Graduation Assignments mmWave Radar

Graduation Assignments mmWave Radar
Back to vacancies overview

Graduation Assignments mmWave Radar

  • Location Groenlo
  • Business unit Healthcare

It is becoming increasingly important to provide care while staying in the comfort of your own home. Determining whether that is possible in a safe and reliable manner, while at the same time reducing the number of care visits, is one of the big challenges that the healthcare sector is facing. One of the possible solutions is to monitor the well-being of a person at home in a non-invasive manner. This would allow authorised healthcare workers and family members to check-up on a person from a distance. Naturally, this raises privacy concerns, which we would like to address with the help of privacy-friendly technology.

We have selected mmWave radar as a candidate technology for addressing these problems, because it produces a point cloud of ~30 points for the average human which is quite sparse and considerably less intrusive than a regular optical camera. We are doing research in the area of mmWave radar sensors in order to quantify behavioural patterns and extract vital signs. Our mmWave radar systems operate in the standard range of ~60Ghz and ~77Ghz and allow for many applications with the advantage of being insensitive to lighting conditions and other situations where optical technologies struggle. 

Research area's

In the area of mmWave we are looking at a set of related problems that require a background in signal processing. Our research areas include:

  • Long range vital sign extraction (>3m) including HRV as physiological markers of stress (empirical-mode-decomposition / sfft, antenna design, adaptive movement compensation)
  • Robust and viewpoint independent activity recognition for both coarse-grained (falling, jumping-jacks) and fine-grained activities (writing, eating, reading, watching relevants)
  • Person identification, tracking and distinguishing between humans and pets (dogs, cats, etc)
  • Robustly dealing with reflection and (non-)stationary sources of noise in realistic home layouts (e.g. environments more complex than an empty rectangular room)
  • Multi-task classification and transfer-learning between tasks in ML-based systems. For instance, to deeply integrate the localisation & activity recognition ML components to reinforce that certain types of behaviours are more likely to occur in certain spatial locations.
  • Active learning & continuous learning with the use of human feedback (from healthcare workers) to fine-tune classification in a specific deployment
  • Unsupervised pre-training on anonymised point clouds from multi-site deployments
  • mmWave antenna design & PCB layout

Each individual research area can be considered to be a graduation assignment and/or internship on its own.

Your team

You'll primarily work in the Tech Exploration Team within Nedap Healthcare. Their mission is to find and apply new technology to solve problems in the Healthcare sector. You'll also work closely with the Data Science team and (technical) physicians in order to get rapid feedback about your work. Naturally, you will get access to hardware and rooms at our campus in Groenlo to conduct experiments.

Our offer

You can work from home or at our cool campus in Groenlo, the choice is yours. Naturally, you will receive a monthly internship allowance and are welcome to attend the 'Nedap Student Day' which is organized every six months. This is an excellent way to get in touch with interns, graduates and working students from other market groups within Nedap.

Required experience and skills

We're looking for someone with a background in electrical engineering, embedded systems development or machine learning. Some prior knowledge about mmWave radar is a plus, but in our experience can be learned as part of the assignment.

Nedap, Healthcare

About Nedap Healthcare

We help care givers and nurses to save time on administrative tasks. With our technology, they are able to spend less time on registering, planning, reporting, and drawing up care plans, allowing them to devote more time to their patients. The family can also be informed immediately about the patient’s wellbeing at the touch of a button.

Application procedure

Interested? Apply directly by using the application button and upload your CV and motivation letter to Jaap Grondman. If you have any questions please do not hesitate to contact us at jaap.grondman@nedap.com

  • 1. Your application
  • 2. First interview
  • 3. Next meet & greet
  • 4. Offer
  • 5. High five!
Step 1
Your application

You’ve applied? Awesome! We will reply to your application as fast as possible, but at least within 10 workdays via e-mail.

We respect your privacy, therefore you can only apply via our website. Applications via e-mail will not be accepted.

Questions? Get in touch with our recruiter