➀ Ceva recently unveiled the XC21 and XC23 DSP cores for advanced wireless and edge AI processing;
➁ The XC21 is optimized for low-power, high-performance applications in cost-sensitive markets;
➂ The XC23 provides significant performance improvements for high-end communications with AI capabilities.
➀ AONDevices specializes in super-low-power Edge AI solutions with a focus on energy efficiency and high accuracy.
➁ The AONx360 platform simplifies ML model creation and deployment.
➂ AONDevices differentiates itself through an end-to-end approach combining ML chips, models, and software.
➀ Lantronix has launched the Open-Q 8550CS System-on-Module (SOM), built on the Qualcomm Dragonwing QCS8550 processor, offering low-power on-device AI and ML capabilities.
➁ The SOM is designed for demanding AI/ML applications in extreme Edge computing environments, supporting advanced video processing, AI inference, and Edge AI gateway integration.
➂ Key features include low power consumption with a 4nm process, dual eNPU for AI acceleration, and support for high-speed connectivity and multiple display interfaces.
➀ Synaptics has introduced the SR-Series adaptive microcontrollers (MCUs) to expand its Astra AI-Native platform for Edge AI applications, offering three performance tiers: performance (100 GOPS), efficiency, and ultra-low-power always-on (AON) processing.
➁ These MCUs feature an Arm Cortex-M55 core combined with an Arm Ethos-U55 neural processing unit (NPU), along with multiple camera interfaces, secure memory, and accelerators, targeting applications such as battery-powered security cameras, sensors, and smart appliances.
➂ The SR-Series supports the Astra Machina Micro development kit and an open-source SDK, enabling developers to create context-aware cognitive IoT devices with adaptive vision, audio, and voice processing capabilities.
➀ AI正逐渐向边缘迁移，这是好事；
➁ 设计师必须为巨大的电力需求做好准备；
➂ 准备合适的硬件和策略以优化效率。
➀ Synaptics has expanded its Veros connectivity portfolio with the SYN461x family;
➁ The SYN461x is designed for embedded Edge AI IoT and offers ULP Wi-Fi, Bluetooth 6.0, and BLE;
➂ It is optimized for power, integration, size, and market time while providing high performance and interoperability.
➀ AONDevices, a leader in super-low-power, high-accuracy edge AI solutions, has announced a strategic partnership with Faraday Technology Corporation.
➁ The collaboration aims to enhance AONDevices' production capabilities for its advanced AI processors.
➂ Faraday Technology brings its comprehensive manufacturing expertise to the partnership.
The Fraunhofer Institute for Integrated Circuits IIS has developed an AI chip for processing Spiking Neural Networks (SNNs). The SENNA spiking neural network inference accelerator, inspired by brain function, consists of artificial neurons and can process electrical impulses (spikes) directly. Its speed, energy efficiency, and compact design enable the use of SNNs directly where data is generated: in edge devices.
SNNs consist of a network of artificial neurons connected by synapses. Information is transmitted and processed in the form of electrical impulses, allowing pulsing networks to be the next step in artificial intelligence: faster, more energy-efficient, and closer to the processing method of the human brain. To bring these advantages into application, small, efficient hardware that mimics a structure of neurons and synapses is needed. For this, the Fraunhofer IIS has developed the neuromorphic SNN accelerator SENNA as part of the Fraunhofer project SEC-Learn.
SENNA is a neuromorphic chip for fast processing of low-dimensional time series data in AI applications. The current version consists of 1024 artificial neurons on less than 11 mm² of chip area. Its low reaction time down to 20 nanoseconds ensures precise timing in time-critical applications at the edge. This makes it particularly strong in real-time event-based sensor data processing and in closed control systems, such as the control of small electric motors with AI. With SENNA, AI-optimized data transmission can be realized in communication systems. There, the AI processor can analyze signal streams and adjust transmission and reception methods as needed to improve efficiency and performance.
➀ YorChip, in collaboration with SiliconIPs, has developed a 50nS latency 100G Ultra Ethernet ready Mac/PCS IP core;
➁ The IP core is optimized for size, power, and latency;
➂ It targets edge AI applications and Data Center Accelerator developers requiring low power and high performance.
➀ NXP is acquiring Kinara, a specialist in programmable discrete neural processing units (NPUs).
➁ The acquisition is valued at $307 million and is expected to close in the first half of 2025.
➂ Kinara's NPUs and software enable energy-efficient AI performance across various neural networks.
➀ The rise of edge AI has spurred semiconductor designers to build accelerators for performance and low power, leading to a proliferation of NPUs among in-house, startup, and commercial IP product portfolios.
➁ The complexity of software and hardware around neural network architectures, AI models, and base models is exploding, requiring sophisticated software compilers and instruction set simulators.
➂ The hardware complexity of inference platforms is evolving, with a focus on performance and power efficiency, especially for edge applications.
➃ The combination of tensor engines, vector engines, and scalar engines in multiple clusters to address the challenges of acceleration is complex and costly.
➄ The supply chain and ecosystem for NPUs are becoming increasingly complex, with intermediate manufacturers and software companies having limited resources to support a wide range of platforms.
