Iridium has unveiled a three-radio module for global IoT deployments. The Iridium 9604 integrates Iridium SBD satellite, LTE-M cellular, and multi-constellation GNSS, with commercial availability set for June 2026.
MIT has demonstrated paired PUFs using standard CMOS fabrication steps. A new “twin fingerprint” method lets two chips authenticate each other directly, avoiding external storage of secret fingerprint data, and targeting power-constrained paired devices such as medical sensors.
Ericsson is pushing AI inference deeper into 5G radio hardware. Ahead of MWC 2026, the company has launched AI-ready Massive MIMO radios with neural network accelerators, plus RAN software updates for beamforming, positioning, and coverage prediction.
Danisense has widened its DN1000ID range with higher isolation distance. The DN1000ID-CP02 extends creepage and clearance to 38 mm and targets high-current, high-voltage power measurement, including EV chargers and energy storage hardware, with 1 ppm linearity at 1000 A continuous.
Niobium has lined up Samsung Foundry for FHE silicon production. A partnership with SEMIFIVE covers design, packaging, and test for an 8nm 8LPU accelerator intended to run fully homomorphic encryption workloads at practical speeds in cloud and AI infrastructure.
Mouser is shipping u-blox MAYA-W3 Wi-Fi 6 modules now globally. The host-based range targets healthcare and industrial systems with SISO Wi-Fi 6/6E, Bluetooth Low Energy 5.4 with LE Audio support, multiple antenna options, and evaluation kits intended to accelerate integration.
AI data centres are turning NAND into a scarce resource. Phison is warning of a 2026 squeeze as enterprise SSD demand accelerates for inference, vector search, and cold-tier storage. With suppliers prioritising margin and process transitions limiting bit growth, embedded buyers may find eMMC and UFS harder to source.
Samsung has begun shipping HBM4 memory for AI accelerators commercially. The company says production uses a 10 nm-class DRAM process with a 4 nm logic base die, delivering 11.7 Gbps per pin, up to 3.3 TB/s per stack, and 24–36 GB capacities.