HENDERSON, Nev. — (BUSINESS WIRE) — October 2, 2019 — Aldec, Inc., a pioneer in mixed HDL language simulation and hardware-assisted verification for ASIC and FPGA designs, will be exhibiting at Arm TechCon (October 8-10, San Jose, California, USA) and demonstrating solutions that stand to aid greatly in the development of Deep Neural Network (DNN) and Machine Learning (ML) applications.

Visitors to Aldec’s booth (#233) will discover how the company’s versatile EDA tools, hardware platforms and reference designs can enable engineers to fast-track their DNN or ML application development.

For instance, Aldec will be demonstrating a traffic detection reference design for its popular TySOM-3A-ZU19EG platform, which features the largest FPGA in the Xilinx® Zynq® UltraScale+ MPSoC family, the ZU19EG. The reference design also includes ready-to-use gesture detection, pedestrian detection, and segmentation detection.

The company will also be demonstrating the recently announced FPGA autopartitioning feature of HES-DVM™, Aldec’s fully automated and scalable hybrid verification environment, as well as third party board support, and how the feature can be used for the optimization of the resources of multiple FPGAs when prototyping ML applications intended for implementation in an ASIC or SoC.

“The programmability and flexibility of hybrid FPGAs that host firmware and software has made them one of the best choices for machine learning and neural network applications,” comments Zibi Zalewski, General Manager of Aldec’s Hardware Division, “and at Aldec we’re committed to providing the most suitable platforms and EDA tools to aid design and verification. Moreover, the growing number of reference designs, many of which are complementary, for our TySOM platforms means engineers are not having to start their projects with a blank sheet.”

Aldec’s demonstrations on booth #233 at Arm TechCon 2019 will be:

• DNN-based Traffic Detection Using Xilinx Zynq US+ FPGA – In this demo, traffic detection is done using a Convolutional Neural Network (CNN) on a TySOM-3A-ZU19EG development board. Deep Learning Processing Units (DPUs) are implemented in the FPGA for the acceleration of object detection and recognition, which results in 45fps for three input channels.
• Hybrid Co-Emulation using ARM Hardware Model - Hardware-software co-design and co-verification are a must for any kind of SoC design. A fast and accurate hybrid emulation platform can be built using the ARM hardware model provided by Xilinx Zynq MPSoC. Aldec will demonstrate how to connect a Xilinx Zynq MPSoC and its ARM Cortex A53/R5 processors with the largest Xilinx UltraScale US440 FPGA on a HES emulation board.
• Automatic Partitioning Design for Multi-FPGA Prototyping - Multi-FPGA partitioning has always been a challenge due to the limited number of FPGA I/Os and FPGA-specific clocking trees. Aldec provides a HES-DVM prototyping toolbox that automates design partitioning for multiple FPGAs and integrates an ultra-fast HES Proto-AXI host bridge. This demo also includes the use of SyntHESer, Aldec’s HDL compiler for HES.

With its focus for Arm TechCon 2019 on DNN and ML application development, one of Aldec’s demonstrations will be of a hybrid emulation platform created through connecting a TySOM board and a HES board.

About Aldec

Established in 1984, Aldec is an industry leader in Electronic Design Verification and offers a patented technology suite including: RTL Design, RTL Simulators, Hardware-Assisted Verification, SoC and ASIC Prototyping, Design Rule Checking, CDC Verification, IP Cores, High-Performance Computing Platforms, Embedded Development Systems, Requirements Lifecycle Management, DO-254 Functional Verification and Military/Aerospace solutions. www.aldec.com

Aldec is a registered trademark of Aldec, Inc. All other trademarks or registered trademarks are the property of their respective owners.

View source version on businesswire.com: https://www.businesswire.com/news/home/20191002005804/en/

Media Contact:
Richard Warrilow
Declaration Limited
T: +44 (0)1522 789 000
Email Contact