Arm® Cortex®-M55 in a nutshell
Introduction:
The Arm® Cortex®-M55 processor brings the benefits of endpoint AI to billions more devices and empowers developers to deliver the next revolution in computing. It is Arm’s first Cortex®-M processor to feature Arm Helium vector processing technology, which brings enhanced, power-efficient DSP and ML performance. The Cortex®-M55 offers an easy way to implement AI for a wide range of IoT use cases (such as earbuds, fingerprint unlock, health trackers, predictive maintenance, smart speakers, and video doorbells) with the ease of use of Cortex®-M, a single toolchain, optimized software libraries, and an industry-leading embedded ecosystem.
Key features of Arm® Cortex®-M55 core
- Armv8.1-M architecture
- Bus interface AMBA 5 AXI5 64-bit master (compatible to AXI4 Ips)
- 4-stages pipeline
- TrustZone® technology security
- 32-bit DSP/SIMD extension
- Helium vector extension
- Floating-point unit (FPU)
- Up to 64 Kbytes with ECC (instruction cache and data cache)
- Up to 16 Mbytes with ECC (instruction TCM and data TCM)
- Up to 480 interrupts + non-maskable interrupt (NMI)
- Internal/external wake-up interrupt controller
- Multiple power domains, sleep modes (sleep and deep-sleep), sleep-on-exit, optional retention support for memories and logic
- Hardware and software breakpoints
- Performance monitoring unit (PMU)
- Optional instruction trace
- Data trace (DWT)
- Instrumentation trace (ITM)
Key advantages of Arm® Cortex®-M55 core
The Cortex®-M55 processor, part of the Armv8.1-M architecture, introduces Helium, a vector instruction set extension that enhances DSP and ML capabilities with over 150 new instructions. It supports efficient 8-bit, 16-bit, and 32-bit fixed-point data processing, improving throughput and maximizing processor resource use, which is crucial for signal and ML processing.
The Cortex®-M55 floating-point unit (FPU) supports vector and scalar half-precision, single-precision, and double-precision floating-point datatypes. Half-precision processes twice the data per clock cycle compared to single-precision, reducing memory footprint. This is ideal for sound and sensor data processing with low resolution but high dynamic range.
Arm® TrustZone® technology in the Cortex®-M55 isolates critical information, reducing software-based attacks. Armv8.1 M adds security enhancements like privileged eXecute never (PXN), Unprivileged Debug Extension (UDE), and TrustZone® improvements for isolated debug permissions, enhancing security and accelerating PSA certified silicon and devices.
The Cortex®-M55 processor includes new debug enhancements like the PMU, direct cache access registers, and the UDE for restricted debug visibility.
Microcontrollers based on the Arm® Cortex®-M55
The STM32N6 is the first STM32 MCU featuring an Arm® Cortex®-M55 core that embeds the ST Neural-ART accelerator™, an in-house developed neural processing unit (NPU) engineered for power efficient edge AI applications.
Series | Speed (MHz) | Performance (CoreMark) | Flash (kbytes) | RAM (kbytes) | Power Supply (V) | Packages | Connectivity | Analog |
STM32N6 | 800 | 3360 | - | 4200 | 1.7 to 3.6 | VFBGA142, VFBGA169, VFBGA178, VFBGA198, VFBGA223, VFBGA264 | CAN FD, USB2.0 D/H HS + HS PHY, UCPD controller, I²C, I3C (SDR), UART, USART, SPI, ULP UART, Gbit Ethernet with TSN | Yes |
Start your development using the Arm® Cortex®-M55 core with our recommended start kit
NUCLEO-N657X0-Q
The STM32 Nucleo-144 development board with the STM32N657X0 MCU supports Arduino, ST morpho, and camera connectivity.