Publications

Passive Monitoring of Dangerous Driving Behaviors Using mmWave Radar

Published in Pervasive and Mobile Computing

Detecting risky driving has been a significant area of focus in recent years. Nonetheless, devising a practical, effective, and unobtrusive solution remains a complex challenge. Presently available technologies predominantly rely on visual cues or physical proximity, complicating the sensing. With this incentive, we explore the possibility of utilizing mmWave radars exclusively to identify dangerous driving behaviors. Initially, we scrutinize the attributes of unsafe driving and pinpoint distinct patterns in range-doppler readings brought about by nine common risky driving manoeuvres. Subsequently, we create an innovative Fused-CNN model that identifies instances of hazardous driving amidst regular driving and categorizes nine distinct types of dangerous driving actions. After conducting thorough experiments involving seven volunteers driving in real-world settings, we note that our system accurately distinguishes risky driving actions with an average precision of approximately 97% with a deviation of ±2%. To underscore the significance of our approach, we also compare it against established state-of-the-art methods. Read more

Recommended citation: Argha Sen, Avijit Mandal, Prasenjit Karmakar, Anirban Das, & Sandip Chakraborty (2024). Passive Monitoring of Dangerous Driving Behaviors Using mmWave Radar. Pervasive and Mobile Computing, 103, p.101949. https://doi.org/10.1016/j.pmcj.2024.101949. https://www.sciencedirect.com/science/article/abs/pii/S1574119224000750

UniPreCIS: A data preprocessing solution for collocated services on shared IoT.

Published in Future Generation Computer Systems

Next-generation smart city applications, attributed to the power of the Internet of Things (IoT) and Cyber–Physical Systems (CPS), significantly rely on sensing data quality. With an exponential increase in intelligent applications for urban development and enterprises offering sensing-as-a-service these days, it is imperative that a shared sensing infrastructure could thwart the better utilization of resources. However, a shared sensing infrastructure that leverages low-cost sensing devices for a cost-effective solution remains unexplored territory. A significant research effort is still needed to make edge-based data shaping solutions more reliable, feature-rich, and cost-effective while addressing the associated challenges in sharing the sensing infrastructure among multiple collocated services with diverse Quality of Service (QoS) requirements. Towards this, we propose UniPreCIS, a novel edge-based data preprocessing solution that accounts for the inherent characteristics of low-cost ambient sensors and their exhibited measurement dynamics concerning application-specific QoS. UniPreCIS aims to identify and select quality data sources by performing sensor ranking and selection that dynamically adapts to the change in sensor attributes. Finally, multimodal data preprocessing is performed in a unified manner to meet heterogeneous application QoS and, at the same time, reduce the resource consumption footprint for the resource-constrained network edge. We study the effectiveness of UniPreCIS on a real-world testbed deployed on our campus. As observed, the processing time and memory utilization of the stakeholder services have been reduced in the proposed approach while achieving up to 90% accuracy, which is arguably significant compared to state-of-the-art sensing techniques. Read more

Recommended citation: Das, Anirban, Navlika Singh, and Suchetana Chakraborty. "UniPreCIS: A data preprocessing solution for collocated services on shared IoT." Future Generation Computer Systems 153 (2024): 543-557. https://www.sciencedirect.com/science/article/abs/pii/S0167739X22001029

mmDrive: mmWave Sensing for Live Monitoring and On-Device Inference of Dangerous Driving

Published in 2023 IEEE International Conference on Pervasive Computing and Communications (PerCom 2023)

Detecting dangerous driving has been of critical interest for the past few years. However, a practical yet minimally intrusive solution remains challenging as existing technologies heavily rely on visual features or physical proximity. With this motivation, we explore the feasibility of purely using mmWave radars to detect dangerous driving behaviors. We first study characteristics of dangerous driving and find some unique patterns of range-doppler caused by 9 typical dangerous driving actions. We then develop a novel Fused-CNN model to detect dangerous driving instances from regular driving and classify 9 different dangerous driving actions. Through extensive experiments with 5 volunteer drivers in real driving environments, we observe that our system can distinguish dangerous driving actions with an average accuracy of > 95%. We also compare our models with existing state-of-the-art baselines to establish their significance. Read more

Recommended citation: Sen, Argha, Avijit Mandal, Prasenjit Karmakar, Anirban Das, and Sandip Chakraborty. "mmDrive: mmWave Sensing for Live Monitoring and On-Device Inference of Dangerous Driving." arXiv preprint arXiv:2301.08188 (2023). https://arxiv.org/pdf/2301.08188.pdf

Where do all my smart home data go? Context-aware data generation and forwarding for edge-based microservices over shared IoT infrastructure

Published in Future Generation Computer Systems

With the explosion of the Internet of Things (IoT) devices, the advent of the edge computing paradigm, and the rise of intelligent applications for smart infrastructure surveillance, in-network data management is gaining a lot of research attention these days. The challenge lies in accommodating multiple application microservices with varying Quality of Service (QoS) requirements to share the sensing infrastructure for better resource utilization. In this work, we propose a novel data collection framework, CaDGen (Context-aware Data Generation) for such a shared IoT infrastructure that enables integrated data filtration and forwarding towards minimizing the resource consumption footprint for the IoT infrastructure. The proposed filtration mechanism utilizes the contextual information associated with the running application for determining the relevance of the data. Whereas the proposed forwarding policy aims to satisfy the diverse QoS requirements for the running applications by selecting the suitable next-hop forwarder based on the microservices distribution across different edge devices. A thorough performance evaluation of CaDGen through a testbed implementation as well as a simulation study for diverse setups reveals promising results concerning network resource utilization, scalability, energy conservation, and distribution of computation for optimal service provisioning. It is observed that the CaDGen can achieve nearly 35% reduction in the generated data for a moderately dynamic scenario without compromising on the data quality. Read more

Recommended citation: Das, Anirban, Sandip Chakraborty, and Suchetana Chakraborty. Where do all my smart home data go? Context-aware data generation and forwarding for edge-based microservices over shared IoT infrastructure. Future Generation Computer Systems, Volume 134, 2022, Pages 204-218, ISSN 0167-739X, https://doi.org/10.1016/j.future.2022.03.027. https://www.sciencedirect.com/science/article/abs/pii/S0167739X22001029

Continuous Multi-user Activity Tracking via Room-Scale mmWave Sensing

Published in 2023 15th International Conference on COMmunication Systems & NETworkS (COMSNETS)

Continuous detection of human activities and presence is essential for developing a pervasive interactive smart space. Existing literature lacks robust wireless sensing mechanisms capable of continuously monitoring multiple users’ activities without prior knowledge of the environment. Developing such a mechanism requires simultaneous localization and tracking of multiple subjects. In addition, it requires identifying their activities at various scales, some being macro-scale activities like walking, squats, etc., while others are micro-scale activities like typing or sitting etc. In this paper, we develop a logistics system called Mars using Commercial off-the-shelf (COTS) Millimeter Wave (mmWave) radar, which employs an intelligent model to sense both macro and micro activities. In addition, it uses a dynamic spatial time-sharing approach to sense different subjects simultaneously. A thorough evaluation of MSARS shows that they can initiate activities an accuracy of > 93% and an average response time of ≈ 2 sec, with 5 subjects and 19 different activities. Read more

Recommended citation: Argha Sen, Anirban Das, Swadhin Pradhan, and Sandip Chakraborty (2024). Continuous Multi-user Activity Tracking via Room-Scale mmWave Sensing. In 23rd ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN 2024) https://dl.acm.org/doi/10.1109/IPSN61024.2024.00018

mmAssist: Passive Monitoring of Driver’s Attentiveness Using mmWave Sensors

Published in 2023 15th International Conference on COMmunication Systems & NETworkS (COMSNETS)

Continuous monitoring of driver attentiveness inside a car has been of significant importance for quite some time. However, the state-of-the-art techniques are primarily inclined toward image-based data, which is invasive and, therefore, could pose challenges in the pervasive adoption of such a system. This work proposes a novel approach for continuous driver attentiveness monitoring, leveraging millimeter Wave (mmWave) sensing to address that. The sensing infrastructure is compact, lightweight, and bears the exclusive potential to be adopted in a pervasive manner due to the continuously increasing popularity of mmWave hardware with 5G technology. We study the driver’s attention as a multi-class problem and address that using Range Doppler information from an mmWave radar. We evaluate the proposed methodologies in a lab and a real-world driving scenario. Within the lab-based setup, we achieved an accuracy of 88%, whereas, in the real-world system, we could achieve an accuracy of up to 79% while monitoring the driver’s activities associated with driving attentiveness. Read more

Recommended citation: Sen, Argha, Anirban Das, Prasenjit Karmakar, and Sandip Chakraborty. "mmAssist: Passive Monitoring of Drivers Attentiveness Using mmWave Sensors." In 2023 15th International Conference on COMmunication Systems & NETworkS (COMSNETS), pp. 545-553. IEEE, 2023. https://ieeexplore.ieee.org/abstract/document/10041297

Leveraging ambient sensing for the estimation of curiosity-driven human crowd

Published in 2022 IEEE International Systems Conference (SysCon), 2022,

Identification and characterization of human crowd formulation have been a topic of immense interest in recent times due to its applicability in a wide range of smart-city applications covering infrastructure automation to targeted advertising. The core idea is to extract the dynamics and associated behavioural patterns of mass gatherings within an environment through a continuous remote monitoring of the crowd. In general, the existing approaches heavily rely on computer vision and image processing based algorithmic tools and techniques to address this problem or mandate the crowd entities to carry a smartphone with them. However, considering the ubiquitous design goals of futuristic smart applications, camera and smartphone driven active sensing is not suitable to honour users’ right to privacy by requiring an active user participation. In this work, we introduce a novel approach towards measuring the spatiotemporal significance of an object in terms of the curious crowd it has attracted over the others. The proposed approach utilizes a set of passive sensors and Wireless signal properties for the necessary estimation. We validate the idea using a room-scale testbed with rigorous experimentation in a real-world scenario. The low cost solution has minimal invasive footprints towards privacy and is capable to reach beyond 90% of accuracy for this measurement. Read more

Recommended citation: A. Das, K. Narayan and S. Chakraborty, "Leveraging ambient sensing for the estimation of curiosity-driven human crowd," 2022 IEEE International Systems Conference (SysCon), 2022, pp. 1-8, doi: 10.1109/SysCon53536.2022.9773844. https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9773844

Experience: Developing a Testbed for Ambient Sensing and in-Network Data Processing

Published in 2022 14th International Conference on COMmunication Systems & NETworkS (COMSNETS)

With ubiquitous adaptation of IoT infrastructure, the smart applications these days are not confined within a single realm. Be it the boundary of a home environment or a city-scale infrastructure, multiple smart applications and associated services tend to serve the users by sharing a common sensing platform. Although the very purpose of this shared infrastructure motivates from the reduced installation and maintenance expenses, it offers a great deal in enabling faster cross-domain inference too. For instance, the same data reflecting room temperature and CO 2 density can be taken into consideration by occupancy estimation service to feed the HVAC system, surveillance system as well as smart resource management system concurrently. Towards faster decision-making and automation it is imperative that the sensory data be pre-processed at the network edge. Now, the existing simulation platforms cannot account for the real-world dynamics affecting the performance of such edge-based sensing and data processing solutions for a shared IoT infrastructure. This work aims to document the practical lessons learnt while developing a testbed to evaluate the performance of such edge-based solutions to support ambient living. Along with the detailed description of system architecture, requirement analysis, design and development arrangement, we also provide a use case demonstration to unfold the inherent challenges and potential solution ideas to setup and conduct experiments on such a testbed environment. Read more

Recommended citation: Das, Anirban, and Suchetana Chakraborty. "Experience: Developing a Testbed for Ambient Sensing and in-Network Data Processing." In 2022 14th International Conference on COMmunication Systems & NETworkS (COMSNETS), pp. 691-699. IEEE, 2022. https://ieeexplore.ieee.org/abstract/document/9668593

A study on real-time edge computed occupancy estimation in an indoor environment

Published in 2022 14th International Conference on COMmunication Systems & NETworkS (COMSNETS)

Sensing the presence of occupants and estimating the occupancy level in an indoor environment are the fundamental requirements for various applications performing remote monitoring, home automation and optimal resource planning. Data generated from a set of passive heterogeneous sensors deployed for this purpose are multimodal and streaming in nature. This work aims to formulate the human occupancy estimation in an indoor environment as a multi-class problem and proposes a edge-based data management framework for human occupancy estimation. The proposed framework is low-cost and light-weight in addition to being capable of performing real-time inference. Also testbed experimentation results is provided to justify the performance of the proposed scheme. Read more

Recommended citation: Das, Anirban, Rohan Gupta, and Suchetana Chakraborty. "A study on real-time edge computed occupancy estimation in an indoor environment." In 2020 International Conference on COMmunication Systems & NETworkS (COMSNETS), pp. 527-530. IEEE, 2020. https://ieeexplore.ieee.org/abstract/document/9027463

Assistant Professor, UPES

Publications

Assistant Professor,
UPES