One of the promises of AI in the Internet of Things (IoT) is that it creates vast new applications that transforms our world. But for the industry to realize this opportunity, we need more effective ways to deliver AI capabilities to devices that are increasingly area, cost, and power-constrained.
We know developers face challenges, including needing extended hardware capabilities and simplified software development. We listened and, in turn, developed Arm Helium technology-enabled processors. Arm Helium is the vector extension for Cortex-M Microcontrollers (MCUs). It provides performance gains in Digital Signal Processing (DSP) and Machine Learning (ML) applications for low-power embedded devices.
The technology is in our mainstream solutions, the Cortex-M55, and Cortex-M85. The response is enthusiastic, fueling innovation in smart camera and other applications.
To deliver greater ML-optimized performance and efficiency, partners want to take the capabilities of Helium technology into smaller devices with smaller footprints and lower power requirements.
Today, Arm announces the Cortex-M52, delivering dramatic increases in DSP and ML performance. And, a range of other features and benefits for applications like, industrial control and predictive maintenance.
The new Cortex-M52 provides a power-efficient implementation that is suitable as a future replacement for mainstream microcontrollers. The Cortex-M52 delivers up to 5.6x performance improvement for ML and up to 2.7x performance uplift for DSP. It matches an improved scalar performance and advanced memory interfaces that ensures it can be designed into appropriate systems.
Security remains critical. Cortex-M52 implements the latest security extensions for the Armv8.1-M architecture. Including, PACBTI (Pointer Authentication and Branch Target Identification), which, combined with TrustZone, offers enhanced software threat mitigation. Cortex-M52 accelerates the route to PSA Certified Level 2 silicon, enabling the next generation of PSA Certified devices.
With Helium technology, Cortex-M52 delivers improvements in DSP and ML processing over its predecessor, the Cortex-M33. Cortex-M52 provides a performance boost over Cortex-M33, improved scaler performance and, energy efficiency to drive AI-powered innovation into new applications.
Developer enablement is critical if we are to see an increase in AI-enabled IoT shipments. With the Cortex-M52, we are delivering the features and capabilities required in a modern development flow today. It brings AI within reach on a unified toolchain and single proven architecture.
The following graphic shows the key stages of an algorithm to identify events. It is a great example of using AI to solve a problem. In this case, it delivers or improving predictive maintenance. DSP tools are used initially for processes such as signal conditioning and feature extraction before a neural network is applied to classify events based on machine learning.
Arm Helium technology enables this with the combination of DSP and ML processing capability that an embedded AI application relies on. Developers can code in a single language against a common API and achieve the performance that they need in both the DSP and ML elements of the application.
There is no need to understand the specific hardware details of the processor. Arm, together with its ecosystem partners, provide the tools and libraries to maximize performance and enable the code to be ported across the range of Helium-enabled Arm microcontrollers.
We continue to invest, alongside our partners, in the right software and tools to make it easier and quicker to develop and maintain IoT applications with reduced risk and cost.
Cortex-M52 is fully software compatible with Cortex-M55 and Cortex-M85. Enabling developers to benefit from and leverage the growing software and tools ecosystem around Helium, as well as free software libraries and an extensive knowledge base from our partner ecosystem.
To help streamline and accelerate the IoT and embedded development process, Cortex-M52 will also be available on Arm Virtual Hardware, our cloud-based offering that enables software development in advance of silicon.
The consistent Helium capability across the Arm Cortex-M microcontrollers also benefits our software ecosystem partners. As they develop their tools, they can be confident that they can address the widest possible market of developers by optimizing to a single architecture. They can develop their optimizations once and see their software and tools deployed against a wide range of hardware offerings in the market.
The unprecedented efficiency levels offered by the Cortex-M52 offers new possibilities to microcontroller developers to develop and deploy ever more demanding use cases on the Cortex-M. Developers benefit from the simple programmer’s model, processing determinism, and low-power schemes, all of which are hallmark characteristics to all Arm Cortex-M processors.
That in turn, is driving new AI-powered innovations into vast new areas of IoT.
Learn more about the Cortex-M52