Kumar Saurabh

Now a days, Industries like manufacturing, power generation, and smart transportation are using advanced internet-connected systems. Unfortunately, this makes them more vulnerable to cyberattacks. To protect against these attacks, policymakers have created rules, implemented security features like secure login and encryption to safeguard these systems. However, the increasing number of cyberattacks like SQL injection attacks, False data Injection attacks, Phishing, etc. shows that these measures are not enough, and have some limitations. The false data injection attack is one of the most accruing attacks possible on Physical and cyber layers both. In Ukraine, there is a total blackout due to false data injection attacks, such case study of FDI attacks in power systems shows the severity of false data injection attacks. In this FDI attack, machines show wrong readings which might lead to economic and life loss, as critical infrastructures like hospitals, smart grids, modern transportation systems, etc. are fully dependent on electricity. In this paper, a few best-performing Machine learning algorithms have been implemented and after Exploratory Data Analysis and Data Preprocessing steps, GridSearchCv is used for hyperparameter tuning, which makes our proposed model outperformer than previous works in FDI attack detection. To evaluate the proposed model’s reliability, the performance evaluation is also done on Power System Attack Datasets provided by Mississippi State University and Oak Ridge National Laboratory. The Decision Tree algorithm stands out as the top performer with a training accuracy of 100% and a testing accuracy of 98%. the other performance metrics like Recall, F1 score, sensitivity training, and testing time also scored well in terms of comparison with other algorithms.

Industrial cyber-physical systems (ICPS) are gradually integrating information technology and automating industrial processes, leading systems to become more vulnerable to malicious actors. Thus, to deploy secure Industrial Control and Production Systems (ICPS) in smart factories, cyber threats and risks must be addressed. To identify all possible threats, Threat Modeling is a promising solution. Despite the existence of numerous methodological solutions for threat modeling in cyber-physical systems (CPS), current approaches are ad hoc and inefficient in providing clear insights to researchers and organizations involved in IIoT technologies. These approaches lack a comprehensive analysis of cyber threats and fail to facilitate effective path analysis across the ICPS lifecycle, incorporating smart manufacturing technologies and tools. To address these gaps, a novel quantitative threat modeling approach is proposed, aiming to identify probable attack vectors, assess the path of attacks, and evaluate the magnitude of each vector. This paper also explains the execution of the proposed approach with two case studies, namely the industrial manufacturing line, i.e., the Internet of Manufacturing (IoM), and the power and industry, i.e., the Internet of Production (IoP).

In the recent years, Distributed Denial of Service (DDoS) attacks on Internet of Things (IoT) devices have become one of the prime concerns to Internet users around the world. One of the sources of the attacks on IoT ecosystems are botnets. Intruders force IoT devices to become unavailable for its legitimate users by sending large number of messages within a short interval. This study proposes NFDLM, a lightweight and optimised Artificial Neural Network (ANN) based Distributed Denial of Services (DDoS) attack detection framework with mutual correlation as feature selection method which produces a superior result when compared with Long Short Term Memory (LSTM) and simple ANN. Overall, the detection performance achieves approximately 99\% accuracy for the detection of attacks from botnets. In this work, we have designed and compared four different models where two are based on ANN and the other two are based on LSTM to detect the attack types of DDoS.

In the recent years, we have witnessed a huge growth in the number of Internet of Things (IoT) and edge devices being used in our everyday activities. This demands the security of these devices from cyber attacks to be improved to protect its users. For years, Machine Learning (ML) techniques have been used to develop Network Intrusion Detection Systems (NIDS) with the aim of increasing their reliability/robustness. Among the earlier ML techniques DT performed well. In the recent years, Deep Learning (DL) techniques have been used in an attempt to build more reliable systems. In this paper, a Deep Learning enabled Long Short Term Memory (LSTM) Autoencoder and a 13-feature Deep Neural Network (DNN) models were developed which performed a lot better in terms of accuracy on UNSW-NB15 and Bot-IoT datsets. Hence we proposed LBDMIDS, where we developed NIDS models based on variants of LSTMs namely, stacked LSTM and bidirectional LSTM and validated their performance on the UNSW\_NB15 and BoT\-IoT datasets. This paper concludes that these variants in LBDMIDS outperform classic ML techniques and perform similarly to the DNN models that have been suggested in the past.

Gait recognition is advanced form of biometric technique that is used to identify a person from far away without the knowledge of that person. Gait recognition has overcome over other forms of recognition techniques like face, finger or iris due its improved technology. This is the reason it is used today in military establishment, bank and airports for security purpose. Gait is the manner or style of walking of human being. It is used to recognize or identify a single person or multiple people at the same time. In this paper we are going to analyze Gait to identify person by using SVM with K-NN and NN. In this paper first video of a person is captured and divided into frames. The frames are pre-processed, and background unwanted part is removed. Third feature of person is extracted using Hanavan's Model i.e., model based approach. Finally the person is searched in gait database by using SVM with K-NN and NN. The steps of loading video, pre-processing, background subtraction and then feature extraction are performed on database videos also. MATLAB Software has been used to implement the work of this paper.