Lattice Semiconductor CrossLinkU-NX System-on-Module (SoM) Board

Lattice Semiconductor CrossLinkU-NX System-on-Module (SoM) Board is built on the CrossLinkU™‑NX LIFCL‑33U FPGA, optimized for resource‑constrained Edge AI applications. This SoM board enables direct access to CSI‑2 camera input, USB 3.1 device interface, digital I/O, and an Always‑On (AON) feature. This enables advanced functionality with a low power footprint in compact designs. The CrossLinkU-NX SoM board can be paired with a purpose‑built carrier card for Raspberry Pi CM5 connectivity. This SoM board supports essential interfaces such as Gigabit Ethernet, USB 3.1 host, microSD storage, GPIOs, and comprehensive debugging. The CrossLinkU-NX SoM board runs the highly optimized Convolutional Neural Network (CNN), supporting real-time person and gesture detection for up to five individuals. This board features a programming circuit that supports the Radiant programmer through a USB or FTDI interface using JTAG or SPI.

Features

LIFCL-33U FPGA (LIFCL-33U-CTG104):
- MIPI CSI-2 receiver Soft D-PHY interface for camera data
- Support SPI Flash configuration
Board resources:
- On-board JTAG connector header
- 22-pin standard connector for Raspberry Pi (RPI) camera
- On-board 100-pin B2B connector to connect with carrier card
- 64Mb for HyperRAM memory interface with the CrossLinkU-NX device
- On-board USB Type-C connector for USB aggregation
- On-board 512Mb SPI Flash interface with the CrossLinkU-NX device

Programming circuit:
- Supports Radiant programmer through a USB or FTDI interface using JTAG or SPI
- On-board Flash
- On-board JTAG header

Kit Contents

CrossLinkU-NX SoM HW board
Generic carrier HW board
Raspberry PI compute module 5
Raspberry PI compute module 5 passive cooler
IMX219 camera with regular (narrow) FOV
200mm 22-22 RPI 5 camera cable
3.00' USB cable Type-A male to Type-C male
IMX219 camera holder
5V 25W AC/DC external wall mount (Class II) adapter multi-blade (sold separately) input
Input plug for wall adapter