Green Air-Ground Integrated Heterogeneous Network in 6G Era

(Beijing University of Posts and Telecommunications,Beijing 100876,China)

Abstract:The research of three-dimensional integrated communication technology plays a key role in achieving the ubiquitous connectivity,ultra-high data rates,and emergency communications in the sixth generation (6G)networks.Aerial networking provides a prom?ising solution to flexible,scalable,low-cost and reliable coverage for wireless devices.The integration of aerial network and terrestrial network has been an inevitable paradigm in the 6G era.However,energy-efficient communications and networking among aerial net?work and terrestrial network face great challenges.This paper is dedicated to discussing green communications of the air-ground integrated heterogeneous network (AGIHN).We first provide a brief introduction to the characteristics of AGIHN in 6G networks.Further,we analyze the challenges of green AGIHN from the aspects of green terrestrial networks and green aerial networks.Finally,several solutions to and key technologies of the green AGIHN are discussed.

Keywords:air-ground integrated heterogeneous network;6G;green communications

1 Introduction

The improvement of network capacity,coverage,delay,security,etc.has always been a key and core task in the development of ground mobile communication net?works.Three-dimensional integrated communication is one of the key research directions in achieving the ubiquitous connectivity,ultra-high data rates,and emergency communica?tions in the six generation (6G) networks[1].Aerial and space networks facilitate adaptive,flexible,scalable,efficient,and re?liable three-dimensional wireless coverage for wireless termi?nals,which have attracted much attention from industry and ac?ademia societies.The air-ground integrated heterogeneous net?work (AGIHN) integrating various aerial communication plat?forms and terrestrial infrastructures is a cost-efficient paradigm to facilitate extended wireless coverage,ultra-high data rates,post-disaster communication assistance and recovery etc.

A typical AGIHN architecture is shown in Fig.1,where aerial communication platforms such as airships,balloons,and unmanned aerial vehicles (UAVs) acting as the carrier of information collection,transmission and processing can pro?vide broadband wireless communications and supplement ter?restrial networks.Terrestrial networks mainly consisting of heterogeneous cellular networks,wireless local area networks(WLAN),and mobile ad hoc networks (MANET) support vari?ous applications and services in the areas where infrastruc?tures are easy and low-cost to be deployed.

Nowadays,industry and academia societies have started re?search and implementation of AGIHN.For example,in 2016,Nokia Bell Labs demonstrated the world’s first flying cell (FCell) based on UAV,which was powered by solar energy and could wirelessly transmit high-definition video[2].In 2017,EE,the British Telecom Operator,broadcast the mountain bike race live on the mini mobile site“Air Mast”connected to the helium balloon[3].As of December 2019,the FirstNet commu?nications platform jointly built by AT&T and First Responder Network Authority had reached more than 1 million connec?tions[4].Flying Cells on Wings (COWs) in the platform is an ideal choice for wildfire and mountain rescue missions.Dur?ing Hurricane Michael,a COW provided services to first re?sponders on the battered Mexican beach in Florida to support disaster recovery.One Aerostat,which was launched later,can provide more than twice the coverage area compared with COWs,helping responders keep connected in the event of a large-scale catastrophic event.

▲Figure 1.Architecture of air-ground integrated heterogeneous net?work(AGIHN)

1.1 Characteristics of AGIHN

1.1.1 Heterogeneity

In addition to terrestrial heterogeneous cellular networks,high altitude platforms (HAPs) such as airship and balloon and low altitude platforms (LAPs)such as UAVs are employed in the aerial networks to achieve seamless wireless coverage and to meet differentiated data rate requirements.This hetero?geneous integrated network enables diverse systems to cooper?ate,coordinate,and share information for serving mobile ter?minals with individuation service anytime and anywhere.

As shown in Fig.2,AGIHN is a large-scale and multi-layer 3D heterogeneous network.HAPs such as airships and bal?loons are distributed in remote rural areas with imperfect terrestrial infrastructure or disaster areas,with an altitude of 17–30 km[5].UAVs are used for high-speed services in hot spots or wireless connection in disaster areas,with an altitude below 10 km[6].Terrestrial base stations (BSs) and access points(APs)are typically deployed in the area with an altitude below 1 km.

Terrestrial heterogeneous cellular networks realize coverage optimization and capacity improvement by deploying dense small cells with lower transmission power,such as microcell,picocell and femtocell.Due to economic cost and terrain con?straints,these terrestrial communication infrastructures are es?tablished according to the human habitation and living habits,which makes wireless traffic a stable and periodical spatialtemporal distribution.

▲Figure 2.Heterogeneity of air-ground integrated heterogeneous network(AGIHN)

Aerial networks,as a supplement to terrestrial networks,have to deal with the complex and diverse application scenari?os with uneven spatial-temporal distributed traffic,diverse ser?vice demand,and sudden surge of wireless traffic.Flexible movement is a key feature of AGIHN.HAPs are quasi-static(relative to the ground) platforms.The moving speed of UAVs is 0–460 km/h[7].Moreover,UAVs can move freely in the 3D space with random trajectory.The ground BSs are typically fixed and deployed in buildings or on high mountains.Ground terminals such as vehicles and terminals on high-speed rails typically have speeds of 0–350 km/h and move with relative?ly fixed trajectories[8].Power supply of network nodes in AGI?HN is also diverse from each other.There is no continuous pow?er supply source for aerial nodes.Battery,wind,solar,and other combined power supply are the main energy sources for bal?loons.The endurance of such platforms can reach 150–200 days[9].UAVs generally use battery power supply and the endur?ance is only about half an hour to 24 hours[10].Ground BSs and APs are driven by grid power system for continuous operation.

For the frequency bands and radio propagations of ground 4G and 5G networks,the frequency resources occupied by 4G system include 1 880–1 900 MHz,2 320–2 370 MHz and 2 575–2 635 MHz,while the frequency resources occu?pied by 5G system include 3.3–4.2 GHz,4.4–5.0 GHz,the millimeter wave band,26 GHz,28 GHz and 39 GHz[5].The aerial nodes such as UAVs,balloons and airships work at the Long Term Evolution (LTE) or Wi-Fi communication bands[11].They can also work in the unlicensed Industrial,Scientific and Medical (ISM) band defined by the ITU Radio?communication Sector (ITU-R)[12].The electromagnetic propa?gation of different frequency bands also differs from each oth?er.Radio attenuation on high frequency bands is more serious.Compared with the electromagnetic fading at 2 GHz,an addi?tional 22.9 dB of fading exists at 28 GHz[13].Good line-of-sight(LoS) transmission links exist in air-to-air,air-to-ground,and ground-to-air channels while the radio propagation in ground transmissions faces more serious signal fading due to rich re?flection,refraction,scattering,etc.

1.1.2 High-Dynamically Changed Network Topology

Flexibility is one of the most key features of aerial net?works.Payload,height,speed and endurance are the four key factors influencing communication performance of aerial plat?forms.Payload represents the maximum carrying weight that the platform can hold.HAPs and LAPs carry different commu?nication equipment with different weight.Height refers to the maximum altitude that the aerial platform can be reached,which is closely related to the coverage of the aerial platform.Endurance refers to the maximum flight duration without charging and refueling.As mentioned above,the height,speed and endurance of difference components in AGIHN differ from each other,which results in the high-dynamic change of the network topology.

Different height and moving speed of diverse platforms make the network topology more stereoscopic.In order to pro?vide flexible services for wireless terminals,communication platforms change their positions and height adaptively.As a result,the network topology changes dramatically with the moving of platforms.The endurance is another key factor hav?ing great impact on the network topology.The network topolo?gy of ground networks changes slightly since ground BSs and APs are generally fixed located and are powered by grid sys?tem while that of aerial networks changes frequently and rapid?ly due to the energy depletion and battery charging.Besides,AGIHN are more vulnerable to malicious attacks such as wire?tapping,hijacking,masking,and jamming,which cause dis?connection and interruption of aerial links and re-connection of surviving nodes.The disconnection and re-connection of networks nodes in AGIHN also contribute to the changes of network topology.

1.1.3 Random Perturbation of Aerial Platforms

Due to the lack of fixed infrastructure,the aircrafts are sus?ceptible to airflow and body vibration,leading to random per?turbation of aerial communication platforms.According to the tests and measurements,the variation of roll angle (i.e.,the el?evation angle in this paper) is ±0.02 rad.The variation of pitch angle (i.e.,the azimuth angle in this paper) is ±0.1 rad[14].The random perturbations of aerial platforms may cause error to the estimation of angle of departure (AOD) and angle of arrival (AOA) between transceivers,further leading to the error of channel state information estimation and distortion of coverage area.Consequently,the perturbation of aerial plat?forms will cause non-robust transmission links,inefficient en?ergy consumption,and serious information leakage,etc.

The perturbation angle of UAV was assumed as high as 10 degrees in Ref.[14].It shows that the jitter of UAV causes in?accurate estimation of deviation angle between the UAV and ground users and increases the error of AOD estimation.The influence of wind on UAVs was then simulated by using onboard sensors in Ref.[15].The maximum amplitude of side?slip angle and trajectory angle jitter was approximately 10 de?grees,which verifies the previous hypothesis.Considering the impact of UAV jitter,energy-saving secure communications in a downlink A2G wiretap system was investigated in Ref.[16].

1.2 Integrating AGIHN in 6G

With the commercialization of 5G networks,various groups from worldwide countries and regions have initialized plans and programs on potential key technologies for 6G networks.Space-air-ground integrated networking (SAGIN) is acknowl?edged as a key direction in achieving global connectivity.AGI?HN,as an important part of SAGIN,is seen as a cost-effective approach to meeting the requirement of ultra-high data rate and ubiquitous coverage.To realize the expected goal,the in?tegration of ground networks and aerial networks has to solve the problem of new network architecture design and the chal?lenge of disruptive technology innovation.

1.2.1 Directions of Network Architecture Design

Network architecture design is the first step to realize the integration of ground networks and aerial networks.Efficient coordination of resources and fully exploitation of cooperation between ground networks and aerial networks are the main goals in network architecture design.In order to solve the com?plex interoperation in the management of heterogeneous net?works,the software defined network (SDN) and network func?tion virtualization (NFV)are applied in 5G networks.SDN and NFV are still seen as efficient solution to the network manage?ment in 6G networks.In the AGIHN with high-dynamically changed topology,SDN and NFV based core network manage?ment architecture can provide distributed and on-demand re?source allocation,service guaranteed network slicing,flexible programming of network functions,and security manage?ment[17–20].SDN and NFV are also seen as promising technolo?gies for providing flexible and reconfigurable green satellite services in space-air-ground integrated networks[21].

Efficient energy utilization and low energy consumption are always key concepts in network architecture design.Green AGIHN architecture design can be carried out from the as?pects of green communications and green computing.For green communications in AGIHN,the aerial platforms can pro?vide cost-effective and energy-saving transmissions for the wireless terminals with appropriate cooperation,trajectory de?sign,user scheduling,power allocation,and combination with improved wireless technologies[22–25].Coordination and cooper?ation architecture of AGIHN and resource management of het?erogeneous network nodes are the keys to green communica?tions of AGIHN.For the green computing,aerial communica?tion platforms with mobile edge computing (MEC) can greatly improve the data rate and latency performance in AGIHN[26].Moreover,distributed cloud architecture can achieve seamless handover and effective task offloading among UAVs and ground terminals[27].Green computing in AGIHN can be real?ized with energy-efficient MEC and green cloud architecture.

Intelligence is a core idea in the 6G era.To realize ubiqui?tous intelligent mobile society in the 6G era,artificial intelli?gence (AI) is expected to fully penetrate the network evolu?tion.With AI applied in AGIHN architecture design,efficient resource management and network optimization can be real?ized by exploiting the potential information in wireless big da?ta and with less or even no human intervention.Moreover,en?hanced privacy preserving can be achieved by leveraging AI in aerial networks[28].

1.2.2 Directions of Key Wireless Technologies

1)Terahertz communications

AGIHN is facing the contradiction between limited spec?trum resources and the rapid growth of high-speed traffic de?mand.Terahertz communications is an important direction to break through the resource limitations in 6G networks[29].The terahertz frequency resources is from 0.1 THz to 10 THz.Tera?hertz communications has the advantages of ultra-low delay,excellent directivity,anti-interference,wide bandwidth,and strong penetration.Moreover,since the terahertz wavelength is greatly reduced,the antenna size can be greatly reduced,which is beneficial to antenna integration.Leveraging tera?hertz technology to aerial networks can further improve the da?ta rate with highly concentrated beams,strong LoS path and wide bandwidth resources.However,the short terahertz wave and the weak diffraction introduce quite high path loss of ra?dio propagation.Thus,denser BSs and APs are required to achieve seamless coverage,which means more energy con?sumption will be introduced.

2)Intelligent reflecting surface

Reconfigurable intelligent reflecting surface (IRS)is attract?ing attention for wireless networks since it can significantly improve the wireless channel quality by adaptively reconfigur?ing wireless propagations with massive low-cost passive re?flecting elements integrated into a planar surface[30].Combin?ing IRS with non-orthogonal multiple access (NOMA) and MEC can greatly improve network throughput and reduce la?tency[31–32].Leveraging IRS in AGIHN,the received signals at the UAV from cellular BSs can be greatly improved by config?uring IRS deployed on building walls.The secrecy rate can be enhanced by jointly optimizing phase shifters of IRS,UAV tra?jectory,and UAV power[33].Moreover,leveraging UAVs with IRS can provide energy-efficient communications,which pro?vides new sights in green communications in 6G networks[34].

3)Spectrum sharing

Spectrum resources are the treasure for wireless communi?cations.In addition to terahertz communications and visible light communications,spectrum sharing is another approach in extending spectrum resources and improving spectrum effi?ciency in 6G networks with flexible and intelligent frequency allocation and reuse[35].Employing blockchain in spectrum sharing can further prevent jamming from malicious users[36].Spectrum reuse between dense ground BSs and flexible-mobil?ity UAVs makes interference management more challenging.With appropriate spectrum sharing between aerial networks and ground networks,the area spectrum efficiency and net?work throughput can be significantly improved.

4)Energy harvesting technologies

With the proliferation of mobile devices and the denser de?ployment of network BSs and APs,prolonged battery life and improved energy harvesting efficiency are the keys to realize green AGIHN.Simultaneous wireless information and power transfer (SWIPT) is one of the popular energy harvesting tech?nologies studied in recent years[37].SWIPT can charge wireless devices while supporting communications and is a promising energy charging technology for sensors nodes.Leveraging SWIPT in AGIHN,the aerial platforms can effectively charge ground terminals and improve energy efficiency by exploiting the benefit of flexible mobility.In addition,SWIPT combined with IRS in aircrafts can further improve the energy harvesting efficiency[38].

2 Challenges of Green AGIHN

According to the forecasts of International Energy Agency(IEA) and Global Electronic Sustainability Initiative (GESI),the information and communication technology (ICT) industry currently consumes 2%–4% of the global energy,which is equivalent to the amount of energy consumption of the avia?tion industry[39].The huge energy consumption not only in?creases operating costs,but also brings series of resource and environment problems.It is requested by the ITU that the global ICT industry reduce greenhouse gas (GHG) emissions by 45% from 2020 to 2030[40].How to improve energy efficien?cy of the communication industry is an urgent problem to be solved in the 6G era.

In AGIHN,aircrafts acting as information carriers can re?duce more energy consumption and provide better energy-effi?cient services compared with terrestrial networks due to the re?duced energy consumption of ancillary facilities such as the air conditioner at BSs.Nevertheless,the green communication in AGIHN still faces many challenges.

2.1 Challenges of Green Terrestrial Networking in AGIHN

2.1.1 Optimized Deployment of BS and AP

In AGIHN of the 6G era,more ground BSs and APs will be deployed to accomplish the requirement of ultra-high data rate services.More infrastructures will be established to support the deployment and operation of BSs and APs.Accordingly,more energy will be consumed.According to energy efficiency requirements for telecommunications proposed by Verizon[41],BSs consume nearly 80% of the energy consumed in cellular networks while the power amplifiers and air conditioners con?sume almost 70% of the total energy at BSs[42].It is reported by Huawei that the maximum energy consumption of a 5G BS is about 11.5 kW,which is 10 times of that of a 4G BS[43].However,according to Daiwa’s prediction,the number of 5G BSs will be four times that of 4G BSs in their respective eras[44].With the employment of advanced wireless technolo?gies,the deployment of BSs will be much denser in 6G net?works.It is foreseen that there will be up to 40 000 sub-net?works per km2in 6G networks[45].Optimizing the deployment of BSs and APs is one of the key approaches in reducing the energy consumption in AGIHN.

In addition,the ground wireless traffic is non-uniformly dis?tributed in both time and space.It is predicted that the data traffic in the downtown of Milan is 4 times of that in Boccono University located in suburb[46].An analysis report of data traf?fic in Shanghai reveals that the data traffic in residential areas is 1.5 times of that in office areas[47].Moreover,the ratio of daytime traffic amount to night-time traffic is around 0.8 in resi?dential areas while it is up to 1.4 in office areas.Although aer?ial networks can provide flexible services for ground terminals in such areas with non-uniform traffic,fixed terrestrial commu?nication infrastructure is still the main approach for providing robust and cost-effective services.To satisfy the requirement of temporal and spatial non-uniform traffic and to simultane?ously save energy,flexible BS sleeping and awake schemes play important roles in saving energy at BSs[42].

2.1.2 Increased Mobile Devices and Wireless Access

According to Cisco,there were 8.8 billion mobile devices and wireless connections in 2018,induduing 4.9 billion smart phones and 1.1 billion IoT devices[48].42% of the devices en?joyed wireless services through 4G cellular networks.Due to the continuous prosperity of sensors,intelligent furniture,In?ternet of vehicles,smart city and medical applications,and the continuous penetration of vertical industry with 5G net?works,there will be 28.5 billion wireless devices by 2022[49],among which 51% (14.6 billion) of the devices are batterypowered machine-to-machine (M2M) devices.It is estimated that by 2025,the global network standby energy consumption of IoT edge devices will approach 46 TWh[50],which is about equivalent to the annual electricity consumption of Portugal in 2019[51].

The massive wireless connectivity brings a great challenge to the wireless access networks due to massive connection re?quests,ultra-heavy traffic load,limited battery of wireless de?vices,and diverse levels of subscribers.Moreover,burst ac?cess attempts may happen due to some unexpected events such as power failure,which will lead to a sharp increase of control signaling,network congestion,and further increased energy consumption.In addition,handover of massive devices among heterogeneous networks introduces complex interopera?tion and resource managements,which also causes the in?crease of energy consumption.Low energy consumption and improved energy efficiency are crucially important for wireless communications in future networks.

2.2 Challenges of Green Aerial Networking

Although aerial platform-based communications can save more energy than ground communications,the explosive growth of aircrafts in wireless communications and the highdynamically changed topology bring new challenges to the green communications.The UAV is the most widely applied aircraft for wireless communications due to its flexible mobil?ity,cost-efficient and rapid deployment,and LoS communica?tion support.The Federal Aviation Administration (FAA)pointed out that,by 2024,the world will see the emergence of 1.48 million recreational Unmanned Aircraft System(UAS)fleets[52].

A small UAV usually needs 20 to 200 W/kg to fly[53].It is usually powered by on-board batteries.Different types of UAVs carry different battery capacities.For example,the Sky?walker fixed-wing UAV carries four 8 500 mAh batteries in series,while the AKS Raven X8 multi-rotor UAV carries two 10 000 mAh batteries[54].The battery power consumption of 30 kg to 35 kg UAVs to complete a flight mission (i.e.,lifting up,hover,flight,and landing) is about 12.53% and 13.82% of the full amount[55].The power consumption of Wi-Fi and GPS communications of a small UAV with the weight of 865 g is about 8.3 W,and that of horizontal flight is 310 W[56].There?fore,in the case of limited battery,reducing the power con?sumption of UAV can provide longer endurance and communi?cation services.

2.2.1 Green Communication Module

Limited battery limits the performance of aerial transmis?sions.Improving the energy efficiency of communication mod?ules of aircrafts and increasing the battery capacity are the two main approaches for green communications of aerial net?working.Aircraft placement,trajectory optimization,power control,flight duration optimization,resource and interference management,and handover in high-dynamic networks are the main factors influencing the energy consumption and energy saving in AGIHN.

Radio propagation and load balance among BSs or APs are closely related with the placement and trajectory design of aer?ial communication platforms.For HAPs,the placement optimi?zation directly influences the load balance between HAPs and ground BSs.Appropriate placement of HAPs can extend wire?less coverage,improve network throughput,and further reduce energy consumption of the communication modules.For LAPs,optimized trajectory design can improve network throughput,reduce energy consumption,and extend flight du?ration.The trajectory design of multi-UAV networking can fur?ther provide seamless connectivity for wireless terminals and extend the coverage area of aerial networks.It should be noted that collision avoidance among multiple UAVs should be con?sidered in the trajectory design.Frequency reuse,spectrum sharing,and power control are key factors in resource and in?terference management between aerial networks and terrestri?al cellular networks.Flexible and adaptive frequency reuse and power control can decrease interference among aerial nodes and ground nodes,which can further improve network throughput and improve energy efficiency.

2.2.2 Green Flight Module

A UAV consumes more propulsion power in flight than in hover[56].LAPs have the advantage of mobility compared with HAPs while they consume more power due to their high-dy?namic mobility.The flight power consumption of a fixed-wing UAV can be expressed as functions of its velocityVand accel?erationa(t)[57]:

whereare two parame?ters(whereρa(bǔ)ndare the air density and zero-lift drag coef?ficient,respectively,SandWare the wing area and aircraft weight,respectively,ande0andARdenote the wing span effi?ciency and aspect ratio of the wing,respectively);gis the grav?itational acceleration.According to Eq.(1),we can see that the energy consumption increases with the increase of velocity and the absolute value of acceleration,which means that more energy will be consumed with higher flight speed and faster change of the speed.Therefore,in order to achieve green flight of UAV and further to achieve green aerial communica?tions,the UAV should move as smoothly as possible while im?proving the transmission performance.The flight power con?sumption of a rotary-wing UAV is a function of the UAV veloc?ity,which is given by Ref.[58]:

whereare two parameters representing the blade profile power and induced power of the UAV in hover,respectively (whereδis the profile drag coefficient,sandAare the rotor stiffness and rotor disc area,respectively,ΩandRdenote the blade angular velocity and rotor radius,respectively,andkdenotes the incremental correction factor to induced power);Utipis the rotor tip veloci?ty;v0is the average rotor induced velocity in hover;d0denotes the fuselage drag ratio.According to Eq.(2),the first and the third terms are the power required to overcome the profile drag of the blades and the fuselage drag,respectively,which increases with the square and cubic of the velocity.The sec?ond term is the power required to overcome the induced drag of the blades,which decreases with the velocity.It is verified in Ref.[58]that the total power consumption first decreases and then increases with the increase of UAV velocity,which means that the energy consumption can be minimized with the optimal UAV velocity.

As a result,the improvement of transmission quality and en?ergy efficiency in AGIHN should take flight consumption into consideration.There are three main research directions for en?ergy-saving UAV communications:the optimization of flight radius and velocity with a fixed trajectory[59];the joint optimi?zation of UAV trajectory,acceleration,and velocity[60];the trade-off between performance improvement and energy con?sumption[61].

3 Key Technologies in Green AGIHN

How to realize energy-efficient network collaboration and integration of heterogeneous networks in AGIHN is one of the keys to SAGIN.In this section,several promising technologies for harvesting energy and reducing energy consumption in AGIHN are analyzed.

3.1 Energy Harvesting Technology

Energy harvesting technologies are promising approaches to prolonging battery life and providing extra power in green com?munications by collecting external energy resources such as light,heat,electromagnetic,and mechanical.Wireless power transfer(WPT)and SWIPT are two energy harvesting technolo?gies for transporting energy with electromagnetic energy.WPT is a basic energy harvesting technology while SWIPT is an en?ergy harvesting technology that integrates WPT and the com?munication function,i.e.,SWIPT can simultaneously transmit information signals while transporting electromagnetic energy.

Leveraging WPT and SWIPT in AGIHN can realize remote charging and mutual charging among aircrafts.However,the mutual energy transport definitely causes increased energy consumption.Traditional energy collection technologies such as solar and wind systems can be combined as a RF energy source to reduce consumption of non-green energy,which is a promising research direction for AGIHN.For example,charg?ing LAPs with solar-powered satellites and HAPs can prolong the flight duration of LAPs and enhance the stability of aerial networking.Moreover,the LAPs can further charge each other by exploiting the benefits of LoS links and charge ground ter?minals with two-hop wireless energy transmission.In addition,leveraging IRS in SWIPT-assisted LAP system can simultane?ously improve the network performance and enhance the ener?gy transmission efficiency[38].

3.2 Cooperative Communications

Cooperative transmission can improve network performance by coordinating multiple diverse nodes to exploit the multi?plexing gain and diversity gain.In AGIHN,coordinating multi?ple aerial nodes as information signal sources can provide sig?nificantly improved capacity and coverage performance with LoS links,flexible-changed topology,and adaptive resource coordination.The transmission power of aerial nodes can be greatly reduced due to the decreased path loss fading.Further,coordinating multiple aerial nodes as electromagnetic energy sources can realize rapid power charging and network restora?tion and reconstruction.Coordinating aerial nodes and ground BSs can provide more energy-efficient transmissions than coor?dinating only ground BSs due to the reduced transmission power of aerial nodes and the exemption of energy consump?tion of BS infrastructures.Especially,the energy efficiency of cell edge users can be significantly improved by deploying pe?riodic UAVs at the edge of ground BSs and periodically coor?dinating UAVs-enabled BSs or relays with the ground BSs[62].

Coordination and cooperation of ground nodes and aerial nodes can realize improved energy efficiency in AGIHN.How?ever,the high-dynamic network topology,non-uniform data traffic,and random perturbation of aerial platforms bring great challenges to the cooperative communications in AGIHN.The high-dynamic network topology introduces frequent disconnec?tion and reconnections,which brings huge signaling overhead.Blockchain based access registration provides a way in reduc?ing the signaling overhead[63].Non-uniform data traffic brings challenges to the user scheduling and traffic load coordination among cooperative nodes,which further causes tradeoff be?tween balanced traffic offloading and efficient energy con?sumption.The random perturbation of aerial platforms causes non-robust and inefficient transmission links between a pair of transceivers.Accurate estimation of platform perturbation,ap?propriate compensation for perturbation,and adaptive power allocation and beam adjusting are required for energy-efficient communications in AGIHN.

3.3 Integrating Intelligence into Green AGIHN

AI technologies,especially machine learning (ML) and big data analysis,are promising and widely acknowledged solu?tions to smart network control and network management.AI has already been applied in mobile networks from physical layer design to network layer control.Leveraging AI in 6G net?works is an inevitable trend.In green AGIHN,AI can be ap?plied in many aspects including intelligent architecture de?sign,real-time network data analysis,flexible aerial platform control,secure aerial platform tracing,smart platform posi?tion,high-efficient energy harvesting,intelligent routing,intel?ligent caching and computing,intelligent sleeping and wakeup mechanisms,adaptive and efficient resource allocation and scheduling,etc.

Leveraging intelligence into AGIHN can realize enhanced energy efficiency by globally optimizing the network control and management.However,mechanisms adopted to realize in?tellectualization will bring additional energy consumption,es?pecially for the intelligent network management through glob?al control or massive data analysis.Therefore,energy-efficient intelligent network control and management mechanisms are required in green AGIHN.

4 Conclusions

Facing the high complexity and diversity of future networks and the proliferation of wireless devices,AGIHN is consid?ered as a cost-effective approach to ubiquitous coverage and ultra-high throughput.With the increasing scarcity of natural resources and complexity of the radio environment,it is urgent to solve the problems of green AGIHN to reduce energy con?sumption and to improve energy efficiency.In this paper,we first introduced the integration of AGIHN and 6G networks.Then,the challenges of green AGIHN were analyzed from the aspects of green terrestrial network and green aerial network.Following the analysis,several promising green technologies that can be employed in AGIHN have been discussed.