1. An intrusion detection system (IDS) (100) for securing Internet of Things (IoT) devices, comprising: a. a data collection module (10) gathers data from IoT devices and networks, where the data includes normal and malicious network traffic patterns, used to train and test the IDS models; b. a pre-processing module (20) pre-processes the collected raw data from IoT devices and networks by the data collection module (10) to remove any inconsistencies, redundancies, and irrelevant features, where the preprocessing involves data cleaning, normalization, feature extraction, and transformation; c. a feature selection module (30) selects relevant features from the preprocessed data using feature selection algorithms to reduce dimensionality and enhance classification performance; d. a machine learning (ML) module (40) employs various machine learning classifiers on the extracted features from the pre-processed data for classification tasks on binary and multiclass data to identify potential security threats; e. a deep learning (DL) module (50) employs convolutional neural networks (CNNs) to analyzes the binary and multiclass data and detects patterns indicative of intrusions; f. a cross-verification module (60) cross verifies the outputs from both ML module (40) and DL module (50) explored on three different datasets to provide a comprehensive IDS solution for securing IoT devices, ensuring robust detection and mitigation of security breaches; g. a performance evaluation module (70) evaluates the performance of the IDS using various metrics such as accuracy, recall, precision, F-score, receiver operating characteristic (ROC), and Area Under the Curve (AUC); h. an alerting module (80) triggers an alert to notify system administrators or automated response mechanisms when an anomaly or potential intrusion is detected, and i. a response module (90) triggers automated responses, such as blocking suspicious network traffic, isolating compromised devices, or escalating the alert to human operators for further investigation depending on the severity of the threat detected.

2. The intrusion detection system (IDS) for securing Internet of Things (IoT) devices as claimed in claim 1, wherein three different datasets comprising NSLKDD, IDS2018 and IOTID20.

3. The intrusion detection system (IDS) for securing Internet of Things (IoT) devices as claimed in claim 1, wherein various machine learning classifiers comprising Decision Tree, Gaussian Naïve Bayes, Logistic regression, and K-Nearest Neighbors (KNN).

4. The intrusion detection system (IDS) for securing Internet of Things (IoT) devices as claimed in claim 1, wherein the feature selection algorithms comprising SELBEST.

5. A method of working of an intrusion detection system (IDS) (100) for securing Internet of Things (IoT) devices as claimed in claim 1, said method comprising steps of: a. gathering data from IoT devices and networks using a data collection module (10), where the data includes normal and malicious network traffic patterns, used to train and test the IDS models; b. pre-processing the collected raw data from IoT devices and networks by a data collection module (20) to remove any inconsistencies, redundancies, and irrelevant features, where the preprocessing involves data cleaning, normalization, feature extraction, and transformation; c. selecting relevant features from the preprocessed data using feature selection algorithms by a feature selection module (30) to reduce dimensionality and enhance classification performance; d. employing various machine learning classifiers on the extracted features from the pre-processed data by a machine learning (ML) module (40) for classification tasks on binary and multiclass data to identify potential security threats; e. employing convolutional neural networks (CNNs) by a deep learning (DL) module (50) to analyzes the binary and multiclass data and detects patterns indicative of intrusions; f. cross verifying the outputs from both ML module (40) and DL module (50) explored on three different datasets by a cross-verification module (60) to provide a comprehensive IDS solution for securing IoT devices, ensuring robust detection and mitigation of security breaches; g. evaluating the performance of the IDS by a performance evaluation module (70) using various metrics such as accuracy, recall, precision, F-score, ROC, and AUC; h. triggering an alert to notify system administrators or automated response mechanisms by an alerting module (80) when an anomaly or potential intrusion is detected, and i. triggering automated responses, such as blocking suspicious network traffic, isolating compromised devices, or escalating the alert to human operators for further investigation depending on the severity of the threat detected by a response module (90).

Description:2. Description:
TITLE: Intrusion Detection System for Securing Internet of Things Devices and Networks and Method thereof.

FIELD OF THE INVENTION
The present invention relates to a system for detecting intrusion. More particularly, the present invention relates to an intrusion detection system (IDS) for securing Internet of Things (IoT) devices and networks which protects IoT devices and networks from various types of attacks and continuously monitors IoT devices and networks for anomalies and suspicious activities, thus ensuring their proper functioning and data security and reducing computational time and produced good accuracy. Advantageously, the present invention detects and prevents potential attacks, thus safeguarding the integrity of the connected devices and the data they process.

BACKGROUND OF THE INVENTION
The Internet of Things, or IoT, is a network of devices that exchanges data among itself over recognized protocols. It can be thought of as an interconnected system. The idea of smartness is being introduced to gadgets, sensors, homes, streets, and even cities thanks to recent developments in IoT. The Internet of Things (IoT) is a rapidly expanding area in contemporary computing and communication technology that has had a significant impact on a number of industries, including the automation of cars and the agriculture industry. These days, the Internet of Everything (IoE), which deals with all linked devices in daily life, is often referred to as IoT.
IoT is made up of multiple levels, one of which being the network layer. The network layer is primarily in charge of sending data packets between hosts and is designed in accordance with the conventional Internet communication layers. Furthermore, a complex and weak point in IoT architecture is the network layer, which can give rise to a number of security problems. To address the security concerns, however, a number of security frameworks are in place. To ensure that the devices function properly and address security risks, these frameworks must be installed in the IoT architecture and/or the devices themselves. Regretfully, the majority of security frameworks demand a significant amount of storage and processing power. To get around the restrictions, though, several strategies including lightweight encryption and authentication methods can be used.
One of the main causes of the security problems is the sheer number of nodes hosts or other IoT-connected devices, because a security breech on one node might cause the entire system to fail. Botnets, ransomware, distributed denial of service (DDoS) attacks, remote recording, routing assaults, and data leakage are the most frequent security concerns that Internet of Things (IoT) systems must contend with. Although using a firewall is thought to be the first line of defence against assaults on IoT devices, the complexity and diversity of IoT systems make this a poor solution.
Recently, intrusion detection systems (IDS) have gained popularity, which is a set of advanced technologies that recognize unpleasant activities to enhance cloud security. There are three types of IDSs: misuse detection, anomaly detection, and hybrid detection. An anomaly IDS is installed to detect attacks based on previously recorded normal behavior. This form of IDS is commonly utilized because it can detect new intrusions by comparing current real-time traffic with recorded regular real-time traffic. However, it registers good false-positive alarms, implying that many regular packets are mistaken for storm packets.
On the contrary, a misuse IDS is used to identify intrusion using a signatures database. It does not cause false alarms but can be passed by a new attack with a unique signature. Moreover, IDS is affected by several constraints that reduce the effectiveness of intrusion detection, such as vast volumes of data, instantaneous detection, the integrity of data, and more.
To address these challenges, researchers and industry practitioners are developing innovative IDS solutions tailored specifically for IoT environments. These solutions often leverage advanced technologies such as Machine Learning (ML), Deep Learning (DL), anomaly detection algorithms, and lightweight cryptography to enhance detection accuracy, efficiency, and scalability.
Some of the prior arts are:
IN202341013028 discloses an Advance Data Security using Machine Learning and Internet of Things comprises over the last decade, IoT platforms have been developed into a global giant that grabs every aspect of our daily lives by advancing human life with its unaccountable smart services. Because of easy accessibility and fast-growing demand for smart devices and network, IoT is now facing more security challenges than ever before. There are existing security measures that can be applied to protect IoT. However, traditional techniques are not as efficient with the advancement booms as well as different attack types and their severeness. Thus, a strong-dynamically enhanced and up to date security system is required for next-generation IoT system. A huge technological advancement has been noticed in Machine Learning (ML) which has opened many possible research windows to address ongoing and future challenges in IoT. In order to detect attacks and identify abnormal behaviours of smart devices and networks, ML is being utilized as a powerful technology to fulfill this purpose. In this survey paper, the architecture of IoT is discussed, following a comprehensive literature review on ML approaches the importance of security of IoT in terms of different types of possible attacks. Moreover, ML-based potential solutions for IoT security has been presented and future challenges are discussed
IN202241023086 discloses a IoT with Artificial intelligence based data authentication from end to end to cyper security of configuration using Deep Learning Techniques comprises Security of the Internet of Things (IoT) is becoming increasingly relevant to people in both academic and industrial settings (IoT). The Internet of Things' security issues range from denial of service (DoS) attacks to network penetration and data leakage, and they can occur at any time (IoT). Machine learning (ML) was used to create a new security architecture that automatically adapts to the growing number of Internet of Things-related vulnerabilities. This research used machine learning to create a new framework (IoT). By leveraging SDN, NFV, and other technologies, this architecture safeguards against a wide range of threats. This artificial intelligence framework allows you to combine a monitoring agent and an AI-based reaction agent, both of which use machine learning models to detect network trends and anomalies in Internet of Things systems. Three things that the framework performs to help it achieve its goals. These are known as supervised learning and neural networks. The testing results show that the suggested strategy is effective. Data mining has shown to be highly effective at quickly and cheaply discovering dangers in order to spread them. An Internet of Things (IoT) anomaly detection system (IDS) was tested in a real-world Smart building scenario.
However, the conventional intrusion detection system as discussed in the prior art need to be retrained whenever there are changes in the network or new types of attacks emerge. This can be a time-consuming process, especially for DL models that require large amounts of data and computational resources. The conventional intrusion detection system often requires significant computational resources, which can be a challenge for resource-constrained IoT devices. The conventional IDS render less effective in protecting IoT networks.
Accordingly, there is a need for an improved intrusion detection system for securing Internet of Things (IoT) devices and networks which protects IoT devices and networks from various types of attacks and continuously monitors IoT devices and networks for anomalies and suspicious activities, thus ensuring their proper functioning and data security and reducing computational time and produced good accuracy.
OBJECT OF THE INVENTION
The principal object of the present invention is to provide an intrusion detection system (IDS) for securing Internet of Things (IoT) devices and networks and reducing computational time and produced good accuracy.
Another object of this invention is to provide an intrusion detection system (IDS) which identifies potential security threats and suspicious activities in real-time, allowing for early detection and prevention of potential attacks
Another object of this invention is to continuously monitor the network and devices, providing around-the-clock protection against potential security breaches.
Another object of the present invention is to optimize the performance of IoT devices and networks, ensuring smooth and efficient operations by detecting and addressing security issues.
SUMMARY OF THE INVENTION
It is a primary aspect of the present invention to provide a An intrusion detection system (IDS) (100) for securing Internet of Things (IoT) devices, comprising: a data collection module (10) gathers data from IoT devices and networks, where the data includes normal and malicious network traffic patterns, used to train and test the IDS models; a pre-processing module (20) pre-processes the collected raw data from IoT devices and networks by the data collection module (10) to remove any inconsistencies, redundancies, and irrelevant features, where the preprocessing involves data cleaning, normalization, feature extraction, and transformation; a feature selection module (30) selects relevant features from the preprocessed data using feature selection algorithms to reduce dimensionality and enhance classification performance; a machine learning (ML) module (40) employs various machine learning classifiers on the extracted features from the pre-processed data for classification tasks on binary and multiclass data to identify potential security threats; a deep learning (DL) module (50) employs convolutional neural networks (CNNs) to analyzes the binary and multiclass data and detects patterns indicative of intrusions; a cross-verification module (60) cross verifies the outputs from both ML module (40) and DL module (50) explored on three different datasets to provide a comprehensive IDS solution for securing IoT devices, ensuring robust detection and mitigation of security breaches; a performance evaluation module (70) evaluates the performance of the IDS using various metrics such as accuracy, recall, precision, F-score, receiver operating characteristic (ROC), and Area Under the Curve (AUC); an alerting module (80) triggers an alert to notify system administrators or automated response mechanisms when an anomaly or potential intrusion is detected, and a response module (90) triggers automated responses, such as blocking suspicious network traffic, isolating compromised devices, or escalating the alert to human operators for further investigation depending on the severity of the threat detected.
Another aspect of the present invention is to provide a method of working of an intrusion detection system (IDS) (100) for securing Internet of Things (IoT) devices as claimed in claim 1, said method comprising steps of:
a. gathering data from IoT devices and networks using a data collection module (10), where the data includes normal and malicious network traffic patterns, used to train and test the IDS models;
b. pre-processing the collected raw data from IoT devices and networks by a data collection module (20) to remove any inconsistencies, redundancies, and irrelevant features, where the preprocessing involves data cleaning, normalization, feature extraction, and transformation;
c. selecting relevant features from the preprocessed data using feature selection algorithms by a feature selection module (30) to reduce dimensionality and enhance classification performance;
d. employing various machine learning classifiers on the extracted features from the pre-processed data by a machine learning (ML) module (40) for classification tasks on binary and multiclass data to identify potential security threats;
e. employing convolutional neural networks (CNNs) by a deep learning (DL) module (50) to analyzes the binary and multiclass data and detects patterns indicative of intrusions;
f. cross verifying the outputs from both ML module (40) and DL module (50) explored on three different datasets by a cross-verification module (60) to provide a comprehensive IDS solution for securing IoT devices, ensuring robust detection and mitigation of security breaches;
g. evaluating the performance of the IDS by a performance evaluation module (70) using various metrics such as accuracy, recall, precision, F-score, ROC, and AUC;
h. triggering an alert to notify system administrators or automated response mechanisms by an alerting module (80) when an anomaly or potential intrusion is detected, and
i. triggering automated responses, such as blocking suspicious network traffic, isolating compromised devices, or escalating the alert to human operators for further investigation depending on the severity of the threat detected by a response module (90).
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 is a block diagram of an intrusion detection system for securing Internet of Things (IoT) devices and networks.
Figure 2 is a framework of a machine learning architecture.
Figure 3 is a framework of a Deep Learning framework for IDS.
DETAILED DESCRIPTION OF THE INVENTION
The present invention as embodied by a “an intrusion detection system for securing Internet of Things (IoT) devices and networks and method thereof” succinctly fulfills the above-mentioned need[s] in the art. The present invention has objective[s] arising because of the above-mentioned need[s], said objective[s] having been enumerated herein above.
The following description is directed to an intrusion detection system for securing Internet of Things devices and networks and method thereof as much as the objective(s) of the present invention are enumerated, it will be obvious to a person skilled in the art that, the enumerated objective(s) are not exhaustive of the present invention in its entirety and are enclosed solely for the purpose of illustration. Further, the present invention encloses within its scope and purview, any structural alternative(s) and/or any functional equivalent(s) even though, such structural alternative(s) and/or any functional equivalent(s) are not mentioned explicitly herein or elsewhere, in the present disclosure. The present invention therefore encompasses also, any improvisation[s]/modification[s] applied to the structural alternative[s]/functional alternative[s] within its scope and purview. The present invention may be embodied in other specific form[s] without departing from the essential attributes thereof.
Furthermore, the terms and phrases used herein are not intended to be limiting, but rather are to provide an understandable description. Throughout this specification, the use of the word "comprise" and variations such as "comprises" and "comprising" may imply the inclusion of an element or elements not specifically recited.
The conventional intrusion detection system need to be retrained whenever there are changes in the network or new types of attacks emerge. This can be a time-consuming process, especially for DL models that require large amounts of data and computational resources. The conventional intrusion detection system often requires significant computational resources, which can be a challenge for resource-constrained IoT devices. The conventional IDS render less effective in protecting IoT networks. But in the case of present invention, an intrusion detection system (IDS) for securing Internet of Things (IoT) devices protects IoT devices and networks from various types of attacks and continuously monitors IoT devices and networks for anomalies and suspicious activities, thus ensuring their proper functioning and data security.
Referring to Figure 1 to 3, in an embodiment of the present invention, provides an intrusion detection system (IDS) for securing Internet of Things (IoT) devices and networks comprising: a data collection module (10) gathers data from IoT devices and networks, where the data includes normal and malicious network traffic patterns, used to train and test the IDS models; a pre-processing module (20) pre-processes the collected raw data from IoT devices and networks by the data collection module (10) to remove any inconsistencies, redundancies, and irrelevant features, where the preprocessing involves data cleaning, normalization, feature extraction, and transformation; a feature selection module (30) selects relevant features from the preprocessed data using feature selection algorithms to reduce dimensionality and enhance classification performance; a machine learning (ML) module (40) employs various machine learning classifiers on the extracted features from the pre-processed data for classification tasks on binary and multiclass data to identify potential security threats; a deep learning (DL) module (50) employs convolutional neural networks (CNNs) to analyzes the binary and multiclass data and detects patterns indicative of intrusions; a cross-verification module (60) cross verifies the outputs from both ML module (40) and DL module (50) explored on three different datasets to provide a comprehensive IDS solution for securing IoT devices, ensuring robust detection and mitigation of security breaches; a performance evaluation module (70) evaluates the performance of the IDS using various metrics such as accuracy, recall, precision, F-score, receiver operating characteristic (ROC), and Area Under the Curve (AUC); an alerting module (80) triggers an alert to notify system administrators or automated response mechanisms when an anomaly or potential intrusion is detected, and a response module (90) triggers automated responses, such as blocking suspicious network traffic, isolating compromised devices, or escalating the alert to human operators for further investigation depending on the severity of the threat detected.
In one embodiment of the present invention, the three different datasets comprising NSLKDD, IDS2018 and IOTID20.
In another embodiment of the present invention, the various machine learning classifiers comprising Decision Tree, Gaussian Naïve Bayes, Logistic regression, and K-Nearest Neighbors (KNN).
In another embodiment of the present invention, the feature selection algorithms comprising SELBEST.
Another embodiment of the present invention is to provide a method of working of an intrusion detection system (IDS) (100) for securing Internet of Things (IoT) devices, said method comprising steps of:
a. gathering data from IoT devices and networks using a data collection module (10), where the data includes normal and malicious network traffic patterns, used to train and test the IDS models;
b. pre-processing the collected raw data from IoT devices and networks by a data collection module (20) to remove any inconsistencies, redundancies, and irrelevant features, where the preprocessing involves data cleaning, normalization, feature extraction, and transformation;
c. selecting relevant features from the preprocessed data using feature selection algorithms by a feature selection module (30) to reduce dimensionality and enhance classification performance;
d. employing various machine learning classifiers on the extracted features from the pre-processed data by a machine learning (ML) module (40) for classification tasks on binary and multiclass data to identify potential security threats;
e. employing convolutional neural networks (CNNs) by a deep learning (DL) module (50) to analyzes the binary and multiclass data and detects patterns indicative of intrusions;
f. cross verifying the outputs from both ML module (40) and DL module (50) explored on three different datasets by a cross-verification module (60) to provide a comprehensive IDS solution for securing IoT devices, ensuring robust detection and mitigation of security breaches;
g. evaluating the performance of the IDS by a performance evaluation module (70) using various metrics such as accuracy, recall, precision, F-score, ROC, and AUC;
h. triggering an alert to notify system administrators or automated response mechanisms by an alerting module (80) when an anomaly or potential intrusion is detected, and
i. triggering automated responses, such as blocking suspicious network traffic, isolating compromised devices, or escalating the alert to human operators for further investigation depending on the severity of the threat detected by a response module (90).

WORKING EXAMPLE
An exemplary embodiment discloses an intrusion detection system (IDS) for securing Internet of things (IoT) devices. The data from IoT devices and networks are gathered using a data collection module (10), where the data includes normal and malicious network traffic patterns, used to train and test the IDS models. The collected raw data is pre-processed from IoT devices and networks by a data collection module (20) to remove any inconsistencies, redundancies, and irrelevant features, where the preprocessing involves data cleaning, normalization, feature extraction, and transformation. The relevant features are selected from the preprocessed data using feature selection algorithms by a feature selection module (30) to reduce dimensionality and enhance classification performance. The various machine learning classifiers on the extracted features from the pre-processed data are employed by a machine learning (ML) module (40) for classification tasks on binary and multiclass data to identify potential security threats. The convolutional neural networks (CNNs) are employed by a deep learning (DL) module (50) to analyze the binary and multiclass data and detect patterns indicative of intrusions. The outputs from both ML module (40) and DL module (50) explored on three different datasets is cross verified by a cross-verification module (60) to provide a comprehensive IDS solution for securing IoT devices, ensuring robust detection and mitigation of security breaches. The performance of the IDS is evaluated by a performance evaluation module (70) using various metrics such as accuracy, recall, precision, F-score, ROC, and AUC. An alert to notify system administrators or automated response mechanisms are triggered by an alerting module (80) when an anomaly or potential intrusion is detected. The automated responses are triggered, such as blocking suspicious network traffic, isolating compromised devices, or escalating the alert to human operators for further investigation depending on the severity of the threat detected by a response module (90).
ADVANTAGE OF THE PRESENT INVENTION
The present invention relates to an intrusion detection system (IDS) for securing Internet of things (IoT) devices and networks which detects suspicious activities or potential threats at an early stage, allowing for timely response and mitigation before they escalate into significant security breaches.
The present invention relates to an intrusion detection system which monitors network traffic and behavior patterns, enabling the detection of abnormal activities that may indicate unauthorized access or malicious intent, even if the specific threat signatures are unknown.
The present invention relates to an intrusion detection system which continuously monitors IoT device communications and network traffic in real-time, providing immediate alerts and notifications of any suspicious or malicious activities, allowing for quick action to be taken.
The present invention relates to an intrusion detection system which detects and prevents zero-day attacks by identifying unusual patterns or behaviors that deviate from normal operations, thus offering protection against previously unknown threats.
The present invention relates to an intrusion detection system which minimizes downtime and financial losses associated with cyber attacks, ensuring continuity of operations for IoT deployments by detecting and mitigating security threats in a timely manner.
The present invention relates to an intrusion detection system which provides valuable information and alerts to incident response teams, facilitating rapid response and containment of security incidents, thereby reducing the impact and severity of cyber attacks.
It will be apparent to a person skilled in the art that the above description is for illustrative purposes only and should not be considered as limiting. Various modifications, additions, alterations, and improvements without deviating from the spirit and the scope of the invention may be made by a person skilled in the art.
LIST OF NUMERALS:
(10). Data collection module
(20). Pre-processing module
(30). Feature selection module
(40). Machine learning (ML) module
(50). Deep learning (DL) module
(60). Cross-verification module
(70). Performance evaluation module
(80). Alerting module
(90). Response module
(100). Intrusion Detection System (IDS)
, Claims:1. An intrusion detection system (IDS) (100) for securing Internet of Things (IoT) devices, comprising:
a. a data collection module (10) gathers data from IoT devices and networks, where the data includes normal and malicious network traffic patterns, used to train and test the IDS models;
b. a pre-processing module (20) pre-processes the collected raw data from IoT devices and networks by the data collection module (10) to remove any inconsistencies, redundancies, and irrelevant features, where the preprocessing involves data cleaning, normalization, feature extraction, and transformation;
c. a feature selection module (30) selects relevant features from the preprocessed data using feature selection algorithms to reduce dimensionality and enhance classification performance;
d. a machine learning (ML) module (40) employs various machine learning classifiers on the extracted features from the pre-processed data for classification tasks on binary and multiclass data to identify potential security threats;
e. a deep learning (DL) module (50) employs convolutional neural networks (CNNs) to analyzes the binary and multiclass data and detects patterns indicative of intrusions;
f. a cross-verification module (60) cross verifies the outputs from both ML module (40) and DL module (50) explored on three different datasets to provide a comprehensive IDS solution for securing IoT devices, ensuring robust detection and mitigation of security breaches;
g. a performance evaluation module (70) evaluates the performance of the IDS using various metrics such as accuracy, recall, precision, F-score, receiver operating characteristic (ROC), and Area Under the Curve (AUC);
h. an alerting module (80) triggers an alert to notify system administrators or automated response mechanisms when an anomaly or potential intrusion is detected, and
i. a response module (90) triggers automated responses, such as blocking suspicious network traffic, isolating compromised devices, or escalating the alert to human operators for further investigation depending on the severity of the threat detected.

2. The intrusion detection system (IDS) for securing Internet of Things (IoT) devices as claimed in claim 1, wherein three different datasets comprising NSLKDD, IDS2018 and IOTID20.

3. The intrusion detection system (IDS) for securing Internet of Things (IoT) devices as claimed in claim 1, wherein various machine learning classifiers comprising Decision Tree, Gaussian Naïve Bayes, Logistic regression, and K-Nearest Neighbors (KNN).

4. The intrusion detection system (IDS) for securing Internet of Things (IoT) devices as claimed in claim 1, wherein the feature selection algorithms comprising SELBEST.

5. A method of working of an intrusion detection system (IDS) (100) for securing Internet of Things (IoT) devices as claimed in claim 1, said method comprising steps of:

a. gathering data from IoT devices and networks using a data collection module (10), where the data includes normal and malicious network traffic patterns, used to train and test the IDS models;
b. pre-processing the collected raw data from IoT devices and networks by a data collection module (20) to remove any inconsistencies, redundancies, and irrelevant features, where the preprocessing involves data cleaning, normalization, feature extraction, and transformation;
c. selecting relevant features from the preprocessed data using feature selection algorithms by a feature selection module (30) to reduce dimensionality and enhance classification performance;
d. employing various machine learning classifiers on the extracted features from the pre-processed data by a machine learning (ML) module (40) for classification tasks on binary and multiclass data to identify potential security threats;
e. employing convolutional neural networks (CNNs) by a deep learning (DL) module (50) to analyzes the binary and multiclass data and detects patterns indicative of intrusions;
f. cross verifying the outputs from both ML module (40) and DL module (50) explored on three different datasets by a cross-verification module (60) to provide a comprehensive IDS solution for securing IoT devices, ensuring robust detection and mitigation of security breaches;
g. evaluating the performance of the IDS by a performance evaluation module (70) using various metrics such as accuracy, recall, precision, F-score, ROC, and AUC;
h. triggering an alert to notify system administrators or automated response mechanisms by an alerting module (80) when an anomaly or potential intrusion is detected, and
i. triggering automated responses, such as blocking suspicious network traffic, isolating compromised devices, or escalating the alert to human operators for further investigation depending on the severity of the threat detected by a response module (90).