Dulal Kar - Academia.edu (original) (raw)
Papers by Dulal Kar
Advances in logistics, operations, and management science book series, Apr 21, 2023
Blockchain technology offers several beneficial features that make it well-suited for supply chai... more Blockchain technology offers several beneficial features that make it well-suited for supply chain management (SCM). Particularly, it can improve transparency, traceability, security, efficiency, and accountability in supply chains (SC). This chapter focuses on how blockchain technology can completely transform SC processes due to its inherent features. By implementing blockchain, businesses can cut costs, improve efficiency, and ensure accountability. The chapter explains the basics of blockchain technology and its potential benefits for SCM. The various beneficial features of blockchain technology such as trust, immutability, reliability, and security are discussed and highlighted in the chapter. The chapter showcases real-world uses of blockchain technology in SCM to depict the possible benefits. Benefits and challenges of incorporating blockchain into SC processes are also discussed in the chapter. Finally, this chapter emphasizes how crucial it is for companies to identify the factors that weaken their SC and to ensure that blockchain is applied to address such issues.
2021 International Conference on Computational Science and Computational Intelligence (CSCI), Dec 1, 2021
International Conference on Wireless Networks, 2010
Environmental monitoring is the process of blending ecosystems, mathematics, and computer science... more Environmental monitoring is the process of blending ecosystems, mathematics, and computer science for the purpose of environmental impact assessments. In the past few years, wireless sensor networks have been emerging as a next generation of computer networks. Applications of wireless sensor networks for environmental monitoring are being studied extensively using low-cost, high-availability, and high-performance sensors. This project focuses on measuring, analyzing, and storing sound information for environmental monitoring using well-known LPC-based algorithms and the Crossbow Imote2 sensor platform. The experimental results show potential environment-monitoring applications using wireless sensor networks.
Today mobile phones provide a wide range of applications that make our daily life easy. With popu... more Today mobile phones provide a wide range of applications that make our daily life easy. With popularity, smartphones have become a target for cybercrime where malicious apps are developed to acquire sensitive information or corrupt data. To mitigate this issue and to improve the security in mobile devices, different techniques have been used. These techniques can be broadly classified as static, dynamic and hybrid approaches. In this paper, a static-based model MalDuoNet is proposed to detect Android malwares, which uses a DualNet framework to analyze the features from the API calls. In the MalDuoNet model, one sub-network is focused to learn the features relevant to malicious behavior and the other sub-network is focused to learn the features in general. Thus it enables the model to learn complementary features which in turn helps get richer features for analysis. Then the features from the two sub-networks are combined in the final fused classifier for the final classification. In addition, each of the feature extractors has a separate classifier so that each sub-network can optimize its performance separately. The experimental results demonstrate that the MalDuoNet model outperforms the two baseline models with single network.
Mobile phones have become a target for cybercrime where malicious apps are developed to acquire s... more Mobile phones have become a target for cybercrime where malicious apps are developed to acquire sensitive information or corrupt data. To mitigate this issue and to improve the security in mobile devices, many machine learning methods have been developed to detect and classify Android malware. One problem with the existing methods is that they all assume a large dataset is available to train the learning models. In this paper, a model named few-shot malware classification (FSMC) is proposed to classify Android apps by using only a few training cases for each class. For three small datasets, the experimental results demonstrate that the FSMC model is able to achieve significantly higher accuracy compared to the existing 2-way 1-shot and 2-way 3-shot malware classification methods.
arXiv (Cornell University), Feb 7, 2022
Energy harvesting technologies offer a promising solution to sustainably power an ever-growing nu... more Energy harvesting technologies offer a promising solution to sustainably power an ever-growing number of Internet of Things (IoT) devices. However, due to the weak and transient natures of energy harvesting, IoT devices have to work intermittently rendering conventional routing policies and energy allocation strategies impractical. To this end, this paper, for the very first time, developed a distributed multi-agent reinforcement algorithm known as global actor-critic policy (GAP) to address the problem of routing policy and energy allocation together for the energy harvesting powered IoT system. At the training stage, each IoT device is treated as an agent and one universal model is trained for all agents to save computing resources. At the inference stage, packet delivery rate can be maximized. The experimental results show that the proposed GAP algorithm achieves ∼ 1.28× and ∼ 1.24× data transmission rate than that of the Q-table and ESDSRAA algorithm, respectively.
IGI Global eBooks, 2010
Significant advancements in hardware technology have propelled the existence of Wireless Sensor N... more Significant advancements in hardware technology have propelled the existence of Wireless Sensor Network (WSN). A WSN consists of simple, low cost yet powerful sensors. Each sensor has the ability to sense, process, and communicate data collected from the environment, in which it is deployed. Sensors usually draw energy from a small battery, and thus energy efficiency emerges as the key issue in any WSN. The basic idea of a
IGI Global eBooks, Jan 18, 2011
IGI Global eBooks, May 8, 2013
2022 27th Asia and South Pacific Design Automation Conference (ASP-DAC)
Energy harvesting technologies offer a promising solution to sustainably power an ever-growing nu... more Energy harvesting technologies offer a promising solution to sustainably power an ever-growing number of Internet of Things (IoT) devices. However, due to the weak and transient natures of energy harvesting, IoT devices have to work intermittently rendering conventional routing policies and energy allocation strategies impractical. To this end, this paper, for the very first time, developed a distributed multi-agent reinforcement algorithm known as global actor-critic policy (GAP) to address the problem of routing policy and energy allocation together for the energy harvesting powered IoT system. At the training stage, each IoT device is treated as an agent and one universal model is trained for all agents to save computing resources. At the inference stage, packet delivery rate can be maximized. The experimental results show that the proposed GAP algorithm achieves ∼ 1.28× and ∼ 1.24× data transmission rate than that of the Q-table and ESDSRAA algorithm, respectively.
Coastlines are an important yet volatile part of the environment and frequently change over time.... more Coastlines are an important yet volatile part of the environment and frequently change over time. Studying changes in coastlines can give important insights about the environment, so using efficient means to observe them is important. UAVs are cheap, fast and effective tools in gathering information, and segmentation can be used to detect and record the coastline. This paper proposes the use of CNN models to assist UAVs in detecting coastlines. Specifically, our research focuses on the combination of a neural network model, such as UNet, with transfer learning using a pre-trained model. The resulting model achieves semantic segmentation of coastlines to distinguish water and land. This allows UAVs to better recognize coastlines and reinforce their role in remote sensing image processing. 2. Detection of GPS Spoofing in UAVs Using Deep Learning and Machine Learning REU Participants: Jaron Burns and Dimitris Amiridis Mentor: Dr. Longzhuang Li Graduate Assistant: Rohith Mandala Abstrac...
Abstract:- This paper describes the development of a prototype that takes in an analog National T... more Abstract:- This paper describes the development of a prototype that takes in an analog National Television System Committee video signal generated by a video camera and data acquired by a microcontroller and display them in real-time on a digital panel. An 8051 microcontroller is used to acquire power dissipation by the display panel, room temperature, and camera zoom level. The paper describes the major hardware components and shows how they are interfaced into a functional prototype. Test data results are presented and discussed. Key-Words: data acquisition, overlaying text over video, real-time video 1
2017 International Conference on Computational Science and Computational Intelligence (CSCI), 2017
Growth of information technology led to the increasing need of computing and storage. Cloud servi... more Growth of information technology led to the increasing need of computing and storage. Cloud services is one such technology with high demand and hence requires more computing resources. Cloud data centers consume huge amount of energy and there by emitting carbon dioxide to the environment. This work proposes an approach for energy efficient resource management. Earlier approaches do not focus on the variations of workloads and lack in examining the effect of algorithms on performance. Virtual machine configuration also plays a vital role for reducing energy consumption and resource wastage, but is not given much importance. To address these weaknesses, this work proposes a novel approach to map groups of tasks to customized virtual machine types. Mapping of the tasks is based on task usage patterns\textemdash length, file size, bandwidth etc. Data is clustered in to group of tasks and is mapped to the suitable virtual machine based on the configuration. Virtual machine migration is employed to balance the load by calculating the load using MIPS, RAM and Bandwidth. Complete end-end architecture is proposed in this work with clustering of tasks, allocation of tasks to virtual machines and virtual machine migration techniques. The results of this work show that the energy consumption is decreased compared to the earlier approaches, which uses traditional virtual machine migration techniques.
2017 International Conference on Computational Science and Computational Intelligence (CSCI), 2017
One of the main challenges in cloud computing is limited availability of resources. As the number... more One of the main challenges in cloud computing is limited availability of resources. As the number of requests for cloud services increases, it becomes necessary for the system to balance the load and serve user requests at stipulated times. Load Balancing is a well-known NP-Complete Problem. This work proposes two variants of fusion-based task scheduling algorithm; both the approaches exploit two existing load balancing algorithms—the traditional round-robin algorithm (RRA) and the priority-based genetic algorithm (PGA), to improve the performance of the system in terms of the completion time. The idea of fusion lies in considering the variable amount of user requests to the cloud system. The first variant i.e. fusionbased load-aware resource allocation algorithm (FLA) uses PGA when there is relatively light load and RRA when the system encounters heavy load. The algorithm determines the intensity of the current load on the system, whether it is light or heavy. In the second variant...
There is increasing demand world-wide, from government agencies and the private sector for cuttin... more There is increasing demand world-wide, from government agencies and the private sector for cutting-edge biometric security technology that is difficult to breach but userfriendly at the same time. Some of the older tools, such as fingerprint, retina and iris scanning, and facial recognition software have all been found to have flaws and often viewed negatively because of many cultural and hygienic issues associated with them. Comparatively, mapping veins as a human barcode, a new technology, has many advantages over older technologies. Specifically, reproducing a three-dimensional model of a human vein system is impossible to replicate. Vein map technology is distinctive because of its state-of-the-art sensors are only able to recognize vein patterns if hemoglobin is actively flowing through the person's veins. Additionally, each individual's vein map is unique, even in the case of identical twins. The combinations of these factors provide vein map authentication an edge ove...
2021 RIVF International Conference on Computing and Communication Technologies (RIVF)
A tool for mobile devices is presented to detect a human face in an image and analyze the detecte... more A tool for mobile devices is presented to detect a human face in an image and analyze the detected face for its aesthetical quality. For the purpose of face detection, the tool uses two well-known classifiers: Haar feature-based cascade classifier and Local Binary Patterns (LBP) cascade classifier. Once a face is detected, the tool uses the Active Shape Model (ASM) algorithm to identify facial feature points and obtain raw data accordingly for aesthetical quality analysis. The analysis is carried out on the notion of Golden Ratio by finding ratios on the facial feature points obtained by using the ASM algorithm. To determine its effectiveness and limitations the tool is tested on animal faces, beautiful faces, ordinary faces, and unattractive faces. The statistics of results exposes distinctions between unattractive faces and beautiful/ordinary faces. However, no such distinction exists to separate beautiful faces from ordinary faces.
Energy efficiency is a primary concern in Wireless Sensor Networks (WSN). This is due to the fact... more Energy efficiency is a primary concern in Wireless Sensor Networks (WSN). This is due to the fact that WSNs are powered battery, and hence the life of WSNs becomes limited by the battery life. Efforts to replace depleted batteries are not feasible if WSNs are often deployed with thousands of sensor nodes, possibly in inaccessible, hostile, hazardous, or remote territories. Though WSNs are energy-constrained, adequate level of security is often desired in many applications of WSNs that guarantees data integrity and confidentiality. However, security protocols introduce extra energy overhead to a sensor node due to additional processing and communications associated with the protocols. Thus it is important to seek security protocols and solutions for WSNs that are energyefficient but also effective. Security is commonly implemented through symmetric key cryptography which requires a key exchange mechanism to establish a key between the communicating parties. One well known key exchang...
IGARSS 2020 - 2020 IEEE International Geoscience and Remote Sensing Symposium, 2020
Bathymetry, or the measurement of depth in any body of water, has been an area of research since ... more Bathymetry, or the measurement of depth in any body of water, has been an area of research since man began to venture out onto the open waters. Historically, researching the near-shore surf zone has been a time consuming and expensive process. The tools and methods used to gather data points in the surf zone are either time inefficient, expensive, or both. This is an issue considering how dynamic the surf zone environment can be. It is possible that by the time the surf zone bathymetry measurements have been completed, they are already out of date. This project utilizes unmanned aerial systems (UAS) to gather high-quality video of the near-shore surf zone waves crest. This footage is then processed using particle image velocimetry (PIV), a method for determining the velocity of particles in sequential images. This velocity is then processed using linear-wave theory shallow water approximations for calculating wave celerity from depth, but ran in reverse, to obtain the bathymetry its...
Advances in logistics, operations, and management science book series, Apr 21, 2023
Blockchain technology offers several beneficial features that make it well-suited for supply chai... more Blockchain technology offers several beneficial features that make it well-suited for supply chain management (SCM). Particularly, it can improve transparency, traceability, security, efficiency, and accountability in supply chains (SC). This chapter focuses on how blockchain technology can completely transform SC processes due to its inherent features. By implementing blockchain, businesses can cut costs, improve efficiency, and ensure accountability. The chapter explains the basics of blockchain technology and its potential benefits for SCM. The various beneficial features of blockchain technology such as trust, immutability, reliability, and security are discussed and highlighted in the chapter. The chapter showcases real-world uses of blockchain technology in SCM to depict the possible benefits. Benefits and challenges of incorporating blockchain into SC processes are also discussed in the chapter. Finally, this chapter emphasizes how crucial it is for companies to identify the factors that weaken their SC and to ensure that blockchain is applied to address such issues.
2021 International Conference on Computational Science and Computational Intelligence (CSCI), Dec 1, 2021
International Conference on Wireless Networks, 2010
Environmental monitoring is the process of blending ecosystems, mathematics, and computer science... more Environmental monitoring is the process of blending ecosystems, mathematics, and computer science for the purpose of environmental impact assessments. In the past few years, wireless sensor networks have been emerging as a next generation of computer networks. Applications of wireless sensor networks for environmental monitoring are being studied extensively using low-cost, high-availability, and high-performance sensors. This project focuses on measuring, analyzing, and storing sound information for environmental monitoring using well-known LPC-based algorithms and the Crossbow Imote2 sensor platform. The experimental results show potential environment-monitoring applications using wireless sensor networks.
Today mobile phones provide a wide range of applications that make our daily life easy. With popu... more Today mobile phones provide a wide range of applications that make our daily life easy. With popularity, smartphones have become a target for cybercrime where malicious apps are developed to acquire sensitive information or corrupt data. To mitigate this issue and to improve the security in mobile devices, different techniques have been used. These techniques can be broadly classified as static, dynamic and hybrid approaches. In this paper, a static-based model MalDuoNet is proposed to detect Android malwares, which uses a DualNet framework to analyze the features from the API calls. In the MalDuoNet model, one sub-network is focused to learn the features relevant to malicious behavior and the other sub-network is focused to learn the features in general. Thus it enables the model to learn complementary features which in turn helps get richer features for analysis. Then the features from the two sub-networks are combined in the final fused classifier for the final classification. In addition, each of the feature extractors has a separate classifier so that each sub-network can optimize its performance separately. The experimental results demonstrate that the MalDuoNet model outperforms the two baseline models with single network.
Mobile phones have become a target for cybercrime where malicious apps are developed to acquire s... more Mobile phones have become a target for cybercrime where malicious apps are developed to acquire sensitive information or corrupt data. To mitigate this issue and to improve the security in mobile devices, many machine learning methods have been developed to detect and classify Android malware. One problem with the existing methods is that they all assume a large dataset is available to train the learning models. In this paper, a model named few-shot malware classification (FSMC) is proposed to classify Android apps by using only a few training cases for each class. For three small datasets, the experimental results demonstrate that the FSMC model is able to achieve significantly higher accuracy compared to the existing 2-way 1-shot and 2-way 3-shot malware classification methods.
arXiv (Cornell University), Feb 7, 2022
Energy harvesting technologies offer a promising solution to sustainably power an ever-growing nu... more Energy harvesting technologies offer a promising solution to sustainably power an ever-growing number of Internet of Things (IoT) devices. However, due to the weak and transient natures of energy harvesting, IoT devices have to work intermittently rendering conventional routing policies and energy allocation strategies impractical. To this end, this paper, for the very first time, developed a distributed multi-agent reinforcement algorithm known as global actor-critic policy (GAP) to address the problem of routing policy and energy allocation together for the energy harvesting powered IoT system. At the training stage, each IoT device is treated as an agent and one universal model is trained for all agents to save computing resources. At the inference stage, packet delivery rate can be maximized. The experimental results show that the proposed GAP algorithm achieves ∼ 1.28× and ∼ 1.24× data transmission rate than that of the Q-table and ESDSRAA algorithm, respectively.
IGI Global eBooks, 2010
Significant advancements in hardware technology have propelled the existence of Wireless Sensor N... more Significant advancements in hardware technology have propelled the existence of Wireless Sensor Network (WSN). A WSN consists of simple, low cost yet powerful sensors. Each sensor has the ability to sense, process, and communicate data collected from the environment, in which it is deployed. Sensors usually draw energy from a small battery, and thus energy efficiency emerges as the key issue in any WSN. The basic idea of a
IGI Global eBooks, Jan 18, 2011
IGI Global eBooks, May 8, 2013
2022 27th Asia and South Pacific Design Automation Conference (ASP-DAC)
Energy harvesting technologies offer a promising solution to sustainably power an ever-growing nu... more Energy harvesting technologies offer a promising solution to sustainably power an ever-growing number of Internet of Things (IoT) devices. However, due to the weak and transient natures of energy harvesting, IoT devices have to work intermittently rendering conventional routing policies and energy allocation strategies impractical. To this end, this paper, for the very first time, developed a distributed multi-agent reinforcement algorithm known as global actor-critic policy (GAP) to address the problem of routing policy and energy allocation together for the energy harvesting powered IoT system. At the training stage, each IoT device is treated as an agent and one universal model is trained for all agents to save computing resources. At the inference stage, packet delivery rate can be maximized. The experimental results show that the proposed GAP algorithm achieves ∼ 1.28× and ∼ 1.24× data transmission rate than that of the Q-table and ESDSRAA algorithm, respectively.
Coastlines are an important yet volatile part of the environment and frequently change over time.... more Coastlines are an important yet volatile part of the environment and frequently change over time. Studying changes in coastlines can give important insights about the environment, so using efficient means to observe them is important. UAVs are cheap, fast and effective tools in gathering information, and segmentation can be used to detect and record the coastline. This paper proposes the use of CNN models to assist UAVs in detecting coastlines. Specifically, our research focuses on the combination of a neural network model, such as UNet, with transfer learning using a pre-trained model. The resulting model achieves semantic segmentation of coastlines to distinguish water and land. This allows UAVs to better recognize coastlines and reinforce their role in remote sensing image processing. 2. Detection of GPS Spoofing in UAVs Using Deep Learning and Machine Learning REU Participants: Jaron Burns and Dimitris Amiridis Mentor: Dr. Longzhuang Li Graduate Assistant: Rohith Mandala Abstrac...
Abstract:- This paper describes the development of a prototype that takes in an analog National T... more Abstract:- This paper describes the development of a prototype that takes in an analog National Television System Committee video signal generated by a video camera and data acquired by a microcontroller and display them in real-time on a digital panel. An 8051 microcontroller is used to acquire power dissipation by the display panel, room temperature, and camera zoom level. The paper describes the major hardware components and shows how they are interfaced into a functional prototype. Test data results are presented and discussed. Key-Words: data acquisition, overlaying text over video, real-time video 1
2017 International Conference on Computational Science and Computational Intelligence (CSCI), 2017
Growth of information technology led to the increasing need of computing and storage. Cloud servi... more Growth of information technology led to the increasing need of computing and storage. Cloud services is one such technology with high demand and hence requires more computing resources. Cloud data centers consume huge amount of energy and there by emitting carbon dioxide to the environment. This work proposes an approach for energy efficient resource management. Earlier approaches do not focus on the variations of workloads and lack in examining the effect of algorithms on performance. Virtual machine configuration also plays a vital role for reducing energy consumption and resource wastage, but is not given much importance. To address these weaknesses, this work proposes a novel approach to map groups of tasks to customized virtual machine types. Mapping of the tasks is based on task usage patterns\textemdash length, file size, bandwidth etc. Data is clustered in to group of tasks and is mapped to the suitable virtual machine based on the configuration. Virtual machine migration is employed to balance the load by calculating the load using MIPS, RAM and Bandwidth. Complete end-end architecture is proposed in this work with clustering of tasks, allocation of tasks to virtual machines and virtual machine migration techniques. The results of this work show that the energy consumption is decreased compared to the earlier approaches, which uses traditional virtual machine migration techniques.
2017 International Conference on Computational Science and Computational Intelligence (CSCI), 2017
One of the main challenges in cloud computing is limited availability of resources. As the number... more One of the main challenges in cloud computing is limited availability of resources. As the number of requests for cloud services increases, it becomes necessary for the system to balance the load and serve user requests at stipulated times. Load Balancing is a well-known NP-Complete Problem. This work proposes two variants of fusion-based task scheduling algorithm; both the approaches exploit two existing load balancing algorithms—the traditional round-robin algorithm (RRA) and the priority-based genetic algorithm (PGA), to improve the performance of the system in terms of the completion time. The idea of fusion lies in considering the variable amount of user requests to the cloud system. The first variant i.e. fusionbased load-aware resource allocation algorithm (FLA) uses PGA when there is relatively light load and RRA when the system encounters heavy load. The algorithm determines the intensity of the current load on the system, whether it is light or heavy. In the second variant...
There is increasing demand world-wide, from government agencies and the private sector for cuttin... more There is increasing demand world-wide, from government agencies and the private sector for cutting-edge biometric security technology that is difficult to breach but userfriendly at the same time. Some of the older tools, such as fingerprint, retina and iris scanning, and facial recognition software have all been found to have flaws and often viewed negatively because of many cultural and hygienic issues associated with them. Comparatively, mapping veins as a human barcode, a new technology, has many advantages over older technologies. Specifically, reproducing a three-dimensional model of a human vein system is impossible to replicate. Vein map technology is distinctive because of its state-of-the-art sensors are only able to recognize vein patterns if hemoglobin is actively flowing through the person's veins. Additionally, each individual's vein map is unique, even in the case of identical twins. The combinations of these factors provide vein map authentication an edge ove...
2021 RIVF International Conference on Computing and Communication Technologies (RIVF)
A tool for mobile devices is presented to detect a human face in an image and analyze the detecte... more A tool for mobile devices is presented to detect a human face in an image and analyze the detected face for its aesthetical quality. For the purpose of face detection, the tool uses two well-known classifiers: Haar feature-based cascade classifier and Local Binary Patterns (LBP) cascade classifier. Once a face is detected, the tool uses the Active Shape Model (ASM) algorithm to identify facial feature points and obtain raw data accordingly for aesthetical quality analysis. The analysis is carried out on the notion of Golden Ratio by finding ratios on the facial feature points obtained by using the ASM algorithm. To determine its effectiveness and limitations the tool is tested on animal faces, beautiful faces, ordinary faces, and unattractive faces. The statistics of results exposes distinctions between unattractive faces and beautiful/ordinary faces. However, no such distinction exists to separate beautiful faces from ordinary faces.
Energy efficiency is a primary concern in Wireless Sensor Networks (WSN). This is due to the fact... more Energy efficiency is a primary concern in Wireless Sensor Networks (WSN). This is due to the fact that WSNs are powered battery, and hence the life of WSNs becomes limited by the battery life. Efforts to replace depleted batteries are not feasible if WSNs are often deployed with thousands of sensor nodes, possibly in inaccessible, hostile, hazardous, or remote territories. Though WSNs are energy-constrained, adequate level of security is often desired in many applications of WSNs that guarantees data integrity and confidentiality. However, security protocols introduce extra energy overhead to a sensor node due to additional processing and communications associated with the protocols. Thus it is important to seek security protocols and solutions for WSNs that are energyefficient but also effective. Security is commonly implemented through symmetric key cryptography which requires a key exchange mechanism to establish a key between the communicating parties. One well known key exchang...
IGARSS 2020 - 2020 IEEE International Geoscience and Remote Sensing Symposium, 2020
Bathymetry, or the measurement of depth in any body of water, has been an area of research since ... more Bathymetry, or the measurement of depth in any body of water, has been an area of research since man began to venture out onto the open waters. Historically, researching the near-shore surf zone has been a time consuming and expensive process. The tools and methods used to gather data points in the surf zone are either time inefficient, expensive, or both. This is an issue considering how dynamic the surf zone environment can be. It is possible that by the time the surf zone bathymetry measurements have been completed, they are already out of date. This project utilizes unmanned aerial systems (UAS) to gather high-quality video of the near-shore surf zone waves crest. This footage is then processed using particle image velocimetry (PIV), a method for determining the velocity of particles in sequential images. This velocity is then processed using linear-wave theory shallow water approximations for calculating wave celerity from depth, but ran in reverse, to obtain the bathymetry its...