More Like this

1. Location-aware smart campus security application

   https://doi.org/10.1109/UIC-ATC.2017.8397588  

   Liu, Kaikai ; Warade, Navjot ; Pai, Tejas ; Gupta, Keertikeya ( August 2017 , 2017 IEEE Smart City Innovation (SCI))

   One of the biggest challenges that Universities face today is the safety of its people on campus from crimes like mugging, battery and even shooting in or around the campus area. Using SJSU campus as an example, over 50 alert cases of burglaries, thefts, batteries, sexual assaults and other incidents have been reported in and around the SJSU campus over the last year. We have Bluelight emergency telephones placed all over the campus, in all buildings, elevators and on the campus grounds. These phones can be used to report emergency situations, suspicious activities, request escorts etc. However, there is a huge delay between the occurrence of incidents and the arrival of the policeman at the site. There is a critical need for a system that would allow the authorities to locate victims and respond faster to these incidents. To reduce the delay in reporting incidents and their occurrence time, we have developed a mobile application that will let users send alerts along with their real-time location to the UPD directly from their mobile phones. However, finding the position of a victim in a building is the most important challenge we are facing. Many existing systems do not work in indoor environment, and the state-of-the-art localization systems are either inconvenience to use or inaccurate enough to pin-point user's locations inside the building. In this paper, we propose a fine-grained location-aware smart campus security systems that leverages hybrid localization approaches with minimum deployment cost. Specifically, we effectively combines the Wi-Fi fingerprinting localization approach with the Bluetooth beacon based trilateration approach, and improves the location accuracy to the meter-level with low cost. 

   more » « less
2. Toward Real-Time, Robust Wearable Sensor Fall Detection Using Deep Learning Methods: A Feasibility Study

   https://doi.org/10.3390/app13084988  

   Yhdego, Haben ; Paolini, Christopher ; Audette, Michel ( April 2023 , Applied Sciences)

   Real-time fall detection using a wearable sensor remains a challenging problem due to high gait variability. Furthermore, finding the type of sensor to use and the optimal location of the sensors are also essential factors for real-time fall-detection systems. This work presents real-time fall-detection methods using deep learning models. Early detection of falls, followed by pneumatic protection, is one of the most effective means of ensuring the safety of the elderly. First, we developed and compared different data-segmentation techniques for sliding windows. Next, we implemented various techniques to balance the datasets because collecting fall datasets in the real-time setting has an imbalanced nature. Moreover, we designed a deep learning model that combines a convolution-based feature extractor and deep neural network blocks, the LSTM block, and the transformer encoder block, followed by a position-wise feedforward layer. We found that combining the input sequence with the convolution-learned features of different kernels tends to increase the performance of the fall-detection model. Last, we analyzed that the sensor signals collected by both accelerometer and gyroscope sensors can be leveraged to develop an effective classifier that can accurately detect falls, especially differentiating falls from near-falls. Furthermore, we also used data from sixteen different body parts and compared them to determine the better sensor position for fall-detection methods. We found that the shank is the optimal position for placing our sensors, with an F1 score of 0.97, and this could help other researchers collect high-quality fall datasets.

    

   more » « less
3. Household Water and Food Insecurity Are Positively Associated with Poor Mental and Physical Health among Adults Living with HIV in Western Kenya

   https://doi.org/10.1093/jn/nxab030  

   Miller, Joshua D ; Frongillo, Edward A ; Weke, Elly ; Burger, Rachel ; Wekesa, Pauline ; Sheira, Lila A ; Mocello, A Rain ; Bukusi, Elizabeth A ; Otieno, Phelgona ; Cohen, Craig R ; et al ( March 2021 , The Journal of Nutrition)

   null (Ed.)

   ABSTRACT Background Household food insecurity (FI) and water insecurity (WI) are prevalent public health issues that can co-occur. Few studies have concurrently assessed their associations with health outcomes, particularly among people living with HIV. Objectives We aimed to investigate the associations between FI and WI and how they relate to physical and mental health. Methods Food-insecure adult smallholder farmers living with HIV in western Kenya were recruited to participate in a cluster-randomized controlled trial of a multisectoral agricultural and asset loan intervention. We used baseline data on experiences of FI (using the Household Food Insecurity Access Scale, range: 0–27) and WI (using a modified scale developed for this region, range: 0–51) in the prior month (n = 716). Outcomes included probable depression (using the Hopkins Symptom Checklist), fatigue and diarrhea in the prior month, and overall mental and physical health (using the Medical Outcomes Study HIV Health Survey, range: 0–100). We first assessed Pearson correlations between FI, WI, and sociodemographic characteristics. We then developed 3 regressions for each health outcome (control variables and FI; control variables and WI; control variables, FI, and WI) and compared model fit indexes. Results Correlations between household FI, WI, and wealth were low, meaning they measure distinct constructs. FI and WI were associated with numerous physical and mental health outcomes; accounting for both resource insecurities typically provided the best model fit. For instance, when controlling for FI, each 10-point higher WI score was associated with a 6.42-point lower physical health score (P < 0.001) and 2.92 times greater odds of probable depression (P < 0.001). Conclusions Assessing both FI and WI is important for correctly estimating their relation with health outcomes. Interventions that address food- and water-related issues among persons living with HIV concurrently will likely be more effective at improving health than those addressing a single resource insecurity. This trial was registered at clinicaltrials.gov as NCT02815579. 

   more » « less
4. ViTag: Online WiFi Fine Time Measurements Aided Vision-Motion Identity Association in Multi-person Environments

   https://doi.org/10.1109/SECON55815.2022.9918171  

   Cao, Bryan Bo ; Alali, Abrar ; Liu, Hansi ; Meegan, Nicholas ; Gruteser, Marco ; Dana, Kristin ; Ashok, Ashwin ; Jain, Shubham ( September 2022 , 2022 19th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON))

   In this paper, we present ViTag to associate user identities across multimodal data, particularly those obtained from cameras and smartphones. ViTag associates a sequence of vision tracker generated bounding boxes with Inertial Measurement Unit (IMU) data and Wi-Fi Fine Time Measurements (FTM) from smartphones. We formulate the problem as association by sequence to sequence (seq2seq) translation. In this two-step process, our system first performs cross-modal translation using a multimodal LSTM encoder-decoder network (X-Translator) that translates one modality to another, e.g. reconstructing IMU and FTM readings purely from camera bounding boxes. Second, an association module finds identity matches between camera and phone domains, where the translated modality is then matched with the observed data from the same modality. In contrast to existing works, our proposed approach can associate identities in multi-person scenarios where all users may be performing the same activity. Extensive experiments in real-world indoor and outdoor environments demonstrate that online association on camera and phone data (IMU and FTM) achieves an average Identity Precision Accuracy (IDP) of 88.39% on a 1 to 3 seconds window, outperforming the state-of-the-art Vi-Fi (82.93%). Further study on modalities within the phone domain shows the FTM can improve association performance by 12.56% on average. Finally, results from our sensitivity experiments demonstrate the robustness of ViTag under different noise and environment variations. 

   more » « less
5. Adapting Convolutional Neural Networks for Indoor Localization with Smart Mobile Devices

   Mittal, A. ; Tiku, S. ; Pasricha, S. ( January 2018 , ACM Great Lakes Symposium on VLSI (GLSVLSI))

   Indoor localization is emerging as an important application domain for enhanced navigation (or tracking) of people and assets in indoor locales such as buildings, malls, and underground mines. Most indoor localization solutions proposed in prior work do not deliver good accuracy without expensive infrastructure (and even then, the results may lack consistency). Ambient wireless received signal strength indication (RSSI) based fingerprinting using smart mobile devices is a low-cost approach to the problem. However, creating an accurate ‘fingerprinting-only’ solution remains a challenge. This paper presents a novel approach to transform Wi-Fi signatures into images, to create a scalable fingerprinting framework based on Convolutional Neural Networks (CNNs). Our proposed CNN based indoor localization framework (CNN-LOC) is validated across several indoor environments and shows improvements over the best known prior works, with an average localization error of < 2 meters. 

   more » « less