Sanat Sarangi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Sanat Sarangi has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Subrat Kar
Department of Electrical Engineering, IIT Delhi, New Delhi, India
4 shared publications Subrat Kar
Bharti School of Telecommunication Technology And Management, IIT Delhi, Hauz Khas, New Delhi - 110016, INDIA
3 shared publicationsA. K. Ray
Department of Geology Presidency College Calcutta India
1 shared publication- Akshat Bisht
Department of Electrical, Computer and Software Engineering, University of Auckland, Auckland, New Zealand
1 shared publication - Anand Prakash Ruhil
NDRI, Karnal, Haryana, India-132001
1 shared publication - Bodhibrata Mukhopadhyay
Department of Electronics and Communication, Indian Institute of Technology Roorkee, Roorkee, Uttarakhand, India
1 shared publication
Latest Publications
- 21 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Letters
- Vol. 10 (3) , 1-4
Quality Assessment and Valuation of Made-tea Using ROI Segmentation and Spectral–TDS Fusion
Accurate and consistent grading is important for quality control, but manual tasting is subjective and hard to scale. We present a compact, fully automated system that predicts a two-digit valuation grade: the first digit is Body (liquor strength) and the second is Zing (briskness), each scored 0–5. It combines spectral imaging with a total dissolved solids (TDS) reading to capture both physical and chemical cues. We improve data quality by processing images in stages: segmenting the sample at a reference wavelength using adaptive K-means, applying a circular mask, running a second pass, and removing low-confidence boundary pixels. To capture clean local signals, we introduce an automatic non-overlapping bounding-box method for particulate made-tea valuation with spectral imaging. We fuse per-box spectra with TDS and train machine learning models; on a test set, a multilayer perceptron reaches 95.2% accuracy and a support vector machine performs similarly. Compared to fixed-region baselines, signal-to-noise ratio rises by 12.4 dB, within-class variance falls by 18.7%, background contamination drops from 14.6% to 0.9%, and rescan repeatability improves ($r=0.97$ versus 0.91; all $p< 0.01$). The system runs in 402 ms per sample on a desktop-class CPU, suiting factory use. Strong region of interest isolation and low-noise features boost classifier performance, enabling accurate, repeatable, and scalable grading.
- 21 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Letters
- Vol. 10 (3) , 1-4
Quality Assessment and Valuation of Made-tea Using ROI Segmentation and Spectral–TDS Fusion
Accurate and consistent grading is important for quality control, but manual tasting is subjective and hard to scale. We present a compact, fully automated system that predicts a two-digit valuation grade: the first digit is Body (liquor strength) and the second is Zing (briskness), each scored 0–5. It combines spectral imaging with a total dissolved solids (TDS) reading to capture both physical and chemical cues. We improve data quality by processing images in stages: segmenting the sample at a reference wavelength using adaptive K-means, applying a circular mask, running a second pass, and removing low-confidence boundary pixels. To capture clean local signals, we introduce an automatic non-overlapping bounding-box method for particulate made-tea valuation with spectral imaging. We fuse per-box spectra with TDS and train machine learning models; on a test set, a multilayer perceptron reaches 95.2% accuracy and a support vector machine performs similarly. Compared to fixed-region baselines, signal-to-noise ratio rises by 12.4 dB, within-class variance falls by 18.7%, background contamination drops from 14.6% to 0.9%, and rescan repeatability improves ($r=0.97$ versus 0.91; all $p< 0.01$). The system runs in 402 ms per sample on a desktop-class CPU, suiting factory use. Strong region of interest isolation and low-noise features boost classifier performance, enabling accurate, repeatable, and scalable grading.
- 8 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 38-45
Assessment of Emissions with Carbon-smart Farming Practices and Participatory Sensing in Rice
Agriculture sector is a significant contributor to greenhouse gas (GHG) emissions especially in the conventional rice ecosystem with carbon-insensitive practices. In this study, we assess the carbon footprint of selected farms based on GHG emissions and carbon sequestration from recommended agricultural practices with a human participatory sensing approach. A set of ten selected farmers was split into two groups and asked to follow carbon-smart crop protocols (CSCP) called CSCP-1 and CSCP-2. With the digitally captured record of operations, process modelling was used to simulate the CSCP scenarios followed on the ground, and a classification model was developed to estimate the Nitrogen uptake to improve fertilizer utilization for farmers. For various potential scenarios involving variation in irrigation and fertilizer application, impact on GHG emissions and SOC dynamics was evaluated. Results showed that CSCP-1 farmers emitted more GHGs when compared to CSCP-2 farmers while they also sequestrated more carbon in comparison with no significant difference in Net GWP (Global Warming Potential). CSCP farmers with both flood irrigation and furrow irrigation sequestrated more carbon than farmers who would follow conventional practices. Net GWP of CSCP farmers was significantly lower than conventional farmers indicating carbon-smart practices can indeed make a significant difference in sustainability initiatives.
- 8 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 38-45
Assessment of Emissions with Carbon-smart Farming Practices and Participatory Sensing in Rice
Agriculture sector is a significant contributor to greenhouse gas (GHG) emissions especially in the conventional rice ecosystem with carbon-insensitive practices. In this study, we assess the carbon footprint of selected farms based on GHG emissions and carbon sequestration from recommended agricultural practices with a human participatory sensing approach. A set of ten selected farmers was split into two groups and asked to follow carbon-smart crop protocols (CSCP) called CSCP-1 and CSCP-2. With the digitally captured record of operations, process modelling was used to simulate the CSCP scenarios followed on the ground, and a classification model was developed to estimate the Nitrogen uptake to improve fertilizer utilization for farmers. For various potential scenarios involving variation in irrigation and fertilizer application, impact on GHG emissions and SOC dynamics was evaluated. Results showed that CSCP-1 farmers emitted more GHGs when compared to CSCP-2 farmers while they also sequestrated more carbon in comparison with no significant difference in Net GWP (Global Warming Potential). CSCP farmers with both flood irrigation and furrow irrigation sequestrated more carbon than farmers who would follow conventional practices. Net GWP of CSCP farmers was significantly lower than conventional farmers indicating carbon-smart practices can indeed make a significant difference in sustainability initiatives.
- 1 February 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Performance evaluation of localization techniques in wireless sensor networks using RSSI and LQI
Low-cost precise localization is crucial for wireless sensor networks. RSSI based localization is cost effective when compared to TOA, AOA, TDOA, ultrasonic and acoustic localization as it does not require any extra hardware, power or bandwidth. The radio of sensor nodes provides information about both the RSSI and LQI of a received radio signal. Localization error can be decreased by simultaneously observing both RSSI and LQI. We propose two novel techniques for localizing a target node using RSSI+LQI. They are Recursive Bayesian-RSSI-LQI (RB-RSSI-LQI) and Maximum a posteriori-RSSI-LQI (MAP-RSSI-LQI). A comparison between these techniques is done with the existing Mean-RSSI technique. We show that MAP-RSSI-LQI gives the best results in terms of localization error and computational complexity. The root mean square error of the RB-RSSI-LQI is 53.35% less than Mean-RSSI in case of stationary target node. The root mean square error of MAP-RSSI-LQI is 52.25% and 58.88% less than Mean-RSSI in case of stationary and mobile target nodes. A combination of simulation and experimental evaluation is used to develop and validate the proposed techniques.
- 1 February 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Performance evaluation of localization techniques in wireless sensor networks using RSSI and LQI
Low-cost precise localization is crucial for wireless sensor networks. RSSI based localization is cost effective when compared to TOA, AOA, TDOA, ultrasonic and acoustic localization as it does not require any extra hardware, power or bandwidth. The radio of sensor nodes provides information about both the RSSI and LQI of a received radio signal. Localization error can be decreased by simultaneously observing both RSSI and LQI. We propose two novel techniques for localizing a target node using RSSI+LQI. They are Recursive Bayesian-RSSI-LQI (RB-RSSI-LQI) and Maximum a posteriori-RSSI-LQI (MAP-RSSI-LQI). A comparison between these techniques is done with the existing Mean-RSSI technique. We show that MAP-RSSI-LQI gives the best results in terms of localization error and computational complexity. The root mean square error of the RB-RSSI-LQI is 53.35% less than Mean-RSSI in case of stationary target node. The root mean square error of MAP-RSSI-LQI is 52.25% and 58.88% less than Mean-RSSI in case of stationary and mobile target nodes. A combination of simulation and experimental evaluation is used to develop and validate the proposed techniques.
- 1 December 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 37 (21531676) , 1-6
Development of a wireless sensor network for animal management: Experiences with Moosense
A wireless sensor network based precision animal management system developed by us called Moosense is presented. Moosense is deployed in the Livestock Research Centre (LRC) of National Dairy Research Institute (NDRI) Karnal. It is spread across a cluster of sensor networks where each cluster has a set of specialized systems, installed on the ground or mounted on the animals, which are responsible for live-monitoring of a variety of cattle-health parameters. Each cluster has a gateway to which the devices are connected either directly or through a set of relay nodes. The gateways are connected to a central station where all sensor data of LRC is archived and analysed. Through Moosense, we share our experiences with the development of structurally and functionally complex sensor networks for challenging environments such as in animal husbandry. Moosense is one of the first operational multi-objective wireless sensor networks in India being actively used for precision dairy research.
- 1 December 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 37 (21531676) , 1-6
Development of a wireless sensor network for animal management: Experiences with Moosense
A wireless sensor network based precision animal management system developed by us called Moosense is presented. Moosense is deployed in the Livestock Research Centre (LRC) of National Dairy Research Institute (NDRI) Karnal. It is spread across a cluster of sensor networks where each cluster has a set of specialized systems, installed on the ground or mounted on the animals, which are responsible for live-monitoring of a variety of cattle-health parameters. Each cluster has a gateway to which the devices are connected either directly or through a set of relay nodes. The gateways are connected to a central station where all sensor data of LRC is archived and analysed. Through Moosense, we share our experiences with the development of structurally and functionally complex sensor networks for challenging environments such as in animal husbandry. Moosense is one of the first operational multi-objective wireless sensor networks in India being actively used for precision dairy research.
- 1 August 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1058-1063
On implementing graded precision localization with sensor networks in indoor spaces
Graded precision localization is the process by which mobile heterogeneous sensor nodes localize themselves to varying levels of precision, even though they co-exist and operate in a common framework. Some nodes may only need to compute their approximate location at infrequent intervals. Others may need to be continuously and precisely localized. Since GPS is not available indoors, we need to provision a generic ubiquitous localization infrastructure that services specific localization requirements of each mobile sensor node on a best-available-information basis. We assume that each mobile node has an associated user module that gets localized. For implementing graded-precision localization in indoor spaces, we propose a modular architecture for the user module, site-specific deployment strategies, and a framework for evaluating localization performance. The user module architecture is realized using an Android phone and a Cricket sensor node. We report our experimental and simulation results for various operational scenarios.
- 1 August 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1058-1063
On implementing graded precision localization with sensor networks in indoor spaces
Graded precision localization is the process by which mobile heterogeneous sensor nodes localize themselves to varying levels of precision, even though they co-exist and operate in a common framework. Some nodes may only need to compute their approximate location at infrequent intervals. Others may need to be continuously and precisely localized. Since GPS is not available indoors, we need to provision a generic ubiquitous localization infrastructure that services specific localization requirements of each mobile sensor node on a best-available-information basis. We assume that each mobile node has an associated user module that gets localized. For implementing graded-precision localization in indoor spaces, we propose a modular architecture for the user module, site-specific deployment strategies, and a framework for evaluating localization performance. The user module architecture is realized using an Android phone and a Cricket sensor node. We report our experimental and simulation results for various operational scenarios.
- 1 August 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1035-1040
A scriptable rapid application deployment framework for sensor networks
Wireless sensor networks are finding application in heterogeneous scenarios with varied deployment constraints and consumer requirements. Effective instrumentation, visualization, archival, analysis, and third-party notification are the cornerstones of any sensor network deployment. Unfortunately, the manifestation of these generalized needs can differ widely between deployments leading to creation of multiple specialized applications. We present RAPIDSNAP-a flexible, generic and scriptable rapid sensor network application deployment framework that can be quickly customized to realize end-user centric sensor network deployments. Algorithms for routing, clustering and localization have been implemented on sensor nodes and integrated with RAPIDSNAP. Along with standard components for control and visualization, RAPIDSNAP allows each deployment specific extension to be captured in a JavaScript based script called a Rapidapp. RAPIDSNAP acts as an execution platform for Rapidapps and an abstraction layer for them to transparently communicate with nodes in the network.
- 1 August 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1035-1040
A scriptable rapid application deployment framework for sensor networks
Wireless sensor networks are finding application in heterogeneous scenarios with varied deployment constraints and consumer requirements. Effective instrumentation, visualization, archival, analysis, and third-party notification are the cornerstones of any sensor network deployment. Unfortunately, the manifestation of these generalized needs can differ widely between deployments leading to creation of multiple specialized applications. We present RAPIDSNAP-a flexible, generic and scriptable rapid sensor network application deployment framework that can be quickly customized to realize end-user centric sensor network deployments. Algorithms for routing, clustering and localization have been implemented on sensor nodes and integrated with RAPIDSNAP. Along with standard components for control and visualization, RAPIDSNAP allows each deployment specific extension to be captured in a JavaScript based script called a Rapidapp. RAPIDSNAP acts as an execution platform for Rapidapps and an abstraction layer for them to transparently communicate with nodes in the network.
- 1 January 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Wireless Sensor Knowledge Archive
Wisekar - the WIreless SEnsor Knowledge ARchive is presented. It is a centralized web repository for collection, management and distribution of events from sensor-enabled devices towards realization of the Internet of Things (IOT). Wisekar offers flexible multimodal event contribution methods adapted to a variety of user skill and requirement levels. Along with rich visualization capabilities, it is versatile in terms of its ability to extend and decorate events with XML based meta-data and images. This allows enforcement of rigorous standards-compliant formats like the Open Geospatial Consortium (OGC) defined SensorML and TransducerML.
- 1 January 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Wireless Sensor Knowledge Archive
Wisekar - the WIreless SEnsor Knowledge ARchive is presented. It is a centralized web repository for collection, management and distribution of events from sensor-enabled devices towards realization of the Internet of Things (IOT). Wisekar offers flexible multimodal event contribution methods adapted to a variety of user skill and requirement levels. Along with rich visualization capabilities, it is versatile in terms of its ability to extend and decorate events with XML based meta-data and images. This allows enforcement of rigorous standards-compliant formats like the Open Geospatial Consortium (OGC) defined SensorML and TransducerML.
- 1 December 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Genetic algorithm based mobility aware clustering for energy efficient routing in wireless sensor networks
Clustering in wireless sensor networks improves lifetime of the network by load distribution, data aggregation and use of clusterheads to efficiently route messages to the central gateway. Parameters governing creation of optimal clusters are often numerous and domain-specific. The challenges increase with mobile nodes due to faster temporal change in one or more parameter values like speed and distance to the gateway. We propose a Genetic Algorithm based algorithm GAROUTE that performs energy-efficient mobility-aware clustering for wireless sensor networks. GAROUTE uses mobility awareness of each node to adaptively create stable clusters for energy-efficient routing of information.
- 1 December 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Genetic algorithm based mobility aware clustering for energy efficient routing in wireless sensor networks
Clustering in wireless sensor networks improves lifetime of the network by load distribution, data aggregation and use of clusterheads to efficiently route messages to the central gateway. Parameters governing creation of optimal clusters are often numerous and domain-specific. The challenges increase with mobile nodes due to faster temporal change in one or more parameter values like speed and distance to the gateway. We propose a Genetic Algorithm based algorithm GAROUTE that performs energy-efficient mobility-aware clustering for wireless sensor networks. GAROUTE uses mobility awareness of each node to adaptively create stable clusters for energy-efficient routing of information.
- Open Access
- 9 July 2010
- Published by Academy and Industry Research Collaboration Center (AIRCC) in International journal of Computer Networks & Communications
- Vol. 2 (4) , 150-159
Performance Analysis Of An Improved Graded Precision Localization Algorithm For Wireless Sensor Networks
In this paper an improved version of the graded precision localization algorithm GRADELOC, called IGRADELOC is proposed. The performance of GRADELOC is dependent on the regions formed by the overlapping radio ranges of the nodes of the underlying sensor network. A different region pattern could significantly alter the nature and precision of localization. In IGRADELOC, two improvements are suggested. Firstly, modifications are proposed in the radio range of the fixed-grid nodes, keeping in mind the actual radio range of commonly available nodes, to allow for routing through them. Routing is not addressed by GRADELOC, but is of prime importance to the deployment of any adhoc network, especially sensor networks. A theoretical model expressing the radio range in terms of the cell dimensions of the grid infrastructure is proposed, to help in carrying out a deployment plan which achieves the desirable precision of coarse-grained localization. Secondly, in GRADELOC it is observed that fine-grained localization does not achieve significant performance benefits over coarse-grained localization. In IGRADELOC, this factor is addressed with the introduction of a parameter that could be used to improve and fine-tune the precision of fine-grained localization.Comment: 10 pages, 10 figure
- 9 July 2010
- Published by Academy and Industry Research Collaboration Center (AIRCC) in International journal of Computer Networks & Communications
- Vol. 2 (4) , 150-159
Open AccessPerformance Analysis Of An Improved Graded Precision Localization Algorithm For Wireless Sensor Networks
In this paper an improved version of the graded precision localization algorithm GRADELOC, called IGRADELOC is proposed. The performance of GRADELOC is dependent on the regions formed by the overlapping radio ranges of the nodes of the underlying sensor network. A different region pattern could significantly alter the nature and precision of localization. In IGRADELOC, two improvements are suggested. Firstly, modifications are proposed in the radio range of the fixed-grid nodes, keeping in mind the actual radio range of commonly available nodes, to allow for routing through them. Routing is not addressed by GRADELOC, but is of prime importance to the deployment of any adhoc network, especially sensor networks. A theoretical model expressing the radio range in terms of the cell dimensions of the grid infrastructure is proposed, to help in carrying out a deployment plan which achieves the desirable precision of coarse-grained localization. Secondly, in GRADELOC it is observed that fine-grained localization does not achieve significant performance benefits over coarse-grained localization. In IGRADELOC, this factor is addressed with the introduction of a parameter that could be used to improve and fine-tune the precision of fine-grained localization.Comment: 10 pages, 10 figure