Smart and secure

  

Component firmware is an increasingly popular focus for cyberattacks. In 2018, security weaknesses rendered more than 3 billion chips (in systems of all types) vulnerable to data theft via the exploitation of firmware vulnerabilities. Unsecured firmware also exposes OEMs to the financial and brand reputation risks associated with device hijacking (for use in DDoS attacks) and device tampering or destruction.

Looking to address these risks, Lattice has announced the MachXO3D FPGA which features root-of-trust capability. When used to implement system control functions, MachXO3 FPGA devices are normally the “first-on/last-off” component on circuit boards. By integrating security and system control functions, the MachXO3D becomes the first link in chain of trust that protects entire systems.

Gordon Hands, Director of Marketing at Lattice, said, “We’re seeing a rising concern over equipment hijacking, design theft, data corruption and data theft, and counterfeiting and overbuilding. People are starting to think about security across the whole product lifecycle.”

With MachXO3D, Lattice is enhancing the device configuration and programming steps in the manufacturing process. These enhancements, in combination with MachXO3D’s security features, are designed to protect systems throughout the component’s entire lifecycle by securing communication between the MachXO3D and legitimate firmware providers.

“System developers commonly take advantage of FPGA flexibility to enhance system functions after deployment,” added Hands. “With MachXO3D we took care to retain that flexibility while adding a secure configuration block to deliver the industry’s first control-oriented FPGA compliant with NIST’s Platform Firmware Resilience specification.”

Along with securer devices, Lattice is also looking to enable smarter devices, and had announced an update to its sensAI solution stack, designed to address the growing need for edge devices with AI support. The stack provides a hardware and software solution for implementing low power (1mW-1W), always-on AI functionality in smart devices operating at the edge and includes a 10x performance boost over the previous version.

Lattice’s sensAI is designed to enable OEMS to update their existing designs with low power AI inferencing optimised for their application requirements. According to Hands, operating such local intelligence also lowers expenses related to Cloud-based analytics by only sending relevant data for further processing. The update also accelerates the design cycle with several new features. For example, expanding neural network and machine learning frameworks support including Keras.