Multiplexer | Demultiplexer | FDM | TDM | PAM | Applications

Multiplexing Principles

Multiplexer: Multiplexing is the process of simultaneously transmitting two or more individual signals over a single communication channel, cable, or wireless. In effect, it increases the number of communication channels so that more information can be transmitted. Often in communication, it is necessary or desirable to transmit more than one voice or data signal simultaneously. An application may require multiple signals, or cost savings can be gained by using a single communication channel to send multiple information signals. Four applications that would be prohibitively expensive or impossible without multiplexing are telephone systems, telemetry, satellites, and modern radio and TV broadcasting.

The greatest use of multiplexing is in the telephone system, where millions of calls are multiplexed on cables, long-distance fiber-optic lines, satellites, and wireless paths. Multiplexing increases the telephone carrier’s ability to handle more calls while minimizing system costs and spectrum usage.

In telemetry, the physical characteristics of a given application are monitored by sensitive transducers, which generate electric signals that vary in response to changes in the status of the various physical characteristics. The sensor-generated information can be sent to a central location for monitoring, or can be used as feedback in a closed-loop control system. Most spacecraft and many chemical plants, e.g., use telemetry systems to monitor characteristics, such as temperature, pressure, speed, light level, flow rate, and liquid level.

The use of a single communication channel for each characteristic being measured in a telemetry system would not be practical, because of both the multiple possibilities for signal degradation and the high cost. Consider, e.g., monitoring a space shuttle flight. Wire cables are obviously out of the question, and multiple transmitters impractical. If a deep-space probe were used, it would be necessary to use multiple transducers, and many transmitters would be required to send the signals back to earth. Because of cost, complexity, and equipment size and weight, this approach would not be feasible. Clearly, telemetry is an ideal application for multiplexing, with which the various information signals can all be sent over a single channel.

Finally, modern FM stereo broadcasting requires multiplexing techniques, as does the transmission of stereo sound and color in TV. Digital TV is multiplexed.

Multiplexing is accomplished by an electronic circuit known as a multiplexer. A simple multiplexer is illustrated in Fig. 10-1. Multiple input signals are combined by the multiplexer into a single composite signal that is transmitted over the communication medium. Alternatively, multiplexed signals can modulate a carrier before transmission. At the other end of the communication link, a demultiplexer is used to process the composite signal to recover the individual signals.

Multiplexer | Demultiplexer | FDM | TDM | PAM | Applications

The two most common types of multiplexing are frequency-division multiplexing (FDM) and time-division multiplexing (TDM). Two variations of these basic methods are frequency-division multiple access (FDMA) and time-division multiple access (TDMA). In general, FDM systems are used for analog information and TDM systems are used for digital information. Of course, TDM techniques are also found in many analog applications because the processes of A/D and D/A conversion are
so common. The primary difference between these techniques is that in FDM, individual signals to be transmitted are assigned a different frequency within a common bandwidth. In TDM, multiple signals are transmitted in different time slots on a single channel.

Another form of multiple access is known as code-division multiple access (CDMA). It is widely used in cell phone systems to allow many cell phone subscribers to use a common bandwidth at the same time. This system uses special codes assigned to each user that can be identified. CDMA uses a technique called spread spectrum to make this type of multiplexing possible. Spread spectrum is covered.

SPATIAL MULTIPLEXING

Spatial multiplexing is the term used to describe the transmission of multiple wireless signals on a common frequency in such a way that they do not interfere with one another. One way of doing this is to use low-power transmissions so that the signals do not interfere with one another. When very low power is used, the signals do not travel very far. The transmission distance is a function of the power level, frequency, and antenna height. For example, these factors may be used to ensure that the signals do not travel more than, say, 3 mi. Beyond 3 mi, these same frequencies may be used again to carry different signals.

Another technique is to use carefully controlled antenna radiation patterns to direct the signals to different locations in such a way that signals sharing the same frequency channel do not interfere with one another. Special antennas using multiple transmit and receive elements and phase-shifting circuits form the beams of radio energy in such a way as to minimize or in some cases completely eliminate interference from nearby signals on a common channel.

Spatial multiplexing is sometimes referred to as frequency reuse. This technique is widely used in satellite and cellular telephone systems.

Frequency-Division Multiplexing (FDM)

In frequency-division multiplexing (FDM), multiple signals share the bandwidth of a common communication channel. Remember that all channels have specific bandwidths, and some are relatively wide. A coaxial cable, e.g., has a bandwidth of about 1 GHz. The bandwidths of radio channels vary and are usually determined by FCC regulations and the type of radio service involved. Regardless of the type of channel, a wide bandwidth can be shared for the purpose of transmitting many signals at the same time.

Multiplexer | Demultiplexer | FDM | TDM | PAM | Applications

Transmitter-Multiplexers (FDM)

Fig. 10-2 shows a general block diagram of an FDM system. Each signal to be transmitted feeds a modulator circuit. The carrier for each modulator (fc) is on a different frequency. The carrier frequencies are usually equally spaced from one another over a specific frequency range. These carriers are referred to as subcarriers. Each input signal is given a portion of the bandwidth. The resulting spectrum is illustrated in Fig. 10-3. Any of the standard kinds of modulation can be used, including AM, SSB, FM, PM, or any of the various digital modulation methods. The FDM process divides up the bandwidth of the single-channel into smaller, equally spaced channels, each capable of carrying information in sidebands.

Multiplexer | Demultiplexer | FDM | TDM | PAM | Applications
Multiplexer | Demultiplexer | FDM | TDM | PAM | Applications

The modulator outputs containing the sideband information are added algebraically in a linear mixer; no modulation or generation of sidebands takes place. The resulting output signal is a composite of all the modulated subcarriers. This signal can be used to modulate a radio transmitter or can itself be transmitted over a single communication channel. Alternatively, the composite signal can become one input to another multiplexed system.

Receiver-Demultiplexers (FDM)

The receiving portion of an FDM system is shown in Fig. 10-4. A receiver picks up the signal and demodulates it, recovering the composite signal. This is sent to a group of bandpass filters, each centered on one of the carrier frequencies. Each filter passes only its channel and rejects all others. A channel demodulator then recovers each original input signal.

frequency-division multiplexing (FDM) Applications

Telemetry

As indicated earlier, sensors in telemetry systems generate electric signals that change in some way in response to changes in physical characteristics. An example of a sensor is a thermistor, a device used to measure temperature. A thermistor’s resistance varies inversely with temperature: As the temperature increases, the resistance decreases. The thermistor is usually connected into some kind of a resistive network, such as a voltage divider or bridge, and to a dc voltage source. The result is a dc output voltage, which varies in accordance with temperature and which is transmitted to a remote receiver for measurement, readout, and recording. The thermistor becomes one channel of an FDM system.

Other sensors have different kinds of outputs. Many have varying dc outputs, and others have ac output. Each of these signals is typically amplified, filtered, and otherwise conditioned before being used to modulate a carrier. All the carriers are then added to form a single multiplexed channel.

Multiplexer | Demultiplexer | FDM | TDM | PAM | Applications

The conditioned transducer outputs are normally used to frequency-modulate a subcarrier. The varying direct or alternating current changes the frequency of an oscillator operating at the carrier frequency. Such a circuit is generally referred to as a voltage controlled oscillator (VCO) or a subcarrier oscillator (SCO). To produce FDM, each VCO operates at a different center or carrier frequency. The outputs of the subcarrier oscillators are added. A diagram of such a system is shown in Fig. 10-5.

Most VCOs are astable multivibrators whose frequency is controlled by the input from the signal conditioning circuits. The frequency of the VCO changes linearly in proportion to the input voltage. Increasing the input voltage causes the VCO frequency to increase. The rectangular or triangular output of the VCO is usually filtered into a sine wave by a bandpass filter centered on the unmodulated VCO center frequency. This can be either a conventional LC filter or an active filter made with an op amp and RC input and feedback networks. The resulting sinusoidal output is applied to the linear mixer.

The linear mixing process in an FDM system can be accomplished with a simple resistor network. However, such networks greatly attenuate the signal, and some voltage amplification is usually required for practical systems. A way to achieve the mixing and amplification at the same time is to use an op-amp summer, such as that shown in Fig. 10-5. Recall that the gain of each input is a function of the ratio of the feedback resistor Rf to the input resistor value (R1, R2, etc.). The output is given by the expression Vout = -[V1(Rf /R1) + V2(Rf /R2) + V3(Rf /R3) + . . . + Vn(Rf /Rn)].

In most cases, the VCO FM output levels are the same, and all input resistors on the summer amplifier are therefore equal. If variations do exist, amplitude corrections can be accomplished by making the summer input resistors adjustable. The output of the summer amplifier does invert the signal; however, this has no effect on the content.

The composite output signal is then typically used to modulate a radio transmitter. Again, most telemetry systems use FM, although it is possible to use other kinds of modulation schemes. A system that uses FM of the VCO subcarriers, as well as FM of the final carrier, is usually called an FM/FM system.

Multiplexer | Demultiplexer | FDM | TDM | PAM | Applications

The receiving end of a telemetry system is shown in Fig. 10-6. A standard superheterodyne receiver tuned to the RF carrier frequency is used to pick up the signal. An FM demodulator reproduces the original composite multiplexed signal, which is then fed to a demultiplexer. The demultiplexer divides the signals and reproduces the original inputs.

The output of the first FM demodulator is fed simultaneously to multiple bandpass filters, each of which is tuned to the center frequency of one of the specified subchannels. Each filter passes only its subcarrier and related sidebands and rejects all the others. The demultiplexing process is, then, essentially one of using filters to sort the composite multiplex signal back into its original components. The output of each filter is the subcarrier oscillator frequency with its modulation.

These signals are then applied to FM demodulators. Also known as discriminators, these circuits take the FM signal and recreate the original dc or ac signal produced by the transducer. The original signals are then measured or processed to provide the desired information from the remote transmitting source. In most systems, the multiplexed signal is sent to a data recorder where it is stored for possible future use. The original telemetry output signals can be graphically displayed on a strip chart recorder or otherwise converted to usable outputs.

The demodulator circuits used in typical FM demultiplexers are of either the phase-locked loop (PLL) or the pulse-averaging type. PLL circuits have superior noise performance over the simpler pulse-averaging types. A PLL discriminator is also used to demodulate the receiver output.

FDM telemetry systems, which are inexpensive and highly reliable, are still widely used in aircraft and missile instrumentation and for monitoring of medical devices, such as pacemakers.

Cable TV

One of the best examples of FDM is cable TV, in which multiple TV signals, each in its own 6-MHz channel, are multiplexed on a common coaxial or fiber-optic cable and sent to nearby homes. TV signals include video and audio to modulate carriers using analog methods. Each channel uses a separate set of carrier frequencies, which can be added to produce FDM. The cable box acts as a tunable filter to select the desired channel.

Fig. 10-7 shows the spectrum on the cable. Each 6-MHz channel carries the video and voice of the TV signal. Coaxial and fiber-optic cables have an enormous bandwidth and can carry more than one hundred TV channels. Many cable TV companies also use their cable system for Internet access. A special modem (modulator-demodulator) permits computer data to be transmitted and received at very high data speeds. You will learn more about cable modems.

FM Stereo Broadcasting

In recording the original stereo, two microphones are used to generate two separate audio signals. The two microphones pick up sound from a common source, such as a voice or an orchestra, but from different directions. The separation of the two microphones provides sufficient differences in the two audio signals to provide a more realistic reproduction of the original sound. When the stereo is reproduced, the two signals can come from a cassette tape, a CD, or some other source. These two independent signals must somehow be transmitted by a single transmitter. This is done through FDM techniques.

Fig. 10-8 is a general block diagram of a stereo FM multiplex modulator. The two audio signals generally called the left (L) and right (R) signals, originate at the two microphones shown in the figure. These two signals are fed to a combining circuit, where they are used to form sum (L + R) and difference (L – R) signals. The L + R signal is a linear algebraic combination of the left and right channels. The composite signal it produces is the same as if a single microphone were used to pick up the sound. It is this signal that a monaural receiver will hear. The frequency response is 50 Hz to 15 kHz.

The combining circuit inverts the right channel signal, thereby subtracting it from the left channel to produce the L – R signal. These two signals, the L + R and L – R, are transmitted independently and recombined later in the receiver to produce the individual right- and left-hand channels.

The L – R signal is used to amplitude-modulate a 38-kHz carrier in a balanced modulator. The balanced modulator suppresses the carrier but generates upper and lower sidebands. The resulting spectrum of the composite modulating signal is shown in Fig. 10-9. As shown, the frequency range of the L + R signal is from 50 Hz to 15 kHz. Since the frequency response of an FM signal is 50 to 15 kHz, the sidebands of the L – R signal are in the frequency range of 38 kHz ± 15 kHz or 23 to 53 kHz. This DSB suppressed carrier signal is algebraically added to, and transmitted along with, the standard L + R audio signal.

Also transmitted with the L + R and L – R signals is a 19-kHz pilot carrier, which is generated by an oscillator whose output also modulates the main transmitter. Note that the 19-kHz oscillator drives a frequency doubler to generate the 38-kHz carrier for the balanced modulator.

Some FM stations also broadcast one or more additional signals, referred to as subsidiary communications authorization (SCA) signals. The basic SCA signal is a separate subcarrier of 67 kHz, which is frequency-modulated by audio signals, usually music. SCA signals are also used to transmit weather, sports, and financial information

Multiplexer | Demultiplexer | FDM | TDM | PAM | Applications


Multiplexer | Demultiplexer | FDM | TDM | PAM | Applications

Special SCA FM receivers can pick up these signals. The SCA portion of the system is generally used for broadcasting background music for elevators, stores, offices, and restaurants. If an SCA system is being used, the 67-kHz subcarrier with its music modulation will also be added to the L – R and L+ R signals to modulate the FM transmitter. Not all stations transmit SCA, but some transmit several channels, using additional, higher-frequency subcarriers.

Another alternative service provided by some FM stations is called the Radio Data System (RDS). It is widely used in car radios and some home stereo receivers. It allows digital data to be transmitted to the FM receiver. Some examples of the types of data transmitted include station call letters and location, travel and weather data, and short news announcements. A popular use of RDS is to transmit the name and artist of music selections being played by the station. The transmitted data is displayed on a liquid crystal display (LCD) in the receiver.

The data to be transmitted is used to modulate another subcarrier at 57 kHz. This is the third harmonic of the 19-kHz pilot carrier and so helps prevent interaction with the stereo signals. A form of phase modulation called quadrature phase-shift keying (QPSK) is used to modulate the subcarrier. The serial data rate is 1187.5 bits per second (bps). As in other FDM systems, all the subcarriers are added with a linear mixer to form a single signal (Fig. 10-9). This signal is used to frequency-modulate the carrier of the broadcast transmitter. Again, note that FDM simply takes a portion of the frequency spectrum. There is sufficient spacing between adjacent FM stations so that the additional information can be accommodated. Keep in mind that each additional subcarrier reduces the amount by which the main L + R signal can modulate the carrier since the maximum total modulation voltage is determined by the legal channel width.

At the receiving end, the demodulation is accomplished with a circuit similar to that illustrated in Fig. 10-10. The FM superheterodyne receiver picks up the signal, amplifies it, and translates it to an intermediate frequency, usually 10.7 MHz. It is then demodulated. The output of the demodulator is the original multiplexed signal. The additional circuits now sort out the various signals and reproduce them in their original form.

The original audio L + R signal is extracted simply bypassing the multiplex signal through a low-pass filter. Only the 50- to 15-kHz original audio is passed. This signal is fully compatible with monaural FM receivers without stereo capability.

Multiplexer | Demultiplexer | FDM | TDM | PAM | Applications

In a stereo system, the L + R audio signal is fed to a linear matrix or combiner where it is mixed
with the L – R signal to create the two separate L and R channels.

The multiplexed signal is also applied to a bandpass filter that passes the 38-kHz suppressed subcarrier with its sidebands. This is the L – R signal that modulates the 38-kHz carrier. This signal is fed to a balanced modulator for demodulation. The 19-kHz pilot carrier is extracted by passing the multiplexed signal through a narrow bandpass filter. This 19-kHz subcarrier is then fed to an amplifier and frequency doubler circuit, which produces a 38-kHz carrier signal that is fed to the balanced modulator. The output of the balanced modulator, of course, is the L – R audio signal. This is fed to the linear resistive combiner along with the L + R signal.

The linear combiner both adds and subtracts these two signals. Addition produces the left-hand channel: (L + R) + (L – R) = 2L. Subtraction produces the right-hand channel: (L + R) – (L – R) = 2R. The left- and right-hand audio signals are then sent to separate audio amplifi ers and ultimately the speaker.

If an SCA signal is being used, a separate bandpass filter centered on the 67-kHz subcarrier will extract the signal and feed it to an FM demodulator. The demodulator output is then sent to a separate audio amplifier and speaker.

If the RDS signal is used, a 57-kHz bandpass filter selects this signal and sends it to a QPSK demodulation. The recovered digital data is then displayed on the receiver’s LCD. Typically, the recovered data is sent to the receiver’s embedded control microprocessor that also controls the LCD display where the data is conditioned prior to display. All the circuitry used in the demultiplexing process is usually contained in a single IC. In fact, most FM receivers contain a single chip that includes the IF, demodulator, and demultiplexer. Note that the multiplexing and demultiplexing of FM stereo in a TV set are exactly as described above, but with a different IF.

Example 10.1
A cable TV service uses a single coaxial cable with a bandwidth of 860 MHz to transmit multiple TV signals to subscribers. Each TV signal is 6 MHz wide. How many channels can be carried?

Total channels = 860/6 = 143.33 or 143

Time-Division Multiplexing ( TDM )

In FDM, multiple signals are transmitted over a single channel, each signal being allocated a portion of the spectrum within that bandwidth. In time-division multiplexing (TDM), each signal occupies the entire bandwidth of the channel. However, each signal is transmitted for only a brief time. In other words, multiple signals take turns transmitting over the single-channel, as diagrammed in Fig. 10-11. Here, each of the four signals being transmitted over a single channel is allowed to use the channel for a fixed time, one after another. Once all four have been transmitted, the cycle repeats. One binary word from each source creates a frame. The frames are then repeated over and over again.

TDM can be used with both digital and analog signals. For example, if the data consists of sequential bytes, 1 byte of data from each source can be transmitted during the time interval assigned to a particular channel. Each of the time slots shown in Fig. 10-11 might contain a byte from each of four sources. One channel would transmit 8 bits and then halt, while the next channel transmitted 8 bits.

The third channel would then transmit its data word, and so on. The cycle would repeat itself at a high rate of speed. By using this technique, the data bytes of individual channels can be interleaved. The resulting single-channel signal is a digital bit stream that is deciphered and reassembled at the receiving end.

The transmission of digital data by TDM is straightforward in that the incremental digital data is already broken up into chunks, which can easily be assigned to different time slots. TDM can also be used to transmit continuous analog signals, whether they are voice, video, or telemetry-derived. This is accomplished by sampling the analog signal repeatedly at a high rate and then converting the samples to proportional binary numbers and transmitting them serially.

Sampling an analog signal creates pulse-amplitude modulation (PAM). As shown in Fig. 10-12, the analog signal is converted to a series of constant-width pulses whose amplitude follows the shape of the analog signal. The original analog signal is recovered by passing it through a low-pass filter. In TDM using PAM, a circuit called a multiplexer (MUX or MPX) samples multiple analog signal sources; the resulting pulses are interleaved and then transmitted over a single channel.

PAM (Pulse Amplitude Modulate) Multiplexers

The simplest time multiplexer operates as a single-pole multiple-position mechanical or electronic switch that sequentially samples the multiple analog inputs at a high rate of speed. A basic mechanical rotary switch is shown in Fig. 10-13. The switch arm is rotated by a motor and dwells momentarily on each contact, allowing the input signal to be passed through to the output. It then switches quickly to the next channel, allowing that channel to pass for a fixed duration. The remaining channels are sampled in the same way. After each signal has been sampled, the cycle repeats. The result is that four analog signals are sampled, creating pulse-amplitude-modulated signals that are interleaved with one another. The speed of sampling is directly related to the speed of rotation, and the dwell time of the switch arm on each contact depends on the speed of rotation and the duration of contact.

Fig. 10-14 illustrates how four different analog signals are sampled by this technique. Signals A and C are continuously varying analog signals, signal B is a positive-going linear ramp, and signal D is a constant dc voltage.

Commutator Switches

Multiplexers in early TDM/PAM telemetry systems used a form of rotary switch known as a commutator. Multiple switch segments were attached to the various incoming signals while a high-speed brush rotated by a dc motor rapidly sampled the signals as it passed over the contacts. (Commutators have now been totally replaced by electronic circuits, which are discussed in the next section.)

In practice, the duration of the sample pulses is shorter than the time that is allocated to each channel. For example, assume that it takes the commutator or multiplexed switch 1 ms to move from one contact to another. The contacts can be set up so that each sample is 1 ms long. Typically, the duration of the sample is set to be about one-half the channel period value, in this example 0.5 ms.

One complete revolution of the commutator switch is referred to as a frame. In other words, during one frame, each input channel is sampled one time. The number of contacts on the multiplexed switch or commutator sets the number of samples per frame. The number of frames completed in 1 s is called the frame rate. Multiplying the number of samples per frame by the frame rate yields the commutation rate or multiplex rate, which is the basic frequency of the composite signal, the final multiplexed signal that is transmitted over the communication channel.

In Fig. 10-14, the number of samples per frame is 4. Assume that the frame rate is 100 frames per second. The period for one frame, therefore, is 1/100 = 0.01 s = 10 ms. During that 10-ms frame period, each of the four channels is sampled once. Assuming equal sample durations, each channel is thus allotted 10/4 = 2.5 ms. (As indicated earlier, the full 2.5-ms period would not be used. The sample duration during that interval might be, e.g., only 1 ms). Since there are four samples taken per frame, the commutation rate is 4 x 100 or 400 pulses per second.

Electronic Multiplexers

In practical TDM/PAM systems, electronic circuits are used instead of mechanical switches or commutators. The multiplexer itself is usually implemented with FETs, which are nearly ideal on/off switches and can turn off and on at very high speeds. A complete four-channel TDM/PAM circuit is illustrated in Fig. 10-15.

The multiplexer is an op-amp summer circuit with MOSFETs on each input resistor. When the MOSFET is conducting, it has a very low on-resistance and therefore acts as a closed switch. When the transistor is off, no current flows through it, and it, therefore, acts as an open switch. A digital pulse applied to the gate of the MOSFET turns the transistor on. The absence of a pulse means that the transistor is off. The control pulses to the MOSFET switches are such that only one MOSFET is turned on at a time. These MOSFETs are turned on in sequence by the digital circuitry illustrated.

All the MOSFET switches are connected in series with resistors (R1–R4); this, in combination with the feedback resistor (Rf) on the op amp circuit, determines the gain. For the purposes of this example, assume that the input and feedback resistors are all equal in value; in other words, the op-amp circuit has a gain of 1. Since this op-amp summing circuit inverts the polarity of the analog signals, it is followed by another op-amp inverter which again inverts, restoring the proper polarity.

All the circuitry shown in Fig. 10-15 is usually contained on a single IC chip. MOSFET multiplexers are available with 4, 8, and 16 inputs, and these may be grouped to handle an even larger number of analog inputs.

The digital control pulses are developed by the counter and decoder circuit shown in Fig. 10-15. Since there are four channels, four counter states are needed. Such a counter can be implemented with two flip-flops, representing four discrete states—00, 01, 10, and 11—which are the binary equivalents of the decimal numbers 0, 1, 2, and 3. The four channels can therefore be labeled 0, 1, 2, and 3.

A clock oscillator circuit triggers the two flip-flop counters. The clock and flip-flop waveforms are illustrated in Fig. 10-16. The flip-flop outputs are applied to decoder AND gates that are connected to recognize the four binary combinations 00, 01, 10, and 11. The output of each decoder gate is applied to one of the multiplexed FET gates. The one-shot multivibrator diagrammed in Fig. 10-15 is used to trigger all the decoder AND gates at the clock frequency. It produces an output pulse whose duration has been set to the desired sampling interval, in this case, 1 ms.

Each time the clock pulse occurs, the one-shot generates its pulse, which is applied simultaneously to all four AND decoder gates. At any given time, only one of the gates is enabled. The output of the enabled gate is a pulse whose duration is the same as that of the one-shot.

When the pulse occurs, it turns on the associated MOSFET and allows the analog signal to be sampled and passed through the op-amps to the output. The output of the final op-amp is a multiplexed PAM signal like that in Fig. 10-14. The PAM output is used to modulate a carrier for transmission to a receiver. FM and PM are common modulation methods.

Demultiplexer Circuits

Once the composite PAM signal is recovered at the receiver, it is applied to a demultiplexer (DEMUX). The demultiplexer is, of course, the reverse of a multiplexer. It has a single input and multiple outputs, one for each original input signal. Typical DEMUX circuitry is shown in Fig. 10-17. A four-channel demultiplexer has a single input and four outputs.

Most demultiplexers use FETs driven by a counter-decoder. The individual PAM signals are sent to op-amps, where they are buffered and possibly amplified. They are then sent to low-pass filters, where they are smoothed into the original analog signals.

The main problem encountered in demultiplexing is synchronization. That is, for the PAM signal to be accurately demultiplexed into the original sampled signals, the clock frequency used at the receiver demultiplexer must be identical to that used at the transmitting multiplexer. In addition, the sequence of the demultiplexer must be identical to that of the multiplexer so that when channel 1 is being sampled at the transmitter, channel 1 is turned on in the receiver demultiplexer at the same time. Such synchronization is usually carried out by a special synchronizing pulse included as a part of each frame. Some of the circuits used for clock frequency and frame synchronization are discussed in the following sections.

Clock Recovery

Instead of using a free-running clock oscillator set to the identical frequency of the transmitter system clock, the clock for the demultiplexer is derived from the received PAM signal itself. The circuits shown in Fig. 10-18, called clock recovery circuits, are typical of those used to generate the demultiplexer clock pulses.

In Fig. 10-18(a), the PAM signal has been applied to an amplifier/limiter circuit, which first amplifies all received pulses to a high level and then clips them off at a fixed level. The output of the limiter is thus a constant-amplitude rectangular wave whose output frequency is equal to the commutation rate. This is the frequency at which the PAM pulses occur and is determined by the transmitting multiplexer clock.

The rectangular pulses at the output of the limiter are applied to a bandpass filter, which eliminates the upper harmonics, creating a sine wave signal at the transmitting clock frequency. This signal is applied to the phase detector circuit in a PLL along with the input from a voltage-controlled oscillator (VCO). The VCO is set to operate at the frequency of the PAM pulses. However, the VCO frequency is controlled by a dc error voltage applied to its input. This input is derived from the phase detector output, which is filtered by a low-pass filter into a dc voltage.

The phase detector compares the phase of the incoming PAM sine wave to the VCO sine wave. If a phase error exists, the phase detector produces an output voltage, which is filtered to provide a varying direct current. The system is stabilized or locked when the VCO output frequency is identical to that of the sine wave frequency derived from the PAM input. The difference is that the two are shifted in phase by 90°.

If the PAM signal frequency changes for some reason, the phase detector picks up the variation and generates an error signal that is used to change the frequency of the VCO to match. Because of the closed-loop feature of the system, the VCO automatically tracks even minute frequency changes in the PAM signal, ensuring that the clock frequency used in the demultiplexer will always perfectly match that of the original PAM signal.

The output signal of the VCO is applied to a one-shot pulse generator that creates rectangular pulses at the proper frequency. These are used to step the counter in the demultiplexer; the counter generates the gating pulses for the FET demultiplexer switches.

A simpler, open-loop clock pulse circuit is shown in Fig. 10-18(b). Again, the PAM signal is applied to an amplifier/limiter and then a bandpass filter. The sine wave output of the bandpass filter is amplified and applied to a phase-shift circuit, which produces a 90° phase shift at the frequency of operation. This phase-shifted sine wave is then applied to a pulse generator, which, in turn, creates the clock pulses for the demultiplexer. One disadvantage of this technique is that the phase-shift circuit is fixed to create a 90° shift at only one frequency, and so minor shifts in input frequency produce clock pulses whose timing is not perfectly accurate. However, in most systems in which frequency variations are not great, the circuit operates reliably.

Frame Synchronization

After clock pulses of the proper frequency have been obtained, it is necessary to synchronize the multiplexed channels. This is usually done with a special synchronizing (sync) pulse applied to one of the input channels at the transmitter. In the four-channel system discussed previously, only three actual signals are transmitted. The fourth channel is used to transmit a special pulse whose characteristics are unique in some way so that it can be easily recognized. The amplitude of the pulse can be higher than the highest-amplitude data pulse, or the width of the pulse can be wider than those pulses derived by sampling the input signals. Special circuits are then used to detect the sync pulse.

Fig. 10-19 shows an example of a sync pulse that is higher in amplitude than the maximum pulse value of any data signal. The sync pulse is also the last to occur in the frame. At the receiver, a comparator circuit is used to detect the sync pulse. One input to the comparator is set to a dc reference voltage that is slightly higher than the maximum amplitude possible for the data pulses. When a pulse greater than the reference amplitude occurs, i.e., the sync pulse, the comparator immediately generates an output pulse, which can then be used for synchronization. Alternatively, it is possible not to transmit a pulse during one channel interval, leaving a blank space in each frame that can then be detected for the purposes of synchronization.

When the sync pulse is detected at the receiver, it acts as a reset pulse for the counter in the demultiplexer circuit. At the end of each frame, the counter is reset to zero (channel 0 is selected). When the next PAM pulse occurs, the demultiplexer will be set to the proper channel. Clock pulses then step the counter in the proper sequence for demultiplexing.

Finally, at the output of the demultiplexer, separate low-pass filters are applied t each channel to recover the original analog signals. Fig. 10-20 shows the complete PAM demultiplexer. Keep in mind that such circuitry today is deep within an integrated circuit and cannot be accessed. Knowing the concept of internal operations is helpful, but generally, you will only be concerned with the input and output signals.

Digital Codes | Hartley’s Law | ASCII | Asynchronous | Encoding Methods ( Multiplexer | frequency division multiplexer (FDM) | Time division multiplexer ( TDM ) | Pulse Amplitude multiplexer( PAM ) | De multiplexer | Commutator | Applications )

FSK | PSK | DPSK | QPSK | QAM | Spectral Efficiency | Modem Concepts ( Multiplexer | frequency division multiplexer (FDM) | Time division multiplexer ( TDM ) | Pulse Amplitude multiplexer( PAM ) | De multiplexer | Commutator | Applications )

PCM ( Pulse Code Modulation ) | T-Carrier Systems| Duplexing ( Multiplexer | frequency division multiplexer (FDM) | Time division multiplexer ( TDM ) | Pulse Amplitude multiplexer( PAM ) | De multiplexer | Commutator | Applications )

Receiver and Transceiver | AM | FM | SW Radio | SDR | Wi-Fi ( Multiplexer | frequency division multiplexer (FDM) | Time division multiplexer ( TDM ) | Pulse Amplitude multiplexer( PAM ) | De multiplexer | Commutator | Applications )

IF Amplifiers | RF Input Amplifiers | Squelch Circuits | Controlling Gain ( Multiplexer | frequency division multiplexer (FDM) | Time division multiplexer ( TDM ) | Pulse Amplitude multiplexer( PAM ) | De multiplexer | Commutator | Applications )

Digital to Analog Converters | Analog to Digital Converters ( Multiplexer | frequency division multiplexer (FDM) | Time division multiplexer ( TDM ) | Pulse Amplitude multiplexer( PAM ) | De multiplexer | Commutator | Applications )

CLICK HERE TO LEARN MORE ( Multiplexer | frequency division multiplexer (FDM) | Time division multiplexer ( TDM ) | Pulse Amplitude multiplexer( PAM ) | De multiplexer | Commutator | Applications )

CLICK HERE TO LEARN ( Multiplexer | frequency division multiplexer (FDM) | Time division multiplexer ( TDM ) | Pulse Amplitude multiplexer( PAM ) | De multiplexer | Commutator | Applications )

Reference : Electronic communication by Louis Frenzel

Leave a Reply

%d bloggers like this: