What you will do
Millimeter-wave radar is increasingly used for contactless sensingand monitoring because it can operate independently of ambient illumination anddirectly capture range, velocity, and motion information. However, conventionalcommercially available MIMO radars have limited angular resolution, restrictingtheir ability to reconstruct detailed spatial representations of pedestrians.
Inverse Synthetic Aperture Radar (ISAR) provides an alternativeapproach to high-resolution imaging by exploiting the relative motion between atarget and a stationary radar. Previous work at imec has demonstrated aMIMO-FMCW ISAR pipeline for reconstructing high-resolution radar images ofpedestrians. The approach combines target tracking, motion compensation andautofocus, time-window selection, MIMO phase compensation, and image formation.Experimental measurements with a 60 GHz radar demonstrated pedestrian imagingat distances from approximately 6 to 11 metres.
Despite these advances, high-quality ISAR reconstruction remainschallenging because human motion is complex and non-rigid. In particular,imperfect motion compensation can introduce phase errors, artefacts, and imagedefocusing. Recent developments in deep learning and deep unfolding offerpromising ways to combine radar-domain knowledge with trainable components forimproved image formation and autofocus.
In this internship, you will investigate machine-learning techniquesfor high-resolution radar imaging of pedestrians, building on imec's existingFMCW/MIMO ISAR processing framework.
Your research activities
Possible extension: Depending on progress, the work may be extended toward 3-D radarimaging using multichannel or interferometric ISAR concepts. The core objectiveremains machine-learning-based autofocus and image reconstruction forpedestrian ISAR.
What we expect from you
Research outcome
The internship will offer hands-on experience at the intersection ofradar sensing, computational imaging, and machine learning. The objective is todevelop and validate new algorithms capable of extracting significantly richerspatial information about pedestrians than conventional radar processing.Subject to the achieved results, the work may provide a basis for a scientificpublication or a follow-up research project.
Type of internship: PhD internship, Master internship
Duration: 1 year
Required educational background: Computer Science, Physics, Electrotechnics/Electrical Engineering
Supervising scientist(s): For further information or for application, please contact Hamed Javadi (< email supprimé pour raison de sécurité >) and Sadat Hossain (< email supprimé pour raison de sécurité >)
The reference code for this position is 2026-INT-339. Mention this reference code in your application.
Imec allowance will be provided for students studying at a non-Belgian university.
Applications should include the following information:
Incomplete applications will not be considered.
Millimeter-wave radar is increasingly used for contactless sensingand monitoring because it can operate independently of ambient illumination anddirectly capture range, velocity, and motion information. However, conventionalcommercially available MIMO radars have limited angular resolution, restrictingtheir ability to reconstruct detailed spatial representations of pedestrians.
Inverse Synthetic Aperture Radar (ISAR) provides an alternativeapproach to high-resolution imaging by exploiting the relative motion between atarget and a stationary radar. Previous work at imec has demonstrated aMIMO-FMCW ISAR pipeline for reconstructing high-resolution radar images ofpedestrians. The approach combines target tracking, motion compensation andautofocus, time-window selection, MIMO phase compensation, and image formation.Experimental measurements with a 60 GHz radar demonstrated pedestrian imagingat distances from approximately 6 to 11 metres.
Despite these advances, high-quality ISAR reconstruction remainschallenging because human motion is complex and non-rigid. In particular,imperfect motion compensation can introduce phase errors, artefacts, and imagedefocusing. Recent developments in deep learning and deep unfolding offerpromising ways to combine radar-domain knowledge with trainable components forimproved image formation and autofocus.
In this internship, you will investigate machine-learning techniquesfor high-resolution radar imaging of pedestrians, building on imec's existingFMCW/MIMO ISAR processing framework.
Your research activities
- Familiarizeyourself with FMCW/MIMO radar signal processing, range-Doppler processing,tracking, and ISAR image formation.
- Processand analyse measured radar data from pedestrians performing different motions.
- Investigatethe limitations of conventional autofocus and motion-compensation techniquesfor human targets.
- Developmachine-learning methods that improve one or more stages of the radar imagingpipeline, with emphasis on autofocus, motion compensation, imagereconstruction, or image enhancement.
- Explorephysics-informed, model-driven, or deep-unfolding approaches that combine knownradar processing models with trainable components.
- Investigatetraining strategies that combine simulated radar returns with experimentalmeasurements.
- Quantitativelycompare the developed techniques with conventional signal-processing baselinesusing suitable image-quality and reconstruction metrics.
- Validatethe resulting algorithms on experimental pedestrian radar measurements.
Possible extension: Depending on progress, the work may be extended toward 3-D radarimaging using multichannel or interferometric ISAR concepts. The core objectiveremains machine-learning-based autofocus and image reconstruction forpedestrian ISAR.
What we expect from you
- You arepursuing an MSc or PhD in electrical engineering, telecommunications, computerscience, artificial intelligence, signal processing, or a closely relatedfield.
- You have astrong interest in radar signal processing, machine learning, or computationalimaging.
- You haveexperience with Python and preferably a deep-learning framework such asPyTorch.
- Knowledgeof digital signal processing, Fourier transforms, estimation theory,optimisation, or array processing is highly valuable.
- Previousexperience with radar, synthetic aperture imaging, computer vision, orcomplex-valued neural networks is an advantage but is not mandatory.
- You arecomfortable working independently on a research-oriented problem and evaluatingalgorithms experimentally.
- You havegood written and spoken English.
Research outcome
The internship will offer hands-on experience at the intersection ofradar sensing, computational imaging, and machine learning. The objective is todevelop and validate new algorithms capable of extracting significantly richerspatial information about pedestrians than conventional radar processing.Subject to the achieved results, the work may provide a basis for a scientificpublication or a follow-up research project.
Type of internship: PhD internship, Master internship
Duration: 1 year
Required educational background: Computer Science, Physics, Electrotechnics/Electrical Engineering
Supervising scientist(s): For further information or for application, please contact Hamed Javadi (< email supprimé pour raison de sécurité >) and Sadat Hossain (< email supprimé pour raison de sécurité >)
The reference code for this position is 2026-INT-339. Mention this reference code in your application.
Imec allowance will be provided for students studying at a non-Belgian university.
Applications should include the following information:
- resume
- motivation
- current study
Incomplete applications will not be considered.
