A
Hybrid Energy and Spectrum Efficient Wireless Network Design for 5G/6G And Wi-Fi 7/8 ApplicationsFanuel.elias@stu.mmu.ac.uk; S.ekpo@mmu.ac.uk
Zacheous Aasa1[0009−0007−9010−4936], Fanuel Elias2[0009−0007−7684−0339] and Sunday C. Ekpo2[0000–0001−9219−3759]
1Department of Electrical and Electronics Engineering, Ladoke Akintola University of Technology, LAUTECH, Ogbomoso, Nigeria
zacheous.aasa@gmail.com
2 Communication and Space Systems Engineering Research Team, Manchester Metropolitan University, Manchester, M1 5GD, UK
* Corresponding author: Zacheous Aasa
Abstract.
The migration of wireless communications from 5G to the sixth generation (6G) of cellular communications and next-generation wireless local area networks (Wi-Fi 7/8) has emerged to support ultra-high data rate, massive connectivity, and ultra-low latency. Nonetheless, these needs exacerbate both energy and spectral resource challenges, especially in ultra-dense and heterogeneous networks. The available techniques have focused dominantly on energy efficiency or spectral efficiency, making them less effective in multi-radio access technology (multi-RAT) networks. In this paper, a hybrid framework that simultaneously leverages energy and spectral efficiency for converged 5G/6G communications and Wi-Fi 7/8 networks based on the integration of resource control based on artificial intelligence (AI), cognitive spectrum management, and reconfigurable intelligent surfaces (RIS) is proposed. A model of a multi-RAT system assisted by reconfigurable intelligent surfaces has been established, and an optimisation problem that maximises energy efficiency while satisfying quality-of-service constraints has been formulated. The challenge of the resulting mathematical model being a non-convex optimisation problem has been resolved using a controller designed based on deep reinforcement learning to dynamically adjust both the power allocation to all the radios involved and the reconfigurable intelligent surfaces' phases. Simulation results have confirmed that the proposed scheme can obtain energy efficiency improvement of up to 35% compared to the typical scheme, and simultaneously, the spectral efficiency gain has also significantly improved when operating at a high frequency. The results have confirmed that integrated energy and spectral resource optimisation could provide a new solution for a sustainable 6G and Wi-Fi 8 network.
Keywords:
Artificial Intelligence
Cognitive Spectrum Management
Energy efficiency
Reconfigurable Intelligent Surfaces
Spectral efficiency
5G/6G
Wi-Fi
A
1 Introduction
The rapid growth in mobile data traffic, device density, and latency-sensitive services related to extended reality, autonomous systems, smart cities, and industrial automation has significantly changed the design of modern wireless networks. Fifth generation (5G) systems meet these demands by providing enhanced mobile broadband, massive machine-type communications, and ultra-reliable low-latency communications [14]. These performance enhancements, however, come at a high cost: substantially increased energy consumption due to heavy reliance on massive multiple-input multiple-output (MIMO), dense base-station deployment, broad bandwidth utilisation, and sophisticated signal processing [15–17]. Notably, base stations account for approximately 70–80% of cellular networks' total power consumption, raising serious sustainability concerns [2], [3]. Beyond 5G, the sixth generation of the network shall support terabit-per-second data rates, latency as low as sub-milliseconds, very high device density, and integrating terrestrial and non-terrestrial networks [9], [11].
In parallel, the IEEE 802.11 wireless local area networks are evolving: Wi-Fi 7 is enabling multi-gigabit throughput through multi-link operation and ultra-wide channels, while the target for Wi-Fi 8 is ultra-high-density scenarios with deterministic latency and improved energy efficiency [1], [26]. Even though such advances significantly enhance the network's capability, they make the energy-efficiency and spectrum utilisation challenges even worse, especially in the increasingly used millimetre-wave and terahertz frequency bands [27]. These include traditional approaches such as static power control, hardware optimisation, and isolated sleep mechanisms that are increasingly ineffective in dynamic and heterogeneous environments [3]. Conventional spectrum management can also hardly address coexistence challenges posed by multiple radio access technologies operating in overlapping frequency bands [18], [19]. Limitations in this regard assure the importance of integrated frameworks for runtime joint optimisation of energy consumption and spectrum utilisation. In particular, this paper proposes a unified hybrid energy-spectrum-efficient framework driven by AI-based resource optimisation, RIS-assisted propagation control, cognitive spectrum access, and multi-RAT coordination for converged 5G/6G and Wi-Fi 7/8 networks. Unlike existing works that consider these technologies in isolation, the proposed approach explicitly leverages their interaction to realise improvements in both energy and spectral efficiency.
This study proposes an innovative hybrid energy and spectrum-efficient network design for integrated 5G/6G and Wi-Fi 7/8 systems. A single architecture has been created to enable joint energy-spectrum optimisation. This includes AI-driven control, Reconfigurable Intelligent Surfaces (RIS), cognitive spectrum access, and coordinated multi-Radio Access Technologies (multi-RAT). A RIS-assisted multi-RAT system model has been developed, and an energy-efficiency maximisation problem has been established, constrained by specific quality-of-service (QoS) requirements. A control method utilising deep reinforcement learning (DRL) has been implemented, enabling real-time joint optimisation of transmit power distribution and RIS phase adjustments. The proposed approach has been shown to improve energy efficiency by up to 35% compared to baseline systems, with just a slight loss in spectrum efficiency. Thorough simulations showed this. The goal of these contributions is to help create intelligent, sustainable, and adaptable wireless networks that will perform well in the future, when there are many people and a lot of data.
The paper is then organised as follows: Section 2 presents the proposed hybrid energy-spectrum-efficient system architecture and its functional components. Section 3 discusses the key design challenges that motivate the synergistic framework. Section 4 introduces the system model and formulates the energy-efficiency optimisation problem. Section 5 explains the interaction of the system components under AI control. Section 6 presents simulation results along with performance analysis, and Section 7 presents a comparative discussion. Finally, Section 8 concludes the paper and suggests future research directions.
2 Proposed Hybrid Energy-Spectrum Efficient System Architecture
This section introduces the architecture of the proposed hybrid energy- and spectrum-efficient framework, along with the roles of each component in making it functional. The proposed framework is concerned with energy consumption as well as spectrum, along with coexistence with heterogeneous networks in 5G, 6G, and Wi-Fi 7/8 networks, because energy consumption, spectrum, and heterogeneous networks have been addressed as major weaknesses of next-generation wireless networks [2], [9], [11].
2.1 Role of Hybrid Renewable Energy in the Proposed Framework
In the proposed architecture, the base stations and access points are powered by a hybrid power supply system comprising grid power, alternative energy sources such as solar and wind, and battery energy storage systems [20, 21]. The proposed hybrid power supply system ensures adaptive network operation in response to traffic demand and power supply, which plays a key role in sustainable 6G networks [2][8],[22].
Unlike conventional grid-dependent networks, the proposed framework integrates the resource optimisation process with considerations of energy availability. During peak hours when renewable energy is abundant, the framework helps support increased traffic volume. The strategies for managing energy scarcity are triggered to conserve energy consumption. This design is consistent with the concept of green networks for next-generation networks such as 6G [2], [3].
2.2 AI-Driven Resource Optimisation and Control
The formulated framework's control mechanism is Artificial Intelligence (AI). The AI controller actively collects real-time data, such as traffic demand, channel state information, mobility patterns, available spectrum, and the status of renewable energy supplies. Decisions are made on the adaptive configuration regarding the RIS configuration settings and the transmission power level.
Machine learning and deep reinforcement learning methods have been shown to effectively address the optimisation challenges with high-dimensional, non-convex structures commonly observed in next-generation wireless communication systems [4], [6]. This is due to the control strategy's ability to learn an optimal control policy by interacting with the environment to optimise the energy spectrum resource.
2.3 RIS-Assisted Propagation Control
Reconfigurable Intelligent Surfaces (RISs) are integrated into the proposed framework to optimise wireless transmission and improve link quality with minimal power overhead [23]. RIS is composed of a large number of passive reflecting elements with adjustable phase values to control wireless channels [5], [10].
The AI controller may determine the optimal phase-shift pattern for the RIS based on CSI and traffic requirements within the given architectural framework. This can assist in directing the signal to the correct receivers, lessen the effects of blockages, and increase interference suppression, especially at terahertz and millimeter-wave frequencies when route loss is high. It can help reduce the transmit power of the base station and the Access Point in wireless networks through the RIS's passive beamforming, thereby enhancing link quality [22, 24].
2.4 Cognitive Spectrum Management and Multi-RAT Coordination
To address spectrum scarcity and coexistence in heterogeneous networks, the proposed system model introduces cognitive spectrum management and multi-RAT coordination. Cognitive spectrum helps efficiently use the spectrum by identifying and opportunistically exploiting available spectrum, thereby reusing it in spectrally congested scenarios [18].
The AI controller can control spectrum access and the RAT together, allowing seamless traffic steering between the Cellular networks (5G/6G) and WLAN networks (Wi-Fi 7/8) operating in overlapping frequency bands [1], [19], [26],[22].
2.5 Integration of Non-Terrestrial Networks
Non-terrestrial networks (NTNs), consisting of low-earth orbit (LEO) satellites and high-altitude platform systems (HAPS), are introduced to provide additional support to these networks through complementary access layers in this architecture design. NTNs improve rural service delivery and network reliability when NTrS are unavailable or inaccessible [7–8]. In the newly proposed framework, the role of NTNs is considered supplementary to connectivity resources rather than a fundamental aim of network optimisation. The decision to hand over communication tasks to NTNs, based on energy, latency, and spectrum resources, is made by the AI controller to make a proper combination of communication layers on both Earth and non-Earth for next-generation wireless communication networks, namely, 6G, efficiently within the framework of [7], [9],[22].
3 Design Challenges and Problems
This section presents the relevant design considerations for our proposed system model and optimization approach.
3.1 Energy Supply Uncertainty in Hybrid-Powered Networks
The effect of renewable energy sources, such as solar and wind power, on wireless infrastructure introduces uncertainty due to the intermittent and time-variant nature. The amount of stored energy and battery life limits the long-term availability of base stations/access points [2], [8]. The aforementioned requirements make energy-efficient resource allocation in the wireless network (RAN) imperative imperative and hence, the need for an energy constraints investigation in the proposed framework.
3.2 Computational Cost of AI-Based Control
Optimisation using artificial intelligence helps manage resources in a smart, adaptive way in a dynamic wireless environment. However, the use of AI algorithms, specifically deep reinforcement learning, increases computational complexity and energy consumption at the control level and on edge devices [4], [6]. Otherwise, the additional cost of training, inference, and control in the energy domain could counteract the advantages in the other domain. In ultra-dense 6G and Wi-Fi 7/8 environments, the state and control space complexity increases significantly due to the presence of a large number of users and radio access technologies, as well as control parameters such as transmission power and RIS phase shift values. This results in significant real-time computational complexity involved in decision-making. In this regard, the current demand for lightweight learning models and convergence-oriented training algorithms is necessary to maintain the energy efficiency of AI control while ensuring a high level of optimisation performance. This challenge of control and computational complexity is an important design consideration in the suggested framework.
3.3 Propagation Challenges for Millimeter-Waves and Terahertz
Since the millimetre-wave and terahertz bands are used in 6G and next-generation Wi-Fi networks to support ultra-high data rates and large bandwidths, the propagation conditions encountered by these bands are more challenging. The millimetre-wave and terahertz bands exhibit high path loss, high sensitivity to blockages, high atmospheric absorption, and lower diffraction capacity [14], [27][28]. Conventional power control and beamforming approaches may not be practical at counteracting these impairments without resorting to a substantial increase in transmission power. It was the reason for including reconfigurable intelligent surfaces (RIS) in the proposed model. The use of RIS can mitigate impairments in the wireless channel through passive beam steering and signal reflection. As a result, the need to counter impairments in high-frequency transmission becomes a challenge that directly supports the integration of transmission with RIS in the proposed model [5], [10], [12].
3.4 Multi-RAT Coexistence and Spectrum Sharing
The coexistence of cellular networks and WLANs demands smart coordination to manage interference and efficiently use the spectrum [1], [19]. Operation of different RATs in an autonomous fashion would lead to an inefficient spectrum use and higher power consumption. These reasons provide a rationale for using the cognitive spectrum management technique and artificial intelligence-powered multi-RAT coordination in the proposed system [18].
3.5 Integration of Non-Terrestrial Networks
Non-terrestrial networks offer advantages in terms of coverage extension and ruggedness. However, there are concerns regarding propagation delay, Doppler shift, and dynamic topological changes associated with these types of networks [7], [9]. Consequently, NTNs are used as supplementary connectivity services in the proposed framework, which involves AI control of the optimisation methodology [8].
4 The Synergistic Framework
A synergistic framework for 6G and Wi-Fi 8 will be needed to achieve the high energy efficiency (EE) and spectral efficiency (SE) required to meet the demand for both technologies [29].
Fig. 1
The Proposed Synergistic Framework represents the proposed synergistic framework with a closed-loop integration of the AI Controller/Orchestrator, Cognitive Spectrum Management (CSM), Hybrid Energy Storage, and Reconfigurable Intelligent Surfaces (RIS).
A synergistic framework (Fig. 1) has been constructed around the tight coupling of AI, cognitive spectrum management (CSM) and RIS with the goal of creating a self-optimizing system (via an optimization loop) to allow for the real-time management of the network conditions as opposed to relying on static resource allocation.
4.1 System Model and Problem Formulation
The system model is defined with a focus on the two leading performance indicators, Spectral Efficiency (SE) and Energy Efficiency (EE), to test the proposed synergistic framework. Imagine a hybrid network with K users that is served by a multi-RAT architecture (6G and Wi-Fi 8) and a Reconfigurable Intelligent Surface (RIS) with N reflecting elements.
Spectral Efficiency (SE)
The Shannon capacity formula, adapted for RIS-assisted channels, is used to determine the feasible data rate Rk for the k-th user using bandwidth B. This is based on the model in [10]. The Signal-to-Interference-plus-Noise Ratio (SINR), represented as
, is specified in (1).
where:
Φ = diag(ejθ1,..., ejθN) is the phase-shift matrix of the RIS, where θn is the phase shift of the n-th element.
The system Spectral Efficiency (SE) in bits/s/Hz is defined as in (2):
Energy Efficiency (EE)
Energy efficiency is the ratio of total network throughput to the total power consumed. The total power consumption Ptotal is modeled as the sum of the transmission power, the hardware circuit power Pcirc, and the power consumed by the RIS elements PRIS, as in [3].
Where ξ is the power amplifier efficiency, N is the number of RIS elements, and Pn is the power consumption of a single active RIS element. Consequently, the global Energy Efficiency (EE) in bits/Joule is formulated as (4).
Optimization Problem
The optimisation task in this study aims to enhance the energy efficiency (EE) of a hybrid wireless network that supports both 6G and Wi-Fi 8 applications while satisfying Quality of Service (QoS) requirements. The goal is to find the best combination of transmit power allocation P and phase shift configuration Φ for the RIS to get the most energy efficiency. The optimisation problem is formulated with respect to two sets of variables and is given by (5).
This is done with three main restrictions. To start, a Quality of Service (QoS) constraint makes sure that each user gets a minimum needed data rate, Minimum Rate per user (
which is defined as in (6).
Second, a power limit is put in place to keep the overall transmit power for all users below a certain level, so that the total transmission power limits to stay within the hardware or regulatory constraints, given by (7).
Finally, a unit-modulus constraint on the RIS elements, 𝜙𝑛, is enforced to show that passive RIS elements can only change the phase of a signal without increasing its amplitude, as given in (8).
These limits work together to create a feasible, realistic approach to improving the energy efficiency of the next-generation intelligent wireless network. This non-convex optimisation problem motivates the use of the AI-driven controller described in Section 5, which utilises Deep Reinforcement Learning (DRL), dynamically adjusting the power allocation matrix P and RIS phase shifts Φ in real-time.
4.2 Interaction and System Components
The framework integrates the different core and auxiliary connectivity technologies in a closed optimization loop.
AI as the Central Controller
The AI engine is the primary and ongoing decision-maker for the entire system. It collects real-time data on the availability of energy from hybrid renewable sources, user mobility, transportation demand, and channel quality indicators obtained using a spectrum-sensing mechanism. The AI makes decisions regarding adaptive energy management (i.e., low-demand access points entering sleep mode) and dynamically manages spectrum resources, maximising efficiency across the entire network [2],[6][30].
Cognitive Spectrum Management (CSM): As one of its main functional layers, the proposed architecture includes Cognitive Spectrum Management (CSM). Cognitive Spectrum Management enables radios to sense their surroundings and use available spectrum effectively in real time. Cognitive Spectrum Management uses the concept of Cognitive Radio and Dynamic Spectrum Access in its framework work [18]. Dynamic Spectrum Sharing (DSS) is used as an execution layer within Cognitive Spectrum Management at this level. DSS enables adaptability in sharing available spectrum between co-existing 6G and Wi-Fi 7/8 networks based on real-time communication demands, channel conditions, and available energy resources [6], [18]. Moreover, the coordination between the multi-RAT and DSS increases this efficiency by providing smooth traffic steering and handover between cellular and Wi-Fi communications. Instead of considering the DSS and the multi-RAT technologies independently, this new framework combines these two techniques under the management of AI (which uses learning algorithms to estimate traffic, mobility, and channel availability to achieve maximum energy and spectral efficiency simultaneously) [1], [4], [6].
With the integration of DSS and Multi-RAT technologies into cognitive spectrum management, this framework not only supports adaptable spectrum reuse but also helps to achieve better QoS and lower energy consumption in heterogeneous networks. In the 6G networks, intelligent, flexible, and energy-aware management of the spectrum will be an essential requirement [9], [11].
RIS Enhanced Propagation Control
This uses RIS panels in the environment to change the wireless channel [5], [12]. The AI controller uses the perceived Channel State Information (CSI) and the planned data path (or beamforming direction) to determine the ideal phase-shift setting for the RIS. This lets the system get around signal blockage, reduce interference in busy locations, and passively direct energy to end devices. This reduces the power required to send signals through central base stations and makes links more reliable [5], [12].
The system proposed continuously solves the dynamic optimisation problem of maximising network throughput while minimising total energy use.
5 Simulation Results and Performance Analysis
To test the efficacy of the proposed framework, simulations have been conducted based on the system model described in Section 4. The parameters considered in the simulation include a 6G small-cell environment at 28 GHz with a system bandwidth of 400 MHz.
The simulation environment consists of a Base Station (BS) at (0, 0, 10) m, an RIS at (50, 20, 10) m, and K = 10 randomly distributed Users in a cluster with a 20m radius around the RIS. The BS-RIS and Users-RIS channels are modelled using the Rayleigh/Rician Channel Model. The Signal Noise Density is set to -174 dBm/Hz.
5.1 Spectral Efficiency as a Function of the Number of RIS Elements
The first performance evaluation considers the effect of the Number of Elements in the RIS on Spectral Efficiency. Here, Spectral Efficiency is calculated as a function of the number of elements (N) in the RIS. The proposal is then compared with the baseline schemes with respect to Spectral Efficiency. Figure 2 shows the performance evaluation results.
Fig. 2
Spectral Efficiency(bits/s/Hz) vs Number of RIS Elements(N)
As depicted in Fig. 2, the Spectral Efficiency of the system is observed to improve logarithmically with the increase in the Number of Elements in the RIS. The proposed framework clearly outperforms the baseline schemes. Hence, the effect of optimized phase values on improving the path loss in 6G frequency band when using the RIS technique cannot be overemphasized. The proposed technique considers 64 antenna elements to improve the performance of the random phase shift scheme by 35%, thereby affirming that optimized phase values play an integral role in reducing the effect of the path loss in 6G radio access networks (RANs).
5.2 Energy Efficiency as a Function of Transmit Power
The performance evaluation considers the effect of the Transmit power on Energy Efficiency. Here, the performance evaluation as a function of the Transmit power is presented. Fig, 3 shows the comparative performance evaluation results.
Fig. 3
Energy Efficiency (Mbits/Joule) vs Transmit Power(dBm)
The performance of the proposed technique is observed to possess higher peak values of energy efficiency than the conventional methods, (see Fig. 3). Hence, the combination of the 6G frequency-band path loss and the optimised phase values in the RIC technique can operate in the green mode without impairing connectivity.
6 Comparative Analysis
The main emerging/enabling technologies for next-generation wireless communication are summarised in Table 1, which highlights their applications, advantages, and drawbacks. RIS technology enables low-power, high-gain cognitive access; however, it requires an expensive setup and installation [5], [12]. AI optimisation improves QoS and reduces energy use, but it requires substantial computing power [6]. NTNs offer a wider range and offload spectrum; however, they induce delay and Doppler shifts [7], [8]. Cognitive radios enable effective spectrum reuse; yet they require intricate processing [15]. Hybrid renewables provide clean energy but depend on the weather [2], [3]. Multi-RAT communication enables fast, flexible communication, but the biggest challenge is the cost of integrating multiple technologies [1][20],[21].
|
Technology
|
Purpose
|
Benefits
|
Limitations
|
Source
|
||
|---|---|---|---|---|---|---|
|
RIS
|
Cognitive Access
|
Low power, high gain
|
Cost of installation
|
[5], [12]
|
||
|
AI Optimization
|
Cognitive Control
|
Reduced Energy Cost, Improved Quality of Service
|
High computation
|
[6]
|
||
|
NTNs
|
Extended Coverage
|
Spectrum Offload
|
High Latency (GEO/MEO) or Doppler Shift / Platform Instability
|
[7], [8]
|
||
|
Cognitive Radios
|
Opportunistic access
|
Efficient reuse of spectrums
|
Complex sensing
|
[18]
|
||
|
Hybrid Renewables
|
Sustainable source of energy
|
Lowering of OPEX
|
Dependence on weather
|
[2], [3]
|
||
|
Multi-RAT
|
Flexible access
|
High data rate
|
Integration cost
|
[1]
|
||
The Wi-Fi standards in Table 2 show that spectral efficiency is improving rapidly due to better modulation techniques, broader bandwidth allocations, and multi-link operations. Wi-Fi 7 and the upcoming Wi-Fi 8 standards are far better than older versions. They allow high-speed throughput over short distances, making them ideal for crowded indoor spaces. However, their efficiency gains depend strongly on the availability of channel width, which may vary across regions.
|
Reference
|
Standard
|
Frequency Range, GHZ
|
Energy
Efficiency
|
Spectrum
Efficiency
|
Peak Data Rate
|
Channel Width, MHz
|
|---|---|---|---|---|---|---|
|
[19]
|
Wi-Fi 4
|
2.4 / 5
|
Low
|
Moderate
|
≤ 0.6 Gbps
|
20/40
|
|
[19]
|
Wi-Fi 5
|
5
|
Moderate
|
High
|
≤ 6.9 Gbps
|
20–160
|
|
[19]
|
Wi-Fi 6
|
2.4 / 5
|
High
|
Very High
|
≤ 9.6 Gbps
|
20–160
|
|
[1]
|
Wi-Fi 6E
|
2.4 / 5 / 6
|
High
|
Very High
|
≤ 9.6 Gbps
|
20–160
|
|
[1]
|
Wi-Fi 7
|
2.4 / 5 / 6
|
Very High
|
Extremely High
|
≤ 46 Gbps
|
≤ 320
|
|
[1], [6],[8]
|
Wi-Fi 8
|
2.4 / 5 / 6+
|
Ultra-High
|
Ultra-High
|
≥ 46 Gbps
|
TBD
|
In contrast, 4G/5G/6G cellular networks (Table 3) place greater emphasis on wide coverage and mobility support, thus offering comparatively lower peak data rates but much higher scalability. Enhanced by reconfigurable intelligent surfaces and terahertz bands, 6G systems will achieve unprecedented spectral and energy efficiency, although at the cost of shorter cell radii due to propagation limitations. In summary, this comparison shows that Wi-Fi provides the highest targeted throughput. In contrast, 6G provides much greater flexibility in wireless coverage, underscoring the need for integrated hybrid architectures that jointly exploit both technologies for sustainable performance.
The proposed hybrid Energy- and spectrum-efficient wireless network design for 5G/6G and Wi-Fi 7/8 ecosystem will enable future-generation deployments of full duplex net zero / green satellite-cellular/Wi-Fi data communications for ultra-low carbon reconfigurable transceivers [30–33] and regenerative transponders [30, 34–38]; advanced distributed MIMO antenna [21, 29, 39], hybrid RF-solar energy harvesting [20, 21, 40] and holographic metasurface beamforming technologies. It promises to enable full digital circular economy uses case and/or applications spanning reconfigurable variable latency and data rate communications [41]; multi-purpose wireless 5G/6G/Wi-Fi-based biosensing, mobile emergency hospital and telemedicine [42–45]; autonomous vehicle-to-infrastructure communication [41]; supply chain inventory and building occupancy monitoring [20]; LEO-based remote monitoring [46]; and integrated smart cities internet of things sensing and communication [20].
|
Reference
|
Standard
|
Frequency Range
|
Energy
Efficiency
|
Spectrum
Efficiency
|
Peak Data Rate
|
Cell
Radius
|
|
|---|---|---|---|---|---|---|---|
|
[25]
|
4G / LTE
|
Sub-6 GHz
|
Moderate
|
Moderate
|
≤ 1 Gbps
|
up to 1 km
|
|
|
[9] [11]
|
5G
|
Sub-6 / mmWave
|
High
|
Very High
|
≤ 20 Gbps
|
≤ 500 m
|
|
|
[9][11][27]
|
6G (target)
|
mmWave / THz
|
Ultra-High
|
Ultra-High
|
≥ 1 Tbps
|
≤ 200 m
|
|
7 Conclusion
This paper provides a comprehensive, novel hybrid energy- and spectrum-efficient framework that can converge 5G/6G and Wi-Fi 7/8 by integrating AI-controlled radio operations, cognitive radio spectrum, and propagation technologies enabled by RISs. A model was designed that provided an RIS-assisted multi-RAT framework. A deep reinforcement learning strategy was provided that enabled the maximisation of energy efficiency with AI while maintaining service quality. Results showed that the designed framework increased energy efficiency by 35% compared to the existing baseline solutions while improving spectral efficiency in high-frequency bands. These simulations prove that energy/power allocation, spectrum utilisation, and propagation technologies should be optimised together to provide environmentally friendly, future-proof wireless networks. Future simulations will proceed to decrease the complexity of AI systems while also validating their integrity in real-world environments.
A
Author Contribution
Conceptualization, Zacheous Aasa and Fanuel Elias; Data curation, Zacheous Aasa and Fanuel Elias; Formal analysis, Zacheous Aasa, Fanuel Elias and Sunday Ekpo; Investigation, Fanuel Elias, Zacheous Aasa and Sunday Ekpo; Methodology, Zacheous Aasa, Fanuel Elias and Sunday Ekpo; Project administration, Fanuel Elias, Zacheous Aasa and Sunday Ekpo; Resources, Sunday Ekpo, Fanuel Elias; Software, Zacheous Aasa and Fanuel Elias; Supervision, Sunday Ekpo; Validation, Zacheous Aasa, Fanuel Elias and Sunday Ekpo; Visualization, Fanuel Elias and Zacheous Aasa; Writing – original draft, Fanuel Elias and Zacheous Aasa; Writing – review & editing, Fanuel Elias, Zacheous Aasa and Sunday Ekpo.
A
Funding
This research received no external funding.
A
Data Availability
The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding author.
The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding author.
Declarations
Ethics approval and consent to participate
Not applicable, this research does not contain any studies involving human participants or animals performed by any of the authors.
A
Clinical trial
Not applicable
Policy statement
The findings and interpretations presented are solely those of the author and do not reflect the views of any person, agency, or institution. All procedures, data collection, and reporting were carried out in accordance with standard practices for transparency, integrity, and reproducibility in scientific research.
Consent for publication
Not applicable. This research does not contain any person’s data in any form.
Competing interests
All authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
References
1.
Yang M, Li B. Survey and perspective on extremely high throughput (EHT) WLAN IEEE 802.11be. Mob Networks Appl. 2020;25(5):1765–80. https://doi.org/10.1007/s11036-020-01567-7.
2.
Aarti S, Gowroju SI, Ansarullah S. A comprehensive review of developments and challenges in the 6G Internet of Things. Discover Networks. 2025;1:13. https://doi.org/10.1007/s44354-025-00013-y.
3.
Lazrek H, El Ferindi H, Zouiten M, Moumen A, et al. Enhancing energy efficiency in 5G networks through AI-driven dynamic discontinuous reception. Discover Comput. 2025;28:245. https://doi.org/10.1007/s10791-025-09765-1.
4.
Jagannath J, Polosky N, Jagannath A, Restuccia F, Melodia T. Machine learning for wireless communications in the Internet of Things: A comprehensive survey. Ad Hoc Netw. 2019;93:101913. https://doi.org/10.1016/j.adhoc.2019.101913.
5.
Renzo MD, Debbah M, Phan-Huy DT et al. Smart radio environments empowered by reconfigurable AI meta-surfaces: an idea whose time has come. J Wireless Com Network 2019, 129 (2019). https://doi.org/10.1186/s13638-019-1438-9
6.
Cui QM, You XH, Wei N, et al. Overview of AI and communication for 6G network: Fundamentals, challenges, and future research opportunities. Sci China Inform Sci. 2025;68:171301. https://doi.org/10.1007/s11432-024-4337-1.
7.
Matin MA, Goudos SK, Wan S, Sarigiannidis P, Tentzeris EM. Artificial intelligence (AI) and machine learning (ML) for beyond 5G/6G communications. Journal of Wireless Communications and Networking 2023, 22 (2023). https://doi.org/10.1186/s13638-023-02212-z
8.
Nguyen CT, et al. Emerging Technologies for 6G Non-Terrestrial-Networks: From Academia to Industrial Applications. IEEE Open J Commun Soc. 2024;5:3852–85. 10.1109/OJCOMS.2024.3418574.
9.
Tataria H, Shafi M, Molisch AF, Dohler M, Sjöland H, Tufvesson F. 6G Wireless Systems: Vision, Requirements, Challenges, Insights, and Opportunities, in Proceedings of the IEEE, vol. 109, no. 7, pp. 1166–1199, July 2021, 10.1109/JPROC.2021.3061701
10.
Wu Q, Zhang S, Zheng B, You C, Zhang R. Intelligent Reflecting Surface-Aided Wireless Communications: A Tutorial. IEEE Trans Commun. May 2021;69(5):3313–51. 10.1109/TCOMM.2021.3051897.
11.
Zhang Z et al. Sept., 6G Wireless Networks: Vision, Requirements, Architecture, and Key Technologies, in IEEE Vehicular Technology Magazine, vol. 14, no. 3, pp. 28–41, 2019, 10.1109/MVT.2019.2921208
12.
Zhang H, Di B, Song L, Han Z. Introductions and Basics. Reconfigurable Intelligent Surface-Empowered 6G. Wireless Networks. Cham: Springer; 2021. https://doi.org/10.1007/978-3-030-73499-2_1.
A
13.Lynn T, Endo PT, Ribeiro AMNC, Barbosa GBN, Rosati P. The Internet of Things: Definitions, Key Concepts, and Reference Architectures. In: Lynn T, Mooney J, Lee B, Endo P, editors. The Cloud-to-Thing Continuum. Palgrave Studies in Digital Business & Enabling Technologies. Cham: Palgrave Macmillan; 2020. https://doi.org/10.1007/978-3-030-41110-7_1.
14.
Jiang D, Liu G. An Overview of 5G Requirements. In: Xiang W, Zheng K, Shen X, editors. 5G Mobile Communications. Cham: Springer; 2017. https://doi.org/10.1007/978-3-319-34208-5_1.
15.
Elias F, Ekpo S, Alabi S, Olasunkanmi N, Unnikrishnan R, Enahoro S, Uko M, Ijaz M, Ji H, Wu Z. Comparative Analysis of MIMO-Based Rectenna Configurations for Energy Harvesting in Ultra-Low Power Applications. In: Ekpo SC, editor. The Third International Adaptive and Sustainable Science, Engineering and Technology. ASSET 2024. Signals and Communication Technology. Cham: Springer; 2025. https://doi.org/10.1007/978-3-031-89537-1_7.
16.
Enahoro S, Ekpo S, Uko M, Alabi S, Elias F, Unnikrishnan R. A metamaterial-grounded ultra-wideband cross-fractal MIMO antenna for K, Ka, and mmWave applications. In: Ekpo SC, editor. The Third International Adaptive and Sustainable Science, Engineering and Technology. ASSET 2024. Signals and Communication Technology. Cham: Springer; 2025. https://doi.org/10.1007/978-3-031-89537-1_23.
17.
Ghosh S, Saha D, Chakraborty A, Chakraborty S, Ekpo SC, Elias F. Design and Analysis of mm-Wave MIMO SIW Antenna for Multibeam 5G Applications, 2023 IEEE-APS Topical Conference on Antennas and Propagation in Wireless Communications (APWC), Venice, Italy, 2023, pp. 154–159. 10.1109/APWC57320.2023.10297489
18.
Haykin S. Cognitive radio: brain-empowered wireless communications, in IEEE Journal on Selected Areas in Communications, vol. 23, no. 2, pp. 201–220, Feb. 2005, 10.1109/JSAC.2004.839380
19.
Khorov E, Kiryanov A, Lyakhov A, Bianchi G. A Tutorial on IEEE 802.11ax High Efficiency WLANs, in IEEE Communications Surveys & Tutorials, vol. 21, no. 1, pp. 197–216, Firstquarter 2019, 10.1109/COMST.2018.2871099
20.
Uko M, Elias F, Ekpo S, Saha D, Ghosh S, Ijaz M, Chakraborty S, Gibson A. Hybrid Wireless RF-Perovskite Photovoltaic Energy Harvester Design Consideration for Low-Power Internet of Things, 2023 IEEE-APS Topical Conference on Antennas and Propagation in Wireless Communications (APWC), Venice, Italy, 09–13 Oct. 2023, pp. 173–176. https://10.1109/APWC57320.2023.10297436
21.
Elias F, Ekpo S, Alabi S, Uko M, Enahoro S, Ijaz M, Ji H, Unnikrishnan R, Olasunkanmi N. Design of Multi-Sourced MIMO Multiband Hybrid Wireless RF-Perovskite Photovoltaic Energy Harvesting Subsystems for IoTs. Appl Smart Cities Technol. 2025;13(3):92. https://doi.org/10.3390/technologies13030092.
22.
Zhang H, Ganchev I, Ji Z, O’Droma M. (2017). A Hybrid Service Recommendation Prototype Adapted for the UCWW: A Smart-City Orientation. https://doi.org/10.1155/2017/6783240
23.
Enahoro S, Ekpo SC, Uko M, Elias F, Unnikrishnan R, Alabi S, Olasunkanmi NK. (2025). Sustainable THz SWIPT via RIS-Enabled Sensing and Adaptive Power Focusing: Toward Green 6G IoT. Sensors, 25(15), 4549. https://doi.org/10.3390/s25154549
24.
Li Z, Dubey A, Shen S, Kundu NK, Rao J, Murch R. (2024). Radio Tomographic Imaging with Reconfigurable Intelligent Surfaces. https://doi.org/10.1109/twc.2024.3433011
25.
Dahlman E, Parkvall S, Sköld J. LTE evolution and advanced mobile broadband, Telecommunication Systems, Springer, vol. 66, no. 3, pp. 369–381, 2017.
26.
He S, Peng C, Huang W, An Z, Qian Y, Liu L. Toward Wi-Fi 8 Standard: A Survey of State-of-the-Art Technologies. IEEE Open J Commun Soc. 2025;6:10150–70. 10.1109/OJCOMS.2025.3637585.
27.
Sudhamani C, Roslee M, Ismail A, et al. Effects of Atmospheric Gases and Rain Intensity on Terahertz Wave Propagation in 6G Wireless Networks. Wirel Pers Commun. 2025;142:263–86. https://doi.org/10.1007/s11277-025-11807-2.
28.
Abedeen Z, Ekpo S, Elias F, Ijaz M, Raza U, Alabi S, Han L. (2024). Path Loss Prediction of 5G in the 24.25–27.5 GHz Band Based on Machine Learning In: Ekpo, S.C, editors The Second International Adaptive and Sustainable Science, Engineering and Technology Conference. ASSET 2023. Signals and Communication Technology. Springer, Cham. https://doi.org/10.1007/978-3-031-53935-0_3
29.
Fanuel Elias SC, Ekpo S, Alabi D, Saha S, Chakraborty S, Ghosh MO, Uko M, Ijaz U, Raza. (2024). Rectifier and Reconfigurable Impedance-Matching Network Analysis for Wireless Sub-6 GHz 5G/Wi-Fi 6/6E Energy Harvester. In: Ekpo, S.C, editors The Second International Adaptive and Sustainable Science, Engineering and Technology Conference. ASSET 2023. Signals and Communication Technology. Springer, Cham. https://doi.org/10.1007/978-3-031-53935-0_8
30.
Ekpo S, George D. Impact of Noise Figure on a Satellite Link Performance. IEEE Commun Lett. June 2011;15(9):977–9. https://doi.org/10.1109/LCOMM.2011.072011.111073.
31.
Sunday C, Ekpo. Rupak Kharel and Mfonobong Uko, A Broadband LNA Design in Common-Source Configuration for Reconfigurable Multi-standards Multi-bands Communications, in Proc., ARMMS RF & Microwave Society Conference, Double Tree by Hilton Oxford Belfry, Thame, UK, 01 & 02 April 2018, pp. 1–10.
32.
Ekpo S, Kettle D. ‘‘mm-Wave LNAs design for Adaptive small Satellite Applications,’’ in Proc. Joint 5th ESA Workshop Millimetre Wave 31st ESA Antenna Workshop, 2009, pp. 843–847.
33.
Ekpo S, George D. ‘‘4–8 GHz LNA design for a highly adaptive small satellite transponder using InGaAs pHEMT technology,’’ in Proc. IEEE 11th Annu. Wireless Microw. Technol. Conf. (WAMICON), Apr. 2010, pp. 1–4.
34.
Ekpo SC, George D. A system engineering consideration for future-generations small satellites design, 2012 IEEE First AESS European Conference on Satellite Telecommunications (ESTEL), Rome, Italy, 2012, pp. 1–6. 10.1109/ESTEL.2012.6400067
35.
Uko M, Ekpo S, Elias F, Enahoro S, Ukommi U, Unnikrishnan R, Iwok UU, Inyang A. (2025). Artificial neural network modelling and characterization of a 3.2 to 3.8 GHz low noise amplifier for sub-6GHz applications In: Ekpo, S.C, editors The Third International Adaptive and Sustainable Science, Engineering and Technology. ASSET 2024. Signals and Communication Technology. Springer, Cham. https://doi.org/10.1007/978-3-031-89537-1_17
36.
Ekpo S, George D. A System-based Design Methodology and Architecture for Highly Adaptive Small Satellites, 2010 IEEE International Systems Conference, San Diego, CA, USA, 2010, pp. 516–519. 10.1109/SYSTEMS.2010.5482323
37.
Ekpo S, George D. Reconfigurable Cooperative Intelligent Control Design for Space Missions. Recent Pat Space Technol. April 2012;2(1):2–11. https://doi.org/10.2174/1877611611202010002.
38.
Ekpo S, George D. A Deterministic Multifunctional Architecture Design for Highly Adaptive Small Satellites, International Journal of Satellite Communication Policy and Management, Vol. 1, No. 2/3, pp. 174–194, August 2012; https://doi.org/10.1504/IJSCPM.2012.049543
39.
Sunday C, Ekpo F, Elias MC, Uko S, Enahoro S, Alias M, Ijaz R, Unnikrishnan, Olasunkanmi N. Multi-Mode Multi-Source Electrical Power Subsystem Design for CubeSats-Internet of Things Missions. IEEE Access J. 2025;13:164965–84. https://doi.org/10.1109/ACCESS.2025.3612339. 19 September 2025.
40.
Zafar M, Ekpo S, George J, Sheedy P, Uko M, Gibson A. Hybrid Power Divider and Combiner for Passive RFID Tag Wireless Energy Harvesting, in IEEE Access, 10, pp. 502–15, 04 January 2022; https://doi.org/10.1109/ACCESS.2021.3138070
41.
Ekpo S, Adebisi B, Wells A. Regulated-element Frost Beamformer for Vehicular Multimedia Sound Enhancement and Noise Reduction Applications. IEEE Access J. December 2017;5:27254–62. https://doi.org/10.1109/ACCESS.2017.2775707.
42.
Iain Lau SC, Ekpo M, Zafar M, Ijaz, Gibson A. IEEE Access. May 2023;11:42850–61. https://doi.org/10.1109/ACCESS.2023.3270777. Hybrid mmWave-Li-Fi 5G Architecture for Reconfigurable Variable Latency and Data Rate Communications;.
43.
Jeena George M, Uko S, Ekpo F, Elias. Design of an Elliptically-slotted Patch Antenna for Multi-purpose Wireless Wi-Fi and Biosensing Applications, e-Prime – Advances in Electrical Engineering. Electron Energy J. Dec. 2023;6:1–38. https://doi.org/10.1016/j.prime.2023.100368.
44.
Ansari U-E-H, Ekpo S, Uko MC, Altaf A, Zafar M, Enahoro S, Okpalugo O. 5G-enabled Mobile Operating Hospital and Emergency Care Service in Proc. 21st IEEE Annual Wireless & Microwave Conference, Sand Key, Florida, USA (Online/Virtual), 28–29 April 2021, pp. 1–6; https://doi.org/10.1109/WAMICON47156.2021.9443613
45.
Jeena George MC, Uko SC, Ekpo M, Ijaz R, Kharel Q, Wang, Ji H. Design of a Multiband RF Slotted-Antenna for Biosensing Applications, in Proc., 12th IEEE/IET International Symposium on Communication Systems, Networks and Digital Signal Processing Conference, Porto, Portugal (Online/Virtual), 20–22 July 2020, pp. 1–6; https://doi.org/10.1109/CSNDSP49049.2020.9249616
46.
Ekpo SC, Adebisi B, George D, Kharel R, Uko M. A System-level Multicriteria Modelling of Payload Operational Times for Communication Satellite Missions in LEO. Recent Progress Space Technol. June 2014;4(1):67–77. https://doi.org/10.2174/2210687104666140620221119.