Wireless, Mobile, and IoT Networking

Introduction

Wireless networking has changed the design assumptions of computer networks. A wired link normally provides a relatively stable medium with predictable attenuation and limited interference. A wireless link is shared, variable, exposed to interception, and affected by propagation, mobility, and radio-frequency (RF) regulations. Mobile and Internet of Things (IoT) networks extend these challenges across large geographic areas and devices with very different capabilities.

This chapter examines the principles needed to design and operate secure wireless local area networks (WLANs), mobile networks, and IoT connectivity. It connects RF behavior with Wi-Fi performance, explains roaming and controller-based WLAN management, compares 4G and 5G, and evaluates protocols for constrained devices and edge systems.

Learning Objectives

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

1. Explain RF propagation, channel use, interference, and antenna behavior.
2. Compare major Wi-Fi generations and select suitable WLAN configurations.
3. Plan WLAN coverage, capacity, roaming, authentication, and controller architecture.
4. Evaluate the architecture and capabilities of 4G LTE and 5G networks.
5. Select appropriate protocols for constrained IoT devices.
6. Design secure, manageable, and resilient wireless and edge-connected systems.
7. Analyze practical trade-offs involving performance, energy, privacy, cost, and regulation.

1. Wireless Networking as a Shared Medium

A WLAN uses radio waves rather than a physical cable. Devices transmit electromagnetic energy through an antenna, and receivers recover information from the received signal. Because multiple devices may transmit in the same frequency range, wireless networks require rules for channel access, addressing, retransmission, and coexistence.

Important terms include:

• Frequency: The number of waveform cycles per second, measured in hertz (Hz).
• Wavelength: The physical distance represented by one cycle of a wave.
• Channel: A defined frequency range used for communication.
• Bandwidth: The width of a channel, measured in hertz.
• RSSI: A receiver-specific indication of received signal strength.
• SNR: Signal-to-noise ratio, the relationship between desired signal power and noise power.
• Interference: Unwanted energy from other transmitters or electronic sources.
• AP: Wireless access point that provides radio connectivity to client devices.
• BSS: Basic Service Set, normally one AP and its associated stations.
• ESS: Extended Service Set, multiple coordinated BSSs presenting one WLAN service.

A wireless LAN is not simply a wired Ethernet network with radio replacing cable. The medium is half-duplex: a radio generally cannot transmit and receive on the same channel simultaneously. Devices must also account for hidden nodes, fading, contention, and changing channel conditions.

2. Radio-Frequency Principles

2.1 Frequency, wavelength, and propagation

Frequency and wavelength are related by:

Here, λ is wavelength in metres, c is the speed of light in a vacuum, approximately 3 × 10⁸ m/s, and f is frequency in hertz. This relationship assumes propagation in free space; signals travel slightly more slowly in materials.

For example, a 2.4 GHz signal has an approximate wavelength of:

Thus, the wavelength is about 12.5 cm. Wavelength influences antenna dimensions, diffraction, penetration, and multipath behavior.

2.2 Free-space path loss

In ideal free space, received power decreases as distance and frequency increase. A commonly used expression is:

In this form, distance d is in kilometres and frequency f is in megahertz. The constant is appropriate for these units. The formula assumes unobstructed line-of-sight propagation and does not account for walls, foliage, antenna gains, polarization mismatch, or multipath.

A more general link budget is:

P_r is received power, P_t is transmit power, G_t and G_r are transmitter and receiver antenna gains, L_p is path loss, and L_c represents cable, connector, and other losses. All power and loss terms must use compatible decibel units.

A receiver can decode a frame only when the received signal is sufficiently above the noise floor and interference. The basic relationship is:

For example, a signal of − sixty dBm and noise of − ninety dBm gives an SNR of 30 dB. The required SNR depends on modulation, coding, channel width, implementation, and the target error rate. Higher data rates usually require better SNR.

2.3 Reflection, diffraction, and multipath

Walls, floors, vehicles, furniture, and people reflect or absorb RF energy. Signals arriving through several paths can combine constructively or destructively. This produces multipath fading, in which a small movement can significantly change received power. Diversity techniques, multiple antennas, beamforming, and modern coding reduce these effects but do not eliminate them.

Lower-frequency bands generally penetrate obstacles better and travel farther, while higher-frequency bands can support wider channels and higher capacity but are more affected by blockage and attenuation. This is a design tendency, not an absolute rule: antenna design, building materials, transmit power, and regulation also matter.

2.4 Antennas and polarization

Omnidirectional antennas radiate approximately around a horizontal plane, whereas directional antennas concentrate energy in a selected direction. Antenna gain does not create energy; it redistributes it spatially. Directional antennas are useful for point-to-point links and targeted coverage but may create unintended dead zones.

Transmitters and receivers should normally use compatible polarization. Polarization mismatch reduces received power. In enterprise WLANs, diversity antennas and multiple-input multiple-output (MIMO) techniques exploit spatially separate signal paths to improve reliability and throughput.

3. Wi-Fi Standards and Medium Access

Wi-Fi is standardized by the IEEE 802.11 family. Important generations include:

Actual performance depends on client capability, channel width, signal quality, contention, protocol overhead, and regulatory domain. A theoretical physical-layer rate is not equivalent to application throughput.

IEEE 802.11 uses carrier sense multiple access with collision avoidance (CSMA/CA). A station senses the channel, waits when it is busy, and uses a randomized backoff before transmitting. Acknowledgements and retransmissions improve reliability. The protocol cannot detect collisions as efficiently as traditional wired Ethernet because a station’s own transmission overwhelms its receiver.

The hidden-node problem occurs when two stations cannot hear each other but can both reach the AP. They may transmit simultaneously and collide at the AP. Request-to-send/clear-to-send mechanisms can reduce this problem, although they add overhead and are not universally beneficial.

A wider channel can increase peak data rate but also consumes more spectrum, raises the probability of overlapping interference, and may reduce the number of independent channels available. In dense environments, narrower channels often provide greater aggregate capacity.

4. WLAN Design: Coverage, Capacity, and Roaming

4.1 Coverage versus capacity

Coverage planning asks, “Where can a device maintain an acceptable connection?” Capacity planning asks, “Can the network serve the expected number and type of devices at the required performance?” A single high-power AP may provide broad coverage but poor capacity because clients compete for one radio and distant clients consume excessive airtime.

Design begins with requirements:

• number and location of users;
• device types and expected traffic;
• applications and latency requirements;
• voice, video, or industrial reliability needs;
• building materials and floor plans;
• interference sources;
• security and identity requirements;
• regulatory restrictions and available channels.

A useful planning approximation is:

where N_AP is the minimum number of APs based on demand, D_total is the total offered demand, and D_AP is the usable capacity of one AP under the actual environment. This is an engineering estimate, not a universal law. D_AP must account for protocol overhead, contention, retransmissions, airtime allocation, and application mix. Coverage and channel reuse may require more APs than this calculation suggests.

Worked example: capacity estimation

A learning laboratory has 36 students. Each student device is expected to generate an average of 2 Mb/s during a collaborative exercise. The design target is 60 Mb/s of usable aggregate application throughput per AP after accounting for contention and protocol overhead. Offered demand is:

The demand-based estimate is:

Therefore, at least two APs are required by this simplified capacity estimate. A survey may still require additional APs because of walls, channel reuse, voice roaming, or uneven device distribution. Increasing transmit power alone would not solve the capacity problem.

4.2 Channel planning

In the 2.4 GHz band, overlapping channels can cause substantial co-channel and adjacent-channel interference. Many deployments use a small set of non-overlapping 20 MHz channels, subject to the regulatory domain. The 5 GHz and 6 GHz bands generally provide more channel choices, but available channels and permitted transmit power vary by country.

Channel reuse should be planned so neighboring APs do not unnecessarily contend on the same channel. However, reusing a channel too aggressively can cause co-channel contention, while using too many channels with weak coverage can increase adjacent-channel problems or roaming instability. Automatic radio resource management can assist, but it should be monitored rather than trusted blindly.

4.3 Roaming

Roaming occurs when a client moves from one AP or BSS to another while retaining network service. The client usually makes the final roaming decision, although WLAN infrastructure can provide measurements and recommendations.

Good roaming requires:

• consistent SSID and security configuration across APs;
• appropriate overlap between cells;
• correctly designed power levels;
• timely client scanning and reassociation;
• suitable authentication performance;
• stable Layer 2 and Layer 3 design;
• application tolerance for brief interruptions.

802.11k can provide neighbor information, 802.11v can support network-assisted transition decisions, and 802.11r can reduce authentication delay through fast transition mechanisms. These features must be tested with actual client devices because compatibility varies.

Excessive cell overlap can cause sticky clients to remain attached to a distant AP. Insufficient overlap produces dead zones. Roaming design is therefore a balance between continuity, interference, and client behavior.

5. WLAN Security and Management

5.1 Authentication and encryption

Open WLANs provide no meaningful confidentiality. WPA2-Enterprise and WPA3-Enterprise commonly use IEEE 802.1X with an authentication server, often RADIUS. The 802.1X model separates the supplicant (client), authenticator (AP or switch), and authentication server.

Enterprise authentication may use certificates, protected password methods, or tunneled methods. Certificate validation is essential; disabling validation creates exposure to credential interception. WPA3 improves protections in several deployment modes, but migration must consider older clients and operational capability.

Personal modes use a shared passphrase. They are appropriate for small or controlled environments but create management problems when users share one credential. A compromised passphrase may require replacing it on every device.

Security controls should include:

• strong cryptography and current firmware;
• separate staff, guest, voice, and IoT networks;
• least-privilege firewall policies;
• protected management interfaces;
• certificate lifecycle management;
• logging and time synchronization;
• rogue AP detection;
• regular validation of configuration and client compatibility.

SSID hiding is not a security control. MAC filtering is weak because MAC addresses can be observed and spoofed. Network segmentation and identity-based access are substantially more useful.

5.2 WLAN controllers

A controller-based WLAN centralizes configuration, radio management, policy, authentication integration, firmware coordination, monitoring, and sometimes tunneling. Lightweight APs commonly establish control relationships with controllers, while cloud-managed systems provide similar functions through a vendor-hosted management plane.

Centralization improves consistency and visibility but introduces dependency on controller availability, licensing, provider reliability, and management-plane security. A sound architecture considers redundant controllers, local survivability, backup configuration, out-of-band recovery, and secure administrative access.

A controller should not be treated as a substitute for RF engineering. It can optimize channels and power within configured constraints, but it cannot understand every physical obstruction or application requirement without measurement and validation.

6. Cellular Networking: 4G and 5G

6.1 4G LTE

4G LTE uses an evolved packet core and radio access network. Important components include the user equipment (UE), eNodeB radio base station, mobility management functions, serving and packet gateways, and policy and subscriber systems. LTE uses OFDMA in the downlink and SC-FDMA in the uplink in order to manage spectral efficiency and terminal power considerations.

LTE supports mobility across cells, authentication through subscriber identity systems, quality-of-service bearers, and packet-based services. Its performance depends on spectrum, cell loading, signal conditions, backhaul, and operator configuration.

6.2 5G

5G introduces a 5G core, gNodeBs, service-based core functions, improved virtualization, and support for diverse service categories. Commonly discussed use cases are:

• Enhanced mobile broadband: high-throughput consumer and enterprise access.
• Ultra-reliable low-latency communications: demanding industrial or control applications.
• Massive machine-type communications: large populations of low-data-rate devices.

These categories describe design goals rather than guaranteed results. Latency includes radio scheduling, transport, processing, queuing, and application delay. A 5G logo alone does not guarantee ultra-low latency.

5G may use sub-1 GHz, mid-band, and high-band spectrum. Lower bands provide broad coverage, mid-band often provides a useful capacity/coverage balance, and millimetre-wave bands can provide high capacity over short distances but are more sensitive to blockage. Network slicing can create logically separated service configurations, but slices still depend on shared physical infrastructure and require strong isolation and governance.

6.3 Private cellular networks

Private LTE and 5G networks can provide controlled coverage, identity management, mobility, and predictable policy in factories, campuses, ports, and mines. They may be preferable to Wi-Fi for wide-area mobility, outdoor coverage, SIM-based identity, or specialized reliability requirements. They also involve spectrum licensing, core-network operations, device compatibility, and higher technical complexity.

7. IoT Networking and Constrained Devices

IoT systems connect sensors, actuators, gateways, platforms, and applications. Devices may have limited memory, processing power, energy, bandwidth, and physical protection. An IoT architecture normally includes:

1. Device layer: sensors, actuators, embedded processors.
2. Connectivity layer: Wi-Fi, Ethernet, Bluetooth Low Energy, IEEE 802.15.4, LPWAN, LTE-M, or NB-IoT.
3. Edge or gateway layer: protocol translation, filtering, local control, and buffering.
4. Platform layer: device registry, message ingestion, storage, analytics, and identity.
5. Application layer: monitoring, automation, alerts, and business integration.

7.1 Protocol selection

MQTT uses a broker. Publishers send messages to topics, and subscribers receive messages according to subscriptions. Quality of Service levels should be selected carefully: at-most-once minimizes overhead, at-least-once may deliver duplicates, and exactly-once has greater protocol and processing cost. Application logic should remain safe if messages are duplicated or delayed.

CoAP commonly uses confirmable messages when delivery acknowledgement is needed. DTLS or OSCORE can protect suitable deployments, while TLS is common for MQTT over TCP. Security is not achieved merely by selecting a protocol name; credentials, authorization, update mechanisms, and key storage are decisive.

7.2 Energy and edge computing

Battery-powered devices should minimize radio-on time, message size, retransmissions, and unnecessary reporting. Sampling locally and transmitting only meaningful changes can reduce energy use, provided that local processing does not compromise safety or data quality.

Edge computing places computation near the data source. It can reduce backhaul traffic, support local operation during cloud disconnection, reduce latency, and improve privacy. Edge systems also create additional assets that require patching, monitoring, identity management, and physical protection.

8. Secure IoT and Wireless Architecture

A defensible architecture applies security throughout the lifecycle:

• assign a unique identity to every device;
• use hardware-backed keys where practical;
• authenticate devices mutually with gateways or services;
• authorize actions by role and least privilege;
• encrypt data in transit and protect sensitive data at rest;
• disable unused services and change default credentials;
• use signed, verified, and rollback-capable firmware updates;
• segment IoT networks from administrative and user networks;
• monitor unusual traffic and device behavior;
• define secure decommissioning and credential revocation.

Threats include rogue access points, evil-twin attacks, credential theft, denial of service, jamming, firmware compromise, botnets, insecure APIs, physical extraction of keys, and privacy leakage. Availability and safety are especially important in healthcare, manufacturing, transport, and building control.

Privacy governance should address data minimization, purpose limitation, retention, consent or lawful basis, access rights, and cross-border transfer. Sensor data can reveal occupancy, health, work patterns, or personal behavior even when it appears technically non-sensitive.

9. Applied Case: Connected University Campus

The following is an illustrative case.

Background

A university plans to support 18,000 students and staff across teaching buildings, residences, laboratories, and outdoor areas. Requirements include secure student access, high-density lecture halls, roaming for voice services, environmental sensors, and reliable access for security staff.

Problem

The existing WLAN has many isolated APs, inconsistent authentication, overlapping channels, and a shared password for building sensors. Students report poor service in lecture halls, while facilities staff cannot distinguish failed sensors from wireless outages.

Analysis

The university separates requirements into traffic classes. Lecture halls require capacity rather than merely broad coverage. Security staff need authenticated mobility. Environmental sensors transmit small messages infrequently and should not share unrestricted access with student devices. A survey identifies concrete walls, dense seating, and interference from neighboring networks.

Approach and solution

The design uses centrally managed Wi-Fi 6/6E where permitted, 20 MHz channels in dense areas, and additional APs based on capacity and measured attenuation. Staff and students use 802.1X with institution-managed identity. Guests receive an isolated service with restricted access. Voice devices use tested fast-transition features and quality-of-service policies.

Sensors use individual credentials or certificates, a separate IoT SSID and VLAN, restrictive firewall rules, and MQTT over TLS to an edge broker. The broker forwards selected data to the campus platform. Local rules continue basic environmental control if the cloud service is unavailable.

Controllers are deployed redundantly, logs are sent to a security monitoring platform, and periodic surveys validate RSSI, SNR, channel utilization, roaming delay, and application performance. The university also defines a firmware update process and a retirement procedure for sensors.

Outcome

The design improves lecture-hall capacity through additional radios and controlled channel reuse, reduces credential-sharing risk, and allows facilities personnel to identify device, RF, and application failures separately. The main cost is greater operational complexity: identity services, certificates, controllers, brokers, and edge systems all require skilled administration.

Learning points

1. Coverage, capacity, and roaming are different design problems.
2. Security must include identity, segmentation, updates, and monitoring.
3. IoT devices require lifecycle management, not merely network connectivity.
4. Edge processing can improve resilience but creates more infrastructure to secure.
5. Measurements and client testing are essential; theoretical data rates are insufficient.

10. Challenges, Governance, and Contemporary Developments

Wireless systems operate within national spectrum rules. Engineers must observe permitted bands, transmit-power limits, indoor/outdoor restrictions, dynamic frequency selection requirements, and certification conditions. Unauthorized operation can interfere with safety, scientific, or licensed services.

Artificial intelligence and machine learning are increasingly used for anomaly detection, RF optimization, predictive maintenance, and device classification. These systems can assist engineers but may produce false positives, hide reasoning, or make unstable configuration changes. Automated optimization should use bounded policies, audit logs, human review for high-impact changes, and safe rollback.

Contemporary deployments increasingly combine Wi-Fi, private 5G, LPWAN, edge computing, and cloud services. The correct choice depends on mobility, latency, energy, coverage, throughput, device density, identity model, and total cost—not on selecting the newest technology.

Best practices include:

• begin with application and risk requirements;
• perform predictive and post-installation surveys;
• design for airtime and capacity, not only signal strength;
• use modern enterprise authentication and certificate validation;
• segment users, guests, voice, management, and IoT;
• document regulatory assumptions;
• monitor utilization, retries, roaming, and authentication failures;
• test firmware, clients, and failover before production;
• plan device replacement and credential revocation;
• retain human accountability for automated decisions.

Chapter Summary

Wireless, mobile, and IoT networks use shared and variable communication media. RF frequency, wavelength, path loss, multipath, antenna gain, SNR, and interference determine whether a radio link is usable. Wi-Fi design must distinguish coverage from capacity and must consider channel reuse, airtime contention, roaming, client behavior, and application requirements. Enterprise security commonly uses 802.1X, strong encryption, identity-based policy, segmentation, and centralized management.

4G LTE provides mature cellular mobility and packet services, while 5G expands flexibility through new spectrum, a service-based core, private networks, and support for diverse performance objectives. IoT design requires matching protocols and link technologies to device constraints, energy budgets, coverage, and message patterns. MQTT, CoAP, cellular IoT, Bluetooth Low Energy, Thread, and LoRaWAN each occupy different design spaces.

Secure architecture is lifecycle-based: unique identity, least privilege, encryption, secure updates, monitoring, privacy governance, and safe decommissioning are all necessary. Controllers, cloud platforms, AI tools, and edge systems improve manageability or performance but introduce dependencies and new attack surfaces. Sound engineering combines standards knowledge, measurement, risk analysis, and operational discipline.

Key Terms

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

Wavelength: Distance represented by one cycle of a wave.

Bandwidth: Width of a communication channel in hertz.

SNR: Difference between signal power and noise power, expressed in decibels.

Path loss: Reduction in received signal power caused by propagation and obstacles.

Multipath fading: Signal variation caused by multiple reflected propagation paths.

MIMO: Use of multiple antennas to improve throughput or reliability.

CSMA/CA: Wi-Fi channel-access method using carrier sensing and collision avoidance.

BSS: Basic Service Set consisting of an AP and associated stations.

ESS: Extended Service Set formed from multiple coordinated BSSs.

Roaming: Transfer of a client connection between APs or cells while maintaining service.

802.1X: Port-based access-control framework commonly used with enterprise WLAN authentication.

RADIUS: Protocol commonly used to exchange authentication and authorization information.

WLAN controller: System that centrally manages AP configuration, policy, and radio functions.

LTE: Fourth-generation cellular technology based on the 3GPP Evolved Packet System.

5G: Fifth-generation cellular system supporting enhanced broadband, IoT, and specialized services.

MQTT: Lightweight publish/subscribe messaging protocol often used for telemetry.

CoAP: Constrained Application Protocol designed for resource-limited devices.

Edge computing: Processing performed near data sources rather than exclusively in a distant cloud.

LPWAN: Low-power wide-area networking category optimized for long-range, low-rate communication.

Review Questions

Short Answer Questions

1. Why does a higher frequency often provide greater capacity but reduced obstacle penetration?
2. Distinguish WLAN coverage planning from WLAN capacity planning.
3. What is the purpose of 802.1X in an enterprise WLAN?
4. Name two differences between MQTT and HTTP in IoT deployments.
5. Why is a shared IoT password poor practice for a large installation?

Long Answer Questions

1. Explain how reflection, diffraction, absorption, and multipath affect WLAN design.
2. Compare Wi-Fi 6, Wi-Fi 6E, and 5G for a dense university environment.
3. Describe a controller-based WLAN architecture, including its advantages and risks.
4. Explain the major security controls required throughout an IoT device lifecycle.
5. Discuss how private 5G and enterprise Wi-Fi can complement rather than simply replace one another.

Analytical / Application Questions

1. A warehouse has mobile scanners, fixed sensors, voice handsets, and autonomous vehicles. Propose a connectivity architecture and justify the choice of Wi-Fi, private cellular, LPWAN, or other technologies for each class.
2. An office has excellent RSSI but poor application throughput. Analyze at least five possible causes and propose measurements to distinguish them.
3. Design a segmented network for smart-building sensors. Specify authentication, messaging, edge processing, update management, monitoring, and failure-handling controls.

Multiple Choice Questions with Answers

Question 1

Which relationship correctly connects wavelength, propagation speed, and frequency?

A. λ = c × f

B. λ = c / f

C. λ = f / c

D. λ = c + f

Correct Answer: B. λ = c / f

Question 2

What is the main reason that a theoretical Wi-Fi physical-layer rate is higher than application throughput?

A. Application traffic is always encrypted twice

B. Protocol overhead, contention, retransmissions, and signal conditions reduce usable throughput

C. APs transmit only one frame per minute

D. Wi-Fi does not use acknowledgements

Correct Answer: B. Protocol overhead, contention, retransmissions, and signal conditions reduce usable throughput

Question 3

Which problem occurs when two stations cannot hear one another but both can reach the same AP?

A. Broadcast storm

B. Hidden-node problem

C. DNS poisoning

D. Route aggregation

Correct Answer: B. Hidden-node problem

Question 4

Which mechanism is most appropriate for enterprise WLAN identity-based authentication?

A. SSID hiding

B. MAC filtering alone

C. 802.1X with an authentication server

D. A single shared guest password

Correct Answer: C. 802.1X with an authentication server

Question 5

A WLAN has 72 Mb/s of offered demand and one AP provides 60 Mb/s of usable capacity. What is the minimum AP count from the simplified demand estimate?

A. One

B. Two

C. Three

D. Four

Correct Answer: B. Two

Question 6

Which Wi-Fi feature is associated with providing neighbor information to assist client roaming decisions?

A. 802.11k

B. 802.11e only

C. 802.11af

D. 802.11h only

Correct Answer: A. 802.11k

Question 7

Which IoT protocol uses a broker and a publish/subscribe model?

A. MQTT

B. FTP

C. ARP

D. ICMP

Correct Answer: A. MQTT

Question 8

Which cellular generation introduces a service-based 5G core and supports network slicing concepts?

A. 2G

B. 3G

C. 4G GSM

D. 5G

Correct Answer: D. 5G

Question 9

Which statement about edge computing is correct?

A. It always eliminates the need for cloud services

B. It can reduce latency and backhaul traffic but creates additional systems to secure

C. It is used only for web browsing

D. It prevents all device failures

Correct Answer: B. It can reduce latency and backhaul traffic but creates additional systems to secure

Question 10

Which practice best protects a large population of IoT devices?

A. Use one permanent password for every device

B. Disable firmware updates to preserve stability

C. Use unique identities, least privilege, secure updates, and network segmentation

D. Rely on SSID hiding

Correct Answer: C. Use unique identities, least privilege, secure updates, and network segmentation