Small FPGAs are energy-efficient for always-on monitoring
Adding to the Lattice Semiconductor iCE40 Ultra family, the iCE40 UltraPlus FPGAs deliver eight times more memory (1.1 Mbit RAM), twice the DSPs (eight) and improved I/Os, compared with previous generations. They are available in multiple package sizes, and are suitable for smartphones, wearables, drones, 360 cameras, human-machine interfaces (HMIs) and industrial automation, as well as security and surveillance products. They can be used for voice recognition, gesture recognition, image recognition, haptics, graphics acceleration, signal aggregation, I³C bridging (the MIPI standard increases bandwidth) for devices that need to be always-on, always listening and ready to instantly process commands locally without going to the cloud.
Heterogeneous computing allows processors to offload power-hungry applications to reduce the power budget. The increased number of DSPs offers the ability to compute higher quality algorithms, says the company, while increased memory allows data to be buffered for longer low-power states. The flexible I/Os enable a more distributed heterogeneous processing architecture for innovations to be introduced to the end product.
The FPGAs provide always-on sensor buffer and distributed processing at sub-1mW power consumption and enables functions such as gesture detection, facial recognition, audio enhancement, audio beam forming, phrase detection, double tap, shake-to-wake and pedestrian dead reckoning (PDR). There is also frame buffer and graphics acceleration and an always-on display while the application processor is in sleep mode. Multiple microphones reduce background noise reduction and perform audio equalisation. The array of GPIOs, which includes SPI, UART, I²C and I³C signals, over a single PCB trace eliminates routing contention issues to reduce system cost and simplify designs, adds the company.
Other key features are the 1.1Mbit SRAM, up to 5K LUTs and non-volatile configuration memory (NVCM) for instant-on applications. MIPI-I³C support for low-resolution, always-on camera applications and sub 100microW of standby power consumption are also beneficial in wearable and IoT design.
Evaluation samples and boards are available now.