PaperlayerView exact revision on arXiv

Source-linked AI summary

Survey of Important Issues in UAV Communication Networks

Lav Gupta, Raj Jain, Gabor Vaszkun

arXiv:1603.08462v1cs.NI

TL;DR

UAV networks have distinctive routing and connectivity requirements, while existing surveys and modified MANET/VANET protocols leave key UAV-specific issues underexplored. This survey synthesizes work on network characterization, routing, SDN, handovers, and energy efficiency, finding that native UAV protocols are still absent and seamless handover remains little studied.

  • Problem

    UAV-specific routing requirements and broader networking issues remain underexplored, with a need for reliable communication protocols suited to UAV networks.

    Source: B. Applicability of existing routing… · p. 9, §VI · p. 28

  • Method

    The survey categorizes UAV networks and reviews research on network characterization, routing, SDN, seamless handovers, and energy efficiency.

    Source: §VI · pp. 27–28

  • Results

    Native UAV network protocols have yet to emerge, and seamless handover in UAV environments has hardly been studied.

    Source: §VI · p. 27

  • Takeaways & Limitations

    The survey expects its synthesis of understudied issues and open research questions to spur further work on multi-UAV networks.

    Source: §VI · p. 28

Abstract

from arXiv · show

Unmanned Aerial Vehicles (UAVs) have enormous potential in the public and civil domains. These are particularly useful in applications where human lives would otherwise be endangered. Multi-UAV systems can collaboratively complete missions more efficiently and economically as compared to single UAV systems. However, there are many issues to be resolved before effective use of UAVs can be made to provide stable and reliable context-specific networks. Much of the work carried out in the areas of Mobile Ad Hoc Networks (MANETs), and Vehicular Ad Hoc Networks (VANETs) does not address the unique characteristics of the UAV networks. UAV networks may vary from slow dynamic to dynamic; have intermittent links and fluid topology. While it is believed that ad hoc mesh network would be most suitable for UAV networks yet the architecture of multi-UAV networks has been an understudied area. Software Defined Networking (SDN) could facilitate flexible deployment and management of new services and help reduce cost, increase security and availability in networks. Routing demands of UAV networks go beyond the needs of MANETS and VANETS. Protocols are required that would adapt to high mobility, dynamic topology, intermittent links, power constraints and changing link quality. UAVs may fail and the network may get partitioned making delay and disruption tolerance an important design consideration. Limited life of the node and dynamicity of the network leads to the requirement of seamless handovers where researchers are looking at the work done in the areas of MANETs and VANETs, but the jury is still out. As energy supply on UAVs is limited, protocols in various layers should contribute towards greening of the network. This article surveys the work done towards all of these outstanding issues, relating to this new class of networks, so as to spur further research in these areas.

I. INTRODUCTION … C. Motivation and key Issues

UAV networks promise useful public and civil applications, but their fluid topology, intermittent links, failures, and limited energy make reliable communications difficult. The survey identifies distinct requirements for self-organization, routing, disruption tolerance, handover, SDN control, and energy efficiency.

A. The growing importance of UAV networks

B. The challenges of UAV networks

  • Fluid topology, intermittent links, and node failures make stable UAV communications challenging and require designs beyond ordinary ad hoc mesh approaches.Nodes and links change as UAVs move, fail, or disappear, and the inter-UAV backbone may need to reorganize repeatedly.Source: B. The challenges of UAV networks · p. 2

C. Motivation and key Issues

  • UAVs move in two or three dimensions, sometimes in organized swarms, so MANET and VANET mobility models do not adequately describe them.Source: C. Motivation and key Issues · p. 2
  • UAV networks need energy-efficient routing and seamless handover as power-limited nodes disappear, links change, and users move between UAVs.Small UAVs may have power for only about thirty minutes of flight; the most effective routing protocol and handover procedure remain unsettled.Source: C. Motivation and key Issues · pp. 2–3
  • Frequent partitioning and long disconnections motivate delay- and disruption-tolerant networking with store-carry-forward capabilities.Deterministic protocols may depend on global network knowledge, while assuming random node behavior raises scalability issues.Source: C. Motivation and key Issues · p. 3
  • Autonomous multi-UAV networks require robust cooperation and self-organization despite link or node failures, alongside MAC and network-layer changes, SDN control, delay tolerance, and energy saving,.Source: C. Motivation and key Issues · p. 3
  • The survey addresses self-organization, disruption tolerance, SDN control, seamless handover, and energy efficiency, issues not emphasized by an earlier UAV flying ad hoc network survey.Source: C. Motivation and key Issues · p. 3

II. CHARACTERIZING THE UAV NETWORK · A. Multi-UAV network · B. Features of the UAV networks

Characterizing UAV networks requires accounting for their changing topology, links, mobility, and power constraints. The section describes multi-UAV benefits and complexities, then contrasts UAV network roles, architectures, and disruption conditions.

II. CHARACTERIZING THE UAV NETWORK

  • UAV network characterization considers topology change, partitioning, network lifetime, architecture, self-organization, protocols, and link quality.Source: §II · p. 3

A. Multi-UAV network

  • Multi-UAV systems use coordinated, smaller UAVs as relays or cooperative service providers, but mobility, failures, and power changes require links to reconfigure automatically.Source: A. Multi-UAV network · p. 3
  • Redundancy lets multi-UAV networks reorganize after a UAV fails; selective connections to control and servers reduce other UAVs’ communication burden.The passage also describes multi-UAV systems as less expensive to acquire, maintain, and operate than larger single-UAV counterparts.Source: A. Multi-UAV network · pp. 3–4
  • Effective multi-UAV task planning and coordination are outside this study’s scope; connectivity-constrained planning is NP-hard, with a polynomial-time heuristic proposed.Source: A. Multi-UAV network · p. 4

B. Features of the UAV networks

  • UAV networks may act like infrastructure with UAV base stations or form highly mobile ad hoc networks, and UAV nodes commonly serve as routers or sensor-data relays.Source: B. Features of the UAV networks · p. 4
  • Star topologies route UAV traffic through ground nodes, risking blockage, higher latency, and loss of inter-UAV communication if the ground center fails; mesh links can shorten paths.Source: B. Features of the UAV networks · pp. 4–5
  • Adapted mesh networks may reconfigure around broken paths, but UAV-specific mobility and link conditions can make MANET routing algorithms unreliable.Mesh networks interconnect nodes over multiple links and hops, while routing protocols must select among multiple routes.Source: B. Features of the UAV networks · p. 5
  • Disruptions are more likely with fast UAV mobility than hovering coverage, while poor link quality or unavailable end-to-end paths can delay data transmission.Source: B. Features of the UAV networks · p. 5

C. Categorization of UAV Networks · D. Self-organization in networks

The paper categorizes UAV networks by application-driven requirements and examines how mesh self-organization can maintain connectivity as nodes and links change.

C. Categorization of UAV Networks

  • UAV networks should be categorized by distinct applications and the mobility, architecture, routing, and control properties needed to support them.The paper points to Table IV as a generic categorization based on this application-first approach.Source: C. Categorization of UAV Networks · pp. 5–6
  • Communication-infrastructure UAVs may hover as stationary aerial access points, while sensing UAVs may move and military UAVs may move rapidly into enemy territory.The paper focuses mainly on infrastructure applications and, to some extent, sensing applications.Source: C. Categorization of UAV Networks · p. 6
  • For communication infrastructure, UAVs may be directly ground-controlled or mesh-linked through one or two control-center UAVs, acting as servers; sensing or attack UAVs act as clients.Source: C. Categorization of UAV Networks · p. 6
  • Stationary, coordinated infrastructure networks have lower delay or disruption probability, while node failures require reconfiguration and seamless session transfer and dynamic applications need disruption-aware routing.More dynamic applications may have intermittent links and require special routing, reconfiguration, and disruption-handling features.Source: C. Categorization of UAV Networks · p. 6

D. Self-organization in networks

  • Mesh networks self-form, reorganize after node failures, tolerate faults, and accept new nodes, supporting easy deployment and resilience.Source: D. Self-organization in networks · p. 7
  • Self-organization uses neighbor discovery, reorganization, connectivity and path establishment, service recovery, and energy management to maintain active-node connectivity over a multi-hop mesh.Node additions or removals prompt message exchanges that can cause medium-access collisions; energy management balances forwarding responsibilities and reduces battery consumption.Source: D. Self-organization in networks · p. 7
  • Link failures from interference, mobility, or bandwidth demand require reconfiguration, but autonomous reconfiguration also consumes computation and bandwidth and remains a major research challenge.Nodes monitor links and failures trigger reconfiguration; battery drain or communication failures can take UAVs out of service, requiring remaining nodes to reorganize and restore communication.Source: D. Self-organization in networks · p. 7

E. SDN-automating UAV network control · III. ROUTING · A. Routing issues to be resolved

The paper presents SDN as a way to make UAV networks more flexible to manage, while emphasizing that their routing must handle distinctive mobility, link, resource, and quality-of-service demands.

E. SDN-automating UAV network control

  • Limited communication resources, intermittent connectivity, impaired channels, and incompatible higher-layer protocols complicate UAV resource planning and cross-network operation.The passage notes that even nodes with the same access technology may not work across networks when routing protocols and network-management operations differ.Source: E. SDN-automating UAV network control · p. 7
  • SDN can program protocol stacks for different environments and help manage changing topology, node failures, intermittent links, and new services,.This flexibility is especially relevant when nodes using the same access technology cannot interoperate because their higher-layer protocols differ.Source: E. SDN-automating UAV network control · p. 7
  • SDN deployment in infrastructure-less wireless networks still lacks a clear, comprehensive account of its advantages and needed adaptations,.Prior concerns include separating controller and forwarding devices, security, control balance, and cooperation between controllers; SDN may help select paths and channels to reduce interference and improve resource use.Source: E. SDN-automating UAV network control · p. 7
  • OpenFlow is claimed to optimize resource use, automate operations, enable finer control and global policies, and speed new services; it separates forwarding from control,.Programmable switches use flow tables and communicate with controller(s).Source: E. SDN-automating UAV network control · pp. 7–8
  • SDN control may be centralized, distributed, or hybrid, with controllers updating switch flow tables and hybrid control delegating packet processing to local agents.Centralized control may be ground-based or airborne; distributed control places control on UAVs.Source: E. SDN-automating UAV network control · p. 8, Figure 3 · p. 8

III. ROUTING

  • UAV routing must handle networks ranging from slow-moving to fast-moving, node failures and replacement, intermittent or impaired links, and diverse reliability, delay, jitter, and bandwidth needs.Radios may switch off to conserve power at low load; earthquake data may require fully reliable transport, while related pictures and video may accept lower reliability but need stricter delay and jitter.Source: §III · p. 8

A. Routing issues to be resolved

  • Airborne routing adds location- and energy-awareness and robustness to intermittent links and topology change; existing ad hoc protocols may not fit, leaving a need for UAV-tailored protocols,.Protocols must address high mobility, dynamic topology, and different routing capabilities while improving delivery ratio and reducing delay and resource consumption; scalability, loop freedom, and energy conservation also remain issues.Source: A. Routing issues to be resolved · p. 9

B. Applicability of existing routing protocols

MANET and VANET routing protocols have been considered for UAV networks, but UAV-specific mobility, localization, intermittent links, power limits, and QoS needs motivate new routing algorithms for reliable communication. The section reviews static, proactive, reactive, hybrid, and geographic approaches and their limitations or potential adaptations.

Static Routing Protocols

  • Static routing has limited UAV applicability because fixed tables cannot adapt to changing conditions; LCAD suits delay-tolerant bulk transfer, while cluster-head changes can burden hierarchical routing.LCAD uses a UAV to carry data and can scale throughput with multiple UAVs, but single-UAV delivery delays are longer. Data-centric routing supports one-to-many transmission in cluster topologies.Source: B. Applicability of existing routing… · pp. 9–10

Proactive routing protocols

  • Proactive routing offers current routes but requires frequent updates that consume bandwidth and may react slowly to topology changes.OLSR's multipoint relays reduce control flooding, but route recalculation remains a concern; DSDV retains substantial update, computing, and storage burdens. BABEL failed to deliver in a UAV evaluation, while B.A.T.M.A.N. results vary by setting and comparator,,,.Source: B. Applicability of existing routing… · pp. 10–11

Reactive routing protocols

  • Reactive routing avoids periodic messaging but route discovery can add latency; AODV delays and bandwidth demands worsen with failures and intermittent links.RGR combines greedy geographic forwarding with reactive end-to-end paths and outperformed AODV in search missions for delay and packet delivery ratio. In one comparison, DSR had the least routing overhead while OLSR had the highest throughput.Source: B. Applicability of existing routing… · pp. 11–12

Hybrid routing protocols

  • Hybrid routing combines proactive intra-zone and reactive inter-zone routing, but dynamic UAV links complicate maintaining routing information and adjusting strategies.ZRP's efficiency depends strongly on zone radius and may add complexity for only slight improvement; TORA limits control-message propagation and finds new routes quickly after broken links, but can produce temporary invalid routes,.Source: B. Applicability of existing routing… · p. 12

Geographic 2-dimension and 3-dimension protocols

  • In 3D UAV networks, geographic routing's recovery from local minima is harder than in 2D; proposed methods use hull-based or randomized recovery, or triangulation.GHG routes on the hull, while tested randomized recovery methods favored bounded DFS in sparse networks and random walks in denser ones [133]. GDSTR-3D guarantees delivery with hop stretch close to 1; MDT offers guaranteed delivery on general 3D connectivity graphs and low routing stretch.Source: B. Applicability of existing routing… · pp. 12–13

C. Routing in networks prone to delays and disruptions

UAV networks face intermittent links and long partitions that can defeat TCP and conventional end-to-end routing. Delay-tolerant routing aims to maximize delivery probability while minimizing delay, using store-carry-forward and methods suited to changing contacts and limited resources.

Delay- and disruption-tolerant routing

  • Because intermittent links and long partitions exceed TCP thresholds and cause drops, UAV routing needs buffered store-carry-forward, selecting likely carriers until forwarding opportunities arise,,,.Multi-step request-response, acknowledgements, and timed-out transmission are unsuitable for long delays and disconnections.Source: C. Routing in networks prone to delays… · pp. 13–14

Location-aware and other routing considerations

  • Because UAV storage, power, and bandwidth are limited, geographic and beacon-less routing can conserve resources; Direct Delivery uses single-copy forwarding.Beacon messages consume bandwidth and energy, while stable routes may be cached to avoid unnecessary exchanges; delays of hours or days make this architecture unsuitable for real-time content.Source: C. Routing in networks prone to delays… · p. 14

Deterministic Routing

  • With predictable trajectories, deterministic routing uses single-copy unicast and fewer resources than stochastic methods, but is unsuitable for frequent topology changes [56].It selects time-dependent paths using known or learned node and link profiles.Source: C. Routing in networks prone to delays… · pp. 14–15

Estimation Based Approach

  • Estimation-based methods use encounter history or contact probabilities to choose when and where to forward; SEPR scales poorly at O(n^2), while RAPID accounts for bandwidth and storage [58],,.PROPHET assumes repeated past contacts make future contacts more likely.Source: C. Routing in networks prone to delays… · pp. 15–16

Stochastic Routing

  • Epidemic flooding spreads copies rapidly and robustly but consumes buffers, bandwidth, and power; Spray-and-Wait limits replication, while ant routing approaches epidemic delivery,.Epidemic routing performs well with large buffers, long contacts, or low network load; ant routing explores routes and exploits promising ones.Source: C. Routing in networks prone to delays… · p. 15

Node-Movement-Control Based Approaches

  • Movement-control methods reduce delays by changing trajectories or using ferry and DataMule carriers; UAV ferries can be directed to connect disaster-recovery partitions,.Ferries may follow preplanned routes or respond to service requests.Source: C. Routing in networks prone to delays… · p. 16

Coding Based Approaches

  • Network coding combines packets, while erasure coding adds decodable redundancy where retransmission is impractical; Hybrid Erasure Coding uses contact time to mitigate node-failure black holes,,.Erasure coding can improve worst-case delay but requires additional processing and power.Source: C. Routing in networks prone to delays… · p. 16

Social Networks Based Routing

  • Social routing can exploit groups, communities, popularity, and known locations, but popular nodes may be overloaded and contention may increase delay; the area needs further work,.Disaster-area or oil-platform communications may suit this approach.Source: C. Routing in networks prone to delays… · p. 16

IV. SEAMLESS HANDOVER

UAV networks need seamless handover as nodes move, fail, conserve power, or are withdrawn, so ongoing communication sessions can continue through network reconfiguration. Handover schemes and IEEE media-independent handover are discussed as aspects of this challenge.

IV. SEAMLESS HANDOVER

  • UAV networks must reconfigure and hand over ongoing voice, video, or data sessions to a working UAV according to predefined criteria.The need arises when nodes go out of service, shut down communication interfaces to conserve power, or are withdrawn to reduce network density.Source: §IV · p. 16
  • Handover is needed in both slow-dynamic disaster-area networks and mobile missions such as crop surveys that require sweeping area coverage.During prolonged missions, UAVs may also lose power or develop faults.Source: §IV · p. 16
  • Handover supports continuity of network communication with only a minor increase in message latency during the process.The section introduces handoff types, applicability of existing schemes and new developments, and IEEE media-independent handover.Source: §IV · p. 16

A. Handoffs in UAV networks

Handover supports continuity of service in wireless mesh networks: soft handover establishes the new link before dropping the old one, while hard handover breaks the old connection first. Handoffs are also classified as horizontal or vertical, and as mobile-device, network, or hybrid controlled.

A. Handoffs in UAV networks

  • Soft handover establishes the new connection before breaking the old one, preserving user sessions during migration; hard handover breaks the old connection first and interrupts active sessions.Figure 5 illustrates hard and soft handover.Source: A. Handoffs in UAV networks · p. 17, Figure 5 · p. 17
  • Horizontal handover moves a device between access points in the same network, whereas vertical handover transfers its connection between networks using different technologies.Figure 6 depicts both types.Source: A. Handoffs in UAV networks · p. 17, Figure 6 · p. 17
  • Hybrid handover involves both the mobile device and network; it has been studied in VANETs and mobile IP networks, but little work is available for UAV networks.Source: A. Handoffs in UAV networks · p. 17

B. Applicability of Existing Handover Schemes

Because UAV handover research is limited, the section considers MANET and VANET schemes while emphasizing that their mobility conditions and protocols do not fully address UAV network mobility. It describes handover mechanisms and identifies evaluation needs for adapting these schemes.

B. Applicability of Existing Handover Schemes

  • Existing handover work is limited even for VANETs: WAVE standards do not address network mobility, and practical studies of mobility protocols using WAVE are absent,.VANETs’ high speed, short communication times, topology changes, and partitioning motivate adapting MANET routing protocols and standards.Source: B. Applicability of Existing Handover… · pp. 17–18
  • VANET mobility can cause frequent route changes, partitioning, and handovers, while random waypoint poorly models vehicular movement; realistic models and further scheme design are needed.Handover latency and packet loss can degrade QoS; models should capture temporal and spatial dependencies, geographical restrictions, and changing node density. STRAW and Manhattan are cited as alternatives.Source: B. Applicability of Existing Handover… · p. 18
  • Proposed approaches include VIP-WAVE with PMIPv6, analytically shown to improve handover performance over standard WAVE, and cross-layer designs targeting packet loss and delay.MIPv4 and MIPv6 have also been applied to VANET handover, while NEMO provides mobility management but does not work well in vehicular networks.Source: B. Applicability of Existing Handover… · p. 18
  • PMIPv6-based mobility management performs better overall than MIPv6, especially with IEEE 802.11p, and can hand over between IEEE 802.11p networks without data loss at moderate speeds.MIPv6 handover latency remains a concern, and MIPv4, MIPv6, and HMIPv6 address terminal rather than network mobility; more rigorous evaluations across topologies and applications are needed,.Source: B. Applicability of Existing Handover… · p. 18
  • UAV handover transfers physical-layer connectivity, state, and user context between UAVs through network discovery, a transfer trigger, and execution.During discovery, a mobile station prioritizes access points using received signal strength; triggering may use criteria such as signal-to-noise ratio, and execution transfers the session to a new access point.Source: B. Applicability of Existing Handover… · p. 18

C. Media Independent Handover

IEEE 802.21 Media Independent Handover (MIH) supports handovers and interoperability across heterogeneous networks through layer-independent services. It could help integrate UAV communications across networks, though deployment requires both mobile devices and networks to implement the standard.

C. Media Independent Handover

  • IEEE 802.21 standardizes MIH services, with the Media Independent Handover Function between layer 2 wireless technologies and IP layer 3 for handovers and interoperability across IEEE and non-IEEE networks.Supported networks can use the same or different media, wired or wireless.Source: C. Media Independent Handover · p. 18
  • MIH supplies link-layer information and triggers to upper layers for handover decisions, but excludes intra-technology handover, policy, security, and link-layer-specific enhancements.The hybrid approach combines mobile-station measurements, such as signal-to-noise ratios, with network information; handover can be initiated by link-layer measurement reports and triggers.Source: C. Media Independent Handover · p. 19
  • For cross-jurisdiction emergencies, MIH’s unified interface and media-specific lower-layer interfaces can support interoperability between UAV networks and other technologies.Search-and-rescue or troop-assistance UAVs could use MIH to provide live video alongside WiFi, WiMAX, and LTE networks.Source: C. Media Independent Handover · p. 19
  • MIH remains nascent and has not been widely deployed or evaluated; operation requires both mobile devices and the network to implement the standard.Source: C. Media Independent Handover · p. 19
  • Combining IEEE 802.21 handover procedures with SDN/OpenFlow can preemptively establish paths for “always best” connectivity and avoid traffic disruption when multiple handover candidates exist.Source: C. Media Independent Handover · p. 19

V. ENERGY EFFICIENCY IN UAV NETWORKS

Communication equipment can substantially reduce UAV flight time or consume payload capacity, making energy efficiency important for network lifetime and useful payload. The section surveys power-saving approaches across protocol layers and notes that energy harvesting remains outside its scope.

Energy consumption

  • Communication equipment sharing the UAV battery can reduce rated flight time by 16%, and adding GPS and sensors can push the reduction beyond 20%.The wireless interface also consumes substantial energy while idle; a falling battery voltage can further inhibit normal UAV function before the battery is fully drained.Source: §V · p. 19

Separate communication battery

  • Eight separate router batteries provided about 8 hours of operation across as many as 19 sorties, but their 92g weight used about 30% of the UAV’s payload capacity.Reducing energy consumption could therefore increase network lifetime or the useful payload carried.Source: §V · pp. 19–20

Power-saving approaches

  • Energy saving can combine transmission-power control and dynamically sleeping unused nodes with physical-, data-link-, and network-layer mechanisms.The survey organizes representative tried or potential UAV protocols by protocol layer and energy-saving strategy, and also considers cross-layer approaches.Source: §V · p. 20

Scope limitation

  • Energy harvesting from kinetic, thermal, or electromagnetic sources is an evolving specialized area and is not included in the survey.Harvested-energy arrivals may require redesigning transmission algorithms, including separate queues for energy and data arrivals.Source: §V · p. 20

A. Energy Conservation in the Network Layer

The survey groups UAV network-layer energy conservation into path-, node-, coordinator-, and sleep-based approaches, while noting that no protocol class is established as best and applicability across UAV scenarios remains unproven.

Path selection based

  • Energy-aware paths include EMM-DSR and minimum-energy multi-hop routing, while FAR assumes static topology and incurs computation costs as UAV positions change.Minimum-energy routing can split high-power hops into smaller low-power hops; FAR requires frequent route computations as topology changes.Source: A. Energy Conservation in the Network… · pp. 20–21

Node Selection based

  • Node-selection schemes can exclude low-energy relays or favor high-energy nodes; DEAR reuses RREQs and simulations report longer network lifetime and better delivery ratio.LEAR drops RREQs when a node's residual battery is below a threshold, while DEAR needs no extra control packets to gather energy information.Source: A. Energy Conservation in the Network… · p. 21

Cluster head or coordinator based

  • CBRP saves energy through coordinator-led clusters, gateway selection accounts for node movement, while GAF risks UAV-network partitioning as nodes move.Gateway selection uses dynamic network partitioning and stability of two-hop neighbors; GAF's connected set may not persist as UAVs move.Source: A. Energy Conservation in the Network… · pp. 21–22

Sleep based

  • CDS keeps a connected dominating set active while other nodes sleep; EAR accepts suboptimal, potentially longer-hop routes to save energy and switch components off.EAR accounts for equipment energy use in routing and traffic engineering, including energy costs alongside QoS constraints.Source: A. Energy Conservation in the Network… · p. 22

B. Energy Conservation in the Data Link Layer

Data-link energy conservation centers on MAC-layer access coordination, where power-saving schedules can reduce energy use but add delivery delay. The surveyed approaches use single- or dual-radio duty cycling, topology control, and clustering to address energy waste from collisions, overhead, listening, and overhearing.

Duty cycle with single radio

  • Single-radio duty-cycling protocols use sleep–activity schedules for both signaling and data, trading lower idle-listening energy for possible throughput or delay costs.S-MAC uses fixed schedules, while T-MAC adapts them after inactivity; SOFA maximizes clients’ sleep time, and MT-MAC uses TDMA slots to reduce collisions and increase throughput,,,,,.Source: B. Energy Conservation in the Data… · p. 23

Duty cycle with dual radio

  • Dual-radio duty cycling separates wake-up signaling from data transfer, with protocols seeking to balance energy efficiency against wake-up latency.STEM keeps its data radio asleep until woken; PTW shortens wake-up time, LEEM synchronizes nodes along a route for sequential waking, and PAMAS turns off radios while the medium is unavailable,,,.Source: B. Energy Conservation in the Data… · p. 23

Topology/power control based

  • Topology-control approaches conserve power by adjusting transmit power or selecting which nodes remain active, while managing connectivity and interference.XTC ranks neighbors by link quality without location data; LFTC chooses power levels to avoid hidden-terminal collisions and is reported to improve hop count and reliability over XTC; PEM schedules transmissions using estimated distances and interference relations.Source: B. Energy Conservation in the Data… · pp. 23–24

Cluster based MAC protocols

  • Cluster-based MAC protocols group nodes under cluster heads or coordinators, enabling non-cluster nodes to use power-saving sleep schedules.LEACH’s heads aggregate cluster data to improve network lifetime, though independent head selection may drain some nodes faster; PEGASIS instead forms a chain of cluster heads for aggregation and selects a new leader each round,.Source: B. Energy Conservation in the Data… · p. 24

C. Energy conservation in the physical layer

The section surveys four physical-layer energy-conservation approaches for UAV networks, where high mobility and three-dimensional propagation complicate reliable communication. It highlights challenges including changing distances and directions, antenna patterns, shadowing, environmental conditions, interference, and jamming.

Physical-layer challenges

  • Three-dimensional UAV mobility complicates physical-layer design, with changing distances and directions, antenna patterns, shadowing, environmental conditions, interference, and jamming affecting communication.The authors suggest spatial and temporal diversity to mitigate high link loss and variation, and call for more research on physical-layer and antenna propagation in 3D environments.Source: C. Energy conservation in the physical… · p. 25

Dynamic voltage control

  • Dynamic voltage control reduces circuit voltage according to traffic or workload to save energy and potentially extend UAV flight time.DVS lowers bias voltage at low traffic loads, yielding proportionally quadratic energy savings; lower voltage can also help payload circuits continue operating as battery voltage drops.Source: C. Energy conservation in the physical… · p. 25

Node level power scheduling

  • Node-level power scheduling saves energy by adjusting transmission power, topology, or device activity while maintaining connectivity.LM-SPT balances energy efficiency and throughput using one-hop neighborhood information; minimum-energy topology reduces power to cover nearest neighbors, while CPLDs can power down idle processors and wake them on beacon detection.Source: C. Energy conservation in the physical… · p. 25

Choosing minimum subset

  • Minimum-subset approaches switch on only the devices or base stations needed to meet coverage, capacity, and traffic demands.Virtual WLAN consolidates hardware into low-power modes, while the time-variable mesh approach dynamically selects active base stations and aims to make energy use proportional to load.Source: C. Energy conservation in the physical… · p. 25

Sleeping schedules with buffering

  • Sleeping schedules with buffering extend node sleep periods by storing packets that arrive while interfaces are inactive.SRA buffers packets during short idle periods so clients can sleep longer; routers and switches coordinate sleep and burst clearing to limit delay penalties and avoid noticeably amplifying loss.Source: C. Energy conservation in the physical… · p. 26

D. Energy Conservation through Cross-Layer Protocols

Protocols optimized within individual layers may conserve energy inefficiently across the network, while cross-layer protocols for UAV networks remain an open issue. The section describes three representative approaches to cross-layer energy efficiency.

D. Energy Conservation through Cross-Layer Protocols

  • Cross-layer energy conservation remains an open issue for UAV networks because optimizing one layer may yield inefficient energy use across layers.Network-layer protocols address connectivity, MAC protocols collisions and duty cycles, and physical-layer protocols device characteristics and power control.Source: D. Energy Conservation through… · p. 26
  • MTEC selects routes using transmission success, node traffic, channel contention, and residual energy to reduce consumption and extend network lifetime.ACW dynamically adjusts back-off time based on transmission success, reducing retransmissions by allowing more successful links greater channel access.Source: D. Energy Conservation through… · p. 26
  • CLEEP combines physical-layer transmission-power and neighbor-connectivity information with network-layer routing to set MAC-layer sleep-activity patterns and maximize sleep time.Source: D. Energy Conservation through… · p. 26
  • CLE2aR2 estimates channel quality from received signal strength and accounts for interference and retransmission energy to find energy-efficient, reliable routes.Source: D. Energy Conservation through… · p. 26

VI. CONCLUSIONS

UAV networks’ distinctive dynamics and constraints make reliable connectivity challenging. This survey classifies UAV networks and reviews routing, SDN, seamless handover, and energy efficiency as important research areas.

VI. CONCLUSIONS

  • UAV networks’ fluid topology, intermittent links, and power and bandwidth constraints distinguish them from previously studied networks and motivate cross-layer design for energy conservation and quality of service.The survey notes that some researchers argue for rebuilding features across the physical, data-link, network, and transport layers.Source: §VI · p. 27
  • The survey addresses understudied UAV-network issues by classifying network characteristics and examining routing, SDN-based control, seamless handover, and energy efficiency.It distinguishes infrastructure and ad-hoc, server- and client-oriented, star and mesh networks, and deployments hardened against delays and disruptions.Source: §VI · pp. 27–28
  • Routing must accommodate UAV mobility, dynamic topology, intermittent links, bandwidth and energy constraints, yet native UAV protocols remain unavailable.Static and proactive protocols have limitations; reactive route discovery can compound delays, while hybrid protocols may add complexity. RGR adaptations have performed better, and delay- and disruption-prone networks may use store-carry-forward routing.Source: §VI · p. 27
  • For partitioned or disruption-prone networks, delay-tolerant store-carry-forward approaches can relay messages when end-to-end connectivity is unavailable.Options include epidemic routing, Spray and Wait, controlled ferrying for disconnected partitions, and social-network models for nodes at known locations.Source: §VI · p. 27
  • UAV seamless handover remains little studied; IEEE 802.21 MIH supports interoperability across IEEE-802 and non-IEEE-802 networks but is nascent and lacks several handover capabilities.MIH does not provide intra-technology handover, handover policies, security, or link-layer enhancements.Source: §VI · p. 27
  • Energy efficiency can extend network lifetime and preserve useful payload through transmission-power control, load distribution, and node sleep strategies.The survey describes reductions at the physical, network, and data-link layers, as well as cross-layer schemes spanning multiple layers.Source: §VI · p. 27

Related papers

Loading 1603.08462v1…