AMD-Xilinx has announced that it has added Versal Premium with AI Engines to the Versal Premium series, providing high-performance signal processing for next-generation wireless and radar systems.

Part of the 7nm Versal Adaptive Compute Acceleration Platform (ACAP) portfolio, Versal Premium with AI Engines has been optimised to meet the demands of signal processing-intensive applications in the aerospace and defence (A&D) and test and measurement (T&M) markets and is able to deliver a 4X increase in signal processing capacity and additional functionality, at reduced power and with a smaller footprint, compared to the prior generation 16nm Xilinx devices. 

Versal Premium with AI Engines covers a wide range of functions, from A&D radar applications such as adaptive beamforming, to signal processing for RF machine learning applications like digital RF memory and direction finding.

In the T&M market, it addresses the growing demand for wireless testers as part of the global 5G rollout, applications such as 5G protocol testers and production testers as well as semiconductor automated test equipment. 

The Versal Premium series with AI Engines eliminates I/O bottlenecks with up to 9Tb/s serial bandwidth, and offers significantly reduced size, weight and power through the heterogeneous, power-optimised integration of hardened, ASIC-like cores such as 100G/600G Ethernet cores, 400G High-Speed Crypto Engines, DDR memory controller, and integrated PCIe Gen5 blocks.

The series combines highly advanced signal processing from the proven Versal AI Core series, with the digital signal processing (DSP) compute capacity and serial bandwidth capacities of the existing Versal Premium series. By combining AI Engines with DSP Engines, users of Versal Premium with AI Engines devices can expect to see significant performance gains, according to AMD-Xilinx, when compared to the prior generation 16nm Xilinx devices and competing products on the market today.

In addition, the Versal Premium series with AI Engines deliver major performance gains and additional functionality at reduced power and with a smaller footprint. In radar beamforming applications, the heterogenous compute engines enable a 67% smaller footprint and up to 43% lower power with 2X beamforming performance.

As with all members of the Versal ACAP portfolio, development tools suited for a wide range of developers are available including: Vivado ML for hardware developers; Vitis and Vitis AI development platforms for software developers and AI and data scientists.

Customers will also have access to AI Engine tutorials and demo designs to help them get up and running.

While the Versal Premium series with AI Engines is expected to begin shipping in the first half of next year, customers can start prototyping using the existing Versal Premium and Versal AI Core Evaluation Kits and devices for which documentation and tools are available.