[month] [year]

IEEE Sensors Journal

The following papers have been accepted in  IEEE Sensors Journal: 

 

  • Deployable Edge AI Solution for Posture Classification using mmWave Radar and Low Computation Machine Learning Model by Dr. Abhishek Srivastava and his student Pranjal Mahajan, MS ECE 2nd Year; Devansh Chaudhary, B.Tech ECE 4th Year, Aligarh Muslim University (AMU); Yash Pratap Singh: B.Tech ECE 4th Year, Aligarh Muslim University (AMU); Aham Gupta: B.Tech ECE 4th Year, Aligarh Muslim University (AMU) and Dr. Mohd Wajid, Associate professor, ECE, Aligarh Muslim University (AMU).  Here is the summary of their research work: 

Identifying correct human postures is crucial in areas like patient care in hospitals. However, the traditional vision-based methods widely used for this purpose raise privacy concerns for the subject, and the other wearable sensor-based approaches are impractical for real-world scenarios. In this paper, we propose a contactless, privacy-conscious, and memory efficient posture classification system based on millimeter wave (mmWave) radar. This system utilizes 3D point-cloud data captured using Texas Instrument’s IWR1843BOOST FMCW radar module to classify the posture of the subject. Two types of datasets are extracted from this radar data: 1. Image dataset derived from the isometric view of the point-cloud data, and (2) spatial coordinates dataset also extracted from the point-cloud data. A low-computational Machine Learning model (TinyML) is employed on the datasets for efficient implementation on embedded hardware, Raspberry Pi 3 B+. The proposed model’s parameters were quantized to 8 bits (int8) which accurately classify four postures i.e., standing, sitting, lying, and bending, with an accuracy of 98.97\% for the image data. However, to make it more computationally efficient, the int8 quantized TinyML model was trained on the spatial coordinates dataset giving an accuracy of 96.12\. This highlights the efficiency and effectiveness of our proposed lightweight model that can be deployed on edge devices for real-world applications.

  • GoldAid: An Integrated Safety System Leveraging IoT and Advanced Algorithms for Rapid Medical Aid in the Golden Hour – Dr. Abhishek Srivastava and his students Anushka Tripathi, MS 2nd year; Arpit Sahni, MS 3rd year ECE , graduated in January 2024; Srikar Somanchi, MS 1st year ECE; Sresthavadhani Mantha, B.Tech Honors, ECE, graduated in 2023 and Mohammed Hammad Khan, MS 1st year, ECE. Here is the summary of the research work as explained by the authors: 

Accidents in industrial environments endanger the lives of workers facing challenging conditions. The major reason for fatal outcomes in such accidents arises from delays in reporting and providing timely medical assistance within the crucial first sixty minutes (golden hour) after the accidents. In this work, we present a safety system GoldAid that is specifically designed for industries to ensure quick incident reporting and expedite medical assistance within the critical golden hour. The proposed GoldAid system presents a significant integration of multiple sensors to provide fall detection, geo-tracking, long-range wireless communication, hazardous gas detection, vital monitoring & SOS functionalities within an Internet-of-Things framework. To cover all possible industrial fall scenarios, this work also presents a Convolutional Neural Network (CNN) model and an acceleration-threshold-based method for low-power edge devices. Moreover, to achieve minimum data loss and low latency in real-time incident reporting in a large industrial setup, a strategic model for placing multiple communication gateways is also proposed in this research. A prototype of the proposed GoldAid system is developed, and experimental results are also presented. Measurement results show that the proposed fall-detection models achieve >98.44% accuracy, and the system achieves a latency of <1 s with bit-error-rate (BER) of <3×10-4 over a communication range of 300 m while reporting a fall incident in an actual thermal power plant.

 

 

June 2024