In this detailed overview, you'll gain a deeper understanding of machine learning, laying the foundations for other courses in the program. With a heavy emphasis on practical application, this course will provide you with essential data cleaning preparation techniques, along with logistic regression, foundational statistics, decision trees, and preliminary exploration of neural networks. This course will be primarily taught using Python, with some additional use of MATLAB.

In this course, you'll take the mathematical theories that underpin the discipline of signal processing, and use them in applied settings, allowing you to analyze, optimize, and adjust a wide range of data types.

You'll learn about the use of:

• Signal filtering
• De-noising
• Signal to noise ratio
• Signal enhancement
• Data compression
• Feature extraction
• Feature engineering
• Discrete and fast Fourier Transform
• Random signals
• Probability distributions

Building on the knowledge gained through the Signal Processing and Machine Learning courses, here you'll learn the basics of natural language processing (NLP)—the linguistic theories underpinning it, the techniques and challenges that define the NLP landscape, and both the current and developing tools used to implement it. You'll also gain a deeper understanding of the principles governing the development of generative AI models.

Prerequisites: Signal Processing, Machine Learning

In this course, you'll take concepts of machine learning and signal processing learned earlier in the program, and learn how these tools can be used to allow computers to extract high-level understanding from visual images. You'll trace the development of machine vision capabilities, from traditional machine vision tools through to the latest neural network algorithm functionality.

Prerequisites: Signal Processing, Machine Learning

This course focuses on the challenges and methods involved in processing sensor data as it streams, as opposed to static datasets. You'll learn about the ways that streaming data is pre-processed, filtered and interpreted, and how cumulative meaning and context can be continually extracted from the data stream. You'll learn about the specific types of neural networks used to process this kind of data, and the real-world challenges such as latency that affect how we use sensors.

Prerequisites: Signal Processing, Machine Learning

You'll learn about the different types of hardware platforms, software tools, and techniques used in the design of intelligent systems. Focusing particularly on the application of microcontrollers, you'll learn how to design, program, test, and debug embedded systems. You'll develop hardware-level device drivers for connected sensors/actuators, implement real-time data processing and control algorithms, and work with communications interfaces.

Prerequisites: C is the primary language used in microcontroller programming. Students should be familiar with the C language and, in particular, with C pointers and structures. Microcontrollers are programmed "at the hardware level"; students should be familiar with basic digital logic concepts including digital I/O, discrete logic gates (AND, OR, NOT), registers, Boolean/hexadecimal number representation, and Boolean arithmetic operations (add, multiply, negate).

In this course, you'll learn how to use FPGA architecture and algorithms for deep neural network learning. You'll gain an overview of the specialized hardware devices being used to implement deep neural networks across a broad range of industries and applications, and why FPGA systems are the natural choice in many of these instances.

In this course, you'll learn how different code needs to be implemented and executed across a variety of platforms, keeping in mind the different capabilities of these platforms, their requirements, and their limitations.

* This course may be taken concurrently with the capstone course.

In this final course, you'll apply everything you've learned and work with your peers on a larger-scale, 'Smart Sensors' project. Your instructors will aim to scaffold your learning by breaking down the project into stages, based on the different subject areas you've already covered. Previous 'Smart Sensors' projects have required students to plan, design and create a mobile sensor device for biomedical application, incorporating multiple course threads such as signal processing, sensor data processing, and NLP keyword processing.