Signals and Data Transmission

Introduction

Data communication is the transfer of information between devices through a transmission medium. Although computers process information as bits, a communication channel carries physical signals: electrical voltages, electromagnetic waves, or pulses of light. Understanding the relationship between data and signals is therefore essential for designing, installing, troubleshooting, and evaluating networks.

This chapter examines analog and digital signals, frequency and bandwidth, spectrum, bit rate and baud rate, channel capacity, transmission modes, and the principal impairments that affect communication quality. These concepts form the foundation for later study of modulation, multiplexing, wired and wireless media, data-link protocols, and network performance.

Learning Objectives

By the end of this chapter, a learner should be able to:

• distinguish data from signals and analog from digital transmission;
• explain amplitude, period, frequency, phase, wavelength, and bandwidth;
• interpret a signal in the time and frequency domains;
• distinguish bit rate from baud rate;
• calculate theoretical channel capacity using Nyquist and Shannon models;
• classify transmission as simplex, half-duplex, or full-duplex;
• identify attenuation, distortion, and noise and explain their effects;
• apply signal-quality measures such as signal-to-noise ratio;
• select suitable methods for improving transmission reliability.

1. Data, Signals, and Transmission

Data is information represented in a form suitable for storage or processing. Text, numbers, images, audio, and video are all forms of data. Signals are physical representations of data that travel through a medium. A binary value may be represented by a voltage level on copper, a light intensity on fibre, or a radio-wave state in a wireless system.

A transmitter converts data into a signal. The signal travels through a channel, where it may be weakened or altered. A receiver detects the signal and reconstructs the original data. In a practical system, coding, modulation, amplification, filtering, error detection, and correction may occur between these stages.

The distinction between data and signals is important. Digital data can be transmitted using digital signals, as in baseband Ethernet, or using analog carrier signals, as in many radio systems. Analog data can also be converted into digital signals through sampling and quantisation, as in digital telephony.

Analog and digital signals

An analog signal varies continuously with time and may take any value within a range. A microphone output is an example: air pressure variations are converted into a continuously varying electrical voltage.

A digital signal has discrete levels or symbols. A simple binary digital waveform may use one voltage range for 0 and another for 1. Real digital signals are not perfectly rectangular because physical channels limit how quickly voltage can change.

Digital transmission does not mean that the physical waveform is mathematically perfect. It means that the receiver interprets the waveform as a set of discrete symbols.

2. Characteristics of Periodic Signals

A basic periodic signal can be represented by a sinusoid:

Here, s(t) is the signal value at time t, A is the peak amplitude, f is frequency in hertz (Hz), and φ is phase angle in radians. The model assumes a single ideal sinusoidal component. Practical signals are often combinations of many such components.

Amplitude

Amplitude is the magnitude or strength of a signal. In an electrical system it may be measured in volts; in an optical system it may be expressed as optical power. Higher amplitude generally makes a signal easier to detect, but excessive amplitude can overload equipment or cause interference.

For a sinusoid, the peak-to-peak value is twice the peak amplitude. When power is related to a voltage across a resistance, the root-mean-square voltage is commonly used. For a sinusoidal voltage with peak value V_p,

This relationship applies to a pure sinusoid measured across a resistive load.

Period and frequency

The period, T, is the time required for one complete cycle, measured in seconds. Frequency, f, is the number of cycles per second, measured in hertz.

A signal with a period of 0.002 seconds has a frequency of 500 Hz. Frequency is not the same as data rate: a signal may carry several bits during one cycle, or a single bit may occupy several cycles, depending on the encoding and modulation scheme.

Phase

Phase describes the relative position of a waveform within its cycle. Two signals with equal amplitude and frequency can differ in phase. Phase shifts are useful in modulation, but unwanted phase changes can make reception more difficult.

Wavelength

The wavelength, λ, is the physical distance travelled by a wave during one cycle:

Here, v is propagation speed in the medium, in metres per second, and f is frequency in hertz. In free space, electromagnetic waves propagate at approximately 3 × 10⁸ m/s, while propagation speed is lower in cables and optical fibre.

3. Spectrum and Bandwidth

The spectrum of a signal describes the frequencies contained in it. A pure sine wave has one frequency. A square wave, a voice signal, or a data pulse contains a range of frequency components.

The bandwidth of a channel is the range of frequencies that it can transmit with acceptable performance. If the lowest relevant frequency is f_L and the highest is f_H, the bandwidth is:

Bandwidth is measured in hertz, not bits per second. In everyday networking discussions, however, “bandwidth” is sometimes used informally to mean data capacity. Technicians should distinguish frequency bandwidth from bit rate.

A low-pass channel begins near 0 Hz and extends to an upper cutoff frequency. A band-pass channel operates between a lower and upper frequency, as in many radio and carrier systems. A channel does not necessarily transmit every frequency equally. Attenuation may be greater at some frequencies, causing distortion.

Time-domain and frequency-domain views

A time-domain graph shows signal amplitude against time and reveals pulses, transitions, delay, and noise. A frequency-domain graph shows the strength of frequency components and reveals occupied bandwidth, filtering, and interference. Both views describe the same physical signal from different perspectives.

A fundamental engineering principle is that rapid changes in a waveform require higher-frequency components. If a channel removes those components, digital pulses become rounded and may spread into neighbouring symbol intervals. This effect contributes to intersymbol interference.

4. Bit Rate, Baud Rate, and Symbol Rate

Bit rate is the number of bits transmitted per second, measured in bit/s or bps. Baud rate, more precisely symbol rate, is the number of signal symbols transmitted per second, measured in baud. A symbol is a distinct signal state or waveform selected from an available set.

If every symbol represents one bit, bit rate and baud rate are equal. If each symbol represents multiple bits, bit rate is higher than symbol rate:

Here, R_b is bit rate in bit/s, R_s is symbol rate in baud, and M is the number of distinct signal states. The formula assumes that all symbols represent the same number of bits and that coding overhead is excluded.

For four signal states, M = 4, so each symbol carries log₂(4) = 2 bits. At 1,000 baud, the gross bit rate is 2,000 bit/s. Error-control coding or framing reduces the useful information rate below the gross rate.

Using a very large number of signal states can increase bit rate without increasing baud rate, but the states become closer together and more vulnerable to noise. Consequently, modulation design involves a trade-off between efficiency, required signal-to-noise ratio, and implementation complexity.

5. Channel Capacity

Channel capacity is the maximum theoretical rate at which information can be transmitted reliably under stated assumptions. It is not necessarily the same as the configured or observed throughput of a network.

Nyquist bit-rate limit for a noiseless channel

For a noiseless baseband channel with bandwidth B hertz and M discrete signal levels, the theoretical maximum bit rate is:

The result is in bit/s. This is a theoretical limit and assumes ideal timing, a noiseless channel, and appropriate pulse shaping. Real channels have noise and other impairments.

Example. A noiseless channel has bandwidth 3,000 Hz and uses four signal levels. Its maximum theoretical rate is:

R_max = 2 × 3000 × log₂(4) = 6000 × 2 = 12000 bit/s.

Thus the Nyquist limit is 12,000 bit/s under the stated assumptions.

Shannon capacity for a noisy channel

For a channel with bandwidth B and signal-to-noise ratio S/N, Shannon capacity is:

Here, C is maximum theoretical reliable bit rate in bit/s. The signal-to-noise ratio must be expressed as a linear power ratio, not in decibels. The formula assumes additive noise and an ideal coding strategy over a sufficiently long transmission. It does not specify a practical modulation or protocol.

If SNR is given in decibels:

Therefore:

Example. A channel has bandwidth 3,000 Hz and SNR of 30 dB. The linear SNR is 10^(30/10) = 1000. Hence:

C = 3000 log₂(1 + 1000) ≈ 3000 × 9.97 ≈ 29,900 bit/s.

This is an upper theoretical limit. Protocol overhead, imperfect equipment, fading, interference, and implementation constraints reduce actual throughput.

Nyquist and Shannon answer different questions. Nyquist relates bandwidth and the number of signal levels under an ideal noiseless assumption. Shannon relates bandwidth and noise and places an upper bound on reliable information transfer regardless of the modulation method.

6. Transmission Modes and Direction of Communication

Transmission mode describes the direction and timing of data flow between devices.

Simplex

In simplex transmission, data travels in one direction only. A broadcast television transmission is a familiar example. The receiver does not send data back through the same communication relationship.

Half-duplex

In half-duplex transmission, both devices can transmit, but not simultaneously. Walkie-talkies use this arrangement: one participant transmits while the other listens. Turnaround time and contention management affect performance.

Full-duplex

In full-duplex transmission, both sides transmit simultaneously. A switched Ethernet link and a telephone conversation are examples. Separate frequency bands, separate conductors, echo cancellation, or other techniques may allow simultaneous transmission.

Transmission mode should not be confused with transmission medium or topology. A fibre link may operate in full-duplex mode, while a shared radio channel may use half-duplex operation.

7. Transmission Impairments

A received signal differs from the transmitted signal because of channel characteristics and external interference. The three fundamental impairments are attenuation, distortion, and noise.

Attenuation

Attenuation is the reduction in signal power as the signal travels through a medium. It is commonly expressed in decibels:

Here, P_T is transmitted power and P_R is received power. For a loss, P_T > P_R, so the result is positive. If voltage measurements are made across equal impedances, the equivalent expression is:

The voltage formula assumes equal source and load impedances. Decibels are logarithmic: a 3 dB power loss is approximately a halving of power, while a 10 dB loss corresponds to one-tenth of the power.

Attenuation is caused by conductor resistance, dielectric absorption, scattering, connector losses, and spreading. It increases with distance and may vary with frequency. Repeaters, amplifiers, higher-quality media, correct termination, and suitable transmit power can reduce its effects. An amplifier also amplifies noise, whereas a digital repeater attempts to reconstruct and regenerate the signal.

Distortion

Distortion occurs when different components of a signal are affected unequally. Amplitude distortion changes the strength of some frequencies more than others. Delay distortion occurs when frequency components arrive at different times. In digital systems, distortion can spread a pulse into adjacent symbol periods, producing intersymbol interference.

Distortion may result from limited channel bandwidth, impedance mismatch, reflections, poor filtering, or unequal propagation speeds. Equalisation, impedance matching, controlled cable geometry, filtering, and appropriate symbol rates are common remedies.

Noise

Noise is unwanted energy added to a signal. Important categories include:

• Thermal noise: caused by random motion of charge carriers; it exists in all electronic equipment.
• Induced or electromagnetic noise: introduced by motors, power cables, radio transmitters, and other electromagnetic sources.
• Crosstalk: unwanted coupling from a nearby communication circuit.
• Impulse noise: short, high-energy disturbances caused by switching, lightning, or electrical faults.
• Intermodulation noise: unwanted frequencies produced when nonlinear devices combine signal components.

Signal quality is often described using signal-to-noise ratio:

Here, P_S is signal power and P_N is noise power, both measured in the same units. A higher SNR generally permits more reliable detection and, according to Shannon’s model, a higher theoretical capacity.

8. Improving Transmission Reliability

Reliable transmission requires coordinated physical and logical measures. Common practices include:

1. Correct medium selection: choose twisted pair, coaxial cable, fibre, or wireless technology according to distance, bandwidth, interference, mobility, and cost.
2. Link-budget planning: verify that received power remains within the receiver’s operating range after cable, connector, splice, and environmental losses.
3. Grounding and shielding: reduce electromagnetic coupling, while avoiding poorly designed grounding arrangements that create ground loops.
4. Impedance matching and termination: reduce reflections on high-speed copper and radio-frequency circuits.
5. Filtering and equalisation: suppress unwanted frequencies and compensate for predictable channel loss.
6. Regeneration and repeaters: restore digital signal levels over long links, while observing maximum segment distances.
7. Error-control coding: add controlled redundancy so that receivers can detect or correct some errors.
8. Appropriate data rate: operate below the channel’s practical capacity rather than relying solely on a theoretical limit.
9. Testing and documentation: use tools such as cable certifiers, optical power meters, spectrum analysers, and packet monitors.

These practices also have operational implications. A network designer must balance capacity, reliability, installation cost, energy use, maintainability, and regulatory requirements.

9. Applied Case: A Noisy Industrial Ethernet Link

Background

This illustrative case concerns a manufacturing plant in which a programmable controller communicates with a remote input/output panel over 100 m of copper cable. The cable route runs parallel to motor power conductors for part of its length. The link experiences intermittent packet errors, especially when large motors start.

Problem analysis

The symptoms suggest impulse and electromagnetic noise, possibly combined with attenuation and crosstalk. A technician observes that the link is configured for full-duplex operation, so a duplex mismatch is unlikely. Cable inspection reveals an unshielded segment installed close to a motor feeder, and one connector is poorly terminated.

A time-domain test identifies reflections from the connector. A spectrum-oriented measurement shows increased interference during motor switching. The issue is therefore not simply insufficient nominal data rate; it is a reduction in the margin between the received signal and noise.

Approach and solution

The plant team takes the following actions:

• replaces the damaged connector and verifies pair termination;
• reroutes the cable away from power conductors where possible;
• uses correctly bonded industrial shielded cable in the high-interference section;
• maintains separation and crosses unavoidable power cables at right angles;
• checks grounding and shield practices according to the equipment manufacturer’s instructions;
• certifies the completed cable at the intended category and frequency;
• monitors error counters after the repair.

Outcome

Packet errors disappear during normal motor operation, and the measured link margin improves. The result is not achieved by increasing the nominal bit rate. Instead, the physical channel is made more suitable for the existing signal, and the receiver is given a clearer distinction between valid symbols and noise.

Learning points

The case demonstrates that network performance depends on the complete transmission path. A high-speed interface cannot compensate indefinitely for poor installation, excessive interference, reflections, or inadequate testing. Troubleshooting should begin with observable symptoms and measurements rather than with assumptions about software or protocol failure.

10. Contemporary Developments and Professional Practice

Modern communication systems increasingly use high-order modulation, adaptive coding, multiple antennas, coherent optical detection, and digital signal processing. These technologies increase information capacity by using bandwidth more efficiently or by improving resistance to noise. They also require better timing recovery, more sophisticated equalisation, and stronger error-control methods.

Software-defined radios can alter modulation and filtering through software, while machine-learning techniques may assist channel estimation, interference classification, and predictive maintenance. Such techniques do not remove physical limits: they still operate within constraints imposed by bandwidth, received power, noise, regulation, and hardware.

Wireless spectrum is a regulated resource. Engineers must observe national frequency allocations, maximum permitted power, emission masks, safety requirements, and licensing conditions. Fibre and copper installations must follow relevant cabling standards, workplace safety rules, and manufacturer specifications. Accurate documentation is an ethical responsibility because incorrect records can lead to unsafe maintenance, service disruption, or interference with other systems.

Challenges and Limitations

The formulas in this chapter are models, not guarantees of practical performance. Nyquist assumes a noiseless channel; Shannon assumes an idealised statistical model and arbitrarily effective coding. Real networks have finite processing speed, protocol overhead, finite block lengths, nonlinear components, changing channel conditions, and hardware limitations.

Likewise, a stated bandwidth does not guarantee a stated throughput. A channel may have adequate frequency range but poor SNR, severe interference, excessive delay spread, or inadequate connectors. Performance must therefore be validated through measurement under realistic operating conditions.

Chapter Summary

Data is information, while a signal is its physical representation for transmission. Analog signals vary continuously; digital signals are interpreted as discrete symbols. Periodic signals are described by amplitude, period, frequency, phase, and wavelength. Frequency spectrum identifies the frequency components of a signal, while bandwidth is the range of frequencies transmitted by a channel.

Bit rate measures bits per second, whereas baud rate measures symbols per second. When symbols represent multiple bits, R_b = R_s log₂(M) relates the two. Nyquist provides a theoretical limit for a noiseless channel with a specified number of signal levels, while Shannon gives a noisy-channel capacity based on bandwidth and SNR. Communication may be simplex, half-duplex, or full-duplex.

Attenuation weakens a signal, distortion changes its form, and noise adds unwanted energy. Decibels and SNR provide useful engineering measures, but practical reliability also depends on installation quality, interference control, equalisation, coding, testing, and regulatory compliance. Effective network professionals understand both theoretical limits and the physical conditions of actual transmission systems.

Key Terms

Amplitude: Magnitude or strength of a signal.

Analog signal: A continuously varying physical signal.

Attenuation: Reduction in signal power during transmission.

Bandwidth: Frequency range a channel transmits with acceptable performance.

Baud rate: Number of symbols transmitted per second.

Bit rate: Number of bits transmitted per second.

Channel capacity: Maximum theoretical reliable information rate of a channel under stated assumptions.

Crosstalk: Unwanted coupling from one communication circuit into another.

Digital signal: A signal interpreted as discrete levels or symbols.

Distortion: Change in signal shape caused by unequal channel effects.

Duplex: A communication arrangement describing direction of transmission.

Frequency: Number of cycles per second, measured in hertz.

Noise: Unwanted energy added to a desired signal.

Phase: Relative position of a waveform within its cycle.

Period: Time required for one complete cycle.

Spectrum: Distribution of signal energy across frequencies.

Symbol: A distinct signal state or waveform used to represent information.

Wavelength: Distance travelled by a wave during one cycle.

Review Questions

Short Answer Questions

1. What is the difference between data and a signal?
2. Define frequency, period, and bandwidth.
3. How does a digital signal differ from an analog signal?
4. Distinguish bit rate from baud rate.
5. Name and briefly explain the three major transmission impairments.

Long Answer Questions

1. Explain the characteristics of a sinusoidal signal and discuss the relationship between frequency, period, and wavelength.
2. Compare the Nyquist and Shannon capacity models, including their assumptions and practical limitations.
3. Describe simplex, half-duplex, and full-duplex transmission with suitable examples.
4. Explain attenuation, distortion, and noise and discuss methods used to reduce their effects.
5. Discuss how a time-domain and frequency-domain analysis can assist in network troubleshooting.

Analytical / Application Questions

1. A noiseless 4 kHz channel uses eight signal levels. Calculate its Nyquist maximum bit rate and explain why the result may not be achieved in practice.
2. A communication channel has bandwidth 2,000 Hz and SNR of 20 dB. Convert the SNR to a linear ratio and calculate its Shannon capacity approximately.
3. An industrial network develops errors only when nearby motors start. Propose a structured diagnostic plan using the concepts of noise, attenuation, distortion, cabling, and measurement.

Multiple Choice Questions with Answers

Question 1

Which statement best distinguishes bit rate from baud rate?

A. Bit rate counts symbols, while baud rate counts bytes

B. Bit rate counts bits per second, while baud rate counts symbols per second

C. Bit rate measures bandwidth, while baud rate measures noise

D. Bit rate and baud rate are always identical

Correct Answer: B. Bit rate counts bits per second, while baud rate counts symbols per second

Question 2

A signal has a period of 0.002 seconds. What is its frequency?

A. 50 Hz

B. 200 Hz

C. 500 Hz

D. 2,000 Hz

Correct Answer: C. 500 Hz

Question 3

According to the Nyquist noiseless-channel formula, what is the maximum rate for a 3,000 Hz channel with four signal levels?

A. 3,000 bit/s

B. 6,000 bit/s

C. 9,000 bit/s

D. 12,000 bit/s

Correct Answer: D. 12,000 bit/s

Question 4

Which impairment is caused by unwanted energy added to a signal?

A. Attenuation

B. Noise

C. Wavelength

D. Phase

Correct Answer: B. Noise

Question 5

A four-level signalling scheme operates at 1,000 baud. Ignoring overhead, what is its bit rate?

A. 500 bit/s

B. 1,000 bit/s

C. 2,000 bit/s

D. 4,000 bit/s

Correct Answer: C. 2,000 bit/s

Question 6

Which transmission mode permits both devices to transmit simultaneously?

A. Simplex

B. Half-duplex

C. Full-duplex

D. Unidirectional

Correct Answer: C. Full-duplex

Question 7

In the Shannon capacity formula, an SNR of 30 dB corresponds to which linear power ratio?

A. 3

B. 30

C. 100

D. 1,000

Correct Answer: D. 1,000

Question 8

Which action most directly reduces reflections on a high-speed copper link?

A. Increasing the number of symbols without testing

B. Correct impedance matching and termination

C. Replacing digital data with analog data

D. Reducing the wavelength to zero

Correct Answer: B. Correct impedance matching and termination

Question 9

What does bandwidth describe in its strict engineering meaning?

A. The number of devices on a network

B. The frequency range transmitted by a channel

C. The number of bits stored in memory

D. The physical length of a packet

Correct Answer: B. The frequency range transmitted by a channel

Question 10

A receiver observes unequal arrival times for frequency components of a signal. Which impairment is most directly indicated?

A. Delay distortion

B. Simplex transmission

C. Thermal capacity

D. Bit quantisation

Correct Answer: A. Delay distortion