An Innovative Architectural Blueprint Towards Sustainable 6G Systems

An Innovative Architectural Blueprint Towards Sustainable 6G Systems Stylianos E. Trevlakis∗, Maria Belesioti†, Harilaos Koumaras‡, Alexandros-Apostolos A. Boulogeorgos§, Ioannis Chochliouros†, and Theodoros A. Tsiftsis¶∗ ∗Department of Research and Development, InnoCube P.C., 17th Noemvriou 79, 55535 Thessaloniki, Greece. †Hellenic Telecommunications Organization (OTE) S.A., 15122 Athens, Greece. ‡National Centre of Scientific Research “Demokritos”, 15310 Athens, Greece. §Department of Electrical and Computer Engineering, University of Western Macedonia, 50100 Kozani, Greece. ¶Department of Informatics & Telecommunications, University of Thessaly, 35100 Lamia, Greece. Emails: tre[email protected], [email protected], [email protected], al.boulogeor[email protected]g, [email protected], [email protected] Abstract—The sixth generation (6G) era aims to push the capabilities of networks towards unprecedented frontiers, while ensuring sustainable design, development, and operation. Motivated by this, this work presents a sustainability-oriented 6G architecture and examines the trends in society, economy, environment, regulations, and technology. The envisioned architecture integrates intelligent, adaptable, energy efficient, transparent and secure elements, while its technology enablers, such as large language models, explainable artificial intelligence, semantic communications, digital twins, human-in-the-loop, blockchain, post-quantum communications, etc. are identified and analyzed. The role of each enabler is clarified. Taking sustainability aspects into consideration from the initial stages of the design process can be the deciding factor for the success or failure of environmentally conscious ecosystems. Index Terms—3D Networking, 6G, Architecture, Artificial Intelligence, Digital twins, Orchestration, Semantic Communications, Sensing, Sustainability. I. INTRODUCTION The evolution of fifth (5G) brings forth enhanced capabilities, increased efficiency, and the introduction of innovative applications for users. These advancements are reshaping the field of communications, while also paving the way for more eco-friendly practices [1], [2]. Mobile network operators (MNOs) are leveraging the attributes of 5G, such as network slicing, disaggregation, and intrinsic design to drive the creation of new applications and business models with a focus on sustainability. The ongoing transition towards digitalization is predicted to generate a significant volume of data while ensuring a sustainable digital footprint. However, these emerging applications demand robust key performance indicators (KPIs) than what current 5G networks offer. MoreThe work of S. E. Trevlakis has been supported by the European Unions Horizon-CL4-2021 research and innovation programme under grant agreement No. 101070181 (TALON). The work of M. Belesioti and I. Chocliouros has been based sypported by the 6G-PATH Project. 6GPATH has received funding from the Smart Networks and Services Joint Undertaking (SNS JU) under Grant Agreement No. 101139172. The work of A.-A. A. Boulogeorgos has been supported by the MINOAS project. The research project MINOAS is implemented in the framework of H.F.R.I call “Basic research Financing (Horizontal support of all Sciences)” under the National Recovery and Resilience Plan “Greece 2.0” funded by the European Union – NextGenerationEU (H.F.R.I. Project Number: 15857).” over, they necessitate exploration into their sustainability aspects, which is still in its early phases. The rollout of 6G networks is expected to enhance communication capabilities between devices and systems enhancing the performance of existing 5G systems to unparalleled levels and enabling the development of novel and more intricate use cases [3], [4]. Considering this viewpoint and acknowledging that 6G services will have an impact on service performance it is vital to evaluate various key value indicators (KVIs) that gauge the impacts of 6G use cases, on the environment, society, and economy. In this narative, sustainability refers to the approach of developing, operating, and incorporating advancements and technologies in an environmentally friendly manner. It involves principles that aim to improve the social and economic impact while maximizing long term benefits. Key elements of these principles include •Taking on responsibility by adopting energy efficient practices and utilizing renewable energy sources to reduce energy consumption, carbon dioxide emissions, and electronic waste. •Efficient use of resources is crucial for safeguarding sustainability. This includes minimizing the use of network resources such as infrastructure and computational power. •Alongside providing connectivity, sustainability fosters economic growth and enhances quality of life. Designing networks that can adapt securely and efficiently to changing environments while meeting communication needs is essential. Establishing an end-to-end (e2e) 6G architecture is necessary, while taking into account sustainability models from the start is will be a significant step towards achieving the United Nations’ sustainability development goals (UN SDGs). In network design, sustainability goes beyond enhancing communication reliability. It involves establishing a network architecture that safeguards the environment while also promoting economic growth. So far, sustainability has been regarded as an afterthought towards optimizing the operation of communication and networking systems. Having identifies this gap, the present contribution This is the accepted manuscript version of the paper prior to IEEE formatting and copyediting. © 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses. •Delves into the vision of future 6G ecosystems and their application scenarios. This analysis highlights the impact of 6G in sustainability that manifests in both indirect ways by shaping innovative use cases that enhance service delivery, while also positively impacting societal, economic, and environmental aspects. •Present a 6G network architecture that takes into consideration the identified sustainability aspects from the ecosystem design phase; thus, ensuring realistic development pathways from key values to KVIs and, eventually, to key performance indicators (KPIs). •Discusses the fundamental enabling technologies of the proposed architecture including 6G sensing, 3D networking, network orchestration, semantic communications, XR, robotic vision, hierarchical digital twins (DTs), post-quantum cryptography (PQC), blockchain, secure lifecycle management, explainable AI (XAI), and human-in-the-loop (HIL). •Picks precision healthcare as an example to illustrate the impact of the proposed architecture and next generation technologies on the upcoming 6G era alongside environmental, societal, and economic sustainability considerations. II. 6G APPLICATION SCENARIOS &REQUIREMENTS The introduction of 6G represents a milestone in the field of telecommunications. This era is characterized by the integration of features like enhanced sensing, artificial intelligence (AI), improved localization, and more. These advancements are pivotal in ushering in a wave of experiences, such as immersive reality while underscoring the critical need for robust privacy protections. It seeks to support the notion that 6G networks will serve as an enabler for driving sustainability across various sectors. While 5G networks have paved the way for creating digital twins (DTs) of individual systems transitioning to 6G is deemed essential for implementing comprehensive e2e DTs. These encompass entire industrial/urban infrastructures and extensive networks to deliver time operational capabilities. The deployment of 6G technology is imperative for establishing connections with minimal latency crucial for seamlessly integrating commercial real-time DTs into production environments. Furthermore, the shift to 6G is expected to enhance network sensing capabilities and location accuracy thereby boosting the performance of existing DT models. Efficient robotic cooperation hinges on sharing of vital local data. Improving communication protocols among connected devices in subnetworks and devising strategies for managing local communication resources are essential tasks. When mobile robots need to stay connected to their surroundings they might need the capability to initiate and coordinate reservation protocols. Recent progress in AI has significantly enhanced the collaborative abilities of robotic entities. Delegating tasks to nearby edge nodes can boost robots efficiency by strengthening their inherent processing capabilities. Future systems are expected to leverage data from sensory mechanisms, diverse applications, or collective knowledge bases to enhance application performance and communication efficiency. Additionally, integrating sensory features into the radio interface is likely to streamline vital data collection for advancements in AI. In the phases of developing 6G technology integrating terrestrial networks (TN) and non-TN (NTN) presents a cost effective strategy to ensure continuous connectivity. The advancement of 6G technology aims to address challenges in areas with infrastructure marked by coverage gaps and limited network services. These regions often struggle to allocate resources. A key aspect of this progress focuses on enhancing video service quality. Implementing 6G technology is crucial for providing access across various conditions offering affordable solutions for essential services. Expected improvements in KPIs such as user data rate, traffic capacity, and accuracy of positioning are projected to exceed the capabilities of current 5G networks. Anticipated advancements in 6G technology include reducing e2e latency, addressing the limitations of 5G for applications requiring ultra reliable low latency communications (URLLC). Utilizing millimeter wave (mmWave) technologies and advancements in sub-THz frequencies is crucial for achieving the desired latency objectives and reducing interference. The integration of 6G plays a significant role in providing dependable services that cater to diverse quality of service (QoS) requirements across various applications while also ensuring the protection of privacy and security is prioritized. Moreover, 6G offers cost connectivity solutions for a wide array of devices. Establishing an infrastructure that is essential for the successful adoption of 6G technology ensuring secure communication channels regardless of the technology providers origin. With its capabilities, 6G ensures secure, affordable, and energy efficient connectivity significantly contributing to sustainability efforts. This impact manifests in both indirect ways by shaping innovative use cases that enhance service delivery, while also positively impacting societal, economic, and environmental sustainability. Furthermore, 6G is poised to provide cellular options suitable for both emerging and extensive network deployments; thus, facilitating broad implementation and acceptance, particularly in economically disadvantaged regions. To ensure the integration of 6G technology, it is crucial to proactively address the aforementioned challenges. The next generation of 6G networks aims to confirm the sustainability aspects of networks through various applications focusing on key areas of sustainable development, including: •Seamless extended reality (XR): Experiencing immersive environments through XR creates a deep feeling of presence and engagement for the users. It fulfills our desire for genuine and self-directed interaction in virtual spaces that encompass visual perception, spatial audio, and haptic feedback. Ensuring data synchronization and communication among participants is necessary for supporting immersive applications, such as telepresenceenabled collaborative work, educational tools, real-time gaming, and development of engaging content. As re- search on XR advances, we can anticipate creative applications that blur the lines between reality and imagination, such as exploring sites, working together with peers, or embarking on virtual adventures against mythical creatures. •Network-assisted mobility: This use case belongs to the family of physical awareness, which integrates sensing, localization, communication, and networking intelligence. This combination allows for features, such as analyzing surroundings, monitoring movement, understanding context, predicting paths, providing navigation assistance, and avoiding collisions. For instance, network-assisted mobility applications supports vehicles, autonomous vehicles, drones, pedestrians, and bicycles by utilizing sensors built into them and information obtained from the network. These applications offer improved safety, effectiveness, and on-the-spot decision making. •Ubiquitous connectivity: The concept of fullconnectivity pertains to the seamless availability of network access and service coverage for all individuals worldwide. It includes high speed mobile broadband services that enable people to communicate, connect to the internet, stream content, and utilize applications from any location. In times of emergencies or disasters, ubiquitous connectivity plays a significant role in facilitating real-time communication, as well as coordinating and sharing information among first responders, affected individuals, and relief organizations. Digital health services also benefit greatly from access by supporting tele-medicine, remote patient monitoring, and various health-related applications. Moreover, in an interconnected landscape, where supply chains depend on the exchange of real-time data, ubiquitous connectivity ensures communication among sensors, tracking devices, and autonomous vehicles to enhance visibility, efficiency, and responsiveness. •Human-centric services: The main goal of humancentric services is to develop improve the quality of life for individuals. These applications are deployed in a plethora of institutions, like hospitals, schools, and elderly care facilities, which aim to offer improved quality of well-being, safety, inclusion, and independence. The technological advances of 6G that are expected to play a key role in such use cases include sensing, AI/ML capabilities, and computational support to establish awareness in context driven actions. Trusted environments form part of a network structure that ensures stringent safeguarding of personal data. •Digital twinning: DTs are virtual replicas of real-world systems or processes that serve a variety of purposes, such as forecasting, maintenance, quality assurance, process optimization, supply chain management, product design, inventory tracking, and ensuring worker safety. An illustrative example describes a factory digital twins that can anticipate when machinery needs maintenance and suggest repairs in due time; thus, ensuring maximum production from the manufacturing line. In essence, DTs link the digital and physical worlds together and at the same time provide useful insights and opportunities for optimization across various fields. •Collaborative & cooperative robots: Collaborative and cooperative robots often referred to as cobots are changing the dynamics of how humans-robots interactions in shared work environments. They are created to operate effectively alongside human employees offering increased flexibility and versatility across various industries. For instance, cobots are utilized in manufacturing and assembly lines to boost efficiency of the procedures and lessen physical strain for humans. Moreover, stateof-the-art cobots play a role in logistics and warehouse operations by managing tasks, such as sorting materials, packing items, and maneuvering through congested areas. In healthcare applications they can provide support in dangerous activities, like disinfection procedures, while working alongside hospital personnel to ensure safety measures are met efficiently. III. NETWORK ARCHITECTURE The envisioned 6G architecture incorporates a structured approach to organize different components and functionalities. This framework, as shown in Fig. 1, is based on the initial 6G e2e system blueprint provided by the HexaX-II initiative and it is build with a specific focus on sustainability [5]. It offers the essential networking and communication resources needed to realize and deploy the use cases mentioned in the previous Section. The required high-level components mainly reside in the applications layer that enables the interaction between service providers and end users. Communication modules, predominantly softwaredriven, belong in the network functions layer, can be customized, and support multiple configurations, while the network applications layer provides the necessary tools for user-oriented network management. Moreover, a plethora of supporting functionalities are specifically designed and developed for covering advanced system optimization and coordination. The layers of the proposed 6G architecture are analyzed below. A. Applications The application layer consists of the components that cater to the specific needs of various applications and may require distinct services from the network. These requirements can vary based on factors such as bandwidth and latency. Control APIs can communicate with the network application layer to allow developers to forward their needs to the network and adjust app expectations based on network conditions. Key players of the application ecosystem like MNOs can access service management and orchestration APIs that are designed with intuitive intent-based approaches and user-friendly interfaces, like CAMARA or NEF. These applications can interact with the control plane of the network functions layer to deliver services to vertical applications, carry out actions or modifications, and optimize network performance. 6G sensing Satellites Cameras XR Sensors Data collection Supporting functionalities Intelligence Vision LLM DT SemCom XAI HIL Security & Privacy Secure lifecycle management Blockchain PQC M&O Orchestration 3D networking Applications Network apps Hardware Apps APIs Integration Transport NFs RAN NFs Core NFs App App Data NFs Servers XR Satellite Mobile Sensors Robot BS Laptop Edge-to-Cloud Continuum Network functions Cooperative & collaborative robots Seemless XR Network-assisted mobility Ubiquitous connectivity Digital twinning Human-centric services Environmental, societal, & economic sustainability Fig. 1. Sustainability-oriented 6G architectural blueprint. B. Network applications By using data gathering, analysis, and AI, network centered apps can improve network efficiency and optimization. These apps prioritize network capabilities and are managed by the network itself. They operate independently from userfocused app layers. This layer offers APIs for developers to access network focused services. These APIs grant access to a range of network functions enabling control of app flows without needing networking expertise. Standard methodologies like CAMARA and NEF have be utilized to standardize these APIs. Aggregation involves merging 6G network capabilities to deliver services efficiently. Enhancing the functionality of these APIs can be achieved by leveraging resources from 6G service providers and integrating with APIs from the following layers. C. Network functions The network functions layer is a system composed of various components, including UEs and subnetwork functions. The UEs protocol stack is controlled by the network through signaling procedures like resource allocation. UEs can collaborate in subnetworks using device-to-device (D2D) communications. A UE gateway is established that serves as a connection point between the subnetwork and the main 6G network. Devices in the subnetwork work either independently or with network based coordination. As a result, some UEs will be under network control while others will be managed by the UE gateway. Another key element is the radio access network (RAN) functions. Their deployment can be customized to meet needs across different scenarios like centralized, distributed, cloud based, and disaggregated setups. These scenarios involve network functions related to RAN that include both physical (PNFs) and virtualized network functions (VNFs). This increased focus on software enables orchestration, programmability, and analytics driven closed-loop resource management. In 5G, the core network (CN) functions have been integrated recently into a service oriented architecture approach, which lays the foundation for fully virtualizing the 6G CN to enhance flexibility and reconfiguration. This implies that the data plane functions can be placed near the network edge, while control plane functions can be located at the edge. By adopting a software defined networking architecture for transport networks it becomes feasible to make the elements within this infrastructure programmable. Virtualized transport network functions can be easily configured, deployed, activated/deactivated, or removed without disruption to services. Future 6G functionalities will not only improve communication capabilities but also introduce new features, like innovative RF-based sensing and 3D networking. Data gathered through network-based sensing can be enriched through intelligent analysis and integrated with information from various network nodes connected to the network. AI related functions will be added as capabilities, alongside 6G positioning and mapping features extending beyond just communications. This integration will lead to services and capabilities being deployed throughout the entire network. While data networks have not typically been viewed as components of mobile networks, the growing reliance on cloud infrastructure in current and future network deployments necessitates their partial integration into operator networks. Organizations like ETSI MEC and O-RAN Alliance aim to establish an ecosystem and value chain in which MNOs could authorize third parties to access their edge computing resources enabling them to deploy applications with augmented flexibility. Effective coordination between the data and control nodes is vital for time-sensitive applications to meet e2e QoS and sustainability needs. D. Supporting functionalities Supporting functionalities are crucial, for maximizing the capabilities of 6G networks. A prime example would be the collection and handling of various data types from different network areas and levels while safeguarding privacy and ownership rights for all parties. Emphasis is placed on complying with security and privacy laws like the General Data Protection Regulation (GDPR) in Europe and the Cloud Act in the United States when handling data movement and integration. By merging data from applications and network levels there are opportunities to enhance network performance and support applications that interpret how the network operates. Another crucial and widely used aspect is the ML operation pipelines that need to be tailored to the system in order to facilitate efficient implementation of advanced AI/ML methods. This encompasses AI systems for specific fields, ones that operate across various sectors, as well as AI integrated into network devices at all protocol levels. Security and privacy play a pivotal role in every facet of the system structure influencing both network design and business activities. It is important to recognize and address security and privacy risks that may arise from novel 6G features in devices, protocols, and architecture level. Moreover, state-ofthe-art methods for risk mitigation are necessary to enhance privacy measures technologies limiting the sharing of data among different entities. Efficient network management will ensure resource coordination across various locations ranging from far edge to central cloud environments. Such techniques need to be designed efficiently to not only manage network capabilities but also consider sensing and computing modalities. The automation features will offer increased autonomy through the use of automated closed-loop control systems empowered by intentdriven orchestration and AI techniques. Finally, continuous integration/continuous development (CI/CD) practices need to be integrated from the start of the software component development process. E. Hardware This layer consist not only of infrastructure components, such as BS, access points, satellites, routers, switches, etc., but also encompasses compute and storage resources along with a range of cloud services. To facilitate data transfer between these resources, transmission must be optimized on a per application basis considering the necessary requirements in terms of latency, capacity, performance, and more. Computing resources, like CPUs and GPUs, as well as storage resource, such as hard drives, are accessible on computing nodes distributed across the e2e chain including end devices, edge nodes, and data centers. 6G envisions a spectrum of resources involving edge devices connected to the cloud. Virtualization technologies enable abstraction of resources and their flexible allocation within the deep-edge-to-cloud continuum to enhance resource utilization. IV. KEY TECHNOLOGY ENABLERS The presented 6G architecture depends on individual and collaborative research breakthroughs in specific technological fields. The most influential technology enablers include 6G sensing, 3D networking, network orchestration, semantic communications, XR, robotic vision, DTs, PQC, blockchain, secure lifecycle management, XAI, and HIL. A. 6G sensing The utilization of location and sensing data in mobile communication systems has been proved very promising for several applications in 6G, such as improving the accuracy of emergency calls, detecting intruders through obstacles, personal navigation, tracking robots and UAVs, and enhancing social networking. Leveraging positioning information can aid in the organization, operation, and enhancement of communication networks. For instance, the authors of [6] provided an overview of how location aware communications are utilized in 5G networks across different levels of the protocol stack. They discussed the patterns, trade-offs, and challenges associated with this technology. Additionally, [7] highlights the potential for network synchronization, while in [8] the authors present an extensive analysis on localization techniques tailored for 5G NR networks with a focus on applications related to IoT. Before delving into practical hurdles faced in implementing location based services, the technologies and characteristics of 5G networks are analyzed. B. 3D networking For mobile networks to ensure connectivity in various situations like emergencies, flights, or high demand applications they need to go beyond the limitations of current land based methods and fully embrace communication systems that operate in the air and space [9]. The wireless communications industry is actively developing innovations to establish a comprehensive network that serves both ground based and aerial needs. Recent approaches involve enhancing existing networks to accommodate users on the air [10] or integrating terrestrial infrastructure, such as low Earth orbit satellite constellations or aerial base stations, for improved performance [11]. Cost considerations could play a role in shaping the development strategy, for future integrated networks. C. Network orchestration Cloud and edge computing alongside various layers of virtualization and flexible resources across different operators and ecosystems has attracted significant attention.As future network environments become intricate traditional control and management solutions for networks fall short. Current solutions lack self-awareness about networks learning abilities and the adaptability to their surroundings [12]. Recently, there have been increasing suggestions advocating for the integration of AI in orchestration services and managing network operations. Prior studies have suggested using a knowledge based network that merges software defined networking (SDN) with network analytics to enable the incorporation of AI techniques in controlling and operating networks efficiently [13]. The ETSI zero touch network and service management group was established with the aim of expediting the development of architectures and solutions required to achieve full automation, in managing networks and services within the realm of 6G technology [14]. The concept of Self Organizing Networks (SON) by 3GPP includes functions like setting up networks optimizing network parameters automatically and dealing with network issues through identification, prevention and correction mechanisms. Despite its advantages SON faces limitations in addressing the upcoming challenges, in future networks. D. Semantic communications With the introduction of 5G, advancements in algorithms have led to an increase in data transmission from user devices to the network. Despite this, the current capacity of 5G uplink may not be sufficient to meet the increasing demands over the decade. Looking ahead to 6G, the increase in the number of intelligent nodes becomes more prevalent, while their capabilities necessitate gathering and analyzing vast amounts of data. Applications like IoT and XR are setting new benchmarks for performance metrics such as latency constraints, consistent packet delivery, reliability, achievable throughput, and system reliability standards. For instance, the adoption of communications utilizing multiple view cameras is projected to require several terabits per second for each connection in both uplink and downlink, along with strict e2e delays to ensure an authentic and seamless remote experience. E. XR The Metaverse has removed the boundaries of distance in networking allowing individuals to engage with one another in an immersive virtual setting regardless of their physical location. Virtual interactions between humans will give rise to a variety of metaverse applications with augmented immersive experiences [14]. Managing vast amounts of data and receiving instantaneous feedback to support XR strains current network capabilities. As a result, 6G needs minimal latency, exceptional reliability, and ample spectrum availability. Moreover, employing holographic displays has proven effective in tasks, such as robot tele-operation. This involves integrating the robots movements into a setting that features a 3D model within an XR context [15]. For instance, a robot could present users with a 3D arrow guide showing both the desired route and any potential obstacles. While the monitoring model, for such tasks works well for interacting with a single robot, it can be mentally challenging when dealing with multiple robots that have diverse goals. F. DTs Conventional methods of network management are struggling to meet the demands of applications such as network overprovisioning and admission control. To address this, overseeing and managing communication networks can be enhanced through DTs, which enable network operators to analyze data, diagnose problems, and plan network optimization through the digital counterpart of the actual network [16]. AI can play a pivotal role in this endeavor by augmenting network modeling, network segmentation, data center traffic optimization, and resource allocation. Despite these advancements, there are unresolved issues related to the integration of DTs in monitoring and control of networks. These challenges include data collection and storage handling as well as uncertainties in developing control strategies and ensuring practicality and scalability. G. Post-Quantum blockchain Quantum computers pose a threat to the security of commonly used encryption methods. While adjustments to its parameters can make symmetric cryptography quantum resistant, public key cryptography remains vulnerable. In 2016 NIST initiated a process to standardize quantum cryptography algorithms resulting in the release of an initial list in 2022. These algorithms are expected to be effective across applications, like blockchain that relies on key cryptography and hash functions for transaction signing purposes. Public key cryptography plays a role in establishing trust among multiple parties within a decentralized system while ensuring the security and resilience of blockchain networks. Although blockchain is considered secure overall potential vulnerabilities arise from quantum threats targeting the wallets. Therefore, post-quantum blockchain emerges as a solution because of its secure and decentralized ledger system that has increased resilience against quantum attacks. H. Secure lifecycle management Recently, there is a focus on developing secure and private applications referred to as confidential computing in the cloud. This will allow users and cloud clients to send their data and computations to the cloud while preventing cloud administrators from accessing any information about the outsourced data and computations. This can be realized through hardware support with trusted execution environments (TEEs), such as Intel’s secure guard extension (SGX), which is also utilized by decentralized platforms like Secret Network [17] and Phala [18] to bolster their security measures. Both industry and academia have made progress, in understanding TEEs concerning speculative execution and conditional branching despite the existence of side channels. However, current decentralized platforms in the commercial sector that use TEEs to protect the privacy of their operational logic are still vulnerable to attacks. Our argument is that the current protective measures for real world systems mainly focus on achieving goals independently while overlooking the fact that the other layers they rely on may change independently either directly or indirectly. I. XAI & HIL Deep learning models are usually complex and difficult to understand often referred to as black box systems because of the challenge in grasping their decision making process. To address this, considerable effort effort has been put towards developing XAI methods that increase transparent and explainable. Despite these advancements, there are still hurdles to overcome, such as the complexity of explaining a model’s reasoning in a human-readable way. Retrospective explanations are sometimes seen as not precise [19]. Current XAI techniques, like perturbation based methods, have broad applications but lack specificity while decision set methods may not consider the frequency of feature values [20]. Another approach to overcome such issues is HIL approaches that involve humans to identify contextual situations and make decisions together with the system, thus assisting in creating action sequences and executing tasks [21]. It is essential for humans to interact effectively and naturally with these integrated systems in order to ensure autonomous operation while considering factors like service quality, reliability and efficiency. V. IMPACT TO 6G ERA &CONTRIBUTION TO KVIS In order to provide a comprehensive example of how 6G is envisioned to reshape the telecommunication ecosystem, the following section will delve deeper in the intricacies of precision healthcare. In this direction, the goal of 6G is to provide care to patients as they move between indoor and outdoor areas. Indicative examples are depicted in Fig. 2. These include locations like urban, suburban, or rural areas, such as homes, medical facilities, schools for children elderly care homes, and workplaces. The main challenges in this use case relate to prioritizing privacy and security as well as ensuring reliability and availability. It’s crucial that sensitive data, sensor data and any resulting information are kept secure within a trusted environment and only accessible for approved purposes. Another hurdle is maintaining services between the local network and the wider network for tasks, like authentication, encryption, or accessing personal data. In the way towards more sustainable 6G networks, the main affected aspects are inclusion and trustworthiness. By ensuring that healthcare services are easily accessible and welcoming to all it aligns with the UN’s SDG#3 of ensuring health and well being for everyone regardless of age. Credibility involves creating systems with reliability, strong security, and high levels of openness, which fosters trust among users. AI can simplify patient interactions with healthcare technology and ensures that all users can depend on and understand the health information provided. 6G also drives innovation and strengthens health systems as described by UN SGD #9 by supporting remote healthcare. It aligns Fig. 2. Precision healthcare use case. with UN SDG #11 by reducing the need for travel, thereby supporting sustainable urban development. Regarding UN SDG #12, 6G-enabled remote healthcare encourages responsible resource use and efficient management of medical waste, promoting sustainable consumption and production. Finally, 6G contributes to UN SDG #13 by reducing the carbon footprint associated with healthcare practices, making progress in climate action. The major key values affected by this use case involve personal health and protection from harm, societal sustainability, trust, as well as privacy and confidentiality. A detailed analysis for each key value is presented in the rest of this section. A. Personal health and protection from harm Major aspects of personal health and protection from harm can be measured through several corresponding KVIs, such as increased access to health, increased guidance, and social inclusion. 6G will enhance access to autonomous health monitoring services by incorporating AI-driven personalized health insights and recommendations, making health monitoring more accessible and informative for elders and/or disabled individuals. To realize this vision, technology enablers, like medical devices with long operation time, 6G sensing, secure and trustworthy AI, and XAI with HIL, will play an instrumental role. Furthermore, the integration of userfriendly interfaces and natural language processing coupled with secure AI to ensure data privacy and trustworthiness will propel efforts to demystify medical data. B. Societal sustainability With regard to societal sustainability, 6G-enabled remote healthcare can provide a reduction in important KVIs, such as reduced average cost of healthcare services, increased societal inclusion, and more. Specifically, 6G can contribute to a decrease of the average cost savings in the healthcare system per patient through intelligent efficient preliminary screening and health data analysis, reducing the need for in-person consultations and diagnostics. Moreover, remote services can be provided to a plethora of handicapped patients; thus increasing societal inclusion. From a technological standpoint, societal sustainability is highly dependent on XR, operational cost efficiency, extended coverage, security lifecycle management, scalable infrastructure, and assured quality of service. C. Trust Since the introduction of AI, the end-users have have been alienated and sceptical with regard to the way networks and communications function. To overcome this issue, 6G focuses on improving trust in next generation networks. In more detail, 6G will elevate reported trust levels in autonomous e-health systems by making AI decisions transparent and understandable, enhancing accuracy in health event identification and decision-making. To achieve this, 6G will be founded on top of secure and trustworthy AI models, XAI with HIL approaches, secure lifecycle management, and more. D. Privacy and confidentiality One of the major concerns with respect to e-health applications is privacy and confidentiality of personal data. To strengthen user control over personal data 6G will enlist transparent data handling and privacy settings, ensuring the privacy and security of health data across storage, transmission, and processing stages. To realize this vision, blockchain, PQC, DIDs for verifiable self-sovereign digital identity, TEEs, and trustworthy AI will play a catalytic role. VI. CONCLUSIONS The emergence of 6G networks opens up a new chapter in communication, where sustainability plays a vital role in shaping the design and operation of future systems. This paper introduced a 6G architecture taht integrates sustainability aspects in an adaptable, sustainable, and secure way. Sustainability transcends solely improving communication reliability; it involves creating a network framework that protects the environment while fostering economic progress. The vision of 6G ecosystems and their potential applications was analyzed, shedding light on how 6G can contribute both directly and indirectly. The proposed 6G network architecture incorporates sustainability considerations from the ecosystem design stage ensuring clear paths from key values to KVIs to KPIs. Finally, this work discussed how the proposed architecture and next-gen technologies will influence the era of 6G with a focus on the end-game environmental, societal, and economic factors. REFERENCES [1] A. Mesodiakaki, A. Kostopoulos, A. Gavras, A. Rahman, B. M. Khorsandi, D. Tsolkas, J. Cosmas, M. Gramaglia, M. Ericson, M. Boldi et al., “The 6g architecture landscape: European perspective,” 2023. [2] S. E. Trevlakis, A.-A. A. Boulogeorgos, D. Pliatsios, J. Querol, K. Ntontin, P. Sarigiannidis, S. Chatzinotas, and M. Di Renzo, “Localization as a key enabler of 6g wireless systems: A comprehensive survey and an outlook,” IEEE Open Journal of the Communications Society, vol. 4, pp. 2733–2801, 2023. [3] S. E. Trevlakis, N. Pappas, and A.-A. A. Boulogeorgos, “Toward natively intelligent semantic communications and networking,” IEEE Open Journal of the Communications Society, vol. 5, pp. 1486–1503, 2024. [4] G. Caso, ¨ O. Alay, A. Brunstrom, H. Koumaras, A. D. Zayas, and V. Frascolla, “Experimentation in 5g and beyond networks: State of the art and the way forward,” 2024. [5] S. Kerboeuf, P. Porambage, A. Jain, P. Rugeland, G. Wikstr¨ om, M. Ericson, D. Thai Bui, A. Outtagarts et al., “Design methodology for 6g end-to-end system: Hexa-x-ii perspective,” IEEE Open Journal of the Communications Society, vol. 5, pp. 3368–3394, 2024. [6] M. Koivisto, A. Hakkarainen, M. Costa, P. Kela, K. Leppanen, and M. Valkama, “High-efficiency device positioning and location-aware communications in dense 5g networks,” IEEE Communications Magazine, vol. 55, no. 8, pp. 188–195, 8 2017. [7] M. Koivisto, M. Costa, J. Werner, K. Heiska, J. Talvitie, K. Leppanen, V. Koivunen, and M. Valkama, “Joint device positioning and clock synchronization in 5g ultra-dense networks,” IEEE Transactions on Wireless Communications, vol. 16, no. 5, pp. 2866–2881, 5 2017. [8] J. Del Peral-Rosado, G. Seco-Granados, R. Raulefs, E. Leitinger, S. Grebien, T. Wilding, D. Dardari, E. Lohan, H. Wymeersch, and J. Floch, “Whitepaper on new localization methods for 5g wireless systems and the internet-of-things,” COST Action CA, vol. 15104. [9] M. Giordani and M. Zorzi, “Non-terrestrial networks in the 6g era: Challenges and opportunities,” IEEE Network, vol. 35, no. 2, pp. 244– 251, 3 2021. [10] M. Chowdhury, “Ensuring reliable connectivity to cellularconnected uavs with uptilted antennas and interference coordination,” ITU J. Future and Evolving Technol, 2021. [11] G. Kurt, “A vision and framework for the high altitude platform station (haps) networks of the future,” IEEE Commun. Surveys Tuts, vol. 23, no. 2, pp. 729–779, 2021. [12] P. Klaine, M. Imran, O. Onireti, and R. Souza, “A survey of machine learning techniques applied to self-organizing cellular networks,” IEEE Communications Surveys & Tutorials, vol. 19, no. 4, pp. 2392–2431, 2017. [13] J. Moysen and L. Giupponi, “From 4g to 5g: Self-organized network management meets machine learning,” Computer Communications, 2018. [14] H. Peng, P.-C. Chen, P.-H. Chen, Y.-S. Yang, C.-C. Hsia, and L.- C. Wang, 6G toward Metaverse: Technologies, Applications, and Challenges. IEEE, 8 2022, pp. 6–10. [15] C. Reardon, K. Lee, and J. Fink, Come See This! Augmented Reality to Enable Human-Robot Cooperative Search. IEEE, 8 2018. [16] P. Almasan, “Network digital twin: Context, enabling technologies, and opportunities,” IEEE Commun. Mag, vol. 60, no. 11, pp. 22–27, 11 2022. [17] L. Fishman, “Secret network documentation: Theoretical attacks,” 2020. [18] Phala, “https://wiki.phala.network/en-us/general/phalanetwork/intro,” 2020. [19] E. Kenny, E. Delaney, D. Greene, and M. Keane, Post-hoc Explanation Options for XAI in Deep Learning: The Insight Centre for Data Analytics Perspective. Springer International Publishing, 2021, pp. 20–34. [20] M. Milad, “Post-hoc explanation of black-box classifiers using confident itemsets,” Expert Systems with Applications, 2021. [21] M. Jirgl, Z. Bradac, and P. Fiedler, “Human-in-the-loop issue in context of the cyber-physical systems,” IFAC-PapersOnLine, vol. 51, no. 6, pp. 225–230, 2018.