Date: 21 May 2025
Time: 4:30 pm - 5:30 pm (GMT+8)
Location: Online
Overview
In this joint presentation, MathWorks and Qualcomm will outline a Model-Based Design (MBD) workflow for efficiently designing complex systems that integrate edge AI and leverage the Qualcomm Hexagon NPU. With MathWorks’ hardware support package for Qualcomm® Hexagon, developers can run simulations to evaluate how their AI models behave and perform within the broader systems context when deployed on Qualcomm NPUs. Afterwards, they can generate hardware-optimized, production code for complete edge AI applications. The generated code for AI models is automatically configured to run on the Hexagon NPU to optimize performance and power.
Using a smart speaker audio processing example, we will demonstrate the following workflow highlights:
Download the support package and related tools from Qualcomm
Simulate the demo example in Simulink using Qualcomm’s hexagon Simulator
Generate optimized target code
Generate the AI model target code and deploy on Qualcomm’s Hexagon NPU
Profile the system performance
The demo will illustrate the transformative role of NPUs for Hexagon, the technology embedded within Qualcomm’s Snapdragon® family of processors in advancing edge computing. The micro NPU on Hexagon is designed specifically for always-on, ultra-low power AI use cases. In addition to NPUs, Hexagon features a DSP, dedicated memory, and always-sensing image signal processors (ISPs). Today, many Snapdragon-powered devices, including smartphones, laptops, autonomous vehicles, XR devices, smart cameras, and IoT devices, are already leveraging the power of the Qualcomm Hexagon NPU.
About the Presenters
Talha Gorsi | Staff Product Manager | Qualcomm
Talha Gorsi is Staff Product Manager with experience of 15 years in the semiconductor industry as Product Manager and System Architect. Currently Talha is responsible for Hexagon Software Products at Qualcomm. Prior to that, he worked as System Architect at Google and Qualcomm. Talha holds master’s degree in electrical engineering from University of Southern California.
Adam Cook | Consulting Engineer | MathWorks
Adam Cook is an Applications Engineering Manager at MathWorks. He manages a team of application engineers focused on Signal Processing, Image Processing, and AI applications. Prior to working at MathWorks, Adam worked as a DSP algorithm developer and design engineer developing professional audio products. Adam has a B.S. in Electrical and Computer Engineering from the University of Rochester.
All rights reserved. Copyright © 2025 TechSource Systems and Ascendas Systems Group