PaperlayerView exact revision on arXiv

Source-linked AI summary

A Comprehensive Survey on 6G Networks:Applications, Core Services, Enabling Technologies, and Future Challenges

Amin Shahraki, Mahmoud Abbasi, Md. Jalil Piran, Amir Taherkordi

arXiv:2101.12475v2cs.NI

TL;DR

Growing IoT deployments and emerging services motivate a survey of how 6G may address network demands. The paper synthesizes 6G requirements, applications, services, enabling technologies, and challenges, concluding with several unresolved challenges for future 6G networks.

  • Problem

    The survey addresses the demands associated with growing IoT deployments and services, including smart environments and massive URLLC applications.

    Source: A. The Key 6G KPIs and Use-cases · p. 7, B. New Service Classes for 6G · p. 9

  • Method

    The paper surveys 6G requirements and trends, research activities, key performance indicators, applications, services, enabling technologies, and challenges.

    Source: §I · p. 2, §VIII · p. 18

  • Results

    The survey presents potential 6G requirements, trends, applications, services, enabling technologies, and several unresolved challenges.

    Source: §VIII · p. 18

  • Takeaways & Limitations

    The paper concludes that future 6G networks have potential unresolved challenges across the areas it surveys.

    Source: §VIII · p. 18

Abstract

from arXiv · show

Cellular Internet of Things (IoT) is considered as de facto paradigm to improve the communication and computation systems. Cellular IoT connects massive number of physical and virtual objects to the Internet using cellular networks. The latest generation of cellular networks, e.g. fifth-generation (5G), use evolutionary and revolutionary technologies to notably improve the performance of wireless networks. However, given the envisioned new use-cases, e.g., holographic communication, and the ever-increasing deployment of massive smart-physical end-devices in IoT, the volume of network traffic has considerably raised, and therefore, the current generation of mobile networks cannot wholly meet the ever-increasing demands. Hence, it is envisioned that the next generation, sixth generation (6G) networks, need to play a critical role to alleviate such challenges in IoT by providing new communication services, network capacity, and ultra-low latency communications (uRLLC). In this paper, first, the need for 6G networks is discussed. Then, the potential 6G requirements and trends, as well as the latest research activities related to 6G are introduced e.g., Tactile Internet and Terahertz (THz). Furthermore, the key performance indicators, applications, new services, and the potential key enabling technologies for 6G networks are presented. Finally, several potential unresolved challenges for future 6G networks are presented.

I. INTRODUCTION

The survey argues that growing IoT traffic and emerging services exceed what current cellular networks can wholly support. It reviews the case for 6G and surveys its requirements, use-cases, enabling technologies, and challenges.

I. INTRODUCTION

  • IoT’s growing device population and data volume motivate new cellular generations,.The introduction cites a Cisco prediction of 14.7 billion IoT-connected devices by 2023.Source: §I · p. 1
  • 5G improves data rates, energy efficiency, reliability, and connection density, but the survey says it may not meet future demands –.The introduction cites growing IoT deployments, new services, and the complexity of future networks among the concerns; it notes that manual optimization and configuration are no longer appropriate, [13]–.Source: §I · pp. 1–2
  • Emerging services such as XR and telemedicine require high transmission rates, reliability, and low latency simultaneously, beyond 5G’s original goals –.The introduction also identifies a need for 10 million connections per km2 and improved coverage as IoT device deployments grow,.Source: §I · p. 1
  • The survey reviews 6G requirements and trends, recent research activities, key performance indicators, applications, service classes, enabling technologies, and unresolved challenges.It also outlines future research directions.Source: §I · p. 2
  • The survey presents its scope as a review of existing 6G work and a discussion of potential requirements, trends, technologies, and future challenges.Its stated contributions include 6G key performance indicators and use-cases such as holographic communication and industrial automation.Source: §I · p. 2

II. RELATED SURVEY ARTICLES

The section reviews prior 6G surveys and compares their coverage across core aspects of the field. The authors position their survey as a broader treatment and organize it around seven questions.

Prior survey coverage

  • Prior 6G surveys cover differing combinations of requirements, applications, enabling technologies, architecture, services, and challenges.The review describes examples spanning user-focused requirements, architecture and AI, applications and open challenges, and core services.Source: §II · p. 3

Guiding questions

  • Seven questions frame the survey, from the need for 6G and its requirements to enabling technologies and open challenges.They also ask about research activity, KPIs, applications, and new service classes.Source: §II · p. 4

Coverage comparison

  • The authors assess existing surveys using seven factors and report that most do not fully cover all 6G aspects.The factors are requirements, trends, vision, service classes, architecture, applications, and challenges.Source: §II · pp. 3–4

Survey scope

  • This survey aims to cover requirements, trends, visions, services, architectures, applications, challenges, and enabling technologies.The authors describe this coverage as extending beyond prior surveys, while aiming to address the aspects to an acceptable extent.Source: §II · p. 4

III. 6G REQUIREMENTS AND TRENDS

The section outlines 6G requirements and trends in response to anticipated demands from future services and increasingly complex networks. It covers spectrum, latency, connectivity, energy, positioning, coverage, and network autonomy.

Spectrum, data rate, and latency

  • 6G is expected to need higher frequency bands, very high opportunistic peak data rates, and end-to-end delay below 1 ms.The listed candidate bands are 73 GHz, 140 GHz, 1 THz, and 3 THz; XR and telepresence are cited as latency-sensitive services.Source: §III · p. 5

Connectivity and reliability

  • Massive machine-type communications must address device scale, connectivity, coverage, and QoS and QoE provisioning.Mission-critical use cases also require fast connectivity and 99.999% reliability or availability, which the paper associates with super MTC (sMTC).Source: §III · pp. 5–6

Self-X network

  • 6G's complexity and flexibility demands motivate ML-supported self-learning, self-configuration, optimization, healing, and protection.The paper presents these capabilities as part of a self-X network trend.Source: §III · p. 6

Precision operation and positioning

  • Super-precision services such as tele-surgery require guaranteed delivery timing and positioning accuracy beyond GPS's meter-scale errors.The paper gives sub-millimeter positioning as a future-service requirement.Source: §III · p. 6

Scalability and energy efficiency

  • The section identifies scalability and energy efficiency as requirements for networks connecting diverse IoT devices, with ML and energy harvesting among the discussed approaches.ML-enabled devices are described as extracting knowledge locally and reducing raw-data transmission; energy harvesting is under investigation for energy efficiency.Source: §III · p. 6

Coverage

  • Three-dimensional coverage and satellite integration are presented as ways to support holographic services and global connectivity.The section describes terrestrial and aerial devices for 3D services, and satellites for localization, broadcast, and Internet connectivity.Source: §III · p. 6

Network management

  • Dynamic network configuration can draw on SDN, NFV, and network slicing over shared physical infrastructure.The section describes slicing as providing on-demand virtual networks.Source: §III · p. 6

IV. 6G RESEARCH ACTIVITIES AND MOTIVATION

The section surveys 6G research initiatives and the requirements of envisioned applications, whose demands include high data rates, reliability, and low latency. It also outlines new use cases and services.

IV. 6G RESEARCH ACTIVITIES AND MOTIVATION

  • Research efforts include ITU’s FG NET-2030, 6Genesis, TERRANOVA, and initiatives in the United States, China, and South Korea.The projects address future requirements and enabling technologies, including THz links.Source: §IV · pp. 6–7

A. The Key 6G KPIs and Use-cases

  • Industrial automation calls for reliability above 1–10−9, latency below 1 ms, multiple links, and, for some control operations, 1µs delay jitter.The section also envisions connected robotics, drones, and autonomous cars.Source: A. The Key 6G KPIs and Use-cases · p. 7
  • Smart environments combine ICT with many IoT devices for services such as security, waste management, transport, and utilities; e-health requires 99.99999% reliability and latency below 1 ms.The text says these applications can be only partially realized in the 5G era because of stringent requirements.Source: A. The Key 6G KPIs and Use-cases · p. 7
  • Tactile Internet enables real-time human-machine exchange of control, touch, and sense data; private cellular networks are proposed for industrial use cases including automation and remote operation.The tactile Internet requires ultra-real-time response, very low latency, and reliable communication; private 6G is described as an evolution path from private 5G.Source: A. The Key 6G KPIs and Use-cases · pp. 7–8
  • Figure 6 schematically classifies envisioned 6G services and use-cases, while Figure 5 represents relative requirement scores for seven applications by ball diameter.Figure 5 is adopted from [48]; Figure 6 is attributed to [31].Source: A. The Key 6G KPIs and Use-cases · p. 7, Figure 6 · p. 8, Figure 5 · p. 8
  • Remote surgery is described as requiring latency below one millisecond, while holographic communication may require up to 4 Tb/s, submillisecond latency, and reliable links.Holographic telepresence adds touch to audiovisual communication across locations.Source: A. The Key 6G KPIs and Use-cases · pp. 7–8, Figure 4 · p. 8

B. New Service Classes for 6G

The section describes new 6G service classes as refinements or extensions of 5G classes, aimed at application demands that combine connectivity, latency, reliability, and data-rate requirements.

B. New Service Classes for 6G

  • Massive URLLC combines 5G URLLC with mMTC to support many devices in mission-critical settings such as autonomous driving.It must jointly support low latency, high reliability, high data rates, massive connectivity, and full mobility; OMA, NOMA, and contention-based access offer different trade-offs.Source: B. New Service Classes for 6G · pp. 8–9
  • Enhanced mobile broadband URLLC targets high-rate, low-latency, reliable service even under high mobility, with energy efficiency, security, privacy, and resource sharing remaining concerns.The passage identifies coexistence of uRLLC and eMBB through resource sharing as a significant technical challenge.Source: B. New Service Classes for 6G · p. 9
  • Massive eMBB is proposed for large-scale IIoT, supporting low-latency connectivity among workers, sensors, and actuators at high connection density.The passage gives Industry 4.0 scenarios as an example, with 100 connections in m3.Source: B. New Service Classes for 6G · p. 9
  • Other proposed 6G classes include computation-oriented communication, event-defined uRLLC, ultra-high-speed low-latency communication, and human-centric services.The cited classifications also include ultra-massive machine-type communications, ultra-low-power transmissions, and control, localization, and sensing,,,.Source: B. New Service Classes for 6G · p. 9

V. EVOLUTIONARY TECHNOLOGIES OF 6G

The section distinguishes evolutionary technologies that extend existing approaches from revolutionary technologies that introduce novel ones. It presents non-terrestrial communications as a way for 6G to address coverage and capacity gaps.

V. EVOLUTIONARY TECHNOLOGIES OF 6G

  • Evolutionary 6G solutions build on existing or previously adopted technologies, whereas revolutionary solutions exploit novel technologies and may fundamentally change network layers.The passage notes that many revolutionary technologies remain at the scientific-investigation stage.Source: §V · p. 9

A. Non-Terrestrial Technologies

  • 6G is expected to integrate UAV-assisted communications and satellites with cellular networks to extend coverage and provide high-capacity connectivity.The stated motivation includes coverage limitations and vulnerability to natural and human-made disasters in legacy terrestrial networks.Source: A. Non-Terrestrial Technologies · p. 9
  • UAVs may serve as aerial users using existing base stations or as aerial base stations and relays that can be deployed quickly for emergencies and rescue operations.Aerial base stations may be more likely than ground-based ones to establish radio links with terrestrial users.Source: A. Non-Terrestrial Technologies · p. 10
  • UAV communications face differing command/control and payload QoS needs, severe line-of-sight-driven air–ground interference, and size, weight, and power constraints.The passage identifies these as significant issues for UAV-based communications.Source: A. Non-Terrestrial Technologies · p. 10
  • Satellite communication is presented as an option for global IoT connectivity, particularly for remote areas beyond terrestrial network coverage.The passage lists reliability, broad coverage, protection, and fast rural deployment among satellite-assisted IoT advantages.Source: A. Non-Terrestrial Technologies · p. 10

B. AI and 6G

The section presents AI as a 6G foundation for network design, optimization, and operations. It describes analytics applications and automated feedback-loop optimization for dynamic networks.

B. AI and 6G

  • 6G is expected to use AI beyond classification and prediction, including for reasoning, decision-making, and design and optimization across network architecture, operations, and protocols.The passage anticipates ubiquitous AI services from the core to end devices.Source: B. AI and 6G · p. 10
  • AI-based approaches are being studied across communication-system layers, with air-interface design and optimization identified as a predicted area of significant impact.The discussion connects this research to anticipated KPIs and 6G challenges.Source: B. AI and 6G · p. 10
  • AI may make network slicing more adaptive because 6G’s heterogeneous network and new radio interfaces exceed what network softwarization alone can address.The text calls for further improvement of SDN, NFV, and network slicing technologies.Source: B. AI and 6G · pp. 10–11
  • AI applications include descriptive, diagnostic, predictive, and prescriptive analytics for network data and operations.The passage describes these as covering historical network conditions, fault and anomaly identification, forecasting, and decisions such as resource allocation and caching placement.Source: B. AI and 6G · p. 11
  • The section expects AI-enabled closed-loop optimization to help address 6G complexity where conventional mathematical and dynamic-programming approaches are described as inapplicable.Deep reinforcement learning can build a feedback loop in which an agent iteratively improves using system feedback.Source: B. AI and 6G · p. 11

C. Energy Harvesting

The section surveys energy harvesting, intelligent surfaces, NOMA, and D2D as approaches to 6G connectivity, efficiency, and service demands. It also identifies practical challenges and potential roles for these technologies.

C. Energy Harvesting

  • 6G energy harvesting must address conversion efficiency and protocol coexistence while supporting massive low-power and batteryless devices.The paper identifies energy efficiency and harvesting or wireless information and power transfer as potential routes; THz directionality may benefit WIPT.Source: C. Energy Harvesting · p. 11

D. Large Intelligent Surfaces (LIS)

  • Intelligent surfaces can alter radio-wave reflections and are proposed to ease the power, processing-complexity, and hardware-cost burdens of deploying massive MIMO with THz.LIS uses artificial electromagnetic metasurfaces as large antennas to increase capacity or support single-anchor localization; its elements need no active power source to transmit data.Source: D. Large Intelligent Surfaces (LIS) · pp. 11–12

F. Non-orthogonal multiple access (NOMA)

  • NOMA uses successive interference cancellation and decoding order to improve spectral efficiency, security, secrecy capacity, and user fairness over OMA.The paper also identifies NOMA as relevant to mMTC and massive URLLC, and notes that NOMA-assisted MEC merits further study.Source: F. Non-orthogonal multiple access (NOMA) · p. 12

G. Device-to-device (D2D) Communication

  • D2D can pool idle UE resources for edge computing and support low-latency links, edge network slicing, and NOMA communication with multiple receivers.The paper envisions near-real-time links between nearby UEs using THz-band D2D.Source: G. Device-to-device (D2D) Communication · p. 12

H. Grant-free Transmission

The section presents grant-free access, sparse signal processing, holographic MIMO surfaces, and THz communications as 6G technologies. It describes both their intended roles and key constraints.

H. Grant-free Transmission

  • Grant-free access lets IoT devices choose time-frequency resources automatically, reducing scheduling-request overhead and latency, while GF-NOMA is proposed for capacity-limited mMTC and URLLC access.Conventional centralized NOMA scheduling can degrade under small, periodic, mass uplink traffic and varied QoS requirements.Source: H. Grant-free Transmission · pp. 12–13

I. Sparse signal processing

  • Sparse signal processing can identify active IoT devices for grant-free decoding and recover sparse THz channels, supporting spectrum and energy efficiency.Approximate message passing is cited as a method investigated for THz channel estimation [85].Source: §I · p. 13

J. Holographic MIMO Surfaces

  • HMIMOS is proposed to address massive-MIMO size, cost, weight, and energy concerns, with possible roles as receiver, transmitter, or reflector.Applications described include outdoor beamforming, user-to-base-station links, physical-layer security and wireless power transfer, plus indoor positioning and coverage enhancement.Source: J. Holographic MIMO Surfaces · p. 13

A. THz Communications

  • The THz band spans 100 GHz to 10 THz and is expected to offer 100 Gbp/s data rates, bandwidth above 50 GHz, dense connectivity, and high-resolution sensing.Long-distance use is constrained by attenuation, channel fluctuation, and propagation loss; large antennas and high computational power are also engineering challenges.Source: A. THz Communications · p. 13

B. Optical Wireless Technology

The section surveys optical wireless communications, 3D coverage, and edge intelligence as 6G technologies. It outlines their potential applications alongside practical constraints and unresolved issues.

B. Optical Wireless Technology

  • Visible-light communication offers unlicensed THz-level bandwidth, avoids electromagnetic interference, and confines transmission indoors, which can support privacy-sensitive short-range settings.It can use illumination sources as base stations without expensive dedicated base stations; reported GaN-based LED rates reach 4 Gbp/s, while hundreds of Gbp/s for VLC are projected.Source: B. Optical Wireless Technology · p. 14

C. 3-Dimensional (3D) Network Architecture

  • 3D network coverage could provide timely, economically efficient connectivity in rural areas and disasters, but real-world deployment requires channel models, trajectory optimization, resource management, and topology solutions.The envisioned architecture integrates terrestrial and non-terrestrial technologies to extend coverage to deep-sea and high-altitude settings.Source: C. 3-Dimensional (3D) Network… · p. 14

D. Edge Intelligence (EI)

  • Edge intelligence combines edge computing and AI to process data near its source, aiming to reduce cloud-transfer costs and delays while addressing privacy and resource constraints.Federated learning trains models on device-resident data and sends model updates to a server for aggregation; Figure 7 depicts a typical FL architecture. FL still faces communication-cost, privacy, security, and scale-related training and inference challenges [101] [102].Source: D. Edge Intelligence (EI) · pp. 14–15, Figure 7 · p. 15

E. Quantum Communications

The section reviews quantum and cell-less communications and identifies energy and AI-related challenges for 6G. It presents potential capabilities alongside resource, privacy, and deployment concerns.

E. Quantum Communications

  • Quantum communications encode data in qubits and may offer secure, high-speed, low-loss links, while quantum repeaters divide long links into shorter segments for error correction.The paper notes investigations of quantum key distribution access networks and quantum switches.Source: E. Quantum Communications · p. 15
  • Quantum computing is presented as a potential way to reduce the energy, time, and resource costs of AI techniques, particularly deep learning.The paper attributes the concern to AI workloads growing faster than traditional chipset capabilities.Source: E. Quantum Communications · p. 15

F. Cell-less architecture

  • Cell-less communication coordinates multiple base stations or access points so users can select among heterogeneous links without conventional handover, potentially reducing handover latency and data loss.Source: F. Cell-less architecture · p. 15

A. Energy Consumption Challenges

  • The expected scale of low-power and batteryless 6G devices makes energy efficiency and power supply mechanisms central challenges, including harvesting, efficient modulation, and dynamic energy management.The paper notes current smartphone battery life is almost one day and proposes investigating energy-management trade-offs between demand and supply.Source: A. Energy Consumption Challenges · p. 16

B. AI-related Challenges

  • AI-related 6G challenges include privacy, security, link capacity and latency, as well as dividing scarce resources between ML workloads and ML-based network management.Federated learning keeps raw data at distributed sites while participants send model updates to a central server; both ML on 6G and ML for 6G are resource-hungry.Source: B. AI-related Challenges · p. 16

C. Terahertz Communications Challenges

The survey identifies technical challenges spanning Terahertz communications, 3D coverage, security, and the Tactile Internet. It also describes research and design needs for addressing them.

C. Terahertz Communications Challenges

  • Terahertz links face severe propagation loss and limited long-distance communication, alongside MAC, hardware, circuit, antenna, and transceiver-design challenges.MAC challenges include deafness, network discovery and coupling, and concurrent transmission scheduling.Source: C. Terahertz Communications Challenges · p. 16

D. 3D Coverage Challenges

  • 3D coverage requires air-to-ground channel modeling and measurement, topology optimization, network planning, and tools for energy, mobility, trajectory, and resource management.Source: D. 3D Coverage Challenges · pp. 16–17

E. Security Challenges

  • 6G security faces fading security boundaries from virtualization and softwarization, edge data risks from AI and analytics, and concerns that traditional IoT protocols may not remain secure.Source: E. Security Challenges · p. 17

F. Tactile Internet Challenges

  • The Tactile Internet remains in its infancy, with open physical-layer, signaling, latency, routing, network-coding, and security challenges.The survey identifies waveform selection, robust modulation, and an intelligent control plane among the issues needing further research.Source: F. Tactile Internet Challenges · p. 17

G. Random Access Protocols

The survey describes challenges in random access, privacy, green networking, and management for 6G and IoT. It concludes by synthesizing 6G requirements, services, technologies, and unresolved challenges.

G. Random Access Protocols

  • Massive IoT access and sporadic, unpredictable traffic make resource allocation difficult, while existing random-access protocols are not optimal for networks with many accessing nodes.The survey identifies cooperation between random access and NOMA, OFDM, massive MIMO, and sparse signal processing as a potential route to more efficient systems.Source: G. Random Access Protocols · p. 17

H. Privacy concerns in the future smart services

  • Smart services may use sensitive household data, making de-identification standards and privacy assessment important; blockchain is a potential 6G privacy solution with few efforts reported.Source: H. Privacy concerns in the future… · p. 17

I. Green 6G

  • 6G aims for ubiquitous connectivity while managing diverse QoS needs, heterogeneous traffic, low power consumption, and energy-efficient communications, computing, and transmitter/receiver technologies.The survey describes green communication and computing as primary goals for reducing power consumption and operating costs.Source: §I · p. 18

J. Network Management and Orchestration (NMO), NTMA, and 6G

  • Research on 6G network management and traffic management analysis remains limited, especially for complex infrastructures such as D2D-enabled and massive-IoT networks.The survey identifies machine learning for these areas as a future research direction.Source: J. Network Management and… · p. 18

VIII. CONCLUSION

  • The paper surveys 6G as a next communication paradigm for IoT, covering its need, requirements, trends, KPIs, applications, services, enabling technologies, and unresolved challenges.Source: §VIII · p. 18

Related papers

Loading 2101.12475v2…