Edge Intelligence Ticks Many Boxes For AI

Find the Latest SoC Solutions
for... Automotive... IoT... Security... Audio... Video

eetimes.com, Feb. 12, 2020 – 

Making the case for edge intelligence: no networks and real-time data.

As the adoption of artificial intelligence (AI) and machine learning (ML) grows, the ability to process large amounts of data in the form of algorithms for computational purposes becomes increasingly important.

To help make the expanding use of data applications across billions of connected devices more efficient and valuable, there is growing momentum to migrate the processing from centralized third-party cloud servers to decentralized and localized on-device, commonly referred to as edge computing. According to SAR Insight & Consulting's latest artificial intelligence/machine learning embedded chips database, the global number of AI-enabled devices with edge computing will grow at a CAGR of 64.2% during the period 2019-2024.

Data computation at the edge, no network needed

Edge AI takes the algorithms and processes the data as close as possible to the physical system, in this case, locally on the hardware device. The advantage is that the processing of data does not require a connection. The computation of data happens near the 'edge' of a network, where the data is being developed, instead of in a centralized data-processing center. Determining the right balance between how much processing can and should be done on the edge will become one of the most important decisions for device, technology and component providers.

Given the training and inferencing engines that produce deep learning predictive models, edge processing usually requires an x86 or an Arm processor from suppliers such as Intel, Qualcomm, Nvidia and Google, an AI accelerator, and be able to handle speeds of up to 2.5 GHz with 10–14 cores.

Click here to read more...

 Back