Communication System Foundations

Introduction

Communication is the transfer of information from a source to a destination through a physical or logical medium. A communication system is therefore more than a transmitter and receiver: it is an engineered arrangement that represents information, conveys it through an imperfect channel, reconstructs it, and evaluates whether the received result is useful.

Examples range from a speech telephone call and a digital television link to satellite navigation, optical fibre networks, cellular systems, industrial sensors, and spacecraft telemetry. Although their technologies differ, they share a common architecture and a common vocabulary.

Learning Objectives

After studying this chapter, a student should be able to:

1. Describe the functional blocks of a communication system.
2. Distinguish information, message, signal, noise, interference, bandwidth, and data rate.
3. Explain the roles of source, transmitter, channel, receiver, destination, and feedback path.
4. Compare analog and digital communication systems.
5. Use decibel notation and a basic link budget correctly.
6. Interpret performance measures such as signal-to-noise ratio, bit error rate, latency, reliability, and capacity.
7. Explain the role of protocols and standards.
8. Relate major historical developments to present-day communication practice.
9. Analyse a simple communication-link design problem.

1. The General Communication Model

A source produces information. The information may be speech, text, an image, a video stream, a sensor reading, or a control command. An input transducer converts a physical phenomenon into an electrical or digital representation when necessary. For example, a microphone converts acoustic pressure into voltage.

The transmitter prepares the message for transmission. It may perform source coding, encryption, channel coding, modulation, frequency conversion, filtering, amplification, and radiation through an antenna or launch into an optical fibre.

The channel is the path between transmitter and receiver. It can be free space, copper cable, optical fibre, or a combination of networks. A channel introduces attenuation, delay, distortion, noise, interference, fading, and sometimes packet loss.

The receiver performs operations approximately inverse to those of the transmitter. It filters and amplifies the received waveform, converts frequency, demodulates or detects the signal, decodes errors, decrypts data, and reconstructs the message.

The destination is the person, machine, application, or control system for which the information is intended. A loudspeaker, display, database, robot actuator, or software process may be the destination interface.

Protocols specify how communicating entities format, exchange, acknowledge, sequence, protect, and interpret information. They are essential when a system contains several devices or network layers.

Figure: Functional architecture of a communication system. Feedback originates at the receiver or destination and travels toward the transmitter.

A system may be simplex, allowing transmission in one direction only; half-duplex, allowing both directions but not simultaneously; or full-duplex, allowing simultaneous two-way communication. A broadcast system has one transmitter and many intended receivers, whereas a point-to-point system connects specified endpoints.

2. Information, Messages, and Signals

Information is the reduction of uncertainty produced by an observation or message. A message is the particular content selected by a source. A signal is the physical representation used to carry that message. The same information can be represented by speech pressure, an analog voltage, a sequence of bits, radio-frequency modulation, or optical intensity.

A deterministic continuous-time signal is often written as x(t). A sampled digital representation is a sequence x[n], where n is an integer sample index. A signal may be:

• continuous-time or discrete-time;
• analog-valued or digital-valued;
• baseband, occupying frequencies near zero, or passband, shifted around a carrier frequency;
• periodic or aperiodic;
• deterministic or random.

The frequency spectrum describes how signal energy or power is distributed with frequency. Bandwidth is the extent of the frequency range significantly occupied by a signal or allocated to a channel. Its exact definition depends on context, such as null-to-null bandwidth, 3-dB bandwidth, or regulatory occupied bandwidth.

Information sources may be classified as speech and audio, still images, video, text and data, machine measurements, or control information. Sources can also be analogue or digital, human-generated or machine-generated, and stationary or time-varying. Source statistics influence compression and error-control design. A highly redundant source can often be compressed more than an already compressed source.

For a discrete source with symbols having probabilities p_i, the average information entropy is:

Here H(X) is measured in bits per source symbol, the sum covers all possible symbols, and ∑p_i = 1. This is a source-model quantity, not automatically the achieved transmission rate. It assumes a specified probability model; correlations between successive symbols require a more complete source model.

3. Transmitter Functions

A practical transmitter commonly includes the following functions:

1. Source encoding: removes redundancy to reduce bit rate or represents an analogue source digitally. Pulse-code modulation, for example, samples, quantizes, and encodes an analogue waveform.
2. Encryption: protects confidentiality and sometimes authenticity. Encryption does not itself correct channel errors.
3. Channel encoding: adds structured redundancy so the receiver can detect or correct errors.
4. Framing and multiplexing: groups data and combines multiple streams where required.
5. Modulation: maps information onto a waveform suitable for the channel. Examples include amplitude shift keying, frequency shift keying, phase shift keying, quadrature amplitude modulation, and orthogonal frequency-division multiplexing.
6. Filtering, frequency conversion, and amplification: shapes the spectrum, moves the signal to its allocated frequency, and supplies transmission power.

In analogue amplitude modulation, a carrier amplitude varies with the message. In digital modulation, discrete symbols select waveform states, although the physical waveform remains continuous in time. This distinction is important: “digital” normally refers to the information representation and decision process, not to an ideal waveform made of discontinuous physical jumps.

4. Channels and Channel Impairments

A channel model expresses how the transmitted signal becomes the received signal. A simple additive-noise model is:

where s(t) is the transmitted waveform, n(t) is additive noise, and r(t) is the received waveform. This model is useful but incomplete when attenuation, delay, multipath, or nonlinear distortion is significant. A more general linear time-varying model may be represented conceptually by a channel response applied to s(t) plus noise.

Important impairments include:

• Attenuation: reduction in signal power or amplitude with distance or components.
• Noise: unwanted random energy, including thermal noise and receiver noise.
• Interference: unwanted energy from another transmitter or system.
• Distortion: alteration of waveform shape due to non-flat amplitude or phase response, nonlinearity, or bandwidth limitation.
• Fading: time-varying received level, often caused by multipath propagation.
• Delay and jitter: time displacement and variation in timing.
• Dispersion: frequency-dependent propagation delay, especially important in optical fibre.
• Packet loss: disappearance or rejection of a complete data unit in a network.

Thermal noise power in bandwidth B is commonly approximated by:

where k = 1.380649 × 10⁻²³ J/K is Boltzmann’s constant, T is absolute temperature in kelvin, and B is bandwidth in hertz. The formula assumes approximately white thermal noise over the relevant bandwidth and a matched resistive system. Receiver noise figure may be included through an equivalent noise temperature or an adjusted noise factor.

5. Receiver and Destination

The receiver selects the desired channel, amplifies it with minimal additional noise, converts frequency when needed, demodulates or detects symbols, synchronizes timing and carrier phase, and decodes the result. In a digital receiver, a decision rule maps noisy observations to likely transmitted symbols. Channel decoding then uses redundancy to identify or correct errors.

The destination interprets the reconstructed information. Quality must be judged at the destination, not merely at an intermediate detector. For voice, intelligibility and delay matter; for a file, bit-perfect delivery matters; for a control loop, latency, bounded jitter, and safety matter.

Feedback may report acknowledgements, retransmission requests, channel quality, synchronization status, or power-control commands. A feedback path can improve reliability, but it consumes capacity and may introduce delay or instability. Systems without a return path must use forward error correction, repetition, diversity, or robust operating margins instead.

6. Protocols and Standards

A protocol is a set of agreed rules for communication. Typical protocol responsibilities include syntax, semantics, timing, addressing, flow control, error control, connection management, and security. Layering separates responsibilities. In a layered model, a lower layer transports symbols or frames while a higher layer interprets packets or application messages.

For example, an application may generate a request, a transport protocol may provide sequencing and reliability, a network protocol may select a route, a data-link protocol may frame and check errors, and a physical layer may modulate bits onto a medium. Encapsulation adds control information at each layer; decapsulation removes it at the receiver.

Standards promote interoperability, spectrum coordination, safety, and economies of scale. Relevant organizations include the International Telecommunication Union, IEEE, ISO, and national regulators. Conformance testing is distinct from merely claiming compatibility.

7. Performance Measures

Signal-to-noise ratio

Signal-to-noise ratio is the ratio of signal power to noise power:

In decibels:

For voltage measurements, the 20-log form is valid only when the compared voltages are measured across equal impedances:

Bit error rate and symbol error rate

Bit error rate is the number of incorrectly received bits divided by the number of received bits examined. It is an empirical estimate and must specify test conditions, modulation, coding, channel, and sample size. Symbol error rate counts incorrect detected symbols instead; one symbol may carry several bits.

Data rate, throughput, and capacity

Gross data rate includes protocol and error-control overhead; throughput is useful delivered information per unit time. They are not interchangeable. Latency is elapsed time from source submission to destination availability. Reliability may be expressed as successful-delivery probability, outage probability, or availability, depending on the application.

For an ideal band-limited additive white Gaussian noise channel, Shannon capacity is:

where C is the theoretical maximum error-free information rate in bits per second, B is bandwidth in hertz, and S/N is received signal-to-noise power ratio. The result assumes the stated idealized channel model, arbitrarily long codes, and acceptable decoding complexity and delay. It is a limit, not a guaranteed operating rate.

Link budget

A link budget accounts for transmitted power, gains, and losses. When absolute powers are referenced to 1 mW, they are expressed in dBm; gains and losses are expressed in dB:

Here P_T and P_R are transmitter output and receiver input powers; G_T and G_R are antenna or directional gains; L_F is propagation or feeder loss; and L_O represents other losses. All terms must refer to compatible points in the system. dBW may be used instead, but all absolute-power terms must then use dBW. A positive loss is subtracted.

For free-space propagation, under ideal line-of-sight assumptions, the path loss is:

where d is separation, λ is wavelength, and both use the same length unit. This formula does not include atmospheric absorption, obstacles, multipath, antenna mismatch, or polarization loss.

Worked example: decibels and a link budget

A transmitter supplies 1 W to a link. The receiver obtains 0.1 W after all effects. The power ratio is 0.1, so the change in power level is:

10 log₁₀(0.1) = −10 dB.

Thus the link has a 10 dB loss, conventionally reported as a positive loss of +10 dB. The negative signed change and positive loss magnitude describe the same physical reduction but must not be confused.

Consider a second link with P_T = 30 dBm, transmitter gain 2 dB, receiver gain 5 dB, feeder loss 1 dB, and propagation loss 100 dB. Then:

P_R = 30 + 2 + 5 − 1 − 100 = −64 dBm.

The result is a received power level referenced to 1 mW. It is not “−64 dB,” because absolute power requires a reference unit.

8. Analog and Digital Evolution

Early communication systems used physical messengers, optical signals, flags, and wired telegraphy. Electrical telegraphy introduced coded symbols over long distances. Telephony enabled analogue speech transmission. Radio broadcasting and amplitude, frequency, and phase modulation extended communication without wires.

Digital techniques developed through pulse-code modulation, digital switching, computers, satellites, fibre optics, integrated circuits, and error-correcting codes. Digital systems offer regeneration, flexible multiplexing, encryption, compression, and strong error control. They are not automatically superior: analogue systems can have low complexity and graceful degradation, while digital systems require synchronization, sampling, quantization, processing, and adequate signal quality for reliable decisions.

A simplified digital conversion chain samples an analogue signal, quantizes each sample, and encodes the quantization index. If a signal is strictly band-limited to a highest frequency f_max, ideal uniform sampling requires:

This Nyquist condition applies to the sampling theorem under its band-limiting and ideal-reconstruction assumptions. It does not by itself specify quantization resolution, anti-alias filtering, or the required signal-to-noise ratio.

Contemporary systems combine analogue front ends with digital signal processing. Software-defined radios, coherent optical systems, massive antenna arrays, adaptive coding and modulation, edge computing, and machine-learning-assisted signal classification are important developments. AI can help estimate channels or detect anomalies, but training data, robustness, explainability, privacy, and adversarial risks must be assessed before deployment.

9. Applied Case: Rural Emergency-Monitoring Link

Background. A health agency needs to collect temperature, oxygen-level, and alarm data from remote clinics. Cellular coverage is intermittent, and the system must protect patient information.

Problem. The link must operate with low power, tolerate fading, deliver urgent alarms promptly, and distinguish a missing packet from a normal sensor reading.

Analysis. Sensor values are digitized and placed in addressed frames with timestamps and a cyclic-redundancy check. Source compression is modest because the data rate is already low. Encryption protects confidentiality. Forward error correction and interleaving address burst errors. An acknowledgement path allows retransmission of critical alarms, while routine data may be buffered and sent later. The radio selects a robust low-order modulation when channel quality falls and a higher-rate mode when margin is available.

Approach and solution. The receiver sends acknowledgements and channel-quality information toward the transmitter. The gateway logs sequence numbers, checks integrity, and forwards accepted measurements over a secure network. Engineers use a link budget including antenna gains, cable losses, fading margin, and receiver sensitivity. They test not only average throughput but also alarm latency, outage probability, battery life, and false alarm rate.

Outcome. The design becomes an end-to-end service rather than merely a radio circuit. A successful packet with a corrupted application value is prevented by integrity checks; a lost routine packet is recovered from buffering; and an urgent alarm is retransmitted within a specified deadline.

Learning points. Requirements determine architecture. Reliability, security, energy, delay, and capacity interact. Feedback is useful only when its overhead and delay are acceptable, and performance must be measured at the destination.

10. Design, Ethical, and Regulatory Considerations

Engineers should document assumptions, reference planes, units, margins, test conditions, and failure modes. Spectrum use requires authorization and coordination. Equipment may need electromagnetic-compatibility, safety, and radio-type approval. Accessibility matters: interfaces should accommodate users with hearing, visual, motor, or cognitive limitations.

Communication systems can expose location, identity, health, and behavioural data. Privacy-by-design, data minimization, authentication, secure updates, and retention controls are engineering requirements, not optional additions. Emergency systems require clear priorities, redundancy, graceful degradation, and human oversight. A technically efficient system that excludes users or creates unsafe failure modes is not a successful communication system.

Chapter Summary

A communication system transfers information through a coordinated source, transmitter, channel, receiver, and destination architecture. Signals represent messages physically; transmitters encode, protect, modulate, and launch them; channels introduce noise, loss, distortion, delay, and fading; receivers recover information; and destinations evaluate its usefulness. Protocols establish interoperable rules across layers.

Analog and digital systems differ mainly in representation and processing, with modern systems commonly combining analogue interfaces and digital signal processing. Important performance measures include bandwidth, data rate, throughput, latency, SNR, BER, reliability, and capacity. Decibel calculations require careful distinction between absolute powers in dBm or dBW and gains or losses in dB. Link budgets, sampling theory, noise models, and Shannon capacity are useful only under their stated assumptions. Practical design also requires standards, security, privacy, regulation, accessibility, and end-to-end testing.

Key Terms

Information: Reduction of uncertainty conveyed by an observation or message.

Message: Particular content selected for communication.

Signal: Physical representation of information.

Source: Origin of information.

Transmitter: Subsystem that prepares and launches a signal.

Channel: Medium and propagation path between endpoints.

Receiver: Subsystem that recovers information from a received signal.

Destination: Intended user, device, or application.

Protocol: Agreed rules governing communication.

Modulation: Mapping information onto a carrier or waveform.

Bandwidth: Frequency range occupied or allocated by a signal or channel.

Noise: Unwanted random energy or disturbance.

Interference: Unwanted energy from another source or system.

Attenuation: Reduction in signal level or power.

SNR: Ratio of signal power to noise power.

BER: Fraction of received bits detected incorrectly.

Throughput: Useful information delivered per unit time.

Feedback: Information sent from receiver or destination toward the transmitter.

Link budget: Accounting of powers, gains, and losses in a link.

Review Questions

Short Answer Questions

1. Distinguish a message, signal, and information source.
2. What functions are commonly performed by a transmitter?
3. Define attenuation, noise, interference, and fading.
4. Why are protocols required in layered communication networks?
5. State the difference between dBm and dB.

Long Answer Questions

1. Explain the complete architecture of a communication system and the role of each functional block.
2. Compare analogue and digital communication systems, including benefits and limitations.
3. Explain how source coding, channel coding, modulation, and demodulation cooperate in a digital link.
4. Discuss SNR, BER, latency, throughput, reliability, and Shannon capacity as performance measures.
5. Explain the assumptions and engineering limitations of a basic link budget and free-space path-loss model.

Analytical / Application Questions

1. A link has 24 dBm transmitted power, 4 dB transmit gain, 3 dB receive gain, 2 dB feeder loss, and 96 dB propagation loss. Calculate received power in dBm and identify two practical losses not represented by the simplified data.
2. Design a communication architecture for a remote environmental sensor. Specify source representation, coding, modulation, feedback policy, performance measures, and security controls.
3. An emergency link has high average throughput but unacceptable alarm delay during fading. Analyse the likely architectural causes and propose changes involving priority, buffering, coding, modulation, retransmission, and measurement.

Multiple Choice Questions with Answers

Question 1

Which block converts information into a form suitable for transmission through a channel?

A. Transmitter

B. Destination

C. Protocol registry

D. Noise source

Correct Answer: A. Transmitter

Question 2

In the model r(t) = s(t) + n(t), what does n(t) represent?

A. The destination

B. Additive noise

C. The carrier frequency

D. The source code

Correct Answer: B. Additive noise

Question 3

A power ratio of 10 corresponds to what gain in decibels?

A. 1 dB

B. 3 dB

C. 10 dB

D. 20 dB

Correct Answer: C. 10 dB

Question 4

A signal power decreases from 1 W to 0.1 W. What is the corresponding positive link-loss magnitude?

A. −10 dB

B. +3 dB

C. +10 dB

D. +20 dB

Correct Answer: C. +10 dB

Question 5

Which quantity is measured as incorrectly received bits divided by examined received bits?

A. Bandwidth

B. Bit error rate

C. Latency

D. Antenna gain

Correct Answer: B. Bit error rate

Question 6

For an ideal AWGN channel, which expression gives Shannon capacity?

A. C = B(S/N)

B. C = B log₂(1 + S/N)

C. C = S + N

D. C = log₁₀(B + S/N)

Correct Answer: B. C = B log₂(1 + S/N)

Question 7

In a feedback communication system, acknowledgements normally travel from the:

A. Transmitter toward the channel only

B. Receiver or destination toward the transmitter

C. Noise source toward the receiver

D. Antenna toward the source only

Correct Answer: B. Receiver or destination toward the transmitter

Question 8

A link budget uses P_T = 30 dBm, gains of 2 dB and 5 dB, and losses of 1 dB and 100 dB. What is P_R?

A. −64 dBm

B. −74 dBm

C. +64 dBm

D. −54 dB

Correct Answer: A. −64 dBm

Question 9

The ideal sampling condition for a strictly band-limited signal with highest frequency f_max is:

A. f_s ≤ f_max

B. f_s = f_max/2

C. f_s ≥ 2f_max

D. f_s ≥ f_max/2

Correct Answer: C. f_s ≥ 2f_max

Question 10

Which statement best distinguishes throughput from gross data rate?

A. Throughput includes no time measurement

B. Throughput is useful delivered information after relevant overhead and losses

C. Gross data rate is always lower than throughput

D. They are different names for antenna gain

Correct Answer: B. Throughput is useful delivered information after relevant overhead and losses