Source-linked AI summary
An Overview of Sustainable Green 5G Networks
Qingqing Wu, Geoffrey Ye Li, Wen Chen, Derrick Wing Kwan Ng, Robert Schober
TL;DR
The paper addresses how 5G can meet stringent traffic and latency requirements without unsustainable energy consumption. It surveys green 5G techniques and energy harvesting, concluding that sustainable networks require an ecosystem of interoperable technologies rather than any single radio access technology.
Problem
5G’s projected traffic growth and one-millisecond latency requirement make limiting its potentially tremendous energy consumption a central design challenge.
Method
The article provides an overview of green 5G technologies and energy harvesting for communication, while identifying technical challenges and potential research topics.
Results
80% of base stations are lightly loaded for 80% of the time but still consume almost their peak power.
Takeaways & Limitations
Sustainable green 5G requires interoperable technologies working as an ecosystem, with energy harvesting underpinning the green vision.
Takeaways & Limitations
Sustainable green 5G cannot be achieved with any particular radio access technology alone.
Abstract
from arXiv · showhide
The stringent requirements of a 1,000 times increase in data traffic and one millisecond round trip latency have made limiting the potentially tremendous ensuing energy consumption one of the most challenging problems for the design of the upcoming fifth-generation (5G) networks. To enable sustainable 5G networks, new technologies have been proposed to improve the system energy efficiency and alternative energy sources are introduced to reduce our dependence on traditional fossil fuels. In particular, various 5G techniques target the reduction of the energy consumption without sacrificing the quality-of-service. Meanwhile, energy harvesting technologies, which enable communication transceivers to harvest energy from various renewable resources and ambient radio frequency signals for communi- cation, have drawn significant interest from both academia and industry. In this article, we provide an overview of the latest research on both green 5G techniques and energy harvesting for communication. In addition, some technical challenges and potential research topics for realizing sustainable green 5G networks are also identified.
I. INTRODUCTION
5G must accommodate explosive traffic and device growth while limiting rising energy consumption. This overview surveys green 5G technologies and energy harvesting as complementary routes toward sustainable networks.
- 5G is expected to support up to a 1,000-fold capacity increase and connect at least 100 billion devices.
- Three throughput paradigms are emphasized: shorter Tx-Rx distances, unused-spectrum exploitation, and massive antenna deployments.These correspond to UDNs and D2D, mmWave and LTE-U, and M-MIMO, respectively.
- These technologies can increase throughput but may raise energy use through antenna-scaled RF-chain power, power-hungry transceivers, and complex signal processing.
- Energy efficiency, measured in bits-per-Joule, is a prominent green design metric motivated by operating cost, terminal lifetime, and environmental concerns.
- EE improvements alone are insufficient for sustainable 5G because Shannon capacity and circuit-power constraints limit further gains amid explosive data-rate requirements.
II. GENERAL EE-SE TRADEOFF
Energy efficiency and spectral efficiency exhibit a fundamental tradeoff shaped by transmit and circuit power. With circuit power included, energy efficiency peaks at a spectral-efficiency threshold, while system configurations and energy harvesting can shift this tradeoff.
- Spectral efficiency can be improved by increasing frequency reuse, signal bandwidth, or the number of spatial beams.These correspond to increasing K, B, or N while considering link distance and SINR.
- Most surveyed 5G technologies reduce transmit power while adding circuit power through hardware expansion or sophisticated signal processing.
- With Pc = 0, energy efficiency decreases monotonically as spectral efficiency increases.
- With Pc > 0, energy efficiency increases with spectral efficiency below a threshold and decreases beyond it.
- As spectral efficiency increases, energy efficiency eventually converges to the same value regardless of circuit power because transmit power dominates.
- Reducing circuit power enlarges the energy-efficiency–spectral-efficiency tradeoff region, which can also be reached through spectrum, reuse, multiplexing, and power-allocation choices.
III. GREEN 5G TECHNOLOGIES: ENLARGING SPECTRUM AVAILABILITY
Expanding spectrum availability through mmWave and LTE-U can increase throughput and spectral efficiency, but energy-efficient designs must address hardware, propagation, and processing costs. The section highlights hybrid transceivers and low-resolution ADCs as approaches with explicit efficiency–performance tradeoffs.
- mmWave and LTE-U expand cellular spectrum by using mmWave bands and unlicensed 5 GHz spectrum, increasing available bandwidth B.
- Millimeter Wave: mmWave systems face high path attenuation, blockage, atmospheric absorption, and rain absorption that complicate guaranteed QoS.
- Millimeter Wave: mmWave’s short wavelengths support compact large antenna arrays and directional beams that provide large array gains.
- Millimeter Wave: Hybrid beamforming combines analog phase-shifter networks with digital precoding to reduce RF-chain demands while retaining multiplexing capability.
- Millimeter Wave: Partially connected hybrid architectures reduce phase-shifter count by a factor equal to the number of RF chains and significantly reduce circuit power.They sacrifice some flexibility in exploiting spatial degrees of freedom.
- Millimeter Wave: Low-resolution ADCs reduce receiver power because ADC dissipation scales linearly with sampling rate and exponentially with bits per sample.Channel-capacity characterization for low-resolution ADCs in mmWave communications remains in its infancy, motivating further study and mixed-resolution designs.
B. LTE-U
LTE-U increases LTE capacity and can improve energy efficiency by operating in unlicensed spectrum, but coexistence with WiFi requires resource-management mechanisms. Listen-before-talk and duty cycling address coexistence differently, with duty cycling offering efficiency benefits but raising fairness concerns.
- LTE-U operates in the unlicensed band to increase LTE capacity and can achieve higher energy efficiency than WiFi by avoiding channel-sensing and backoff waste.
- Harmonious Coexistence: LTE-U’s continuous interference can keep WiFi channels busy, causing prolonged backoff, high energy consumption, and low WiFi energy efficiency without transmission.
- Harmonious Coexistence: Listen-before-talk requires LTE-U devices to verify channel occupancy before transmitting, whereas duty cycling periodically turns LTE transmission on and off.
- Harmonious Coexistence: Increasing contention among WiFi devices raises collision probabilities, while duty cycling may use unlicensed spectrum more efficiently.
- Harmonious Coexistence: Duty cycling assumes carriers control on–off scheduling, conflicting with the common-property nature of unlicensed spectrum and potentially limiting WiFi time windows.
- Harmonious Coexistence: No globally accepted coexistence protocol exists yet, although LTE-U is described as potentially a better WiFi neighbor than WiFi itself.
IV. GREEN 5G TECHNOLOGIES: SHORTENING TX/RX DISTANCE
Shortening transmitter–receiver distances through ultra-dense networks and device-to-device communication improves spectral efficiency and link quality. Energy-efficient deployment additionally depends on user-centric coordination, base-station sleeping, and interference management.
- Short-range communication improves link quality and spatial reuse by increasing K and decreasing d.
- Ultra-Dense Networks: User-centric architectures decouple signaling from data and uplink from downlink, enabling flexible association and lower signaling overhead.
- Ultra-Dense Networks: Separating signaling and data enables base-station sleeping, exploiting traffic variation across time and geography to save energy.Today, 80% of base stations are lightly loaded for 80% of the time yet consume almost peak energy.
- Ultra-Dense Networks: Dense data-base-station deployment can meet peak capacity while allowing lightly loaded stations to sleep or switch off.
- Ultra-Dense Networks: High frequency reuse makes intercell interference a critical concern in ultra-dense heterogeneous networks.Increasing neighboring base-station transmit powers can cancel out without improving throughput, producing low system energy efficiency.
- Ultra-Dense Networks: Power control, resource partitioning, scheduling, cooperative transmission, and interference alignment are needed for energy-efficient ultra-dense deployment.Removing the two or three strongest interferers is expected to bring an order-wise network energy-efficiency improvement.
B. D2D Communications
D2D communication supports energy-efficient 5G through direct reuse, user cooperation, relaying, multicast, and caching. Its practical deployment also raises incentive and mode-selection challenges.
- Mode selection and power control: D2D communication lets users communicate directly without routing through base stations, while reusing assigned or unassigned cellular spectrum.
- Mode selection and power control: Underlay D2D is preferable for energy-efficiency-oriented designs, whereas overlay D2D tends to suit spectral-efficiency-oriented designs.The distinction reflects interference mitigation through transmit-power limitation in energy-efficiency-oriented designs.
- Active user cooperation: D2D devices can act as mobile relays, saving substantial energy for degraded cell-edge uplinks and extending cooperation to multi-hop or multiple-relay scenarios.
- Active user cooperation: D2D devices can serve as multicast cluster heads, using favorable channels to provide spatial diversity or share received content among members.
- Active user cooperation: Wireless caching exploits underused device storage and content reuse to handle asynchronous demands, reduce backhaul burden, and reduce power consumption.
- Active user cooperation: A remaining challenge is motivating self-interested devices to relay, share, or cache data despite sacrificing their limited energy.Rewarding and pricing schemes from economic theory are proposed as possible tools for cooperative protocol design.
V. GREEN 5G TECHNOLOGIES: ENHANCING SPATIAL DEGREES OF FREEDOM VIA M-MIMO
Green M-MIMO improves energy efficiency by exploiting large antenna arrays, simplified processing, and lower transmit power, but circuit power, hardware limits, and pilot overhead constrain deployment.
- M-MIMO can improve multiplexing by increasing the number of users served with the same time-frequency resource.
- Energy-efficient M-MIMO: When transmit power dominates, more antennas improve energy efficiency; when circuit power dominates, additional antennas are no longer energy efficient.
- Energy-efficient M-MIMO: 100-200 antennas can be the energy-efficiency-optimal deployment for serving many terminals with current circuit technology.
- Signal processing and green hardware implementation: Linear MRC and MRT processing can approach optimal throughput while M-MIMO simplifications reduce computation energy relative to sophisticated processing methods.
- Signal processing and green hardware implementation: M-MIMO can operate with RF transmit power in the milliwatt range, producing substantial power-amplifier savings.
- Pilot design: Pilot resources scale with transmit-antenna count, making FDD impractical and motivating low-complexity pilot-beamforming and mitigation schemes.
VI. ENABLER OF SUSTAINABLE 5G COMMUNICATIONS: ENERGY HARVESTING
Energy harvesting is presented as an important enabler of sustainable green 5G, using renewable resources or RF signals to power communication systems. The paper compares these technologies from multiple perspectives.
- Energy harvesting technologies draw energy from renewable resources or RF signals and are important enablers of sustainable green 5G networks.
- Table II compares renewable-resource and RF-energy harvesting from different perspectives.
A. Energy Harvesting from Renewable Resources
Renewable-resource energy harvesting introduces energy-causality, data-causality, and battery-capacity constraints that challenge 5G resource allocation. The paper surveys offline, online, hybrid-supply, and energy-cooperation approaches.
- Renewable harvesting transforms natural sources such as solar and wind into electrical energy for green communication systems.
- Distinctive energy-data causality and battery capacity constraints: Energy harvesting communication is constrained because energy cannot be used before it is harvested, reflecting intermittent renewable availability.
- Distinctive energy-data causality and battery capacity constraints: Data causality and battery capacity constrain transmission because packets must arrive before delivery and harvested energy storage is limited.
- Distinctive energy-data causality and battery capacity constraints: Conventional assumptions of infinitely backlogged data and unlimited energy storage are no longer valid for highly diverse 5G traffic and devices.
- Offline versus online approaches: Offline optimization uses non-causal information to obtain performance upper bounds, whereas online policies use causal information for practical adaptation.
- Offline versus online approaches: When statistical information is unavailable or changes over time, learning-based schemes iteratively discover transmission policies from actions and immediate rewards.
- Hybrid energy supply solutions and energy cooperation: Hybrid harvesting combines sources such as solar, wind, diesel, and grid power to provide uninterrupted service despite intermittent renewable energy.
- Hybrid energy supply solutions and energy cooperation: In dense networks, harvesting can compensate for added circuit power, reduce grid consumption, and enable surplus-energy cooperation between base stations.
B. RF Energy Harvesting
RF energy harvesting research examines architectures that transfer energy through wireless signals while balancing harvested energy, information transmission, and system efficiency. The section identifies efficiency bottlenecks and opportunities from 5G techniques and interference reuse.
- Typical system architectures and extensions: RF energy harvesting follows two canonical architectures: SWIPT splits received signals between information decoding and energy harvesting, while WPCNs separate harvesting and transmission in time.SWIPT optimizes the receiver’s power-splitting ratio; WPCNs design the durations of wireless power transfer and information transmission.
- Improving the efficiency of WPT: WPT performance is constrained by low energy-conversion efficiency and severe path loss.The section motivates using narrow energy beams, shorter transfer distances, and other 5G communication techniques to improve transfer efficiency.
- Improving the efficiency of WPT: Receiver circuit power reduces the net harvested energy available for storage, especially when conventional active mixers consume substantial power.This motivates receiver architectures that avoid active devices.
- Rethinking interference in wireless networks with WPT: WPT reframes co-channel interference as a potential energy source rather than solely a harmful signal.Dense spectrum reuse in UDNs and D2D communications creates opportunities to harvest interference, while interference levels can be optimized for information-energy tradeoffs.
- Rethinking interference in wireless networks with WPT: WPT-enabled devices can harvest energy when interference is strong and decode information when interference is relatively weak.Artificial interference may also benefit overall system performance when receivers need energy for normal operation.
VII. CONCLUSIONS
The paper surveys green 5G and energy-harvesting technologies as components of sustainable networks. It concludes that their complementary advantages should be combined holistically, while diverse scenarios require interoperable technologies rather than one radio access technology.
- VII. CONCLUSIONS: The surveyed technologies are expected to enable sustainable green 5G networks, with Fig. 3 presenting a holistic overview of their components.The conclusion frames the technologies as parts of an integrated network design.
- VII. CONCLUSIONS: Energy harvesting provides green energy, while spectrum-efficient 5G technologies can be tailored to realize energy-efficient wireless networks.The conclusion presents these as complementary contributors to sustainable networking.
- VII. CONCLUSIONS: Improving system energy efficiency requires jointly considering user traffic, channel, power consumption, and content popularity across diversified communication scenarios.The conclusion links these factors to the system EE design problem.
- VII. CONCLUSIONS: Sustainable green 5G networks cannot satisfy diverse applications and heterogeneous user requirements with any particular radio access technology alone.The paper therefore calls for an ecosystem of interoperable technologies whose respective advantages can be combined, while acknowledging new system-design challenges.