Nilesh Kumar Sahu

Bio: Nilesh Kumar Sahu is an academic researcher from Birla Institute of Technology, Mesra. The author has contributed to research in topics: Wireless sensor network & Feature engineering. The author has an hindex of 1, co-authored 6 publications receiving 5 citations. Previous affiliations of Nilesh Kumar Sahu include Birla Institute of Technology and Science.

Machine Learning based anomaly detection for IoT Network: (Anomaly detection in IoT Network)

Abstract: As the world is leading towards having everything smart, like smart home, smart grid smart irrigation, there is the major concern of attack and anomaly detection in the Internet of Things (IoT) domain. There is an exponential increase in the use of IoT infrastructure in every field leads to an increase in threats and attacks too. There can be many types of possible attacks and anomaly that can affect the IoT system which can lead to failure of the IoT system. In this paper, different anomalies are predicted based on a different feature in the data set. Two machine learning classification models are used and comparisons between the performance of these used models are shown. Logistic regression and artificial neural network classification algorithms are applied. Since there are more than 3.5 lakh data set, two different approaches are experimented. In the first case, the classification algorithm stated above is applied on the whole 3.5 lakh dataset, and in the second case, all the classification algorithms are applied after omitting the feature “value” having data as 0 and 1. Data is divided into two sets, training and test set where the training set is 75% of total data available and the rest are test set, 99.4% accuracy is obtained for ANN for the first case while 99.99% accuracy is obtained for the algorithm stated above for the second case. This work can be used for identifying threats and anomaly occurring in a smart device and IoT solutions and prevent attacks.

Data Analytics and Its Applications in Brief

Abstract: With an ever-increasing amount of data created, it has become a major challenge for infrastructures and frameworks to process a lot of information inside stipulated time and resources. So as to effectively extract from this information, organizations are required to discover new devices and strategies in particular for large information preparation. Therefore, data analytics has become a key factor for organizations to uncover hidden data and accomplish the upper hand in the market. As of now, tremendous distributions of larger data and information processing make it hard for experts and specialists to discover points they are keen on and track forward-thinking. This chapter puts forth an outline of data analytics, extension, and discoveries just as opportunities emancipated by analysis of data. The chapter also deals with various applications of data analytics which include applying an algorithmic or mechanical procedure to infer bits of knowledge, for instance, going through a few informational collections to search for significant connections between one another.

Energy conservation in query driven wireless sensor networks

Abstract: Network topology used to connect sensor nodes is important factor in wireless sensor networks (WSNs) that affect the energy usage of any network. Proper topology connection thereby reduces energy usage of network by reducing number of packet transmissions. Therefore node connection must be optimized in wireless sensor networks. Hence we introduce a new multiple tree based architecture of sensor network, comprising of heterogeneous nodes, to develop an energy efficient query processing WSN application. This article proposes Multi-tree and Multiple-tree algorithms for improving network parameters in the usual data collection WSN application scenarios. A multi-tree simply suggests a strategy where similar parameter type sensors are arranged in a tree based topology so as to reduce the energy consumption as well as the end to end delay in data communication. The Multiple-tree exemplifies a more effective way of query result propagation in data centric applications where delay requirements are not very stringent.

Feature Engineering for Various Data Types in Data Science

Abstract: The precision of any machine learning algorithm depends on the data set, its suitability, and its volume. Therefore, data and its characteristics have currently become the predominant components of any predictive or precision-based domain like machine learning. Feature engineering refers to the process of changing and preparing this input data so that it is ready for training machine learning models. Several features such as categorical, numerical, mixed, date, and time are to be considered for feature extraction in feature engineering. Datasets containing characteristics such as cardinality, missing data, and rare labels for categorical features, distribution, outliers, and magnitude are currently considered as features. This chapter discusses various data types and their techniques for applying to feature engineering. This chapter also focuses on the implementation of various data techniques for feature extraction.

Federated-Learning-Based Anomaly Detection for IoT Security Attacks

Abstract: The Internet of Things (IoT) is made up of billions of physical devices connected to the Internet via networks that perform tasks independently with less human intervention. Such brilliant automation of mundane tasks requires a considerable amount of user data in digital format, which, in turn, makes IoT networks an open source of personally identifiable information data for malicious attackers to steal, manipulate, and perform nefarious activities. A huge interest has been developed over the past years in applying machine learning (ML)-assisted approaches in the IoT security space. However, the assumption in many current works is that big training data are widely available and transferable to the main server because data are born at the edge and are generated continuously by IoT devices. This is to say that classic ML works on the legacy set of entire data located on a central server, which makes it the least preferred option for domains with privacy concerns on user data. To address this issue, we propose the federated-learning (FL)-based anomaly detection approach to proactively recognize intrusion in IoT networks using decentralized on-device data. Our approach uses federated training rounds on gated recurrent units (GRUs) models and keeps the data intact on local IoT devices by sharing only the learned weights with the central server of FL. Also, the approach’s ensembler part aggregates the updates from multiple sources to optimize the global ML model’s accuracy. Our experimental results demonstrate that our approach outperforms the classic/centralized machine learning (non-FL) versions in securing the privacy of user data and provides an optimal accuracy rate in attack detection.

An overview of queuing delay and various delay based algorithms in networks

Abstract: In the era of modern technology, the demand for seamless and fast communication service is gradually increasing. Moreover, as time goes on, more devices are getting connected to the network. In the vast network, for quick responses to these devices, the delay becomes one of the important factors to concern about. Among different types, Queuing delay causes more impact on a network compared to other delays. A survey on delay, different models, effects, and management are illustrated here. Queuing models, like Little Theorem, M/M/1, M/M/m, M/G/1, etc, are discussed in detail. We move on explaining the delay effects on different types of the network such as Wireless network, Mobile communication network, SDN, IoT, etc. This paper also focuses on TCP/IP fields like ECN bits, MPLS, and characteristics of flows. Lastly, the effects of delay on other related networking are discussed.

Smart Strawberry Farming Using Edge Computing and IoT

Abstract: Strawberries are sensitive fruits that are afflicted by various pests and diseases. Therefore, there is an intense use of agrochemicals and pesticides during production. Due to their sensitivity, temperatures or humidity at extreme levels can cause various damages to the plantation and to the quality of the fruit. To mitigate the problem, this study developed an edge technology capable of handling the collection, analysis, prediction, and detection of heterogeneous data in strawberry farming. The proposed IoT platform integrates various monitoring services into one common platform for digital farming. The system connects and manages Internet of Things (IoT) devices to analyze environmental and crop information. In addition, a computer vision model using Yolo v5 architecture searches for seven of the most common strawberry diseases in real time. This model supports efficient disease detection with 92% accuracy. Moreover, the system supports LoRa communication for transmitting data between the nodes at long distances. In addition, the IoT platform integrates machine learning capabilities for capturing outliers in collected data, ensuring reliable information for the user. All these technologies are unified to mitigate the disease problem and the environmental damage on the plantation. The proposed system is verified through implementation and tested on a strawberry farm, where the capabilities were analyzed and assessed.

Recent Advancement of Data-Driven Models in Wireless Sensor Networks: A Survey

Abstract: Wireless sensor networks (WSNs) are considered producers of large amounts of rich data. Four types of data-driven models that correspond with various applications are identified as WSNs: query-driven, event-driven, time-driven, and hybrid-driven. The aim of the classification of data-driven models is to get real-time applications of specific data. Many challenges occur during data collection. Therefore, the main objective of these data-driven models is to save the WSN’s energy for processing and functioning during the data collection of any application. In this survey article, the recent advancement of data-driven models and application types for WSNs is presented in detail. Each type of WSN is elaborated with the help of its routing protocols, related applications, and issues. Furthermore, each data model is described in detail according to current studies. The open issues of each data model are highlighted with their challenges in order to encourage and give directions for further recommendation.

Lightweight Internet of Things Botnet Detection Using One-Class Classification

Abstract: Like smart phones, the recent years have seen an increased usage of internet of things (IoT) technology. IoT devices, being resource constrained due to smaller size, are vulnerable to various security threats. Recently, many distributed denial of service (DDoS) attacks generated with the help of IoT botnets affected the services of many websites. The destructive botnets need to be detected at the early stage of infection. Machine-learning models can be utilized for early detection of botnets. This paper proposes one-class classifier-based machine-learning solution for the detection of IoT botnets in a heterogeneous environment. The proposed one-class classifier, which is based on one-class KNN, can detect the IoT botnets at the early stage with high accuracy. The proposed machine-learning-based model is a lightweight solution that works by selecting the best features leveraging well-known filter and wrapper methods for feature selection. The proposed strategy is evaluated over different datasets collected from varying network scenarios. The experimental results reveal that the proposed technique shows improved performance, consistent across three different datasets used for evaluation.