require a definite time to complete, the amplitude of the sampled value is held constant via a so-called Sample-and-Hold stage. This is then followed by the A/D converter and a parallel-serial converter, for the final generation of the PCM signal. Fig. 01: schematic diagram of a PCM modulator In the quantising process, a so-called "quantising error" is produced, caused by the fact that all analog amplitude values that are within one of the 256 quantising intervals can be formed to only one average value corresponding to the quantising size. This error is made evident as non-linear distortion of the signal and is therefore also known

Page 2 of 3 as quantising (or quantisation) noise. The amplitude of the quantising interval is calculated by: U Q = U/265. With an information signal amplitude of U = 10 V pp the noise is then: U Q = 10 V/265 = 39mV. The signal power of the quantising noise is solely dependent on the number of bits in a quantising stage and not on the amplitude of the information (wanted) signal. The signal-to-noise ratio is then given by: S Q = 10 lg P inf/ P Q in db mit S Q = Signal-to-noise ratio P inf = Power of the wanted signal P Q = Power of the quantising noise This means however that with small amplitude wanted signals, the signal-to-noise ratio becomes less. Also, the human ear senses the noise on small signals much more than with larger amplitude wanted signals. To counteract this and still be able to work with the PCM standard method for telephony of an 8-bit resolution, the quantising is realised with a 12-bit A/D converter then compressed via a non-linear, logarithmic characteristic. The result is that small amplitude signals are much finer quantised as large amplitude signals, so that the absolute amount of the quantising noise increases with signal amplitude but the signal-to-noise ratio remains approximately the same. The standard rule for forming the logarithmic characteristic in Europe, is the A-Law characteristic with 13 segments, and in North America and Japan, the m-law characteristic with 15 segments. The exercise assembly here can be used for either characteristic. The word length of the PCM signal for telephony, at a sample frequency of 8 khz, results in a standard bit rate of 8 Bit 8 khz = 64 kbit/s. At the demodulator side, after a parallel-serial conversion, the data is expanded according to the compression process that has been used, so that a 12-bit signal is again available for the D/A converter. This explains the name used for the entire process, of "companding" (= compression + expanding). At the output of the PCM demodulator, the PAM signal is again available that was applied to the PAM demodulator described previously. Differential PAM / Differential PCM (DPAM / DPCM) A considerable part of the static average in speech as well as picture signals, is

Page 3 of 3 redundant, i.e. the signals show little change between each sample value, or there is no change at all. This means however, that for transmitting the information, it is sufficient to transmit only the changes in the signal, that have occurred since the previous sample. Thus, the difference from the previous sample value is added to the fundamental part of the signal that has remained constant (redundant), so that as a result, the original signal is obtained. In this exercise, the DPAM signal is created by differentiating the information signal as a function of time. The DPAM signal is then quantised in the DPCM modulator and coded.

Page 1 of 4 MTI 7602 PCM Modulation and Demodulation Exercise Assembly Insert the PAM / PCM modulator in the left hand side of the Experimenter and the PAM / PCM demodulator in right hand side. Using 2mm connection cables, connect the signal outputs of the PAM modulator to the inputs of the PAM demodulator: CLCK - CLCK, SYNC. - SYNC., PCMout - PCMin. When the power supply of the UniTr@in-I Interface is switched on, all necessary voltage supplies are available and the exercise can commence. Safety Notice: External sources of voltage must not be connected to the measurement points. Such action will cause damage to the components on the panels! Before switching on the power supply, set all the following switches and potentiometers in their initial position: Experimenter card Element Position SO4201-7R Gain potentiometer, channel 1 Fully CCW SO4201-7R Gain potentiometer, channel 2 Fully CCW SO4201-7R Compression mode selector A-Law linear SO4201-7R Channel selector Channel 1

Page 1 of 4 MTI 7602 PCM Modulation and Demodulation Quantising and Errors, without Companding Exercise 1 Examining the quantising when transmitting a DC voltage On the PAM-/PCM modulator Experimenter card (SO4201-7R), connect the "LF1" input of channel 1 to the output of the internal DC source, "DC +5V/-5V". Adjust the DC source voltage potentiometer for an output of +2.5V (check with a multimeter). Connect the "LF2" input for channel 2 to the ground socket "AGND". Set the gain control for channel 2 fully counter-clockwise. Set the "Compression mode" selector on the PAM-/PCM modulator to"8-bit linear, A- Law" (no jumpers inserted) and the LED mode in the setting for channel 1. Proceed in a similar manner for the settings on the PAM-/ PCM demodulator. Connect voltmeter A to the signal output "U1.1" and adjust the gain of channel 1 so that all LED's on the PAM-/PCM modulator card (SO4201-7R) are just at the point of lighting. Now set the bit patterns (data words) as given in table 1 on the worksheet and measure the corresponding input voltage at the output of the Sample and Hold element for channel 1 "U1.1" on the PAM-/ PCM modulator, using voltmeter A as well as the corresponding demodulated output voltage at the output of the Hold element "Hold 1" on the PAM-/ PCM demodulator, using voltmeter B. Note: The input and output voltages are measured at the outputs of the Hold element to reduce the effects of errors in the analog section of the transmission path, as much as possible. For an easier fine adjustment of the voltage value, use the potentiometer for the gain of channel 1 and for selecting the measurement range, use the potentiometer for the DC source. For measurements with the voltmeters, use the Average-Mode "AV" and switch the measurement range according to the value being measured. The values obtained can inserted in the table by a simple 'drag-and-drop' procedure with the mouse. DC in /V 2 7 MSB 2 6 2 5 2 4 2 3 2 2 2 1 2 0 LSB DC out /V 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 0 1 1 0 1 0 0 1 0 1

Page 2 of 4 Table 1: Linear quantising with counting according to A-Law For the bit pattern, 1 0 1 0 0 1 0 1 measure the signals at PCMout on channel A and the SYNC signal on channel B of the oscilloscope, Trigger on SYNC. How is the data transmitted in the data channels. What is the significance of MSB and LSB Calculate the quantising interval. Determine the quantising interval by measuring the difference in the input voltages for two successive bit patterns (e.g..g. 10100110 followed by 10100101), then calculate the interval. For calculation, use the output values of the demodulator. What technical problems are introduced by the transmission of very small signals 1 0 1 0 0 1 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 1 0 0 0 0 1 0 0 1 0 1 0 1 1 1 1 1 0 1 0 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 X = 10 µs/div X/T (B) Chan. A= 2 V/DIV DC Chan. B= 2 V/DIV DC

Page 3 of 4 Fig. 1: Serial PCM signal of 1 0 1 0 0 1 0 1 value Exercise 2 Examining the quantising when transmitting an alternating, sinusoidal voltage On the PAM-/PCM modulator Experimenter card (SO4201-7R), connect the "LF1" input of channel 1 to the output of the internal sinewave source "500Hz". All other settings remain as they were for exercise 1. Measure the signals at PCMout on channel A and the SYNC signal on channel B of the oscilloscope, trigger on SYNC, for the cases when the amplitude of the sinewave signal is firstly, U = 2V pp and secondly, U = 10V pp. What changes can be seen in the PCM signal for the two sinewave signals Why is the display of the MSB, compared to the other bits, relatively constant X = 10 µs/div X/T (B) Chan. A= 2 V/DIV DC Chan. B= 2 V/DIV DC

Page 4 of 4 Fig. 2: PCM signal for a sinewave, 500 Hz at U=2 V pp X = 100 µs/div X/T (B) Chan. A= 1 V/DIV DC Chan. B= 1 V/DIV DC Fig. 3: PCM signal for a sinewave, 500 Hz at U=10 V pp

Page 1 of 7 MTI 7602 PCM Modulation and Demodulation Examining the Quantising with Companding according to the A- Law (13-segment characteristic) and the m-law (15-segment char.) Exercise 1 Examining the data word when transmitting DC voltage, using either method of companding Set the "Compression mode" selector on the PAM-/PCM modulator to "A-Law noninverting" (jumper inserted in position 1) and the LED mode in the setting for channel 1. Proceed in a similar manner for the settings on the PAM-/ PCM demodulator. Connect voltmeter A to the signal output "U1.1" and adjust the gain of channel 1 so that all LED's on the PAM-/PCM modulator card (SO4201-7R) are just at the point of lighting. Now set the bit patterns (data words) as given in table 1 on the worksheet and measure the corresponding input voltage at the output of the Sample and Hold element for channel 1 "U1.1" on the PAM-/ PCM modulator, using voltmeter A as well as the corresponding demodulated output voltage at the output of the Hold element "Hold 1" on the PAM-/ PCM demodulator, using voltmeter B. Note: The input and output voltages are measured at the outputs of the Hold element to reduce the effects of errors in the analog section of the transmission path, as much as possible. For an easier fine adjustment of the voltage value, use the potentiometer for the gain of channel 1 and for selecting the measurement range, use the potentiometer for the DC source. For measurements with the voltmeters, use the Average-Mode "AV" and switch the measurement range according to the value being measured. The values obtained can inserted in the table by a simple 'drag-and-drop' procedure with the mouse. DC in /V 2 7 MSB 2 6 2 5 2 4 2 3 2 2 2 1 2 0 LSB DC out /V 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 0 1 1 0 1 0 0 1 1 1 1 0 1 0 0 1 0 1

Page 2 of 7 Table 1: Quantising according to A-Law Set the "Compression mode" selector on the PAM-/PCM modulator to "µ-law" (jumper inserted in positions 1 and 3) and the LED mode in the setting for channel 1. Proceed in a similar manner for the settings on the PAM-/ PCM demodulator. Now set the bit patterns (data words) as given in table 2 on the worksheet and measure the corresponding input and output voltages, analog to the previous exercise. 1 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 1 0 0 1 0 1 0 0 1 0 0 1 1 1 0 1 1 1 1 1 0 1 0 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 DC in /V 2 7 MSB 2 6 2 5 2 4 2 3 2 2 2 1 2 0 LSB DC out /V 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 1 1 1 0 0 1 0 1 1 1 1 0 0 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 0 1 1 0 1 1 0 0 1 1 1 0 1 1 0 0 1 0 1 0 0 0 0 0 0 1 0

Page 3 of 7 Table 2: Quantising according to µ-law Compare the two methods of companding, A-Law and µ-law. Comment on the quantising intervals for very small, average and large values. 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 Exercise 2 Examining the companding by direct comparison of input and output voltages Use the exercise assembly described above. Set both gain controls for channel 2 fully counter clockwise. Set the "Compression mode" selectors on both Experimenter cards to 8-Bit Linear (no jumpers inserted) and the LED-mode in the setting for channel 1. Measure the input voltage to channel 1 of the PAM-/PCM modulator card (SO4201-7T ) on channel A of the oscilloscope at the signal output "U1.1" and the output voltage of channel 1 at the signal output "Hold 1`on the PAM-/PCM demodulator (SO4201-7T) on channel B of the oscilloscope. Note: Adjust the timebases of channels A and B on the oscilloscope so that the 0V lines are on top of each other. Switch off the trigger since DC voltage signals are being measured. Use the most sensitive measurement range possible, commencing with 0V signal level! Now, compare the output and input voltages by sweeping slowly over the whole transmission area commencing at 0V, for the following cases: 8-Bit linear "Compression mode" selector: No jumpers inserted A-Law without inversion "Compression mode" selector: Jumper inserted in position 1 µ-law "Compression mode" selector: Jumper inserted in positions 1 and 3 Compare and comment on the results. What is the effect on the output voltage, of

Page 4 of 7 varying the input voltage within the limits of a quantising interval What effect has this quantising error, on the quality of the recovered output signal Exercise 3 Displaying the companding characteristics Use the exercise assembly described above. Set both gain controls for channel 2 fully counter clockwise. Set the "Compression mode" selectors on both Experimenter cards to 8-Bit Linear (no jumpers inserted) and the LED-mode in the setting for channel 1. Connect the output of channel 1 on the PAM-/PCM demodulator (SO4201-7T) to channel A of the UniTrain Interface and input "NF1" of channel 1 on the PAM-/PCM modulator (SO4201-7R) to the output of the function generator and to channel B of the UniTrain Interface. Start the Amplitude Response tester (right hand end of the test instrument icons) and load the "Compander" workspace via the "File" / "Load workspace..." menu. Note: The negative inputs A- and B- as well as the earth of the function generator must be connected using short cables! Now, record the transfer characteristics with the Amplitude Response tester, in one chart, for the following cases: 8-Bit linear - "Compression mode" selector SO4201-7R: No jumpers inserted - "Compression mode" selector SO4201-7T: No jumpers inserted A-Law without inversion - "Compression mode" selector SO4201-7R: Jumper inserted in position 1 - "Compression mode" selector SO4201-7T: No jumpers inserted µ-law - "Compression mode" selector SO4201-7R: Jumper inserted in positions 1 and 3 - "Compression mode" selector SO4201-7T: Jumper inserted in position 3 Label the curves recorded. For this, click the right hand mouse button on the characteristic and in the pop-up menu, select the "Add Labels" command which then opens a text box. Double-click on this text box and enter the appropriate name for the

Page 5 of 7 curve. Comment on the results obtained. Fig. 1: Transfer characteristic, linear, A-Law and µ-law Exercise 4 Distortion with a small input signal Use the exercise assembly described above. Set both gain controls for channel 2 fully counter clockwise. Set the "Compression mode" selectors on both Experimenter cards to 8-Bit Linear (no jumpers inserted) and the LED-mode in the setting for channel 1. Set the function generator for a sinewave signal with the following data:

Page 6 of 7 f M = 200 Hz (frequency range 100) Signal shape: Sinewave Level: 1:10, 5% (approx. 80mVpp) Measure the input voltage to channel 1 on the PAM-/PCM modulator card (SO4201-7R) on channel B of the oscilloscope connected to the input "LF1". Measure the output voltage of channel 1 on the PAM-/PCM demodulator card (SO4201-7T) at the "LF1" socket via channel A of the UniTrain Interface, with the oscilloscope. Synchronise the oscilloscope on channel B. Record the result in Fi. 2 and observe the LED indicator. In this state of the exercise assembly, change the setting of the "Compression mode" selectors on both Experiment cards at the same time, to the following positions: A-Law without inversion "Compression mode" selector SO4201-7R: Jumper inserted in position 1 µ-law "Compression mode" selector SO4201-7R: Jumper inserted in positions 1 and 3 What is the result Observe the LED indicator! Comment on the results, including reference to the previous parts of the exercise. What practical significance have logarithmic quantising characteristics, such as A- and µ-law X = 1 ms/div X/T (B) Chan. A= 50 mv/div AC Chan. B= 50 mv/div AC

Page 7 of 7 Fig. 2: Distortion of small signals in digital signal transmission, without companding

Page 1 of 3 MTI 7602 PCM Modulation and Demodulation Principle of Time Division Multiplex in PCM Transmission Lines Exercise 1 Examining the principle of time division multiplex, in the transmission of two DC voltages on a PCM transmission line On the PAM-/PCM modulator card (SO4201-7R), connect the inputs of channel 1 ("LF1") and channel 2 ("LF2") to the output of the internal DC source "DC -5V...+5V". Adjust the DC source voltage potentiometer for an output of +2.5V (check with a multimeter). Set the gain controls for both channels fully counter clockwise. Set the "Compression mode" selectors on both Experimenter cards to 8-Bit linear, A- Law (no jumper inserted) and the LED mode in the setting for channel 1. Now, adjust the gain of channel 2 for a voltage of approximately 3.5V at output "U2.1" (gain for channel 1 remains fully counter clockwise). Compare the bit patterns by switching the LED-mode back-and-forth, between channel 1 and channel 2. On channel A of the oscilloscope, measure the output voltage at the PCMout socket on the PAM-/ PCM modulator card (SO4201-7R). On channel B of the oscilloscope, measure the synchronous signal SYNC on the PAM-/PCM modulator. Trigger on channel B and set the trigger level to 50% of the amplitude of the synchronous signal. Compare the parallel codes with the serial data code on the PCM transmission path. Assign the channels to the oscillograms. Comment on the results. X = 20 µs/div X/T (B) Chan. A= 2 V/DIV DC Chan. B= 2 V/DIV DC

Page 2 of 3 Fig. 1: PCM time division multiplex signal of two DC voltages Exercise 2 Examining the principle of time division multiplex, in the transmission of one DC voltage and one sinusoidal alternating voltage, on a PCM transmission line Retain the settings for channel 1 from the previous exercise. Connect the input to channel 2 "LF2" to the test signal output, SIN 1000 Hz. Set the gain controls for both channels fully counter clockwise. Compare the bit patterns by switching the LED-mode back-and-forth, between channel 1 and channel 2. On channel A of the oscilloscope, measure the output voltage at the PCMout socket on the PAM-/ PCM modulator card (SO4201-7R). On channel B of the oscilloscope, measure the synchronous signal SYNC on the PAM-/PCM modulator. Trigger on channel B and set the trigger level to 50% of the amplitude of the synchronous signal. Then, measure the demodulated output voltages at the sockets LF1 and LF2 on the PAM-/PCM demodulator card (SO4201-7T). Vary the amplitude of the input voltage to both channels and observe the changes at the LF1out and LF2out sockets. Important:

Page 3 of 3 In multiplex operation of the PAM-/PCM demodulator card (SO4201-7T), the channel selector for the LED indicator must be set to the position for both channels! Compare the parallel codes with the serial data code on the PCM transmission path. Assign the channels to the oscillograms. Comment on the results. What practical significance can be derived from this exercise with time division multiplex X = 20 µs/div X/T (B) Chan. A= 2 V/DIV DC Chan. B= 2 V/DIV DC Fig. 2: PCM time division multiplex signal of a DC voltage and a sinewave signal Result: