8440-2088  Today, developers are using secure and performance-enhancing technologies to develop small, low-power embedded systems that enable previously unimaginable AI applications such as voice, vision, and vibration that are changing the world.

The embedded space is undergoing a profound transformation. Connected devices are evolving into systems that make their own decisions based on the data they collect. The ability to process data closer to the source than at the iot gateway or in the cloud promises to speed up decision making, reduce latency, address data privacy concerns, reduce costs, and improve energy efficiency.

Many applications are driving the demand for edge computing in terms of performance and functionality, such as industrial automation, robotics, smart cities and home automation. In the past, the sensors in such systems were much simpler and disconnected, but now artificial intelligence (AI) and machine learning (ML) have increased the level of local intelligence, making decisions on the side, which was not possible with the simple control algorithms used in the past.

The evolution of general-purpose processors in the AI era

Years ago, developers focused on logic and control algorithms as the core of software development, however, with the advent of digital signal processing (DSP) algorithms, many enhanced voice, visual and audio applications are supported.

This shift in application development has entered a new era and is affecting the design of computing architectures. We have now moved to reasoning as the primary core of algorithm development, a phase that brings new or higher demands on computational performance, energy efficiency, latency, real-time processing, and scalability.

8440-2088  The need is not only for new processor accelerators, but also for general processing power to provide the necessary balance for developers and support applications such as feature checking or person detection in live video.

A few years ago, developers had to rely on frequency-based filters to create noise cancellation applications. Today, developers can improve the performance and functionality of their applications by combining filtering with ML/AI models and reasoning. To make these development tasks more efficient and serve users as seamlessly as possible, the need for processors and tools is also increasing.

Promote the intelligence of edge side and end side equipment

This evolution and innovation is driven by ML, but it also faces many technical challenges. Years of trying to create a common development methodology for iot and embedded devices have prompted the industry to transform the way iot is developed to unlock the infinite possibilities of scale.

Today, developers are using secure and performance-enhancing technologies to develop small, 8440-2088  low-power embedded systems that enable previously unimaginable applications such as voice, vision, and vibration that are changing the world. Various versions of programming languages and Transformer models will soon find a place in iot edge devices with new computing capabilities. This certainly opens up more possibilities for developers to dream of.

In the process of development evolution and innovation, in order to meet the needs of developers for hardware, Arm introduced Arm® Helium™ vector processing technology in the Armv8.1-M architecture a few years ago. Helium delivers significant performance gains for ML and DSP applications in small, low-power embedded devices. In addition, it offers Single instruction Multiple Data (SIMD) capability, which takes Arm Cortex®-M devices to a whole new level of performance and supports applications such as predictive maintenance and environmental monitoring.

Helium improves DSP and ML performance, speeding up signal conditioning (such as filtering, noise cancellation, and echo cancellation) and feature extraction (audio or pixel data), which can then be transferred to classifications using neural network processors.