Register for trainings at SPIE Photonics West

At Photonics West 2025, we will be offering hands-on trainings to current and prospective Moku users. Register to learn how to:

  • 10 a.m.: Moku 101: Get started with reconfigurable, FPGA-based instrumentation
  • 11 a.m.: Neural Network: Develop machine learning for real-time signal processing
  • 12 p.m.: Moku 101: Get started with reconfigurable, FPGA-based instrumentation
  • 1 p.m.: Python and MATLAB: Use Moku APIs for automated experiments
  • 2 p.m.: Boxcar Averager: Improve SNR with a custom Boxcar Averager
  • 3 p.m.: Custom FPGA programming: Code, compile, and deploy tools to Moku Cloud Compile

Event details
January 29, 2025
SPIE Photonics West Exhibition | South Lower Mezzanine, room #74
The Moscone Center
San Francisco, California, USA

See below for more details on each training session. Space is limited, so register now to secure your spot for a chance to win a Moku:Go.

Getting started with reconfigurable, FPGA-based instrumentation: Moku 101

Location: Room #74, South Lower Mezzanine

Time: 10 am and 12 pm, Wednesday, January 29th

Abstract

Traditional optical and electronic systems often rely on standalone test devices, leading to challenges such as instability, noise, and complexity. Flexible, digitally implemented instruments offer a powerful solution for optimizing control and characterization systems with enhanced speed and precision. In this session, we present innovative strategies for modernizing experimental setups with agile, reconfigurable, FPGA-based technology through Moku devices. Register to:

  • Understand the advantages of real-time digital signal processing for photonic and electronic experiments.
  • Leverage flexible, FPGA-based processing with a digital-first approach to quickly adapt instrumentation to your measurement requirements.
  • Build sophisticated signal-processing pipelines with multi-instrument capabilities.
Register

Neural Network development for real-time signal processing

Location: Room #74, South Lower Mezzanine

Time: 11 am, Wednesday, January 29th

Materials: Moku:Pro devices will be provided. Please bring your laptop.

Abstract

In this session, learn to deploy a real-time, FPGA-based neural network in-line with other Moku test and measurement instruments. Build, train, and run a signal analysis autoencoder to denoise arbitrary input signals in real time. This session will be a hands-on tutorial walking through Neural network configuration in Python, deploying to Moku, and applying machine learning algorithms to physical systems for real-time data analysis. 

Sign up to learn how to:

  • Access the Moku Neural Network, which offers five dense layers of up to 100 neurons each, and five different activation functions.
  • Generate training data, train your neural network, and run it in real time in Multi-instrument Mode on Moku:Pro.
  • Build a neural network with ready-to-use Python notebooks, including the example shared in this presentation.
Register Download quick start guide

Using Python and MATLAB for automated experiments using Moku APIs

Location: Room #74, South Lower Mezzanine

Time: 1 pm, Wednesday, January 29th

Materials: Moku:Pro, Moku:Lab, and Moku:Go devices will be provided. Please bring your laptop.

Abstract

Automating control of experiments is critical for efficient data taking and repeatable results. To this end, Python has become the programming language of choice for an expansive list of research fields, owing to its ease of use and abundant supporting resources. Likewise, MATLAB is also widely used for experimental control and analysis due to its powerful processing capabilities and toolboxes.

 

During this hands-on training session, we will explain how to use Python and MATLAB to implement an experimental control stack with Moku, a family of reconfigurable, FPGA-based instruments, to maximize efficiency and speed. You’ll learn how to leverage Python and MATLAB for streamlined connection, control and data viewing, which will allow you to easily integrate Moku devices into your experimental stack, alongside other necessary components.

Sign up to learn how to:

  • Install the Python and MATLAB APIs for Moku
  • Connect to and configure your Moku device using the API commands 
  • Control instrument settings and retrieve data collected from the instrument
Register

Implementing a Boxcar Averager with custom FPGA-based tools

Location: Room #74, South Lower Mezzanine

Time: 2 pm, Wednesday, January 29th

Materials: Moku:Pro, Moku:Lab, and Moku:Go devices will be provided. Please bring your laptop.

Abstract

In this session, learn how to deploy a boxcar averager tool alongside Moku instruments like an Oscilloscope, Arbitrary Waveform Generator, or Lock-in Amplifier to enhance signal to noise ratio of low-duty-cycle signals. We will demonstrate how to implement the readily-available Moku Boxcar Averager tool to the FPGA inside Moku devices using a pre-built bitstream, then customize the boxcar parameters to fine tune the result. Sign up to learn how to:

  • Implement a real-time, custom Boxcar Averager for SNR improvement
  • Build multi-instrument signal processing pipelines for complex signal analysis in optics and photonics research
  • Adjust and understand boxcar parameters to achieve ideal results
Register Download quick start guide

Designing and deploying custom tools: How to use Moku Cloud Compile

Location: Room #74, South Lower Mezzanine

Time: 3 pm, Wednesday, January 29th

Materials: Moku:Pro, Moku:Lab, and Moku:Go devices will be provided. Please bring your laptop.

Abstract

Learn to develop and deploy custom functions in minutes with Moku Cloud Compile, making FPGA programming simple with the ability to implement custom functionality on Moku devices. Moku Cloud Compile enables you to run a user-programmable FPGA alongside advanced test and measurement instrumentation to stimulate, analyze, or augment your custom design. In this session, we will walk through a variety of custom functions deployable with Moku Cloud Compile including simple arithmetic, complex math functions like a square root function, and finally a boxcar averager. Sign up to learn how to:

  • Customize existing VHDL examples without the typical overhead of FPGA programming
  • Deploy custom features to the Moku FPGA
  • Build unique signal processing pipelines and analyze results in real time
Register