Hardware Support

Qualcomm Hexagon Hardware Support from Embedded Coder

Generate code optimized for Qualcomm Hexagon NPU, CPU, and LPAI with Embedded Coder

The Qualcomm® Hexagon® Support Package enables you to model advanced embedded applications in MATLAB and Simulink and generate optimized code for Qualcomm Hexagon NPU, CPU and LPAI. You can test and validate your embedded application with processor-in-the-loop (PIL) simulations.

To accelerate design and verification of your embedded application, you can:

  • Use the QHL scalar processor and HVX vector processor libraries
  • Integrate with third-party software like Qualcomm Hexagon SDK and Qualcomm AI Engine
  • Generate optimized code using DSP blocks, DSP objects, math operators, and math functions
The code and corresponding replacements report for Qualcomm Hexagon.

Getting Started with the Support Package

Generate, verify, and profile the optimized code using Embedded Coder Support Package for Qualcomm Hexagon Processors.

Output waveform transitions that slowly fades-in and fades-out.

Deploy Smart Speaker Model on Qualcomm Hexagon eNPU using LPAI SDK Add-on

Deploy a Simulink model designed as smart speaker system that applies Speech command Recognition Network on Qualcomm Hexagon LPAI. 

Deep Learning network classifications

Deploy Smart Speaker Model on Qualcomm Hexagon DSP

Deploy a Simulink model designed as smart speaker system that applies Speech Command Recognition Network on Qualcomm Hexagon DSP. 

Diagram of processor-in-the-loop simulation.

Using Code Verification and Validation with PIL

Use Embedded Coder Support Package for Qualcomm Hexagon Processors when performing code verification and validation with PIL.

Chart compares the output of a 8-bit quantized eAI model with the output of a floating-point MATLAB network

Convert MATLAB Deep Learning Networks to eAI Model Using Qualcomm LPAI SDK

Convert Deep Learning networks developed using MATLAB built-in layers into eAI models for deployment to LPAI using Qualcomm LPAI SDK. 

Display various metrics of the profiled time, such as average execution time.

C Code Optimization Using MATLAB Coder app for Hexagon DSP

Generate optimized C code from MATLAB Coder app for the Qualcomm Hexagon Simulator using the Qualcomm Hexagon QHL Code Replacement Library (CRL).

Supported Hardware

The support package has been tested on many platforms, including (but not limited to):

  • Snapdragon® 8 Elite (SM8750)
  • Snapdragon® 8 Gen 3 (SM8650)
  • Snapdragon® 8 Gen 2 (SM8550)
  • Qualcomm Dragonwing™ RB3 Gen 2 (QCS6490)
  • Qualcomm® Robotics RB5 (QRB5165)