Smart Cities Research
Indoor Air Quality
Poor indoor air quality can have a number of negative health consequences if left undiscovered. Researchers at MSU are working to enhance the capabilities for consumers to monitor the air quality within their homes using connected air quality sensors.
Occupant Detection & Indoor Mapping
Technologies developed by MSU Faculty help Smart Homes map the interior of a space allowing systems to be custom tailored and safer.
- Self-improving indoor localization by profiling outdoor movement on smartphones
- iFrame: Dynamic indoor map construction through automatic mobile sensing
- Cooperation among smartphones to improve indoor position information
- EyeLoc: Smartphone Vision Enabled Plug-n-play Indoor Localization in Large Shopping Malls
- Typical occupancy profiles and-behaviors in residential buildings in the United States
- NestDNN: Resource-aware multi-tenant on-device deep learning for continuous mobile vision
- Interaction effects of building technology and resident behavior on energy consumption in residential buildings
Smart Utilities
Smart Home technologies allow consumers to create a tailored ecosystem designed around their comfort and convenience. MSU faculty are working on developing predictive technologies to help utilities meet power demands.
- Development and analysis of residential change-point models from smart meter data
- Thermal comfort evaluation for mechanically conditioned buildings using response surfaces in an uncertainty analysis framework
- City-scale energy modeling to assess impacts of extreme heat on electricity consumption and production using WRF-UCM modeling with bias correction
- Effect of technology-enabled time-of-use energy pricing on thermal comfort and energy use in mechanically-conditioned residential buildings in cooling dominated climates
- Effective Features to Predict Residential Energy Consumption Using Machine Learning
- Interaction effects of building technology and resident behavior on energy consumption in residential buildings
- Human factors of high performance multifamily housing
- Modeling urban energy dynamics under clustered urban heat island effect with local-weather extended distributed adjacency blocks
Cyber Security
High-assurance systems and cybersecurity at different layers of communications. With more and more of our devices now connected to the internet, our faculty researchers at MSU are working to ensure that our IoT networked devices are safe and secure from outside attacks.
- Detecting wormhole attacks in a 3d wireless network
- Protecting distributed multiagent systems from cyberattack
- Building the next generation of IoT cybersecurity using a cognitive protection system
- Counteracting jamming attackers in wireless networks
- SecWIR: securing smart home IoT communications via Wi-Fi routers with embedded intelligence
- Access control with delegation for smart home applications
- Secure wireless monitoring and control systems for smart grid and smart home
- RELAX: a language to address uncertainty in self-adaptive systems requirement
- A Lightweight Block Validation Method for Resource-Constrained IoT Devices in Blockchain-Based Applications
- CapChain: A Privacy Preserving Access Control Framework Based on Blockchain for Pervasive Environments
Biometrics
Monitoring the identity, health, well-being and cognitive state of a vehicle’s occupants.
- Learning scene-specific pedestrian detectors without real data
- Extended-depth-of-field iris recognition using unrestored wavefront-coded imagery
- Matching highly non-ideal ocular images: An information fusion approach
- Coupled marginal fisher analysis for low-resolution face recognition
- Image super-resolution via deep recursive residual network
- Face alignment across large poses: A 3d solution
- Disentangled representation learning gan for pose-invariant face recognition
- Video-based face recognition using adaptive hidden markov models
- Demographic estimation from face images: Human vs. machine performance
- Generalizing Fingerprint Spoof Detector: Learning a One-Class Classifier
- Low-Quality Video Face Recognition: Multi-mode Aggregation Recurrent Network (MARN) c
- End-to-End Latent Fingerprint Search
- Fingerprint Spoof Buster: Use of Minutiae-centered Patches