One of the biggest new trends we are seeing this year is the growing demand for edge computing systems. The main reason for this is that it is much faster and cheaper to process and analyze data captured by sensors locally on the system rather than send it to the corporate network or cloud over expensive and often low-bandwidth wireless connections.
This is particularly the case for applications that utilize cameras for security and surveillance, facial recognition, and autonomous driving and thus entail handling massive amounts of video and image data. Sending a constant stream of footage from a driverless vehicle, for example, over a wireless network to the cloud for analysis would cost an absolute fortune – not to mention be extremely difficult in areas with spotty coverage. It therefore makes a lot more sense to have in-vehicle systems that process and analyze the captured video locally and only send alerts to the central server when they are triggered by events such as accidents.
Even for less data-intensive applications like the remote monitoring of production machinery for predictive maintenance or air conditioners and heating systems for temperature and power consumption, the same principle applies. There is no point in clogging up the network with a flood of useless non-actionable data when you can use cost-effective edge computing systems to search through the haystack to find the needles of information that are vital for ensuring the smooth-running of your operation.
With over ten years’ experience of developing embedded platforms and systems, we have built up rich expertise in the design and deployment of reliable, low power edge computing systems — long before they were even known by that name. To see our latest solutions for smart industry, smart retail, smart building, and computer vision applications, you are welcome to visit our booth at Embedded World 2018.
For more information, please click here.