Snapdragon X80 5G Modem-RF Platform
There has been a lot of talk about Moore’s Law over the past decade. Most of this is due to either him dying or at the very least slowing down. Clearly, leading companies like Intel and TSMC continue to make advancements in terms of smaller process node geometries. However, it is equally clear that the industry can no longer rely solely on Moore’s Law for performance gains. Some try to achieve more performance by using heterogeneous computing architectures, others use chipsets, and still others use both. At this year’s Mobile World Congress, alongside its established heterogeneous compute architecture approach, Qualcomm demonstrated another tool that can be used: artificial intelligence (AI).
There’s a commonly used aphorism that goes something like “if you have something great and you want to make it better, just add some bacon.” Whether or not one agrees with this, or simply replaces “bacon” with something else, this sentiment remains true in that certain things, when added, only improve the things. For the technology industry, and more specifically the chip industry, AI is shaping up to be just that.
It’s not just generative AI, which has only matured enough in the last couple of years to have a significant impact. In contrast, adding traditional or functional machine learning-based AI to enhance product capabilities has been a differentiation strategy for the past decade. At this year’s MWC, Qualcomm took this potential recipe for success and applied it to 5G with the announcement of its latest 5G offering, the Snapdragon X80.
Making AI the star
The Snapdragon Supporting 3GPP Release 17 as well as planned Release 18 features, the X80 modem enables 6x downlink carrier aggregation, up to 6 receive channels for smartphones, up to 10 Gbit download speed /s, a maximum download speed of 3.5 Gbps, support for non-terrestrial narrowband networks and of course, AI-based 5G optimization.
The baseband is equipped with Qualcomm 2sd Generation 5G AI processor that enables the platform to use AI to improve quality of service and end-user experience by intelligently controlling modem and RF functions. In conjunction with their 3rd generation 5G AI Suite, performance metrics for data speeds, power management and efficiency, coverage, spectrum efficiency, latency and GNSS location are all improved through the use of AI. Additionally, it uses AI processing to facilitate mmWave beam management, which is critical to providing 5G mmWave range extension when used in wireless access Customer Premises Equipment (CPE). fixed (FWA).
These improvements are achieved in part through the use of AI to manage multi-antenna subsystems more intelligently and efficiently. AI is also used to provide contextual input to optimize the radio link by identifying and considering the state of the RF environment and what the user is doing in terms of applications or workload . For example, if the user is doing something that depends on latency, such as a video call, the AI can increase the transmission power to compensate for any degradation in channel quality and prioritize throughput and latency by in relation to the resulting increase in energy consumption.
According to Qualcomm, compared to the previous generation, implementing AI to optimize 5G performance allows them to reduce the time to select the best cell by 20%, reduce link acquisition by up to to 30% and improve localization accuracy to a similar degree. For mmWave applications, it also offers up to 60% faster CPE service acquisition and 10% lower power consumption when connected.
Using AI to improve AI
As noted, Qualcomm hopes to significantly improve user experience with its latest X80 modem-RF platform. While this will always remain a goal with each new generation, it is needed more than ever as advanced use cases and applications, including generative AI, continue to demand higher processing, throughput, and latency. fast, while maintaining or improving power. consumption on these devices.
Besides throughput and latency, user experience also depends on battery life. Tirias Research recently conducted a study on the latest flagship smartphones running generative AI workloads and determined that current battery technologies will need all the help they can get in the AI era.
Applying AI to ensure the best possible combination of modulation coding schemes, transmission power, and antenna array configuration for a given workload is useful in several ways. In addition to maximizing uplink and downlink throughput and minimizing latency, which are already major factors in improving user experience, AI also minimizes the length of time the transmission chain (RF transceiver and power amplifier in the RF front end) is powered on and, when powered on, ensuring that the power used to send the transmission is as strong as necessary to achieve the desired performance. Along with the applications processor and the screen, the transmission chain is one of the largest consumers of energy provided by the battery of a mobile device. By minimizing the time the transmission chain is activated, either by minimizing retransmissions to overcome high error rates, or by maximizing throughput and minimizing latency, the impact on user experience in terms of Longer battery life can be significant.
AI-powered experiences don’t just benefit the end user. When a given transmission can be completed faster with lower transmission power, mobile network operators also benefit from maximizing capacity through a lower effective noise floor, ultimately benefiting the end user by helping to minimize interference and ensuring the best possible downlink and uplink speeds. for the given RF environment.
What’s next for 5G and AI
In addition to its AI-based enhancements, the X80 modem-RF platform also supports features expected in 3GPP Release 18 (Rel 18). As such, Qualcomm claims that this latest 5G offering is “5G Advanced Ready”. Whatever features ultimately become standardized through Release 18, at the very least, OEMs that build the X80 RF-modem into their device lineup will get what Qualcomm believes will be value-added capabilities for the next generation. not only smartphones, but also other types of devices. as well as PCs, XR devices, automotive CPE and FWA. According to Qualcomm, the platform is currently undergoing sampling and is expected to be released in the second half of this year.
Qualcomm’s use of AI to improve product performance is not a new idea and it’s not even the first time Qualcomm or other companies have done it. What makes it remarkable is that this product launch serves to highlight the critical role that traditional AI plays, and will continue to play, in improving existing technologies and products to a level of performance that makes no only possible, but usable workloads such as generative AI. It also demonstrates another weapon in a chipmaker’s arsenal to continue delivering the necessary performance increases with or without Moore’s Law.
Qualcomm isn’t the only company using AI in this way. However, in a world where all the focus is on generative AI, it is often easy to forget the importance of traditional machine learning-based AI product optimizations. Traditional AI used in this way is a powerful differentiator and every opportunity to use it as such should be taken.