Weartech-ICs & Semiconductors

Computing-in-memory technology is poised to eliminate the massive data communications bottlenecks associated with AI speech processing at the network’s edge, said Witinmen. The company has worked with Microchip Technology’s subsidiary Silicon Storage Technology (SST) to develop an embedded memory that simultaneously performs neural network computation and stores weights. Microchip Technology announced that its SuperFlash memBrain neuromorphic memory has been combined with the Witinmem neural processing SoC. The SoC is claimed to be the first in volume production that enables sub-mA systems to reduce speech noise and recognise hundreds of command words, in real time and immediately after power-up.

Microchip has worked with Witinmem to incorporate Microchip’s memBrain analogue in-memory computing, based on SuperFlash technology, into Witinmem’s low-power SoC. The SoC features computing-in-memory technology for neural networks processing including speech recognition, voice-print recognition, deep speech noise reduction, scene detection, and health status monitoring. Witinmem is working with multiple customers to bring products to market during 2022 based on this SoC.

“Witinmem is breaking new ground with Microchip’s memBrain solution for addressing the compute-intensive requirements of real time AI speech at the network edge based on advanced neural network models,” said Shaodi Wang, CEO of Witinmem. “We were the first to develop a computing-in-memory chip for audio in 2019, and now we have achieved another milestone with volume production of this technology in our ultra-low-power neural processing SoC that streamlines and improves speech processing performance in intelligent voice and health products.”

Microchip’s memBrain neuromorphic memory is optimised to perform vector matrix multiplication (VMM) for neural networks. It enables processors used in battery-powered and deeply-embedded edge devices to deliver the highest possible AI inference performance per Watt. This is accomplished by both storing the neural model weights as values in the memory array and using the memory array as the neural compute element. The result is 10 to 20 times lower power consumption than alternative approaches, claims Microchip, and a lower overall processor bill of materials (BoM) costs because external DRAM and NOR are not required. 

Permanently storing neural models inside the memBrain’s processing element also supports instant-on functionality for real time neural network processing. Witinmem has leveraged SuperFlash technology’s floating gate cells’ non-volatility to power down its computing-in-memory macros during the idle state to further reduce leakage power in demanding IoT use cases.

http://www.sst.com

http://www.microchip.com

http://www.witintech.com

Certified security is combined with a tamper status detection mechanism and battery-free sensing in the NTAG 22x DNA StatusDetect IC family by NXP. The ICs measure a change in ambient conditions, such as moisture, liquid fill level or pressure and allow developers to combine secure authentication with opening status detection or condition monitoring of products to help maintain a secure supply chain and product integrity. 

Physical products can be authenticated by leveraging the NTAG 22x DNA IC’s secure unique NFC (SUN) authentication message feature. This allows manufacturers to cost-effectively combat counterfeits and supply chain fraud, said NXP. 

The electronic tamper status detection of the ICs enables manufacturers or product users to verify a product’s unauthorised opening. By measuring capacitive changes in an item’s environmental conditions such as moisture, pressure or fill level, upon a simple tag readout, it is also possible to ensure product quality remains intact or capture digital sensing data for healthcare, retail or industrial applications. 

According to NXP, the inclusion of security-certified NFC sensing turns a tag into a simple battery-less sensing device to detect a physical product’s first opening status, or a change in its specific ambient condition. It can help manufacturers protect product integrity, whilst enabling a new level of intelligence to assure product quality 

The NTAG 22x DNA family is Common Criteria EAL3+ -certified, and features a powerful, cryptographically secure authentication message that dynamically changes on every NFC phone tap, making the taps unclonable, without requiring a user application. 

The NTAG 22x DNA StatusDetect also includes configurable conductive or capacitive tamper detection, with once-open status irreversibly stored and protected in the IC memory without the need for a dedicated app. The conductive mode is suitable for tamper-evident labels and seals fixed on to a product or its package. The capacitive mode is suitable to integrate the tag into a physical product, and is also harder to reconstruct by a fraudster, said NXP.

The StatusDetect ICs can also be used as a passive sensing device to detect an environmental change influencing the capacitance value, interpreted with a mobile or cloud-based application. This facility allows new applications for medical IoT devices, such as a plaster that can detect moisture levels for wound care, fill level sensing for smart injectable dosage devices. It can also be used for consumer products as refill reminders based on package fill levels and leak detection. 

Security features include a 7byte identifier, a SUN message authentication using an AES-128 key and has user memory protected with 32-bit password or with mutual authentication with AES-128 key.

The StatusDetect devices have capacitive measurement with up to 64 granular steps and automated mirroring of UID, NFC counter and status value into IC’s user memory as part of NFC-NDEF message, secured with a SUN message code.

The ICs are available in sawn and bumped wafer format (120 and 75 micron) and with an internal tuning capacitance of 50pF.

http://www.nxp.com