Are AccelerComm, Lockheed Martin, Telesat, and SpaceX about to Disrupt the Communications Industry with Unified RAN?

Starlink has been given the green light to launch 840 direct-to-cell capable satellites over the next six months. These regenerative mode satellites[1] are equipped with new eNodeB modems, giving them 4G Long Term Evolution (LTE) base station capabilities such as Radio Frequency (RF) filtering, frequency conversion, amplification, routing, coding, and modulation. This brings forward the potential for a unified Radio Access Network (RAN) that leverages both terrestrial and non-terrestrial base stations with the constellation’s Mobile Network Operator (MNO) partners.

Starlink is not the only player deploying satellite base stations. The upcoming 198-satellite-strong Low Earth Orbit (LEO) network by Telesat, Lightspeed, will also feature on-orbit data processing in space with regenerative architecture. Additionally, Lockheed Martin is testing a satellite with a 3GPP standards-compliant 5G regenerative payload (gNodeB) alongside partners Accelercomm, Radysis, and Keysight. The satellite is set to launch in 2024 and is powered by Accelercomm’s 3GPP standard-compliant LEOphy Layer 1 solution, which enables critical performance algorithms that enhance the power and minimize the Block Error Rate (BLER) of the satellite system.

[1] Regenerative mode satellites have onboard processing capabilities equivalent to a cellular base station. Hence, the satellite payload may redirect or remodulate the waveform of an incoming signal before re-transmits.

The introduction of eNodeB (LTE) and gNodeB (5G) modems onto satellites to unlock base station capabilities is a major advancement for the space sector’s standardization efforts. These developments help validate the work done by the 3GPP and its New Radio Non-Terrestrial Network (NR NTN) standard, where LEO satellites host gNodeBs and act as part of the 5G-RAN being able to connect directly with smartphones and other terrestrial communications equipment, such as the Internet of Things (IoT). The standardization efforts and real-world deployments underline a transition by satellite networks to unlock greater on-orbit processing and become part of a unified RAN. The result of adopting this model with greater on-orbit processing has several upsides:

• Hybrid Connectivity: Sharing the same protocol stack will allow NTN solutions to facilitate switching between cellular and satellite networks using the same device. This ubiquitous connectivity is one of the key advantages of a standardized solution.
• Inter-Satellite Links: Since regenerative architecture supports inter-satellite links, this architecture offers greater flexibility, coverage, and performance than bent-pipe architecture[1]. As LEO networks such as Starlink and Kuiper already use Inter-Satellite Links (ISLs), there is a benefit from leveraging the existing on-orbit LEO ISL infrastructure.
• Edge Computing: On-orbit processing unlocks the ability to move workloads to space and alleviate the burden on the terrestrial cloud while enhancing resiliency. Furthermore, as data from LEO grows and satellite networks become another node in a unified communications network (e.g., 6G), the demand for breaking down boundaries between verticals and migrating operating environments will become more critical. As such, satellite networks that explore hosting edge cloud-native functions and even distributed cloud represent a strong potential revenue option.

While SpaceX is starting with 4G LTE satellite base stations, the next logical step for the company's planned 840 direct-to-cell constellations is the Gen2 upgrade to 5G satellite base stations. Work done by Lockheed Martin, AccelerComm, Radisys, and Keysight on TacSat shows that 5G NTN Layer 1, 2, and 3 solutions are available to support the specific challenges of LEO NTN. In this respect, AccelerComm’s Layer 1 (physical layer) covers the transmission of radio frequencies over physical components, while Radisys’ Layer 2 (data link layer) covers establishing the link between network nodes, and Layer 3 (network layer) provides the logical address of an endpoint, such as Internet Protocol (IP). As such, the components necessary to build a next-generation 3GPP-compliant 5G NR NTN satellite are commercially available on the market today.

Therefore, the reality of SpaceX and other LEO satellite networks moving to NR NTN may be sooner than what the wider industry expects or is even prepared for. In this way, mobile devices still need to integrate antennas and NR NTN-enabled chipsets with 3GPP Release 17 spectrum-defined functionality to handle Doppler and timing variation effects by LEO satellite channels. Further work on regenerative payloads, end-user terminals with higher output power, and NTN support for 5G RedCap IoT devices is expected with 3GPP’s Release 19 starting in 2024.

[1] Bent Pipe mode satellites relay or “bend” the received ground signal without any onboard processing or capabilities.  

The trend towards more capable on-orbit processing is not going to slow down. New space segments, such as on-orbit Artificial Intelligence (AI)/Machine Learning (ML) or edge computing inherently require greater processing capabilities onboard the satellite. Therefore, the expansion of these new segments will require alignment with 3GPP standards as well as systems designed with NodeB modems and 3GPP NTN Layer 1, 2, and 3 solutions in mind. Satellite operators, solutions providers, and equipment vendors can help develop the market and accelerate the transition to greater on-orbit processing in the following ways:

• 3GPP-Based System Design: Satellite constellation designers, vendors, and operators should consider partnering with some of the companies mentioned in this insight, such as AccelerComm, to help develop 3GPP standards-based NTN systems. These companies already have attractive solutions to help unlock the full capabilities of on-orbit processing and can play a pivotal role in unlocking more critical New Space and 6G functionalities, like edge computing and caching.
• NTN User Equipment: Chipset vendors and antenna manufacturers need to consider adding standards-based solutions to their portfolio. While satellite networks can evolve to support ubiquitous connectivity, the device ecosystem must adopt NTN-capable equipment to connect with satellite or terrestrial networks when needed. ABI Research anticipates that 3GPP-based handsets will become the dominant NTN User Equipment (UE) solution in the coming years, and reach up to 427 million shipments by 2030, representing a 2022-2030 Compound Annual Growth Rate (CAGR) of 100%.