Claudia Carballo González, Sara Pizzi, Maurizio Murroni, Giuseppe Araniti
©SHUTTERSTOCK.COM/ANTLII
Multicast/broadcast delivery is a critical challenge of future 6G mobile networks where massive Internet of Things (IoT) deployment and extended reality multimedia such as teleportation are target application scenarios. Non-terrestrial networks (NTNs) are considered essential for the success of 6G, which aims to provide true “global” services by extending mobile access worldwide, thus overcoming the coverage limit of current terrestrial networks (TNs). This article discusses how the main distinguishing features of NTNs can be effectively exploited for 6G multicasting. Furthermore, in line with the evolution of future 6G networks toward softwarized systems, we evaluate the potential of using the softwarization paradigm in the heterogeneous TN–NTN architecture to deliver multicast services.
The 5G mobile system is being adopted worldwide with significant advances in spectrum usage, system capacity, network performance, and reliability [1]. However, 5G falls short of fulfilling future applications’ stricter requirements and realizing true global connectivity. For this reason, even though 5G is still at its initial stage, the research community has started to focus on its successor: the 6G mobile system.
Expected to be deployed around 2030 [2], 6G will enable a wide range of advanced applications such as haptic communications, full-sensory digital reality, extremely high-definition video, fully automatic driving, deep-sea sightseeing, and massive IoT. These groundbreaking applications envisaged for the incoming decade will be characterized by diverse key performance indicators (KPIs), imposing tight quality of service (QoS) requirements in terms of ultrahigh reliability, data rate, energy efficiency, low latency, and scalability.
In diverse 6G scenarios, the same content could be requested by many users simultaneously, which makes imperative the support of point-to-multipoint (PTM) delivery, also called multicast/broadcast, due to its capability to exploit network resources economically and efficiently. Supporting multicast services from the initial 6G design stages is primarily needed to address the requirements of future IoT deployments, such as massive software updates or multimedia data acquisition beyond augmented and virtual reality, from which arise severe communication challenges in 6G networks. For instance, for teleportation, the data rate requirement of a 3D holographic display producing a raw color hologram, full parallax, and 30 fps is 4.32 Tbps [1]. Additionally, vehicular applications can benefit from multicast transmissions, where terminals involved in the same services (e.g., traffic management) or within the same area (e.g., cars close to the position of an accident) can be grouped to disseminate data among interested vehicles [3].
The Third Generation Partnership Project (3GPP) defined, starting from 2005, the multimedia broadcast and multicast service (MBMS) to optimize the distribution of broadcast and multicast services in cellular networks. While the first versions of the 5G New Radio (NR) development (releases 15 and 16) were only focused on point-to-point (PTP) communications [4], following the increasing interest in PTM delivery, in release 17 [5], the 5G network architecture has been enhanced to support multicast/broadcast services (MBS).
Among the requirements that 6G is claimed to meet, realizing a fully connected digital world is a key requirement that TNs may fail to fulfill due to limitations in terms of deployment and coverage. In the last several years, a booming interest has been devoted to NTNs, also due to their capability to complement terrestrial infrastructure for achieving continuous, ubiquitous, and global connectivity (e.g., through nanosatellite constellations) [6]. Indeed, NTNs are crucial to cover unserved/unconnected areas, complement the TNs’ deployment during overcrowded situations, and serve as backhaul. Thus, the combination of unmanned aerial vehicles (UAVs), high-altitude platforms (HAPs), satellite constellations, and TNs will constitute the 3D ecosystem of 6G [7]. Consequently, the cooperation of TNs and NTNs must be effectively managed to satisfy the “always best connected” paradigm and dynamically switch among unicast/multicast/broadcast service delivery according to the application types and network conditions.
In this complex scenario, the network slicing paradigm arises as a critical piece to introduce flexibility, dynamism, and isolation. Being a significant characteristic of the 5G system [8], it will also be vital for the future 6G deployment [1], together with the software-defined network (SDN) and network function virtualization (NFV) technologies, to efficiently manage network resources in real time. Additionally, artificial intelligence (AI) will complement SDN, NFV, and slicing technologies to dynamically make proactive decisions regarding network control and resource management [9].
NTNs have been the subject of recent survey papers [10]. However, although some works have recently focused on the integration of TN–NTN and the role of NTNs in the future 6G [11], little attention has been devoted to discussing their potential in delivering MBSs. To this aim, after describing the 5G MBS architecture and introducing the basics of NTNs, we examine why NTNs are an up-and-coming solution for 6G multicasting by discussing how their main distinguishing features can be effective for multicast service delivery. Then, in line with the evolution of future 6G networks toward softwarized systems to fulfill the strict applications’ requirements, constantly changing users’ demands, and mobility behavior, we evaluate the potential of exploiting the softwarization paradigm to the heterogeneous TN–NTN architecture in the delivery of multicast services.
The remainder of the article is organized as follows. The section “The MBS Architecture in 5G and Beyond” describes the MBS architecture and the envisaged development. The sections “NTNs” and “Why and How to Use NTNs for Multicast Services Delivery” summarize the basics of NTNs and discuss benefits, challenges, and enhancements in delivering multicast services over NTNs, respectively. The section “Toward a Softwarized Hybrid Terrestrial/Non-terrestrial–MBS Architecture” outlines the key enablers that will drive the development of future integrated TN–NTN over a softwarized environment. Finally, we present the “Conclusions” section.
Since the adoption of evolved Multimedia Broadcast and Multicast Service (eMBMS) in release 9, 3GPP has enhanced the PTM capabilities to support MBS efficiently. However, the first 5G NR release (release 15) only focused on 5G unicast mode, and release 16 only included the specifications for LTE-based 5G broadcast [4]. Starting from 5G release 17, the support of MBSs is being developed over the existing 5G framework, ensuring the smooth introduction of future functionalities and compatibility with legacy MBMS network nodes for service continuity [5].
Figure 1 shows the 5G MBS system architecture composed of new functional components and other elements that must be updated to support MBS. The new multicast/broadcast network functions (MB-NFs), highlighted in Figure 1 (dashed blocks), belong to the 5G core (5GC) and have the following functionalities:
Figure 1 The 5G MBS system architecture. NG-RAN: next-generation RAN; NSSF: network slice selection function; NWDAF: network data analytics function; PCF: policy control function; RAN: radio access network; SMF: session management function; UE: user equipment; UPF: user plane function; 5GC: 5G core.
The following elements enhance their functionalities to support MBS:
Release 17 details two possible delivery methods to transmit the MBS data between 5GC and NG-RAN:
Figure 2 summarizes the evolution in support of MBSs from release 15 to the future release 18, considered the first 5G-advanced standard that will lay the groundwork for future 6G deployment. In the envisaged MBS architecture, decentralized and distributed caching and edge computing capabilities will be critical in reducing service delay and backhaul data traffic. The new specifications must be oriented to increase energy efficiency for MBS transmission and fulfill successful MBS reception of many users that could be distributed in a scattered and wide area. Moreover, 5G-advanced/6G networks require AI for data-driven and intelligent network solutions in a hierarchical and distributed fashion. Integrating these paradigms is essential to increase MBS resource efficiency and improve each MBS session’s QoS in future ultradense heterogeneous environments. For example, the RANs can follow a cooperative ML strategy to determine which of them has the best conditions to satisfy the group of users requesting the same content simultaneously.
Figure 2 The 5G MBS evolution.
According to the definition by the 3GPP [12], an NTN “refers to a network or segment of networks using RF resources on board a satellite (or UAS platform).”
The network elements that constitute an NTN are
The NTN terminal can be either a 3GPP UE or a specific satellite terminal. Terminals may operate in the radio frequency of Ka-band (i.e., 30 GHz in the uplink and 20 GHz in the downlink) or S-band (i.e., 2 GHz). ISLs, relevant in the case of a constellation of satellites and requiring regenerative payloads on board the satellites, may operate in RF frequency or optical bands.
While the radio interface for the service link is 3GPP-defined NR, both 3GPP or non-3GPP-defined radio interfaces may be used for the feeder link and ISLs.
NTN platforms are classified as spaceborne or airborne, depending on their altitude, beam footprint size, and orbit. NTN platforms generate (typically several) beams that can be steerable (i.e., generate fixed beam footprint on the ground) or fixed (i.e., generate moving beam footprint on the ground). The footprints of the beams are typical of an elliptic shape.
Spaceborne platforms can be distinguished as follows:
LEO and MEO satellites are also known as non-GEO satellites for their motion around Earth with a lower period than the Earth’s rotation, ranging from 1.5 to 10 h.
The airborne category encompasses UAS platforms, which are typically placed at an altitude between 8 and 50 km and include HAP systems (HAPSs) at 20 km altitude. As it happens for the GEO satellite, the UAS position can be kept fixed in the sky concerning a given point on the ground. UAS beam footprint size ranges from 5 to 200 km. Additionally, UAVs represent a particular case with lower altitudes (usually around 100 m). They are more flexible regarding coverage and quick deployment perspective.
NTN platforms may also be distinguished according to the carried payload that may be transparent (or bent-pipe) or regenerative. While the transparent payload repeats the received waveform signal unchanged, the regenerative payload has onboard processing. Thus, all (or part) of the base station (BS) (e.g., gNB) functions are on board the satellite (or UAS platform).
As a result of the above-discussed classifications regarding the platform type (GEO/LEO) and the carried payload (transparent/regenerative), 3GPP has identified the six macroscenarios of interest that are reported in Table 1. While scenarios A, C2, and D2 have been considered with higher priority, the possibility of implementing steerable beams (scenarios C1 and D1) has recently received increasing attention.
Table 1 Reference scenarios [12].
The 3GPP has defined the following options for the NR-enabled NTN architecture to minimize the need for new interfaces and protocols in NG-RAN to support NTNs:
Figure 3 illustrates the discussed NTN architectures.
Figure 3 NTN platform types and architecture options.
Future 6G networks are claimed to support a wide range of traffic types, among which the demand for video, currently accounting for about 69% of all mobile data traffic, will continue to be significant as it is forecasted to increase up to 79% in 2027 [13].
A wide range of multicast use cases in various verticals, such as media and entertainment, automotive and safety, industrial IoT, and healthcare, make use of immersive live videos, requiring the delivery of interactive videos together with audio, data transmission, and feedback controls (see Figure 4). These applications demand a highly reliable delivery and a less strict latency than ultrareliable low-latency communications but are more stringent than traditional enhanced mobile broadband services. Thus, in the future transition from 5G to 6G, the demand for network capabilities in terms of capacity and availability will increase significantly, along with the need to reduce the costs of the provided services. This growth in demand is not feasible to be sustained by systems that only leverage unicast transmissions. This has already become clear in 5G systems, where multicasting and broadcasting have gained increasing attention, but places multicasting as a cornerstone technology of 6G systems for meeting the requirements of future applications.
Figure 4 NTNs for multicast use cases. IRS: intelligent reflecting surface.
NTNs represent a key solution for the above-mentioned multicast use cases because of their main distinguishing features:
However, NTNs’ benefits come with some challenges whose severity depends on the NTN platform type, the scenario under investigation, and the considered architecture option. Among the most relevant are the Doppler shift, propagation delay, and round trip time, which may significantly complicate critical procedures at physical (PHY) and medium access control (MAC) levels. In addition, in the case of multicast traffic delivery, the effect of such problems may be very different among UEs belonging to the same multicast group.
To improve the performance of the multicast transmission, the impact of the adverse effects of typical NTN channel impairments on more disadvantaged users should be mitigated to provide benefits to all group members. Possible solutions rely on exploiting a multicast architecture aided by intelligent reflecting surfaces (IRSs), in which the channel conditions of the weakest link can be enhanced by carefully tuning the IRS phase shifts. Significant improvements may also be achieved by relying on a multilayer architecture encompassing ground, air, and space and leveraging multiconnectivity to enhance the user throughput and increase reliability. Finally, network slicing is a crucial feature for NTNs to segment and provide multiple service instances at different altitudes. Therefore, numerous applications can run on different NSs in parallel, receiving differentiated QoS treatment according to the specific traffic performance.
Future wireless networks will evolve into highly complex and ultradense heterogeneous systems. Integrating TNs and NTNs is crucial to extend the service coverage area, increase capacity, and guarantee the “always best connected” vision, thus paving the way toward the true global connectivity promised by 6G systems. Based on the TN–NTN cooperation, the user can be connected to the most suitable access network at each moment, ensuring service continuity and enhanced transmission performance.
In this complex scenario, softwarization and slicing paradigms arise as critical pieces to introduce flexibility, dynamism, and isolation. Figure 5 shows how, with network softwarization, the heterogeneous TN–NTN infrastructure is abstracted as network, computing, and storage resources. In this environment, the SDN controller enforces intelligent decisions taking complete control of the network. At the same time, NFV guarantees the orchestration of computational and storage resources needed to instantiate network functions.
Figure 5 NS high-level architecture. GW: gateway.
The end-to-end (E2E) resources and functions are isolated into multiple NSs to flexibly configure resources according to users’ demands and heterogeneous network conditions. Critical concerns in NS planning are determining how many NSs to deploy and what functions/features to share across multiple NSs. One NS can allocate multiple service instances, which can be associated with several RANs and core network segments [14]. Each instance would be activated based on the NS template’s specification, which includes the configuration of related network functions and E2E resources.
Furthermore, in diverse 6G scenarios, many users could request the same content (e.g., live video at significant events). Thus, the slicing paradigm must be enhanced to support multicast/broadcast capabilities. Based on the application type, the users’ distribution, and network conditions, a common content flow must be dynamically mapped into unicast/multicast/broadcast slice instances, exploiting network and radio resources economically and efficiently. In this context, the cooperation among network operators and content providers is fundamental to managing the E2E system.
In the future ultradense heterogeneous environment, the orchestration and management of the network must be conducted by AI techniques combined with slicing and softwarization approaches. The dynamic and intelligent tasks must be oriented to the access network/NSs selection to satisfy the user petition, multicast group formation, and load balancing, including dynamic adjustment between PTP and PTM and a suitable strategy during an overload situation.
In the remainder of this section, we will evaluate the benefits that the application of the softwarization paradigm could bring to the heterogeneous TN/NTN architecture in delivering multicast services.
The applied algorithm for selecting the most suitable combination of access network/NSs to satisfy the user petition is inspired by [15]. Specifically, the resources are dynamically orchestrated considering the RANs in the coverage area, the available resources, the configured NSs, the users’ QoS parameters, mobility behavior, and tariff plan. The algorithm assigns resources according to the maximum throughput required by the requested services if enough capacity is available. When the network is overloaded, the RANs/NSs exploit a collaborative approach to balance resources, avoid network performance degradation, and meet users’ requirements.
The scenario under analysis comprises a macro-BS, a micro-BS, and a UAV serving as an aerial BS. The network hosts a new user every 2 s. Each user is randomly positioned in the simulated area with a random-way point mobility model. The simulation analyzes the advantage of combining unicast/multicast services in the context of network slicing and softwarization technologies. We consider four services with diverse KPIs, assuming that one out of the four services is multicast. Users can request from one to four services simultaneously, and each NS allocates one of these services (i.e., for a total of four NSs).
In Figures 6 and 7, all results show that integrating TN–NTNs and exploiting unicast/multicast capabilities (solid lines) in a softwarized context outperforms the other three use cases (TN–NTN unicast, TN unicast/multicast, and TN unicast) in terms of
Figure 6 System ADR with unicast and unicast/multicast capabilities.
Figure 7 Thsat average with unicast and unicast/multicast capabilities.
Figures 6 and 7 highlight that the scenario where unicast services are delivered via TNs only has the worst performance, reaching the first saturation point for the smallest number of users (equal to 49). Starting from this point, due to the scarcity of resources, the network splits the resources among active users at the expense of affecting the Th sat performance. The load balancing is a gradual and collaborative process until the users in the network receive the minimum throughput according to services’ constraints. At this point, it is unfeasible to admit new users until, for example, some terminals leave the network and free up resources.
In contrast, the scenarios with TN and TN–NTN exploiting unicast/multicast capabilities provide a better performance w.r.t. unicasting only. All users of the multicast group efficiently utilize the same NS resources assigned for the multicast service, positively impacting the system ADR performance. Additionally, NTNs play a crucial role as an alternative access network to complement the TNs’ deployment by increasing the network capacity. Therefore, combining TN–NTN and unicast/multicast allows the delivery of a higher system ADR and average Th sat w.r.t. the other analyzed cases (i.e., the first saturation point occurs with 117 users).
6G networks are in the sight of the scientific community, with many challenges to be solved to cope with the future advanced applications, impacting the media and entertainment, automotive and safety, industrial IoT, and healthcare. This article analyzed how MBS delivery aided with NS and softwarization paradigms is suitable for handling a wide range of use cases through economic and efficient resource allocation mechanisms. It delved into the architecture of NTNs and identified how multicast/broadcast could be effectively implemented in 6G scenarios, describing the main features and challenges of NTNs. Moreover, the article summarized the state-of-the-art multicast/broadcast techniques and their future development. It outlined the key enablers that will drive the growth of future integrated TN–NTN over a softwarized environment. The presented results demonstrated how integrating TN–NTNs and exploiting unicast/multicast capabilities can efficiently allocate network resources.
Claudia Carballo González (claudia.carballogonz@unica.it) is a Ph.D. student at the Department of Electrical and Electronic Engineering (DIEE/UdR CNIT), University of Cagliari, Cagliari, Sardinia, Italy. Her research interests include enhanced wireless systems, QoS, resource allocation, and AI. She is a Student Member of IEEE.
Sara Pizzi (sara.pizzi@unirc.it) is an assistant professor in telecommunications at Mediterranea University of Reggio Calabria, 89135 Calabria, Italy, where she received her Ph.D. degree (2009) in computer, biomedical, and telecommunication engineering. Her current research interests focus on non-terrestrial networks, radio resource management for multicast service delivery, device-to-device and machine-type communications over 5G/6G networks. She is a Member of IEEE.
Maurizio Murroni (murroni@unica.it) , Ph.D., is an associate professor of communications at the Department of Electrical and Electronic Engineering, University of Cagliari, Cagliari, Sardinia, Italy, 09123 Cagliari, Italy, and member of the Italian National Inter-University Consortium for Telecommunications. His research focuses on broadcasting, signal processing for radio communications, and multimedia data transmission and processing. He is a Senior Member of IEEE.
Giuseppe Araniti (araniti@unirc.it) is an assistant professor of telecommunications at Mediterranea University of Reggio Calabria, 89135 Calabria, Italy, where he received his Ph.D. degree in electronic engineering in 2004. His major area of research is 5G/6G networks and includes personal communications, enhanced wireless and satellite systems, traffic and radio resource management, eMBMS, D2D, and M2M/MTC. He is a Senior Member of IEEE.
[1] M. Giordani, M. Polese, M. Mezzavilla, S. Rangan, and M. Zorzi, “Toward 6G networks: Use cases and technologies,” IEEE Commun. Mag., vol. 58, no. 3, pp. 55–61, Mar. 2020, doi: 10.1109/MCOM.001.1900411.
[2] Z. Zhang et al., “6G wireless networks: Vision, requirements, architecture, and key technologies,” IEEE Veh. Technol. Mag., vol. 14, no. 3, pp. 28–41, Sep. 2019, doi: 10.1109/MVT.2019.2921208.
[3] J. Montalban et al., “Multimedia multicast services in 5G networks: Subgrouping and non-orthogonal multiple access techniques,” IEEE Commun. Mag., vol. 56, no. 3, pp. 91–95, Mar. 2018, doi: 10.1109/MCOM.2018.1700660.
[4] E. Garro et al., “5G mixed mode: NR multicast-broadcast services,” IEEE Trans. Broadcast., vol. 66, no. 2, pp. 390–403, Jun. 2020, doi: 10.1109/TBC.2020.2977538.
[5] “Technical specification group services and system aspects; Architectural enhancements for 5G multicast-broadcast services; Stage 2 (Release 17),” 3rd Generation Partnership Project, Sophia Antipolis, France, 3GPP TS 23.247, 2021.
[6] G. Araniti, A. Iera, A. Molinaro, S. Pizzi, and F. Rinaldi, “Opportunistic federation of CubeSat constellations: A game-changing paradigm enabling enhanced IoT services in the sky,” IEEE Internet Things J., vol. 9, no. 16, pp. 14,876–14,890, Aug. 15, 2022, doi: 10.1109/JIOT.2021.3115160.
[7] G. Araniti, A. Iera, S. Pizzi, and F. Rinaldi, “Toward 6G non-terrestrial networks,” IEEE Netw., vol. 36, no. 1, pp. 113–120, Jan./Feb. 2022, doi: 10.1109/MNET.011.2100191.
[8] A. A. Barakabitze, A. Ahmad, R. Mijumbi, and A. Hines, “5G network slicing using SDN and NFV: A survey of taxonomy, architectures and future challenges,” Comput. Netw., vol. 167, Feb.. 2020, Art. no. 106984, doi: 10.1016/j.comnet.2019.106984.
[9] M. K. Shehzad, L. Rose, M. M. Butt, I. Z. Kovács, M. Assaad, and M. Guizani, “Artificial intelligence for 6G networks: Technology advancement and standardization,” IEEE Veh. Technol. Mag., vol. 17, no. 3, pp. 16–25, Sep. 2022, doi: 10.1109/MVT.2022.3164758.
[10] M. M. Azari et al., “Evolution of non-terrestrial networks from 5G to 6G: A survey,” IEEE Commun. Surveys Tuts., vol. 24, no. 4, pp. 2633–2672, Aug. 2022, doi: 10.1109/COMST.2022.3199901.
[11] Z. Xiangming and C. Jiang, “Integrated satellite-terrestrial networks toward 6G: Architectures, applications, and challenges,” IEEE Internet Things J., vol. 9, no. 1, pp. 437–461, Jan. 2022, doi: 10.1109/JIOT.2021.3126825.
[12] “Solutions for NR to support non-terrestrial networks (NTN),” 3rd Generation Partnership Project, Sophia Antipolis, France, 3GPP TR 38.821, Jan. 2020.
[13] “Ericsson mobility report,” Ericsson, Stockholm, Sweden, Jun. 2022. [Online] . Available: https://www.ericsson.com/en/reports-and-papers/mobility-report/reports/june-2022
[14] “Technical specification group services and system aspects; Service requirements for the 5G system; Stage 1,” 3rd Generation Partnership Project, Sophia Antipolis, France, 3GPP TS 22.261, Mar. 2022.
[15] C. C. González, E. F. Pupo, L. Atzori, and M. Murroni, “Dynamic radio access selection and slice allocation for differentiated traffic management on future mobile networks,” IEEE Trans. Netw. Service Manag., vol. 19, no. 3, pp. 1965–1981, Sep. 2022, doi: 10.1109/TNSM.2022.3150978.
Digital Object Identifier 10.1109/MVT.2022.3232919