Ad Hoc, Mesh, Vehicular, UAV and Satellite Networks

Introduction and Context

Wireless networking is commonly introduced through infrastructure-based systems: mobile stations communicate through access points, base stations, or satellites. In many important environments, however, fixed infrastructure is unavailable, damaged, too expensive, or unable to follow moving users. Ad hoc networks form temporary connections among participating devices; mesh networks use multiple wireless relays to extend coverage and improve path diversity; vehicular networks connect vehicles and roadside infrastructure; UAV networks use drones as mobile communication platforms; and satellite networks provide connectivity through non-terrestrial infrastructure.

These systems differ substantially in mobility, propagation delay, energy availability, topology stability, routing, regulation, and resilience. A routing method that works in a static community mesh may fail in a high-speed vehicular network or a rapidly changing UAV swarm. Similarly, satellite links can offer wide-area coverage but introduce propagation delay and stringent spectrum and orbital constraints.

This chapter compares these networking models and develops design principles for routing, topology control, delay management, resilience, and applications.

Learning Objectives

After studying this chapter, the reader should be able to:

1. Distinguish ad hoc, mesh, vehicular, UAV, and satellite network architectures.
2. Explain proactive, reactive, geographic, opportunistic, and delay-tolerant routing.
3. Analyze topology control using connectivity, interference, energy, and mobility constraints.
4. Compare latency, reliability, resilience, and capacity across network types.
5. Select an appropriate networking model for an application scenario.
6. Evaluate security, safety, regulatory, and ethical considerations.

1. Common Architecture and Terminology

A wireless network can be described as a graph G=(V,E), where V is the set of nodes and E is the set of wireless links. An edge exists when two nodes can communicate under the relevant power, interference, protocol, and propagation conditions. Unlike a wired graph, the edge set changes with time:

Here, t is time. Nodes may join or leave, and links may appear or disappear because of mobility, fading, obstruction, battery depletion, or channel access decisions.

Important terms include:

• Infrastructureless network: A network without mandatory fixed access points or base stations for local communication.
• Multi-hop communication: Delivery through one or more intermediate nodes.
• Forwarding node: A node that relays another node’s packet.
• Control plane: Protocol functions for discovery, routing, addressing, and topology maintenance.
• Data plane: Forwarding of user or application packets.
• Network partition: Separation of nodes into groups with no usable path between them.
• Link lifetime: The duration for which a link remains usable under specified criteria.
• Store-carry-forward: A delay-tolerant method in which a node stores a packet, physically carries it, and forwards it later.
• Non-terrestrial network (NTN): A network using platforms such as satellites or high-altitude systems rather than only terrestrial stations.

A useful performance model separates end-to-end delay into processing, queuing, transmission, propagation, and retransmission components:

The summation is over the traversed links or forwarding stages. This expression is a model rather than a universal exact equation; its terms depend on protocol behavior and traffic conditions. Transmission delay for a packet of length L bits over a link of rate R bits/s is:

Propagation delay is approximately:

where d is path distance in metres and v is propagation speed in metres per second. In free space, v is close to the speed of light, approximately 3 × 10⁸ m/s. For a satellite path, the physical distance can dominate the delay even when transmission is fast.

2. Ad Hoc Networks

An ad hoc network is created dynamically by participating wireless devices. Nodes may simultaneously act as hosts and routers. Examples include emergency-response teams, temporary classrooms, military formations, and sensor deployments.

2.1 Characteristics

Ad hoc networks commonly exhibit:

• no mandatory central controller;
• rapidly changing connectivity;
• multi-hop paths;
• distributed channel access;
• limited battery and processing resources;
• uncertain node trustworthiness; and
• possible network partitions.

The absence of infrastructure improves deployment flexibility but transfers responsibility to distributed protocols. Address allocation, route discovery, service discovery, synchronization, and security must be achieved with limited global knowledge.

2.2 Routing Approaches

Proactive routing maintains routes continuously. Each node periodically exchanges topology information. It provides low route-discovery delay but creates control overhead, especially when mobility is high. Link-state and distance-vector families illustrate this approach.

Reactive routing discovers a route when traffic requires one. Route-request flooding followed by a route-reply is a common pattern. This reduces idle overhead but introduces discovery latency and may cause broadcast congestion.

Hybrid routing uses proactive information within a local region and reactive discovery beyond it. It attempts to balance control overhead and route availability.

Geographic routing forwards packets using node positions, often selecting a neighbor closer to the destination. It can avoid maintaining complete routes, but it requires position information and may encounter local minima, obstacles, or inaccurate location estimates.

Opportunistic routing allows several candidate receivers to cooperate. It can exploit broadcast reception and fading diversity, but duplicate suppression and coordination are necessary.

2.3 Capacity and Energy Considerations

Multi-hop communication extends range but consumes resources at intermediate nodes. A relay must receive and transmit, and it may become a bottleneck. In a simple idealized chain, spatial reuse and interference determine capacity; therefore, capacity does not increase linearly with the number of relays.

For a battery-powered node, the energy budget may be represented as:

The terms denote energy used for transmission, reception, idle operation, and sensing. This accounting identity applies over a chosen observation interval. It does not predict radio energy unless each component is measured or modeled.

3. Wireless Mesh Networks

A mesh network consists of mesh clients, mesh routers, and possibly gateways. Mesh routers forward traffic for other nodes, while gateways connect the mesh to the Internet or another network. Unlike a purely infrastructureless ad hoc network, a mesh may contain relatively stable routers and planned backhaul, although access remains multi-hop and wireless.

3.1 Architecture and Routing

Mesh networks may be:

• infrastructure mesh, with relatively fixed routers;
• client mesh, where end devices also relay traffic; or
• hybrid mesh, combining fixed routers, mobile clients, and gateways.

Stable routers permit more organized routing and channel planning. Routing may use link-quality metrics rather than hop count alone. A path with fewer hops can be inferior if it has low throughput or high loss.

A general additive path metric can be expressed as:

where c_i is the cost of link i. The metric may represent expected transmission count, delay, energy, or an administratively weighted combination. Combining different quantities requires normalization; adding raw milliseconds to raw joules is not physically meaningful without a stated utility model.

3.2 Topology and Channel Planning

Mesh design must address gateway placement, channel assignment, relay density, and interference. Excessive relay density can increase contention, while sparse deployment can create weak links and partitions. Directional antennas, multiple radios, and separate backhaul channels can improve capacity, but increase cost and management complexity.

Mesh networks are valuable in rural broadband, community connectivity, industrial sites, campuses, public safety, and temporary events. Their principal limitation is that traffic near gateways may aggregate, creating a funneling effect and a reliability hotspot.

4. Vehicular Ad Hoc Networks

A vehicular ad hoc network (VANET) connects vehicles, roadside units, traffic controllers, pedestrians, and cloud services. Vehicle-to-vehicle communication is often called V2V; vehicle-to-infrastructure is V2I; vehicle-to-network is V2N; and vehicle-to-everything is V2X.

4.1 Distinctive Properties

Vehicles move rapidly but often follow constrained road geometry. This produces predictable mobility patterns in some contexts, yet link lifetimes can be short. Network density changes sharply between traffic jams and sparse highways. Applications range from safety warnings with strict latency requirements to infotainment with less stringent delay requirements.

Safety messages are typically short and periodic or event-triggered. Their requirements emphasize bounded delay, high delivery probability, congestion control, authentication, and protection against false information. A route-discovery protocol designed for long-lived flows may be unsuitable for a warning that must be delivered within a small time window.

4.2 Routing in VANETs

Topology-based routing may work in dense, slowly changing traffic but can break frequently at high speed. Geographic routing is attractive because vehicles can use position and road-map information. Carry-and-forward routing is useful in sparse networks: a vehicle stores data until it encounters a suitable forwarder.

Roadside units can act as stable anchors, disseminate local traffic information, and provide gateway access. However, infrastructure deployment is expensive and coverage may be incomplete.

4.3 Safety and Security

VANET security must protect message authenticity, integrity, privacy, and availability. Misbehavior detection is especially important because a validly formatted message may still contain false location or hazard information. Privacy mechanisms must balance pseudonymity with the ability of authorized entities to investigate serious incidents.

5. UAV and Flying Ad Hoc Networks

A UAV network uses unmanned aerial vehicles as communication nodes, sensing platforms, or both. A group of UAVs can form a flying ad hoc network (FANET). UAVs may communicate directly, relay traffic from ground users, or connect users to a terrestrial or satellite network.

5.1 Mobility and Design Constraints

FANET mobility is three-dimensional and may be highly dynamic. Nodes have limited flight time, and communication decisions interact with navigation, collision avoidance, payload, weather, and airspace restrictions. A UAV may be a relay, but it is also a moving object whose position must be controlled.

Compared with conventional MANETs, FANETs often have:

• greater relative mobility;
• more frequent topology changes;
• line-of-sight opportunities at altitude;
• limited onboard energy; and
• strong dependence on mission objectives.

5.2 Routing and Placement

Position-based routing can exploit GPS or other localization systems. Predictive routing uses planned trajectories and estimated link lifetimes. Clustering can reduce control overhead by selecting cluster heads, but cluster-head energy depletion and mobility can create instability.

UAV placement is a joint communication-and-control problem. Increasing altitude may enlarge coverage and improve line of sight, but it can also increase path loss, interference footprint, and exposure to wind. A practical controller therefore optimizes coverage, connectivity, energy, and mission safety rather than coverage alone.

5.3 Applications

Applications include disaster-area connectivity, precision agriculture, infrastructure inspection, temporary event coverage, search and rescue, environmental monitoring, and delivery coordination. Autonomous coordination introduces safety requirements: loss of communication must not cause unsafe flight, and the system must define fail-safe behavior.

6. Satellite and Non-Terrestrial Networks

Satellite networks use satellites as relays, access points, or backbone nodes. They may support remote connectivity, navigation augmentation, Earth observation, maritime and aviation communications, disaster recovery, and global broadcast.

6.1 Orbital Classes and Effects

• Geostationary Earth orbit (GEO): A satellite appears fixed relative to the Earth over the equator. It offers broad coverage but has a long propagation path.
• Medium Earth orbit (MEO): Used in several navigation and communications systems, with lower delay than GEO but requiring constellation management.
• Low Earth orbit (LEO): Provides lower propagation delay and often lower path loss than GEO, but satellites move rapidly relative to the ground and require frequent handovers.

Satellite systems may be bent-pipe, where a satellite mainly repeats signals, or regenerative, where onboard processing performs routing, switching, or regeneration. Inter-satellite links can support space-based mesh connectivity but increase routing and synchronization complexity.

6.2 Link Budget Fundamentals

A simplified free-space received-power model is:

where P_r and P_t are received and transmitted power in watts, G_t and G_r are dimensionless antenna gains, λ is wavelength in metres, and d is separation in metres. This equation assumes free-space propagation, aligned polarization, far-field operation, and no additional losses. Real satellite links require atmospheric, rain, pointing, feeder, implementation, and polarization losses.

The corresponding free-space path loss is:

It is dimensionless as a power ratio. In decibels, a ratio X is expressed as 10 log₁₀ X.

6.3 Delay and Handover

A one-way ground-to-satellite-to-ground path is longer than a terrestrial path. In addition to propagation delay, satellite networks experience queueing, gateway processing, link scheduling, and handover delay. LEO systems reduce propagation delay but require tracking and frequent beam or satellite handovers. Applications must distinguish delay from jitter and packet loss; interactive control may tolerate neither large delay nor unpredictable delay.

7. Comparative Routing and Topology Control

Routing selection should follow network conditions rather than convention.

Topology control adjusts transmission power, active links, relay selection, channels, or node positions. Its objectives may conflict:

• Connectivity requires sufficient links.
• Low interference favors fewer simultaneous links.
• Energy conservation favors low power and short forwarding distance.
• Reliability favors path diversity and redundancy.
• Low delay favors short, uncongested paths.

A robust design uses measurements such as received signal strength, signal-to-interference-plus-noise ratio, packet delivery ratio, queue length, residual energy, predicted link lifetime, and geographic position. No single metric is universally optimal.

8. Delay, Reliability, and Resilience

Reliability is not identical to connectivity. A network may have a path but still fail to deliver packets within the application deadline. Useful service measures include delivery probability, deadline-meeting probability, outage duration, route repair time, and recovery after node or link failure.

Resilience is the ability to maintain or restore acceptable service under failures, attacks, congestion, or environmental disruption. Methods include:

1. multipath routing;
2. redundant gateways or satellite links;
3. store-carry-forward operation;
4. local route repair;
5. distributed control without a single point of failure;
6. graceful degradation of application quality; and
7. authenticated rekeying and secure recovery procedures.

Redundancy has a cost. Duplicate transmissions consume bandwidth and energy, and excessive control traffic can worsen congestion. Resilience must therefore be designed against explicit failure scenarios.

9. Worked Example: Choosing a Route for a UAV Relay Network

Consider a three-hop UAV relay path. Each packet has length 1,200 bytes, and every link has a nominal rate of 6 Mbit/s. Ignore processing, queueing, propagation, and retransmission delays for a first estimate.

Packet length in bits is:

Transmission delay per hop is:

For three hops:

This is only the ideal transmission component. If each hop has an average 2 ms queueing delay and 1 ms processing delay, the estimated total becomes:

The example demonstrates why hop count alone is insufficient. A nominally fast route may fail its application deadline if queues, retransmissions, or route repairs are frequent. The assumption of a constant 6 Mbit/s rate is also optimistic; wireless rates vary with distance, interference, and modulation.

10. Applied Case: Disaster-Response Connectivity

Background

This illustrative case concerns a coastal region after an earthquake. Terrestrial cellular towers are partially unavailable, roads are obstructed, and emergency teams must exchange voice, maps, sensor readings, and medical data.

Problem

A single technology cannot satisfy all requirements. Responders need local low-delay communication, coverage across disconnected areas, and a backhaul link to a regional emergency coordination center.

Analysis

A portable mesh can connect teams within a damaged district. UAVs can provide elevated relays over areas where buildings or terrain block ground paths. A satellite terminal can provide backhaul when fiber and cellular infrastructure are unavailable. However, satellite delay and limited capacity make it unsuitable for all local traffic. UAV battery life limits continuous coverage, and mesh gateways may become bottlenecks.

Approach and Solution

The proposed architecture uses:

1. local mesh routing for team-to-team traffic;
2. geographic or link-quality routing to select robust ground relays;
3. UAVs positioned to bridge partitions and restore line-of-sight paths;
4. satellite backhaul for essential external traffic; and
5. traffic classes separating life-safety messages from routine data.

Emergency messages receive priority, authentication, and replication over independent paths where feasible. Nonessential video is rate-limited. UAVs periodically reposition according to coverage gaps, battery state, and predicted team movement.

Outcome

The hybrid system provides local low-delay communication while using the satellite link selectively. Its resilience comes from combining different technologies rather than assuming that any one network remains available. Operational procedures must include manual fallback, battery replacement, spectrum coordination, and safe UAV recovery.

Learning Points

• Network heterogeneity can improve resilience.
• Backhaul capacity must be reserved for critical traffic.
• Mobility and energy decisions are inseparable in UAV systems.
• Routing metrics must reflect application deadlines and link quality.
• Emergency deployments require governance, authentication, and safety planning.

11. Security, Ethics, and Regulation

Infrastructureless and non-terrestrial networks enlarge the attack surface. Threats include eavesdropping, spoofing, Sybil identities, route manipulation, jamming, malware, denial of service, compromised relays, and falsified location information. Security mechanisms should provide confidentiality, integrity, authentication, authorization, availability, and auditability.

A practical design uses authenticated control messages, replay protection, key management, secure boot, software updates, intrusion detection, and least-privilege access. In highly dynamic networks, certificates and revocation mechanisms must operate despite intermittent connectivity.

Regulatory issues include spectrum licensing, satellite coordination, aviation rules, remote identification of UAVs, lawful interception requirements, export controls, and privacy law. Vehicular systems must protect location privacy without weakening emergency accountability. Emergency use does not automatically remove the need for safe flight, spectrum discipline, or data protection.

Artificial intelligence can assist mobility prediction, channel selection, anomaly detection, and UAV placement. Nevertheless, learned decisions require uncertainty handling, adversarial testing, explainable fallback behavior, and human oversight in safety-critical applications. AI should complement, not replace, deterministic safety constraints.

12. Contemporary Developments and Best Practices

Current research and deployment trends include integrated terrestrial–NTN systems, 5G and emerging 6G non-terrestrial integration, edge computing near mobile users, programmable radios, intelligent reflecting surfaces, autonomous UAV coordination, digital twins for network planning, and machine-learning-assisted routing.

Best practices are:

• define application deadlines and reliability targets before selecting a protocol;
• measure link quality rather than relying only on distance or hop count;
• design for partitions, handovers, and loss of positioning data;
• separate control, safety, and bulk data traffic;
• use multipath selectively and quantify its overhead;
• include energy and battery state in UAV decisions;
• test under realistic mobility, interference, weather, and failure conditions;
• secure identity, routing, firmware, and data throughout the lifecycle; and
• provide graceful degradation and manual fallback.

Chapter Summary

Ad hoc networks create temporary multi-hop connectivity without mandatory infrastructure. Mesh networks add more stable relays and gateways, often improving coverage but creating interference and gateway bottlenecks. Vehicular networks exploit road-constrained mobility but require short-delay, authenticated, and safety-aware communication. UAV networks offer flexible three-dimensional coverage, while their routing is coupled to flight energy, trajectory, and mission safety. Satellite and other non-terrestrial networks provide broad reach and disaster resilience, but introduce propagation delay, handover, spectrum, and orbital-management challenges.

Routing may be proactive, reactive, hybrid, geographic, opportunistic, predictive, or delay tolerant. Topology control must balance connectivity, interference, energy, reliability, and delay. Resilience depends on diversity, redundancy, local repair, store-carry-forward operation, and graceful degradation. Effective designs are application-driven, security-aware, regulatory-compliant, and tested under realistic failure and mobility conditions.

Key Terms

Ad hoc network: A dynamically formed network in which devices communicate without mandatory fixed infrastructure.

Mesh network: A multi-hop wireless network using relays and, often, gateways.

MANET: Mobile ad hoc network with moving wireless nodes.

VANET: Vehicular ad hoc network connecting vehicles and roadside systems.

FANET: Flying ad hoc network formed by UAVs.

NTN: Non-terrestrial network using satellites or other aerial platforms.

Multi-hop: Communication through one or more intermediate nodes.

Proactive routing: Routing that maintains paths before traffic needs them.

Reactive routing: Routing that discovers paths when traffic requires them.

Geographic routing: Forwarding based primarily on node or destination position.

Opportunistic routing: Routing that exploits multiple possible receivers or paths.

Store-carry-forward: Delaying transmission while a node physically carries data.

Topology control: Adjustment of links, power, channels, relays, or positions to meet design objectives.

Link lifetime: Time for which a communication link remains usable.

Network partition: A division of nodes with no usable path between groups.

Resilience: Ability to maintain or restore service after disruption.

LEO: Low Earth orbit.

GEO: Geostationary Earth orbit.

V2X: Vehicle-to-everything communication.

Link budget: Accounting of transmitted power, gains, losses, and received power.

Review Questions

5 Short Answer Questions

1. What distinguishes a mesh network from a purely infrastructureless ad hoc network?
2. Why is geographic routing attractive in vehicular networks?
3. State two constraints that make UAV networking different from conventional MANET networking.
4. What are the principal delay sources in a multi-hop wireless path?
5. Why do satellite networks commonly require handover management?

5 Long Answer Questions

1. Compare proactive, reactive, hybrid, and geographic routing, including their strengths and limitations.
2. Explain how topology control balances connectivity, interference, energy, reliability, and delay.
3. Discuss the design requirements of a safety-critical VANET application.
4. Compare LEO, MEO, and GEO systems with respect to delay, mobility, coverage, and operational complexity.
5. Explain how a hybrid mesh–UAV–satellite architecture can improve disaster-response resilience.

3 Analytical / Application Questions

1. A UAV network has two candidate routes: one has fewer hops but weak, unstable links; the other has more hops but high delivery probability. Propose a route-selection metric and justify its components.
2. Design a connectivity plan for a rural community using a wireless mesh and satellite backhaul. Identify likely bottlenecks and mitigation measures.
3. A highway safety application must operate during periods of sparse traffic and intermittent roadside-unit coverage. Compare geographic forwarding, carry-and-forward, and cellular/satellite assistance.

Multiple Choice Questions with Answers

Question 1

Which routing approach discovers a path only when a source has traffic to send?

A. Proactive routing

B. Reactive routing

C. Static routing

D. Floodless routing

Correct Answer: B. Reactive routing

Question 2

In the simplified transmission-delay equation D=L/R, what does L represent?

A. Link distance

B. Received power

C. Packet length in bits

D. Propagation speed

Correct Answer: C. Packet length in bits

Question 3

Which network is most directly characterized by road-constrained mobility and vehicle safety messages?

A. VANET

B. GEO network

C. Sensor-only mesh

D. Fixed backbone network

Correct Answer: A. VANET

Question 4

What is a major advantage of geographic routing?

A. It requires complete global topology knowledge

B. It can forward using position information without maintaining complete routes

C. It eliminates the need for localization

D. It guarantees delivery in every topology

Correct Answer: B. It can forward using position information without maintaining complete routes

Question 5

Which factor is especially important in UAV network design?

A. Orbital inclination only

B. Flight energy and mission trajectory

C. Permanent wired backhaul

D. Fixed vehicle lanes

Correct Answer: B. Flight energy and mission trajectory

Question 6

Which satellite orbit generally has the lowest propagation delay among GEO, MEO, and LEO?

A. GEO

B. MEO

C. LEO

D. All have identical delay

Correct Answer: C. LEO

Question 7

A 1,200-byte packet is transmitted over a 6 Mbit/s link. Ignoring other delays, what is the transmission delay?

A. 0.16 ms

B. 1.6 ms

C. 16 ms

D. 160 ms

Correct Answer: B. 1.6 ms

Question 8

What is the main purpose of topology control?

A. To maximize the number of disconnected components

B. To adjust links or operating conditions to satisfy connectivity and performance objectives

C. To remove all relay nodes

D. To guarantee unlimited bandwidth

Correct Answer: B. To adjust links or operating conditions to satisfy connectivity and performance objectives

Question 9

Which technique is particularly useful when a sparse vehicular network lacks an immediate end-to-end path?

A. Store-carry-forward

B. Permanent circuit switching

C. Frequency elimination

D. Single-hop-only forwarding

Correct Answer: A. Store-carry-forward

Question 10

Which statement best describes network resilience?

A. The ability to increase antenna size indefinitely

B. The ability to maintain or restore acceptable service after disruption

C. The elimination of all redundancy

D. The use of only satellite links

Correct Answer: B. The ability to maintain or restore acceptable service after disruption