An Intelligent Deep Learning Network based on Internet of Things (IoTs): An Enhanced Approach for Network Threat Mitigations, Techniques, Issues and Opportunities

http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 231 An Intelligent Deep Learning Network based on Internet of Things (IoTs): An Enhanced Approach for Network Threat Mitigations, Techniques, Issues and Opportunities Samiullah Aftab AITEECAI Technology Exchange and Engineering Consultancy, Lahore, Punjab, Pakistan Email: samiullahafta[email protected] Majid Ali AITEECAI Technology Exchange and Engineering Consultancy, Lahore, Punjab, Pakistan Email: [email protected] Fasiha Ikram Bahria University, Department of Computer Science, 13 Stadium Road, Karachi, Pakistan Email: [email protected] & fasi[email protected] Arshad Ali Faculty of Computer and Information Systems, Islamic University of Madinah, Al Madinah Al Munawarah, 42351, Saudi Arabia Email: [email protected] Saad Hamayun Department of Robotics, Faculty of Computer Science & IT, Superior University, Lahore, 54000, Pakistan Email: saadhamay[email protected] Farhan Zohaib Department of Robotics, Faculty of Computer Science & IT, Superior University, Lahore, 54000, Pakistan Email: [email protected] Ali Hassan Department of Robotics, Faculty of Computer Science & IT, Superior University, Lahore, 54000, Pakistan Email: [email protected] Ukshah Ejaz Department of Robotics, Faculty of Computer Science & IT Superior University Lahore, 54000, Pakistan Email: [email protected] Hamayun Khan Department of Computer Science, Faculty of Computer Science & IT, Superior University, Lahore, 54000, Pakistan Email: hamayun.kha[email protected] The Internet of Things (IoT) is among the most commonly utilized technologies nowadays, and it has a substantial impact on our lives in many different ways, including social, commercial, and economic ways. Regarding automation, productivity, as well as the comfort of consumers in diverse application segments, including education and smart cities, the current and future IoT technologies promise significant advancement in the overall human life quality. Nevertheless, cyber-attacks and threats have a significant impact on smart applications in the context of the IoT. The conventional IoT security approaches would not suffice, considering the current security issues in light of the advanced burgeoning of various forms of attacks and threats. The solution to providing an updated and dynamic security system to the next- http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 232 generation IoT system is to deploy the artificial intelligence (AI) expertise and, specifically, machine, and deep learning solutions. In the entire article, we show a comprehensive image of IoT security intelligence that is constructed on machine and deep learning technologies that can extract insights from raw data in order to intelligently defend against numerous cyber-attacks on the Internet of Things. Lastly, in line with our research, we bring out the research issues and future directions based on our research. On the whole, the article aims to become a reference point and guide, especially on the technical level, to cybersecurity experts and researchers operating in the field of IoT. Keywords: Internet of Things (IoT), Deep learning, Convolutional neural network, Recurrent neural network, Long short-term memory INTRODUCTION: The Internet of Things (IoT) stands among the most popular technologies nowadays and is commonly referred to as a network of heterogeneous elements that are capable of creating smart systems and services detecting, acquiring, disseminating, and analyzing data [1]. Things in the IoT devices are those smart devices, which are sensors, smart watches, smart refrigerators, smoke readers, radio frequency identification (RFID), heartbeat, accelerometers, smart phones, etc, which gather and send data [2]. IoT systems are growing daily in the number of interrelated things. As an example, the number of connected things in the world will reach approximately 20.4 billion in 2022, as compared to 8.4 billion connected things in 2020 [3, 4]. The IoT has a profound impact on our lives in a diversity of ways, such as social, commercial and economic. The IoT market is expected to expand its revenues by 892 billion in 2018 and reach 4 trillion in 2025 in terms of the development of the digital economy [5, 6]. Eq (1) The IoT facilitates massive technological development and value-added services across many sectors of our lives, such as smart homes, smart cities, transportation, logistics, smart health, retail, agriculture, and business, smart metering, remote monitoring, and automation of processes [7, 8]. Currently and in the future, IoT applications and services have enormous potential in changing the consumer quality of life in terms of automation, performance, and comfort. Nonetheless, when considering IoT, various types of cyber-attacks and threats are considered problematic to the growth of the IoT [9, 10]. Eq (2) This paper is thus based on IoT security intelligence to ensure that systems and applications are well safeguarded against various cyber-attacks and threats in the IoT http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 233 [11, 12]. The simplest requirement in the IoT network is the security of all the systems, applications, and devices connected to the system. The huge scale of IoT networks presents a host of issues in numerous domains such as device management, data management, computing, security, and privacy, among others [13, 14]. With the expansion of the IoT, different security issues are being raised as potential threats. The new IoT applications, including the above-mentioned, will not be in a position to fulfill the needs of people and society and they will not get all the potential when there is no trusted system in place [15, 16]. The IoT systems are usually based on multiple layers, such as the perception or sensing layer, the networking and data communication layer, the middleware or support layer, and the application layer [17, 18]. Section 3 gives a very fleeting discussion of these layers. To every layer, there is a certain set of tasks and applicable technologies to do in an IoT application, and with each layer comes new issues and security concerns [19, 20]. ELU – ELinear Unit with 0 < α is { Eq (3) Latest Statistics About IoT and Denial of Service (DoS) Attacks To illustrate, the most prevalent IoT attacks are denial of service (DoS) attacks, spoofing attacks, jamming, eavesdropping, data tampering, man in the center attacks, malicious, and others [21]. Therefore, based on the type of security concerns, possible IoT security measures, including authentication, access control, threat and risk forecasting, malware analysis, anomaly or intrusion detection, prevention, etc, would be applicable [22, 23]. The traditional methods of addressing security incidents have ceased to work due to the development of high levels of boom in security threats and attacks, and the complexity in security incidents. As such, a smart security system that is grounded on the latest technologies and can deal with these security issues is highly demanded to ensure the security of the next-generation IoT system [24, 25]. One of the most crucial technologies is Artificial Intelligence (AI) in creating intelligent systems, and it is deemed as a component of the Fourth Industrial Revolution (4IR) [27, 28] as well. Therefore, with the use of AI expertise, especially machine and deep learning, we will be able to identify anomalies or unwanted malicious behavior in the IoT, and, consequently, provide a dynamic security solution, which is continuously enhanced and updated [29, 30]. Usually, machine or deep-learning models are a collection of rules or techniques or intricate transfer functions that derive valuable knowledge or interesting data patterns out of the security data [31, 32]. Eq (4) http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 234 In this way, one can use the ensuing security models to condition the machines to anticipate threats or risks early on, or detect the abnormality in the IoT to create a required defensive policy [33]. Eq (5) Figure 1: Configuration Hierarchy for Mitigating Attacks using IoT Devices [34] Internet of Things (IoTs) and M2M Standards One of the recent smart technologies in the Fourth Industrial Revolution (or Industry 4.0), indicating the continuation of the automation of traditional manufacturing and industrial practices, is the Internet of Things (IoT) [34, 35]. In M2M standards cybersecurityhas many renowned techniques UNSW-NB15 [36, 37], CAIDA [38, 39], ISOT10 [40, 41], ISCX12 [42, 43], CTU-13 [44, 45], CIC-IDS [46, 47], CICDDoS2019 [48, 49]. The IoT is a collection of internet-connected and interconnected devices that can gather and transmit data on a wireless network without the involvement of human operators. Several organizations and research groups define IoT and smart environment in several different ways and perspectives. As an example, [50, 51] describe the IoT as an item of hardware and a digital information stream grounded on RFID tags. According to the Institute of Electrical and Electronics Engineers (IEEE), the IoT is something that is described as a mixture of things that have sensors and make a network linked to the Internet [52, 53]. European Telecommunications Standards Institute (ETSI) defines machine-to-machine (M2M) communications as a type of automated system of communications performing decisions and data operations without the direct involvement of human operators [54, http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 235 55]. Cisco, as the global leader in IT, networking, and cybersecurity solutions, has simplified the concept of the IoE (Internet-of-everything) as a network of people, data, things, and processes [56]. √ According to the RFID (Radio Frequency Identification) group, the IoT is the global cyberspace of interconnected objects that can be addressed uniquely in accordance with standard communication protocols [57, 58]. As stated by Cluster of European research projects on the IoT [59] - Things are active participants in business, information and social processes in which they are facilitated to interact and communicate with each other and with the environment by exchanging data and information sensed about the environment and by responding autonomously to the real/physical world events and influencing it by running processes that cause actions and create services with or without human intervention [60]. Figure 2: Generic Network Architecture Mitigating Attacks using Deep Learning Architecture [61] http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 236 Authors in [62] refer to the interconnection of actuating and sensing devices as IoT, which offers the capability to exchange information across platforms using a single framework, creating a common operating picture to facilitate novel applications. According to [63], IoT is described in three paradigms, including internet-oriented (middleware), things-oriented (sensors), and semantic-oriented (knowledge). ∑(  ) ∑  In [64], the authors introduce a number of research challenges and opportunities associated with IoT security, and in that case, they have taken into account the overall security background of the IoT. In [65], the current situation in the field of IoT security studies, including the related tools, such as IoT modelers and simulators, was outlined. The authors in [66, 67] give a summary of the concept of security, technological issues, security issues, practical solutions, and future research that can be done to protect the IoT. In their survey, they provide their analysis of the current state and the problems of IoT security, considering three layers of architecture, which include the perception layer, the network layer, and the application layer. The survey on IoT security provided by the authors of [68] considers the domain of applications, security threats, and solution architecture. It has introduced a taxonomy on IoT vulnerabilities, attack vectors, attacks that take advantage of these vulnerabilities and methodologies to address them in [69, 70]. The authors in [71] provide a study on IoT security, and their particular study concentrates on the latest threats and vulnerabilities of IoT security as discovered through an extensive evaluation of the existing IoT security research. Table 1: Comparative Analysis of Renowned Public Datasets for IoT Security Dataset No. of features No. of instances Name of attacks Separate traintest set Ref BoTNetIoT-L01 23 1,111,864 UDP, Scan, Syn, Ack, TCP, UDP plain, Combo, and Junk Yes [72] NSL-KDD 42 148,517 DoS, Probe, R2L,and U2R Yes [73] KDD99 42 4,886,431 DoS, Probe, R2L,and U2R Yes [74] UNSWNB15 49 1,540,044 DoS, Fuzzers, Backdoors, Worms, Reconnaissance, Analysis, Exploits, Generic, and Shellcode Yes [75] http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 237 BoTNetIoT-L01 23 1,111,864 UDP, Scan, Syn, Ack, TCP, UDP plain, Combo, and Junk Yes [76] √ ∑ (  ) Literature Review The authors investigate the threat model of IoT systems and analyze the security solutions of IoT systems in terms of machine learning, such as supervised learning, unsupervised learning, and reinforcement learning. They discuss ways of data privacy protection based on learning-based IoT authentication, access control, secure offloading, and malware detection. Eq (9) In [74], the authors consider the security requirements, the attack vectors and other discussions related to computer learning in IoT networks. A survey of computer and deep learning methods of IoT security was introduced. The influence of IoT's new capabilities on protection and privacy in the light of emerging threats, the solutions and challenges that exist were discussed in [77]. √ Eq (10) To build data-driven security systems based on machine and deep learning methods, it is worth knowing the nature of data, as well as diverse types of cyber threats and characteristics associated with them. A number of these datasets are available in the field of cybersecurity. Therefore, we have summarized as NSL-KDD [78], UNSWNB15 [79], DARPA [80, 81], CAIDA [82], ISOT10 [83, 84], ISCX12 [85, 86], CTU13 [87, 88], CIC-IDS [89], CIC-DDoS2019 [90, 91], MAWI [92], ADFA IDS [93], CERT [94,] Depending on the domain of the problem, it is possible to construct the machine and deep learning based model using these datasets. As an example, a deep learning model of a neural network is employed to construct a model on intrusion detection using NSL-KDD [95]. The authors apply such NSL-KDD [96], UNSWNB15 [97] and CIC-IDS [98] in [99] when testing their network intrusion detection model based on machine learning when applied to the IoT environment. Robotics is a very Emerging Field in the past few year’s and it is now a part of almost every field, like Healthcare, Manufacturing, Warfare. Countries like China, Japan, US are very advanced in this field and global leaders in this field. So, every day we hear something new has been invented. So, this shows how fast the field of Robotics is growing and this growth is much higher in the coming future [98]. The Internet is also growing well and people use the Internet in different ways to solve their issues or problems like (Playing heavy games on the cloud) so people use internet in different ways. So, what if we also use it with robotics, means we operate a Robot with internet http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 238 it can solve many problems of Humans. This is not just a thought it is really happening like (Doctor doing a remote surgery from one country to in another country) [99]. But controlling or operating a Robot with the internet or on a network is not that easy it has some problems is well. Its biggest issues are Delay, Data Loss and Security Issues, which make this type of technology Unstable and Unsafe [100]. Eq (11) Intelligent Machines in the Network This paper discusses and explains three control models and also has real-life examples is well. In this review paper, we will discuss those issues that people face when they operate a robot with internet and we will also discuss the method and technologies used in this research paper. We also discuss the Strengths and Weaknesses of the research. And what problems we face in this and how can we operate a robot on a network easily [101, 102]. This paper discusses problems and also explains how other researchers try to solve these problems and issues in the past. This paper also shows some real-world examples, which is give very clarity [103]. Those robots which are operate on the internet mainly have these four parts: Intelligent Mobile Robot, Server for a Robot, Web Server and End-devices (Computer/Laptops/Phones). The user gives commands form an End-device and then the web server receives these commands and web server sends them to the Robot server, which give commands to the intelligent robot and the robot follows these commands and performs a given task. In this we have two servers because one deals with robot side and the other deals internet side, which is also the user side. Eq (12) Control Architectures and Time Delay Time Delay is a very important issue that is discussed in this research paper. When we operate a robot on the internet, we face one big problem, which is a Time delay in simple words, this is called a lag. This lag and delay cause some big issues in this type of technology [104]. There are four types of different delay’s which are discussed in this research paper. Delay in sensor Data, Delay in user decision, Delay in commands Transmission and Delay in Robot Actions. Suppose a user needs a camera or sensor data from a robot to send next command but this will take time to reach data from the robot to user which cause a delay. When a user receives camera or sensor data and user will take some time, which will also cause a delay. Now the user sends a command to the robot but we all know that the http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 239 network always takes some time to send a packet, which will again cause a delay. Eq (13) Table 2: Analysis of renowned techniques used for network-based IoT Ref Aspect Dataset Network Architecture Algorithm/Model Used [105] Response Time Reduction Decreased from hours to seconds for incident response Fast Decision Network Twin Model [106] False Positive Rate Reduction Reduction of 4060% in false positives Multi-Area Microgrid QoS Model [107] Monitoring Availability 24/7 continuous monitoring of IoT devices and networks Strategy Network Auto-Regressive Moving Average with Exogenous Input Model Delay in Robot Actions Now the robot receives a command from the user but it also needs some time to perform an action so this will also cause a delay. Autonomy is a very important topic in robotics, which is discussed in this research paper. Autonomy means a smart robot means how much work a robot can done by itself, autonomy [108]. If a robot can do more work on it than this robot is smart and has high autonomy. A smart robot can think on its own and user only need to give command in simple English and rest is done my robot-like user say move 10m East so robot take this command and them its robot calculate 10m distance and find it’s East on it on no more user interaction is need but on the other side a robot with low autonomy always relay on user it can’t anything on its own [107, 108]. Now the question autonomy can do in a Networked Intelligent Robotic system so high autonomy can reduce the delay because when a robot can do a task on its own, then it is not always relay on user commands. Yes, this will not fully solve the problem but it will surely reduce the delay in Networked Intelligent Robotic systems [109, 110]. Now we have an idea how big a problem delays in Networked Intelligent Robotic systems but this delay becomes worse because the internet is unstable. Data that travels on the internet doesn’t have a constant speed sometimes user commands reach the robot fast, sometimes slow so this make it more difficult to operate a robot on the internet [111, 112]. http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 246 Figure 5: Architecture based on Multi robotic Network Method and Materials The different ways of IoT Networks Implementation through control architectures in the internet direct control, supervisory control and learning control. The major issues are addressed at conception, stability problem, task performance and learning capability. Proposed Network Model The proposed Deep Learning based secure Network is an innovative and automatic system that is learnt on the data produced by the host IoT network and identifies the intrusion of the network when properly trained. The proposed Intrusion Detection System (IDS) has a dynamic connector that is used to create the bond between emulated network and requests originating within the IoT network. An interface module of the feature extractor and network classifier communicates with the emulated network. The feature extractor takes out the feature of the network packets, which are the input of the deep neural network of the proposed methodology. The proposed IDS is a dynamic one and is updated on a regular basis depending on the new features that have been discovered using the classifier Updated module. After the intrusion has been detected, the network classifier forwards it to the mitigation step. This stage mitigates the effects of the intrusion. The proposed intrusion detection system has been designed with the consideration of simplicity and easy deployment. This is the reason why the detection system was created in a modular way. The proposed Deep Intrusion Detection (DID) will be made of three major modules. They include Communication Module (CM), Intrusion Detection Module (IDM) and Mitigation Module (MM). These three modules collaborate with each other. Communication module handles the entire communication between the receiver and the sender which is two-way. Communication of any kind goes through the IDM module. In case of any intrusion detection, the packets are sent to the mitigation module by this module. The communication module is operationalized by the use of Dynamic Connection, Network Emulator, and Interface module. Among the fundamental contributions of the proposed system is the system that is not tied to any http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 247 communication protocol. The protocol used as the background of communication between IoT devices is not a limitation to the DID system. It has been achieved by innovative design in the communication module. This subsection has addressed the working principle of the communication module. Figure 6: Secure Network Model to Mitigate Intrusion Detection The Classifier Updated is one of the novel methods of this proposed intrusion detection system. This module has been developed to minimise the need of humans to update the network classifier. This module has tasks. The initial one is to monitor the network signals that do not belong to the training set. And the second step is to check the correctness of the SVM at all times. The Classifier Updated sounds the alarm when the accuracy of the SVM is less than 90% several times. Once it is fed will the labeled dataset, it will update the SVM on the dataset that resulted in performance degradation together with the former dataset. The intrusion detection system proposed recognizes the intrusion with the help of a deep neural network, and it prevents the attack. Simultaneously, it will create a system response to produce an alert. This whole process involves two activities. The initial step is sensing the nature of the attack and the second step triggers a scheme that is relevant to reduce the effects of the intrusion. These two activities have two elements. The first part is the proposed deep neural network to classify the attack and the second part is the controller to select an appropriate option among a set of options to reduce the consequences of the attack. Dataset Preprocessing and Utilization This experiment has prepared a dataset of 25,000 instances that has been used. The ratio of Regular and Malicious instance of the dataset is 82:18. It implies that 82 percent of 25,000 cases are normal cases, and 18 percent are malicious cases. The http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 248 haphazard blend and there is no willful arrangement. The dataset has been split into the training, testing, and validation dataset in the ratio of 70:15:15. The training and validation data has been involved in the training stage. The testing data is not manipulated in the process of training and validating and is only utilized during the experimentation. Two critical steps in the automatic intrusion detection systems include proper feature selection and feature reduction [154, 155]. The IoT system is secure when the ratio between the transmission and reception is near unity. Nonetheless, a large variation in unity implies that the sender of the data is not transmitting to the right recipient or the receiver is not receiving the data to the right source. It signifies spoofing. It is possible that there exists some intermediate device, which is not listed in the network, that violated the network security and has manipulated the sender and the receiver. The mode of transmission is what defines the communication protocol, the present condition of the communication cycle and the next stage of the communication cycle. A Deep Neural Network that has been trained on the nature of communication in the various modes of transmission determines an intrusion when one of the devices behaves differently when it is under a specific mode of transmission. The result of the quantitative analysis is the qualitative results employing which the intrusion in an IoT network is identified. Figure 7: Deep learning Architecture based on Neural Network In this experiment, a six layer deep neural network is optimized and designed to identify intrusion. Our initial network architecture was a 2-hidden layers network. Nonetheless, the experimental observation demonstrates that the network performance cannot be good using 2 hidden layers. This is the reason as to why we tried various number of hidden layers, and the 6-layer deep neural network produces the most appropriate outcome. Fully Connected Feed Forward Neural Network (FCFFN) architecture [156], shown in Figure 7, was used. Each neuron of a layer is linked to all the other neurons of the other layer in FCFFN architecture. The emphasis and omission of connections is not an extra complexity of implementation. The FCFFN architect was employed to make the implementation of the architect simple and easy. Issues and Opportunities: Dynamic Network Connection and Emulator The probe signals are sent by the dynamic connector. It also sends beacons to all the http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 249 IoT equipment in the specified network. It also scans through the broadcast beacons, sniffs the handshake packets and scans the session requests to determine the communication protocols. It has a table of protocols that are in operation in the IoT space. The dynamic connector dynamically keeps functional network interfaces that intercept communication signals of various devices. The dynamic connector converts the packets that it receives on the various network interfaces. It stores the translation packets in the secondary storage as the cache storage. It then transforms the bit stream into network packets through the relevant protocol based on the protocol protocols it has stored in the protocol table when it receives the bit stream of the IoT devices. The protocol table contains the underlying network protocols of the networks in the IoT. All the related information about the network is contained in it. The proposed approach connects with the network using a creative mechanism of staying in touch rather than directly through the network. And that is via a network emulator. It has Virtual Network Client (VNC) module that retrieves the packets stored in the cache memory of the dynamic connector, selects the necessary network protocol within the protocol table, processes the packets based on the protocols, and lastly sends them to the targeted IoT device. Discussion and Analysis Human beings have long fantasized about the invention of a machine that would be able to activate the human brain. The most important solution is Artificial Intelligence (AI), which simulates the working of the human brain and introduces all the capabilities of the human brain into machines to enable them to think and behave like humans. AI is a branch of computer science that aims at creating computers or systems that are capable of performing tasks that would otherwise require human intelligence such as speech recognition, decision making, language translation and visual perception. Artificial intelligence is capable of causing a fundamental change in many spheres and can change our lifestyle and labor. Nevertheless, it also has such moral and social challenges as privacy issues and the displacement of jobs. Machine learning (ML) is a subfield of AI, in which algorithms and statistical models are generated that can help machines to learn through information and make decisions or predictions without being programmed explicitly. Images, text, or sensor readings can serve a wide range of data sources, and these algorithms can be trained to spot patterns and relationships in the data. The overall aim of ML is to design models capable of making predictions or decisions that would extrapolate to unseen and new data. ML algorithms can be of three main types: supervised, unsupervised and reinforcement. In supervised learning, the machine is given a dataset that is labelled and can also learn how to predict the output when given a new input, depending on the examples that have been given to it in the dataset. Even in unsupervised learning, the machine is provided with unlabeled data and it is left to identify patterns or structures in the data. Clustering or anomaly detection are common tasks to which this form of learning is applied, in which one is interested in determining the existence of groups or anomalies in the data. Finally, reinforcement learning is also the process in which the ML interacts with an environment and is rewarded (or punished) to provide feedback. These are the three categories of ML algorithms that are important in ensuring that machines can learn, adapt, and enhance their decision-making skills. As http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 250 the availability of data and computing resources grows, ML is becoming one of the most important parts of various industries, healthcare, finance, transportation, etc. With its ability to make machines learn based on the data, ML can change the way we treat technology and how we solve complicated issues. Network Threat and Intrusions Blackhole Attack, Distributed Denial-of-Service (DDoS) Attack, Opportunistic Service Attacks, Sinkhole Attack, and Wormhole Attack are the most common and harmful attacks of the IoT networks. This is because of the nature of the data, which has changed. The speed of the data flow, the character of the attack, and the target of the attackers have varied. Consequently, conventional systems that are based on rules are no longer as effective as before the IoT revolution. This is why deep neural networks suggested in this paper are the most recent methods of big data processing, pattern identification, and making relevant predictions [157, 158]. Distributed Denial-of-Service (DDoS) and (OSA)Attack A distributed denial-of-service attack or DDoS attack is an ill-intentioned effort to interrupt normal traffic of a particular server, service, or network with large quantities of Internet traffic to hamper or cripple the target or the infrastructure it relies on. This kind of attack is referred to as a distributed denial-of-service attack (DDoS). In an opportunistic service attack, the malicious device gains the trustworthiness gradually by initially giving out services of great trustworthiness but later changing to offer services of lesser quality to the effect of making more profit for itself [159]. Sinkhole Attack (SHA) and Wormhole Attack (WHA) One type of attack is a sinkhole attack in which a compromised node tries to lure network traffic by advertising its own fictitious routing update. Among the adverse impacts of sinkhole attacks, one can mention the fact that they can be employed to carry out additional attacks, including selective forwarding attacks, acknowledgment spoofing attacks, drops, or routing information modifying attacks [160]. In a wormhole tunnel attack, two attacking devices also use a virtual private connection to communicate with one other in order to execute an attack. The packets sent to the victim device are received and first forwarded through the wormhole and it is replayed later on. Table 4: Analysis of Intrusion Detection-based Models Using DL Classifier Delay Attack Accuracy (%) Precision (%) Recall (%) F1-Score (%) IDR RNN t= 5ms Blackhole 52.34 59.89 23.48 62.19 59.89 DT DDoS 54.32 54.32 17.45 57.34 54.32 NBB Sinkhole 57.92 57.92 16.74 57.78 57.92 RF Wormhole 60.21 60.21 24.64 61.28 60.21 SVM Average 53.68 53.68 19.42 55.79 92.41 CNN Opportunistic 54.32 91.07 57.97 17.16 54.72 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 251 Service Figure 8: Performance Analysis of Intrusion Detection based Models Using DL Figure 9: Variation in Performance based on Intrusion Detection Models Using DL Conclusion and Future Directions This article examined how intelligent machines can be controlled through the Internet, and what challenges and opportunities come with this technology. Because Internet access is now available almost everywhere, people can operate robots from far away, http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 252 creating new possibilities in factories, hospitals, homes, security, and even entertainment. However, controlling robots through the Internet is not always smooth—problems like delays, lost data, slow connections, and security risks can affect how well these systems work. Researchers have developed several methods to reduce these problems. These include predicting the robot’s movements ahead of time, using special control signals that stay stable even when the network is slow, and using mathematical techniques to keep the robot steady despite delays. Although these methods help, relying only on low-level, real-time user control is still difficult when the network is unreliable. This is why improving the robot’s own intelligence is important. By giving robots more autonomy such as the ability to avoid network intrusion mitigation, plan routes, understand their surroundings, and make simple decisions—they can continue working even when network communication is slow or interrupted. High-level control methods let users give simple instructions like ―go to this location‖ instead of controlling every movement. At the same time, better user interfaces, including mobile apps and web platforms, make operating these robots much easier. Many real-world examples show how useful networked robots can be: robots that help with surgery, wheelchairs that assist disabled users, museum tour-guide robots, security robots, and robots used for space or scientific missions. These applications demonstrate how robots can safely support humans in places that are dangerous, difficult, or far away. Looking to the future, progress in this field will depend on improving robot intelligence, making communication more reliable, adding better learning abilities, and ensuring stronger security. Trends such as cloud robotics, mobile control interfaces, and human-friendly interaction will help robots become more connected to everyday life. As these technologies develop, robots will not only follow remote commands but also act as smart partners that help humans in many tasks. Overall, although challenges still exist, the development of networked intelligent robots is moving quickly. Enhancing robot autonomy, stabilizing communication, and improving how humans interact with robots will be key to unlocking their full potential. This paper’s findings help support continued research in this important and growing field. DECLARATIONS Acknowledgement: We appreciate the generous support from all the contributor to the research and their different affiliations. Funding: No funding body in the public, private, or nonprofit sectors provided a particular grant for this research. Availability of data and material: In the approach, the data sources for the variables are stated. Authors' contributions: Each author participated equally in the creation of this work. Conflicts of Interest: The authors declare no conflict of interest. Consent to Participate: Yes Consent for publication and Ethical approval: Because this study does not include human or animal data, ethical approval is not required for publication. All authors have given their consent. http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 253 REFERENCES Islam, M. K., Rahman, M. M., Ali, M. S., Mahim, S. & Miah, M. S. Enhancing lung abnormalities diagnosis using hybrid dcnnvitgru model with explainable ai: A deep learning approach. Image Vis. Comput. 142, 104918 (2024). Amanullah, M.A.; Habeeb, R.A.A.; Nasaruddin, F.H.; Gani, A.; Ahmed, E.; Nainar, A.S.M.; Akim, N.M.; Imran, M. Deep learning and big data technologies for IoT security. Comput. Commun. 2020, 151, 495–517. Li, Y.; Zuo, Y.; Song, H.; Lv, Z. Deep learning in security of internet of things. IEEE Internet Things J. 2021, 9, 22133–22146. Alferidah, D.K.; Jhanjhi, N. Cybersecurity impact over bigdata and iot growth. In Proceedings of the 2020 International Conference on Computational Intelligence (ICCI), Bandar Seri Iskandar, Malaysia, 8–9 October 2020; IEEE: New York, NY, USA, 2020; pp. 103–108. [Google Scholar] Yılmaz, H.E.; Sirel, A.; Esen, M.F. The impact of internet of things self-security on daily business and business continuity. In Research Anthology on Business Continuity and Navigating Times of Crisis; IGI Global: Hershey, PA, USA, 2022; pp. 695–712. [Google Scholar] Harada, R.; Shibata, N.; Kaneko, S.; Honda, K.; Terada, J.; Ishida, Y.; Akashi, K.; Miyachi, T. Quick Suppression of DDoS Attacks by Frame Priority Control in IoT Backhaul with Construction of Mirai-based Attacks. IEEE Access 2022, 10, 22392–22399. Batten, L.M.; Moonsamy, V.; Alazab, M. Smartphone applications, malware and data theft. In Computational Intelligence, Cyber Security and Computational Models; Springer: New York, NY, USA, 2016; pp. 15–24. [Google Scholar] Tariq U, Ahmed I, Bashir AK, et al. A critical cybersecurity analysis and future research directions for the internet of things: a comprehensive review. Sensors. 2023;23(8):4117. Kaspersky. IT threat evolution Q3 2023[R/OL]. (2023–12–01)[2024–05–11]. https:// secur elist. com/ ksb-2023-stati stics/ 111156/. Ma B, Zhang Z, Chen Y, Wu JX. The defense method for code-injection attacks based on instruction set randomization. J Cyber secur. 2020;5(4):30–43. Sihao SHAO, Qing GAO, Sen MA, et al. Progress in research on buffer overflow vulnerability analysis technologies. J Softw. 2018;29(5):1179–98. Qiang LIU, Yapin DENG, Zheng XU, et al. Research on hidden trojan horse detection technology. Comput Eng. 2006;32(1):180–2. Xiao-Meng F, Qiu-Ye S, Bing-Yu W, Jia-Wen G. The coordinated cyber physical power attack strategy based on worm propagation and false data injection. Acta Automatica Sinica. 2022;48(10):2429–41. Yadav B, Tokekar S. Recent innovations and comparison of deep learning techniques in malware classification: a review. Int J Inf Secur Sci. 2021;9(4):230–47. Shaukat K, Alam T M, Luo S, et al. 2021 A review of time-series anomaly detection techniques: a step to future perspectives//Advances in Information and Communication: proceedings of the 2021 future of information and communication conference (FICC), Volume 1. Springer International Publishing, 865–77. http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 254 Sung A H, Xu J, Chavez P, et al. 2004 Static analyzer of vicious executables (save)[C]//20th annual computer security applications conference. IEEE, 326– 34. Nataraj L, Karthikeyan S, Jacob G, et al. 2011 Malware images: visualization and automatic classification[C]//Proceedings of the 8th international symposium on visualization for cyber security. 1–7. Smmarwar SK, Gupta GP, Kumar S. Android malware detection and identification frameworks by leveraging the machine and deep learning techniques: a comprehensive review. Telemat Inf Rep. 2024;12:100130 Gao, J., Wang, H., & Shen, H. (2020). Task failure prediction in cloud data centers using deep learning. IEEE Transactions on Services Computing, 1111–1116. https:// doi.org/10.1109/BigData47090.2019.9006011 ... Guo, X., Aviles, G., Liu, Y., Tian, R., Unger, B. A., Lin, Y. H. T., & Kampmann, M. (2020). Mitochondrial stress is relayed to the cytosol by an OMA1–DELE1– HRI pathway. Nature, 579(7799), 427–432. Jindal, A., Aujla, G. S., & Kumar, N. (2019). SURVIVOR: A blockchain-based edgeas-a-service framework for secure energy trading in SDN-enabled vehicle-togrid environment. Computer Networks, 153, 36–48. Khoramshahi, M., & Billard, A. (2019). A dynamical system approach to task adaptation in physical human-Network interaction. Autonomous Networks, 43(4), 927–946. Lin, K., Li, Y., Sun, J., Zhou, D., & Zhang, Q. (2020). Multi-sensor fusion for a body sensor network in a medical human-Network interaction scenario. Information Fusion, 57, 15–26. Manogaran, G., Baskar, S., Hsu, C. H., Kadry, S. N., Sundarasekar, R., Kumar, P. M., & Muthu, B. A. (2021). FDM: Fuzzy-optimized Data Management Technique for Improving Big Data Analytics. IEEE Transactions on Fuzzy Systems, 29(1), 177–185. Manogaran, G., Shakeel, P. M., Priyan, R. V., Chilamkurti, N., & Srivastava, A. (2019). Ant colony optimization-induced route optimization for enhancing the driving range of electric vehicles. International Journal of Communication Systems, e3964. https://doi.org/10.1002/dac.3964 Gesture-based human-Network interaction for human assistance in manufacturing. The International Journal of Advanced Manufacturing Technology, 101(1), 119–135. Nguyen, N. T., Liu, B. H., Pham, V. T., & Huang, C. Y. (2016). Network under limited mobile devices: A new technique for mobile charging scheduling with multiple sinks. IEEE Systems Journal, 12(3), 2186–2196. Preeth, S. S. L., Dhanalakshmi, R., & Shakeel, P. M. (2020). An intelligent approach for energy efficient trajectory design for mobile sink based IoT supported wireless sensor networks. Peer-to-Peer Networking and Applications, 13(6), 2011–2022. Priyan, M. K., & Devi, G. U. (2018). Energy-efficient node selection algorithm based on node performance index and random waypoint mobility model on the Internet of vehicles. Cluster Computing, 21(1), 213–227. Ramprasad, L., & Amudha, G. (2014, February). Spammer detection and tagging based user generated video search system—A survey. In International Conference on Information Communication and Embedded Systems (ICICES2014) (pp. 1–5). http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 255 Imtiaz, M. A., Amir, A., Bakhet, S., Siddique, H., & Rizwan, S. M. (2025). An Optimal Diabetic Retinopathy Detection and Classification Approach based on integrated Hybrid Convolutional Neural Networks (CNNs). Spectrum of Engineering Sciences, 3(2). Ayub, N., Uzair, M., Din, I., Ali, A., Khan, H., Yuldashev, F., & Makhmudov, S. (2025). An Efficient Human Activity Recognition (HAR) Model Based on Convolutional Neural Networks for Computing Devices Aiming to Reduce Latency and Tackle the Inactivity of Gadgets. Engineering, Technology & Applied Science Research, 15(6), 28885-28890. Khan, H. M. S., Hayat, C. M. A., Tayyab, H., & Ali, K. (2024). An Enhanced Cost Effective and Scalable Network Architecture for Data Centers. Spectrum of Engineering Sciences, 2(4), 1-32. Rafay, A., Salman, W., Yahya, G., & Malik, U. (2024). SD Network based on Machine Learning: An Overview of Applications and Solutions. Spectrum of Engineering Sciences, 2(4), 150-165. Javed, M. A., Ahmad, M., Ahmed, J., Rizwan, S. M., & Tariq, A. (2025). An Enhanced Machine Learning based Data Privacy and Security Mitigation Technique: An Intelligent Federated Learning (FL) Model for Intrusion Detection and Classification System for Cyber-Physical Systems in Internet of Things (IoTs). Spectrum of Engineering Sciences, 3(2), 377-401. Aziz, R., Mehmood, A., Tariq, A., Nasim, F., Farooq, U., Naqvi, S. A. A., & Khan, H. (2025). Critical Evaluation of Data Privacy and Security Threats: An Intelligent Federated Learning-based Intrusion Detection System Poisoning Attack and Defense for Cyber-Physical Systems its Issues and Challenges Related to Privacy and Security in IoT. The Asian Bulletin of Big Data Management, 5(1), 73-84. M. Gondal, Z. Hameed, M. U. Shah, H. Khan, "Cavitation phenomenon and its effects in Francis turbines and amassed adeptness of hydel power plant", In 2019 2nd International Conference on Computing, Mathematics and Engineering Technologies (iCoMET), IEEE., pp. 1-9, Mar. 2019 Liang, Y., Ur Rahman, S., Shafaqat, A., Ali, A., Ali, M. S. E., & Khan, H. (2024). Assessing sustainable development in E-7 countries: technology innovation, and energy consumption drivers of green growth and environment. Scientific Reports, 14(1), 28636. Waleed, R., Ali, A., Tariq, S., Mustafa, G., Sarwar, H., Saif, S., ... & Uddin, I. (2024). An Efficient Artificial Intelligence (AI) and Internet of Things (IoT's) Based MEAN Stack Technology Applications. Bulletin of Business and Economics (BBE), 13(2), 200-206. U. Hashmi, S. A. ZeeshanNajam, "Thermal-Aware Real-Time Task Schedulabilty test for Energy and Power System Optimization using Homogeneous Cache Hierarchy of Multi-core Systems", Journal of Mechanics of Continua and Mathematical Sciences., vol. 14, no. 4, pp. 442-452, Mar. 2023 Mumtaz, J., Bakhet, S., Javed, A., Naz, A., Rashail, M., & Khan, H. (2025). An Intelligent Diagnosis and Tumor Segmentation Method based on MRI Images Using Pre-trained Deep Convolutional Neural Networks (CNNs). The Asian Bulletin of Big Data Management, 5(1), 147-163 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 262 insightful Machine Learning based Privacy-Preserving Technique for Federated Learning. The Asian Bulletin of Big Data Management, 4(4), 332343. Farooq, I., Ahmed, S. A., Ali, A., Warraich, M. A., Aqeel, M., & Khan, H. (2024). Enhanced Classification of Networks Encrypted Traffic: A Conceptual Analysis of Security Assessments, Implementation, Trends and Future Directions. The Asian Bulletin of Big Data Management, 4(4), 500-522. Fawy, K. F., Rodriguez-Ortiz, G., Ali, A., Jadeja, Y., Khan, H., Pathak, P. K., ... & Rahman, J. U. (2025). Catalytic exploration metallic and nonmetallic nanocatalysts, properties, role in photoelectrochemistry for sustainable applications. Reviews in Inorganic Chemistry, (0). Liaqat, M. S., Sharif, N., Ali, A., Khan, H., Ahmed, H. N., & Khan, H. (2024). An Optimal Analysis of Cloud-based Secure Web Applications: A Systematic Exploration based on Emerging Threats, Pitfalls and Countermeasures. Spectrum of engineering sciences, 2(5), 427-457. Adil, M. U., Ali, S., Haider, A., Javed, M. A., & Khan, H. (2024). An Enhanced Analysis of Social Engineering in Cyber Security Research Challenges, Countermeasures: A Survey. The Asian Bulletin of Big Data Management, 4(4), 321-331. Maqsood, M., Dar, M. M., Javed, M. A., & Khan, H. (2024). A Survey on the Internet of Medical Things (IOMT) Privacy and Security: Challenges Solutions and Future from a New Perspective. The Asian Bulletin of Big Data Management, 4(4), 355-368. Khawar, M. W., Salman, W., Shaheen, S., Shakil, A., Iftikhar, F., & Faisal, K. M. I. (2024). Investigating the most effective AI/ML-based strategies for predictive network maintenance to minimize downtime and enhance service reliability. Spectrum of Engineering Sciences, 2(4), 115-132. Ahmad, J., Salman, W., Amin, M., Ali, Z., & Shokat, S. (2024). A Survey on Enhanced Approaches for Cyber Security Challenges Based on Deep Fake Technology in Computing Networks. Spectrum of Engineering Sciences, 2(4), 133-149. Ayub, N., Ejaz, A., Hassan, B., Hussain, M. Z., Nadeem, M., Sabir, L., & Fatima, S. (2025). An Efficient Machine Learning And Deep Learning Based Deep Packet Security Framework For Detection Of Computing Network Faults In The Iots. Spectrum of Engineering Sciences, 3(5), 659-674. Ayub, N., Imtiaz, M. A., Ali, E., Alqahtani, A. M., Ali, A., Ashurov, M., ... & Law, F. L. (2025). A Decision Framework for Intra Task Fixed Priority INTEL PXA270 Distributed Architecture for Soft RT-Applications Based on Deep Learning. Engineering, Technology & Applied Science Research, 15(3), 23553-23558. Ayub, N., Waheed, A., Ahmad, S., Akbar, M. H. A., Fuzail, M. Z., & Hashmi, A. H. (2025). Strengthening Network Security: An Efficient DL Enabled Data Protection and Privacy Framework for Threat Mitigation and Vulnerabilities Detection in IoT Network. Annual Methodological Archive Research Review, 3(6), 1-25. Farooq, M., Younas, R. M. F., Qureshi, J. N., Haider, A., & Nasim, F. (2025). Cyber http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 263 security risks in DBMS: Strategies to mitigate data security threats: A systematic review. Spectrum of engineering sciences, 3(1), 268-290. Ayub, N., Habib, Z., Bakhet, S., Riaz, S., Rizwan, S. M., Abid, M., ... & Khan, H. (2025). An Optimal Ai & Deep Learning Mechanism For Mitigating Hacking Threat Identification Using Secure Network Infrastructure Based On Linux And Software-Defined Network (Sdn). Spectrum of Engineering Sciences, 3(5), 675-687. Aslam, I., Tariq, W., Nasim, F., Khan, H., Khawaja, M. F., Ahmad, A., & Nawaz, M. S. (2025). A Robust Hybrid Machine Learning based Implications and Preventions of Social Media Blackmailing and Cyber bullying: A Systematic Approach. Ayub, N., Anwer, M. A., Iqbal, A., Rizwan, S. M., Shahbaz, A., Abid, M. H., & Rafi, S. (2025). Enhanced ML Framework based on Artificial Neural Network for countermeasures of Data Protection and Network Vulnerabilities Detection in Industrial Internet of Things. Annual Methodological Archive Research Review, 3(5), 410-431. Akhtar, M. H., Ali, A., Ali, S., Nasim, F., Aziz, M. H., Khan, H., & Naqvi, S. A. A. (2025). A Novel Machine Learning Approach for Database Exploitation to Enhance Database Security: A Survey. Spectrum of Engineering Sciences, 3(2), 26-57. Jabeen, T., Mehmood, Y., Khan, H., Nasim, M.F. and Naqvi, S.A.A., 2025. Identity Theft and Data Breaches How Stolen Data Circulates on the Dark Web: A Systematic Approach. Spectrum of engineering sciences, 3(1), pp.143-161. Ali, H., Ayub, N., Irfan, A., Fayyaz, S., Masood, H., Ahmad, A., ... & Khan, H. (2025). A Unified AI-powered Social Media Platform for Intelligent Scheduling and Data Driven Analytics Using Multi-Layered Artificial Neural Networks (ANNs): https://doi. org/10.5281/zenodo. 17572988. Annual Methodological Archive Research Review, 3(11), 94-134. Khawar, M. W., Ayub, N., Shaheen, S., Iftikhar, B., Masood, H., Ahmad, A., & Khan, H. (2025). An Efficient system based on Artificial Intelligence for the Detection and Mitigation of network Intrusion using encrypted traffic protocols: A Systematic Approach. Annual Methodological Archive Research Review, 3(11), 32-71. Musharraf, S. T., Masab, M. M., Ayub, N., Murtaza, S., Ullah, H., Ahmad, A., ... & Khan, H. (2025). An Efficient Artificial Intelligence-Based Early Prediction of Heart Attack Using Deep Learning CNN and SVM Models: https://doi. org/10.5281/zenodo. 17551611. Annual Methodological Archive Research Review, 3(10), 265-301. Ayub, N., Alghamdi, T., Din, I., Ali, A., Khan, H., Ganiyeva, O., & Makhmudov, S. (2025). An Enhanced Artificial Intelligence and Deep Learning Assisted Breast Cancer Classification and Diagnosis Based on the Internet of Medical Things (IOMTs). Engineering, Technology & Applied Science Research, 15(6), 30612-30616. Niaz, H. U., Qadeer, Q. B. Q., Niaz, H., Mansib, H., Awais, M., & Khan, H. (2025). Artificial Intelligence Assisted Autonomous Unmanned Aerial Vehicles (UAVs) and Aerial drones based on Machine Vision for Enhancing Remote http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 264 Sensing of Precision crop Health Monitoring. The Asian Bulletin of Big Data Management, 5(4), 155-177. Abdullah, M. M., Ghafoor, U., Qadeer, Q. B., Khadim, F., Khan, H. S., Ahmad, A., & Khan, H. (2025). An Efficient of Artificial Intelligence based Brain Tumor Diagnosis and Classification: An Advance Medical Diagnosis Approach. The Asian Bulletin of Big Data Management, 5(2), 208-242. Bacha, A., Sehar, H., Naseem, S., & Khan, M. I. (2024). FEDERATED LEARNING FOR THREAT INTELLIGENCE SHARING: A PRIVACY-PRESERVING COLLABORATIVE DEFENSE MODEL. Spectrum of Engineering Sciences, 656-664. Mahmood, F., Shehroz, M., Ansari, Z., & Rauf, F. (2024). A Survey of SoftwareDefined Networks Based on Advance Machine Learning Based Techniques. Spectrum of Engineering Sciences, 2(4), 232-257. Aqeel, N., Alam, A., Bhatti, Z., & Amir, A. (2024). A Survey on Tor’s Multi Layer Architecture and Web Implications in Dark Web. Spectrum of Engineering Sciences, 2(4), 212-231. Gul, W., Nawaz, A., Hamaz, M. T., & Khan, H. AN EFFICIENT MODEL FOR THE SELECTION OF LEADERSHIP COMPETENCIES AND PERFORMANCE IMPROVEMENT FOR THE SUCCESS OF TRANSPORTATION PROJECTS, JOURNAL OF MECHANICS OF CONTINUA AND MATHEMATICAL SCIENCES Vol.-16, No.-5, May (2021) pp 49-65 https://doi.org/10.26782/jmcms.2021.05.00005 Hamayun Khan,Sheeraz Ahmed,S. Farhan Haider Shah,Rehan Ali Khan,Zeeshan Najam,Hasnain Abbas,Asif Nawaz,Zubair Aslam Khan, JOURNAL OF MECHANICS OF CONTINUA AND MATHEMATICAL SCIENCES, Vol.- 15, No.-8, August (2020) pp 628-646 https://doi.org/10.26782/jmcms.2020.08.00053 Ali, G., Shahbaz, H., Hassan, M. A., Ahmad, M., & Waleed, M. (2024). An Enhanced Approach of Exploring Digital Economy Using Modern Computer Networks. Spectrum of Engineering Sciences, 2(4), 292-312. Nawaz, S., Salman, W., Shahid, U., Khokhar, M. L., Iqbal, M. Z., & Hamid, A. (2024). A Survey on Latest Trends and Technologies of Computer Systems Network. Spectrum of Engineering Sciences, 2(4), 85-114. Ismaeel, S., Saleemi, H., Amir, U., Ashraf, S., & Hamza, A. (2024). A Detailed Review of latest Trends, Technologies Applications Raza, A., Khan, H., & Rehman, S. U. (2023). Computational Analysis of Nanomaterials for Energy Storage. International Journal of Advanced Sciences and Computing, 143-154. Ali, M., Khan, H., & Rehman, S. U. (2023). Edge Computing for Low-Latency IoT Applications. International journal of advanced sciences and computing, 3849. Malik, A., Khan, H., Ali, A., Nawaz, A., & Ahmad, S. The Impact of Climatic Parameters on the Streamflows & Future Sustainable Hydro-Energy Generation Predicting Streamflows for the 21st Century Under Climate Change Scenarios. Zainab, Khan, H., Din, I. U., Tariq, M. I., Khalid, A., & Naz, A. (2023, May). An http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 265 Efficient Implementation of an IoT-Based Smart Home Security System. In International Conference on Computing & Emerging Technologies (pp. 249-259). Cham: Springer Nature Switzerland. Ali, M., Khan, H., Din, I. U., Tariq, M. I., & Javed, A. Design and Implementation Role of Middleware in Shared Network Environments: A Systematic Review. Securing the Digital Realm, 229-243. Muhammad Anas, Muhammad Atif Imtiaz,Saad Khan,Arshad Ali,Noor Fatima Naghman,Hamayun Khan,Sami Albouq, AN ADVANCED MACHINE LEARNING (ML) ARCHITECTURE FOR HEART DISEASE DETECTION, PREDICTION AND CLASSIFICATION USING MACHINE LEARNING, Cont.& Math. Sci, Vol.20, No.3 https://doi.org/10.26782/jmcms.2025.03.00005 Faruqui, N.; Yousuf, M.A.; Whaiduzzaman, M.; Azad, A.; Barros, A.; Moni, M.A. LungNet: A hybrid deep-CNN model for lung cancer diagnosis using CT and wearable sensor-based medical IoT data. Comput. Biol. Med. 2021, 139, 104961. Wójcicki, K.; Biegańska, M.; Paliwoda, B.; Górna, J. Internet of Things in Industry: Research Profiling, Application, Challenges and Opportunities—A Review. Energies 2022, 15, 1806. Hasan, M.K.; Akhtaruzzaman, M.; Kabir, S.R.; Gadekallu, T.R.; Islam, S.; Magalingam, P.; Hassan, R.; Alazab, M.; Alazab, M.A. Evolution of industry and blockchain era: Monitoring price hike and corruption using BIoT for smart government and industry 4.0. IEEE Trans. Ind. Inform. 2022, 18, 9153– 9161. Zhao, Y.; Lian, Y. Event-driven Circuits and Systems: A Promising Low Power Technique for Intelligent Sensors in AIoT Era. IEEE Trans. Circuits Syst. II Express Briefs 2022, 69, 3122–3128. Soldatos, J.; Gusmeroli, S.; Malo, P.; Di Orio, G. Internet of things applications in future manufacturing. In Digitising the Industry Internet of Things Connecting the Physical, Digital and Virtual Worlds; River Publishers: Delft, The Netherlands, 2022; pp. 153–183. [Google Scholar] Sharma, R.; Arya, R. Security threats and measures in the Internet of Things for smart city infrastructure: A state of art. Trans. Emerg. Telecommun. Technol. 2022, 1, e4571. Rondon, L.P.; Babun, L.; Aris, A.; Akkaya, K.; Uluagac, A.S. Survey on enterprise Internet-of-Things systems (E-IoT): A security perspective. Ad Hoc Netw. 2022, 125, 102728. Williams, R.; McMahon, E.; Samtani, S.; Patton, M.; Chen, H. Identifying vulnerabilities of consumer Internet of Things (IoT) devices: A scalable approach. In Proceedings of the 2017 IEEE International Conference on Intelligence and Security Informatics (ISI), Beijing, China, 22–24 July 2017; IEEE: New York, NY, USA, 2017; pp. 179–181. [Google Scholar] Hilton, S. Dyn Analysis Summary of Friday October 21 Attack. Dyn Blog. 2016. Available online: https://dyn.com/blog/dyn-analysis-summary-of-fridayoctober-21-attack (accessed on 1 January 2023). http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 12 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 266 Nayak, J.; Meher, S.K.; Souri, A.; Naik, B.; Vimal, S. Extreme learning machine and bayesian optimization-driven intelligent framework for IoMT cyber-attack detection. J. Supercomput. 2022, 78, 14866–14891. Husnain, M.; Hayat, K.; Cambiaso, E.; Fayyaz, U.U.; Mongelli, M.; Akram, H.; Ghazanfar Abbas, S.; Shah, G.A. Preventing MQTT Vulnerabilities Using IoT-Enabled Intrusion Detection System. Sensors 2022, 22, 567. Zheng, Y.; Li, Z.; Xu, X.; Zhao, Q. Dynamic defenses in cyber security: Techniques, methods and challenges. Digit. Commun. Netw. 2022, 8, 422–435. Nimbalkar, P.; Kshirsagar, D. Analysis of rule-based classifiers for IDS in IoT. In Data Science and Security; Springer: New York, NY, USA, 2021; pp. 461– 467. [Google Scholar] Al-Garadi, M.A.; Mohamed, A.; Al-Ali, A.K.; Du, X.; Ali, I.; Guizani, M. A survey of machine and deep learning methods for internet of things (IoT) security. IEEE Commun. Surv. Tutor. 2020, 22, 1646–1685.