Programmable silicon, with its ability to adapt and optimize network operations, is emerging as a game-changer for non-terrestrial networks (NTN). While satellite communications have been around for a long time, it is only now that NTN is becoming accessible and affordable for use cases that previously struggled to leverage it. For more on those use cases, check out our previous blog in this series “NTN Use Cases for Programmable Silicon.”
NTN is now becoming more accessible to previously untapped markets. In this blog, we discuss the transformative potential of programmable silicon in revolutionizing non-terrestrial networks.
Enhanced 5G Software
Not all software that has been developed to support the wireless space, particularly 5G and its evolution, is ready to support NTN. Solid-performing lab-tested software may check all the required boxes,but it needs to be field-tested. It can be tough to simulate real-world effects in a lab, such as interference and reflections, that could necessitatetweaks to part of the software or the entire stack.
Even3GPP-compliant code that has been battle tested inotherscenariosmay need adjustments to be5G-NTNready.For instance, adjusting timing relationships and modify timers may be necessary; adapting CSI use for delays; optimize pre compensation for the Doppler shift; and tweak code for better beam and spectrum management.
Programmable software can simplify the type of adjustments needed to prepare for NTN. It also enables simpler testingof these changes on both the base station and user equipment (UE). Doing so when the software implementation and architecture are optimized for the digital signal processor (DSP) makes this process even simpler.
Programmable silicon enables the dynamic allocation and reallocation of network resources based on changing demands. This flexibility empowers network operators to efficiently manage capacity, optimize performance, and adapt to varying user requirements. Moreover, the ability to update and reconfigure network functions through software updates eliminates the need for costly hardware upgrades, providing a cost-effective solution for NTN.
Overall system performance improves when hardware and software are developed in tandem. When deployed over programmable DSPs designed for wireless applications, the instruction set, accelerators, and software combinations typically outperform software ported onto standard chips from third parties by a wide margin.
A system on a chip (SoC) used in a 5G NTN network also should draw sufficiently low power to work in any form factor, whether a handheld or satellite. Chipsets also should be tunable to very low power consumption requirements, whether supporting IoT NTN or new radio (NR) NTN. They should align with form factor requirements, including those of smaller class satellites. And to minimize the charges generated by high energy particles in space, which can impact memory and transistor logic state, any silicon deployed on satellite payloads needs adequate protection from high radiation.
In this blog, we discussed NTN challenges, such as beam management, which call for system-level considerations. Standard interfaces are certainly important for partnering with other components, particularly the analog radio front-end. But in terms of architecture, where the silicon sits in the software stack is a key consideration. In that context, the choice between transparent vs. regenerative mode is highly relevant.
The 3GPP process has initially focused on the transparent mode, in which the satellite acts as a simple relay. The regenerative mode, however, which involves deploying some or all gNodeB (gNB) functionality in the satellite payload, has advantages. (5G NR Open RAN enables centralized or distributed architectures.) The regenerative mode allows you to partition a solution into aspects that are more sensitive to delays and latency versus those that are not. As always, there are tradeoffs: you need more processing power and components on the gNB but benefit from lower latency and better user experience.
The Case for Programmable Silicon
The programmability of silicon in NTN opens doors for rapid prototyping and innovation. Network operators and researchers can develop and test new protocols, algorithms, and services using programmable silicon-based platforms. This accelerates the pace of innovation and enables the exploration of novel solutions for NTN challenges. Additionally, programmable silicon facilitates seamless integration with emerging technologies such as edge computing, Internet of Things (IoT), and 5G, unlocking new possibilities for non-terrestrial network applications.
Interested in learning more?
This is the third blog in a series. Our previous blogs provide a good sampling of the kinds of enhancements to 5G that can facilitate integration with a satellite-based NTN.
To download our full white paper on 5G Non-Terrestrial Networks and Satellites that includes topics discussed in this blog, click here.