Energy-Efficient Hybrid Learning for Secure Wireless Sensor Networks

The rapid growth of the Internet and information technologies and also the diminution of the price of hardware components like wireless sensors leads to the fast growth in Wireless Sensor Network (WSN). The WSN is a bunch/group of sensors that are located across the area, in order to monitor and determine environmental conditions such as temperature, humidity, vibrations, wind, water levels, pollution levels. WSNs are susceptible to major attacks like Denial of Service (DOS), Wormhole attack, Sinkhole attack, etc. The main threat to the WSN is because of the broadcast nature of the nodes in the network. Therefore, the security of WSNs is the essential part that must be done. Hence, to overcome these problems or threats, we are trying to detect it using AI technology. With the expanding fields of Machine Learning and Deep Learning, we can apply various algorithms in order to classify different types of attacks. Once we detect the attack properly, we can prevent it accordingly. We are using WSN-DS. It has 4 classes of attacks which are Grayhole, Blackhole, TDMA(Scheduling), and Flooding which comes under the category of DOS attacks. In this chapter, we have analysed and compared the accuracies of 5 main machine learning classification algorithms. Also, we have analysed the 1 deep learning algorithm. The ANN (Artificial Neural Network) and 5 machine learning algorithms have been trained on the dataset. Furthermore, we have used K-fold cross-validation to get even more accurate predictions. After analysis of these algorithms, we came to the point that, the Machine Learning algorithms like Random Forest, Support Vector Machines and Deep Learning algorithm, namely, Artificial Neural Network can help us for detecting intrusions in the system or network. This will help researchers in designing their own machine learning model on top of our suggested model.

The vast literature on the wireless sensor research community contains many valuable proposals for managing energy consumption, the most important factor that determines sensors’ lifetime. Interesting researches have been facing this requirement by focusing on the extension of the entire network lifetime: either by switching between node states (active, sleep) or by using energy-efficient routing. We argue that a better extension of the network lifetime can be obtained if an efficient combination of management mechanisms can be performed at the energy of each single sensor and at the load distribution over the network. Considering these two accuracy levels (i.e., node and network), this paper presents a new approach that uses cost functions to choose energy-efficient routes. In particular, by making different energy considerations at a node level, our approach distributes routing load, avoiding, thus, energy-compromised hotspots that may cause network disconnections. The proposed cost functions have completely decentralized and adaptive behavior and take into consideration the end-to-end energy consumption, the remaining energy of nodes, and the number of transmissions a node can make before its energy depletion. Our simulation results show that, though slightly increasing path lengths from sensor to sink nodes, some proposed cost functions (1) improve significantly the network lifetime for different neighborhood density degrees, while (2) preserving network connectivity for a longer period of time.

Wireless Sensor Networks (WSN) are composed of small, low cost, resource-constrained computing nodes equipped with low power wireless transceivers. Their unattended nature and possible deployment in hostile environmental conditions poses several challenges in ensuring that a WSN is formed effectively and survives long enough to fulfill its function. WSN nodes are vulnerable to many Denial of Services(DoS) attacks. To secure WSN still against attacks many innovative solutions/protocols have been proposed. However, a number of ambiguities exist in the definition of these DoS attacks. To remove these ambiguities and to clarify our specifications we represented the DoS attacks in a formal modeling notation Z. The specifications are written in a bottom-up approach starting from the basic definitions (like node, message etc), specifying simple operations of WSN (like receive, transmit etc), complex operations (like eavesdrop key, node capture etc) and going all the way up to formally define the DoS attacks. Different WSN routing protocols are then specified to confirm their vulnerability against DoS attacks. Use of Formal methods, thus, confirms how vulnerable a particular routing protocol is against a particular attack. This is a novel work and to the best of our knowledge DoS attacks have not yet been formally defined.

Wireless Sensor Networks (WSN) is a network of sensors, actuators, mobile and wearable devices that have processing and communication modules to monitor physical and environmental conditions. Currently millions of these type of smart devices serving in many fields like military, environment, and health services. Due to their unique deployment places even in hostile territories WSN are subject to various kinds of attacks. Self configuration, autonomous device addition, network connection and resource limitation are the main features of WSN that makes it highly prone to network attacks. Denial of Service (DoS) attacks which targets the availability of a WSN system is one of the most potent threat to which a WSN must be resilient in order to continue operations. This studies aim to analyze and classify the WSN DoS threats and their countermeasures. Based on the survey we present the best approach to designing a WSN resilient against DoS attacks.

As a relatively recent and emerging technology, wireless sensor networks (WSNs) are beginning to be deployed frequently in a wide variety of environments ranging from military and emergency environments to natural and embedded environments. For example, a security system may be deployed using WSNs as outlined in M. Turon, J. Suh (Apr. 2005), and P. Dutta., et al., (Apr. 2005). Furthermore, WSNs are of particular interest to adversaries due to their frequent deployments in open and unprotected environments. Preventive mechanisms can sometimes be applied to protect WSNs from an assortment of attacks. However, in many cases, a more sophisticated methodology needs to be applied for situations in which intrusions or attacks cannot necessarily be anticipated in advance. In such instances, an intrusion detection system is warranted. For WSNs, one of the primary concerns deals with availability of the network and individual nodes. Denial of service (DoS) attacks are a particularly great threat to WSNs. The effects of a DoS attack are described extensively in A.D. Wood and J.A. Stankovic (Oct. 2002),and http://www.tinyos.net. WSNs are particularly vulnerable to failure under such attacks because of their limited energy, processing capacity, and storage. An effective DoS attack merely has to deplete the resources of the nodes to render them unavailable. In this paper we design and implement a preliminary intrusion detection system (IDS) for WSNs that addresses the security concern of DoS attacks and fits the demands and restrictions of WSNs.

At recent years, Wireless Sensor Networks (WSNs) had a widespread range of applications in many fields related to military surveillance, monitoring health, observing habitat and so on. WSNs contain individual nodes that interact with the environment by sensing and processing physical parameters. Sometimes, sensor nodes generate a big amount of sequential tuple-oriented and small data that is called Data Streams. Data streams usually are huge data that arrive online, flowing rapidly in a very high speed, unlimited and can’t be controlled orderly during arrival. Due to WSN limitations, some challenges are faced and need to be solved. Extending network lifetime and reducing energy consumption are main challenges that could be solved by Data Mining techniques. Clustering is a common data mining technique that effectively organizes WSNs structure. It has proven its efficiency on network performance by extending network lifetime and saving energy of sensor nodes. This paper develops a grid-density clustering algorithm that enhances clustering in WSNs by combining grid and density techniques. The algorithm helps to face limitations found in WSNs that carry data streams. Grid-density algorithm is proposed based on the well-Known K-Means clustering algorithm to enhance it. By using Matlab, the grid-density clustering algorithm is compared with K-Means algorithm. The simulation results prove that the grid-density algorithm outperforms K-Means by 15% in network lifetime and by 13% in energy consumption.

Internet of Things (IoT) technology is getting more importance as many organizations from different sectors are moving toward adopting this technology. It introduces an efficient ways of collecting, processing and distributing data. Wireless sensor networks (WSNs) play a vital role in the expanding growth of internet of things (IoT). However, DoS (denial-of-service) attacks are a major threat of WSNs. In this study, graph theory based model has been proposed to analyse the security of WSN under DoS attacks. The proposed model describes and assesses the security of WSN when it encounter DoS attacks by finding the attacks success probability, attack cost, attack impact, mean-time-to-compromise, attack risk and return-on-attack values. The impact of mitigation methods toward strengthening the WSN security has been described. smart home scenario WSN has been tacking as an example. The results showed that the capability of the proposed model to analysis the effects of DoS attacks in WSN. Also it showed the impact of mitigation methods in improving WSN security.

Network reprogramming is a crucial service in wireless sensor networks (WSNs) that relies on epidemic strategy for spreading software updates by just having a local view of the networks. Securing the process of network reprogramming is essential in some certain WSNs applications, state-of-the-art secure network reprogramming protocols for WSNs aim for the efficient source authentication and integrity verification of code image, however, due to the resource constrains of WSNs, existing secure network reprogramming protocols are vulnerable to Denial of Service (DoS) attacks when sensor nodes can be compromised (insider DoS attacks). In this paper, we identify different types of DoS attacks exploiting the epidemic propagation strategies used by Deluge and propose corresponding analysis models to attempt to quantify the cost of these attacks damage. Simulation further shows the impact of insider DoS attacks on network reprogramming in WSNs.

In this paper, we present a novel Energy-Aware Data-Centric Routing algorithm for wireless sensor networks, which we refer to as EAD. We discuss the algorithm and its implementation, and report on the performance results of several workloads using the network simulator ns-2. EAD represents an efficient energy-aware distributed protocol to build a rooted broadcast tree with many leaves, and facilitate the data-centric routing in wireless micro sensor networks. The idea is to turn off the radios of all leaf nodes and let the non-leaf nodes be in charge of data aggregation and relaying tasks. The main contribution of this protocol is the introduction of a novel approach based on a low cost backbone provisioning within a wireless sensor network in order to turn off the non backbone nodes and save energy without compromising the connectivity of the network, and thereby extending the network lifetime. EAD makes no assumption on the network topology, and it is based on a residual power. We present an extensive simulation experiments to evaluate the performance of our EAD forwarding-to-parent routing scheme over a tree created by a single EAD execution, and compare it with the routing scheme over a regular Ad hoc On-Demand Distance Vector (AODV) Protocol. Last but not least, we evaluate the performance of our proposed EAD algorithm and compare it to the Low-Energy Adaptive Clustering Hierarchy (LEACH) protocol, a cluster-based, energy-aware routing protocol specifically designed for sensor networks. Our results indicate clearly that EAD outperforms AODV and LEACH in energy conservation, throughput, and network lifetime extension.

Wireless sensor networks (WSNs) are appearing to be one of the most promising pervasive applications now. In some scenarios such as commercial building surveillance or military reconnaissance, WSNs meet a lot of challenges in security, among which, user authentication is one of the most crucial. Two-factor authentication has been used in WSNs since M. L. Das's scheme in 2009, and has been attracting researchers' interest because of its robustness and flexibility, which suits resource-constrained WSNs very well. However, lots of researchers pointed out diverse security flaws in two-factor authentication scheme and came up with their improved versions. In this paper, we show that few of the existing protocols are resilient to Denial of Service (DoS) attack. And under the scenario of capturing a sensor node, some security pitfalls including gateway impersonation attack and forgery attack still exist in existing protocols. Then we propose an enhanced two-factor user authentication scheme which employs two novel techniques: lightweight pre-authentication based on Merkle hash tree and personalized secret parameters for sensor nodes. Through analysis of security and performance, we show that our proposed scheme is equipped with more security features, especially protection from DoS attack launched not only by adversaries but also by greedy users, and resilience to gateway impersonation and sensor node forgery after sensor nodes are compromised. Moreover proposed scheme maintains an acceptable performance and could adapt dynamically to DoS attacking scenarios for designated applications of WSNs.

Energy management in Wireless Sensor Networks (WSNs) remains a critical challenge, particularly in clustering processes. This article compares three optimization algorithms—Grasshopper Optimization Algorithm (GOA), Bat Algorithm (BA), and Whale Optimization Algorithm (WOA)—to achieve energy-efficient clustering and extend network lifetime. Initial cluster head placement is performed using K-means clustering, and a novel cost function is introduced that considers energy consumption and node distribution, enhancing the network’s efficiency and resilience. The algorithms are evaluated across three scenarios with varying base station (BS) placements. In the simplest scenario, with the BS centrally located, GOA slightly outperforms WOA in extending network lifetime, although WOA remains competitive. BA, while energy-efficient, lags behind GOA and WOA. As complexity increases with BS placement at the edge, WOA demonstrates superior energy management, delaying node death and extending network lifetime more effectively than GOA and BA. In the most challenging scenario, where the BS is placed in a remote corner, WOA emerges as the most effective algorithm, maintaining network performance and balancing energy consumption for the longest duration. GOA, while relatively strong, shows faster network lifetime decline, particularly in later stages, whereas BA faces significant challenges, leading to quicker node failures. Overall, this study highlights the importance of efficient clustering and optimization for prolonging WSN lifetimes. WOA excels in complex scenarios, while GOA leads in simpler environments. Integrating K-means clustering with the novel cost function enhances algorithm performance, contributing to the development of resource-efficient WSNs, especially in resource-constrained settings.

Wireless sensor networks (WSN) is widely utilized for collecting data related to physical parameters from the environment. Security remains a challenging issue in the design of WSN. Security in WSN from Denial of Service (DoS) attack is an important security risk. This study introduces an artificial flora optimization algorithm with functional link neural network (AFOA-FLNN) model for DoS attack classification in WSN. The presented AFOA-FLNN model initially undergoes data pre-processing to transform the data into meaningful way. Secondly, the FLNN model is utilized for the effective recognition and classification of intrusions in WSN. Finally, the AFOA is exploited for optimally tuning the parameters involved in the FLNN model and results in enhanced performance. In order to demonstrate the better outcomes of the AFOA-FLNN model, a wide-ranging experimentation assessment on test data and the results pointed out the improved outcomes of the AFOA-FLNN model.

Wireless Sensor Networks (WSNs) continue to face two major challenges: energy and security. As a consequence, one of the WSN-related security tasks is to protect them from Denial of Service (DoS) and Distributed DoS (DDoS) attacks. Machine learning-based systems are the only viable option for these types of attacks, as traditional packet deep scan systems depend on open field inspection in transport layer security packets and the open field encryption trend. Moreover, network data traffic will become more complex due to increases in the amount of data transmitted between WSN nodes as a result of increasing usage in the future. Therefore, there is a need to use feature selection techniques with machine learning in order to determine which data in the DoS detection process are most important. This paper examined techniques for improving DoS anomalies detection along with power reservation in WSNs to balance them. A new clustering technique was introduced, called the CH_Rotations algorithm, to improve anomaly detection efficiency over a WSN’s lifetime. Furthermore, the use of feature selection techniques with machine learning algorithms in examining WSN node traffic and the effect of these techniques on the lifetime of WSNs was evaluated. The evaluation results showed that the Water Cycle (WC) feature selection displayed the best average performance accuracy of 2%, 5%, 3%, and 3% greater than Particle Swarm Optimization (PSO), Simulated Annealing (SA), Harmony Search (HS), and Genetic Algorithm (GA), respectively. Moreover, the WC with Decision Tree (DT) classifier showed 100% accuracy with only one feature. In addition, the CH_Rotations algorithm improved network lifetime by 30% compared to the standard LEACH protocol. Network lifetime using the WC + DT technique was reduced by 5% compared to other WC + DT-free scenarios.

As the deficiencies of present Wireless Sensor Network Routing Protocol in extending network lifetime and improving network performance, a kind of Routing Algorithm based on Multi-optimization Function is proposed in order to achieve the optimization goals of balancing network node energy consumption and extending network lifetime. By introducing parameters like available energy of nodes, routing hops and physical distance between nodes into routing selection function, optimum path is set up. Therefore, the integrated optimization of Wireless Sensor Network performance is realized. As NS2 emulation result shows, this algorithm has achieved higher transmission reliability, lower energy consumption and much longer network lifetime. Compared with traditional directed diffusion algorithm, the rate of data transmission success has risen by 15.3%, utilization ratio of network energy by 9.7%, and network lifetime by 37.8%. Routing Algorithm based on Multi-optimization Function has more advantages in Wireless Sensor Network.

In wireless sensor networks, different applications feature different requirements in terms of such performance metrics as sensing coverage and data reporting latency. In most applications, it is usually sufficient to provide a Desired Sensing Coverage (DSC) lower than full coverage at any instance with the guarantee that the whole area will eventually be covered within a specified delay bound. Due to the fact that these applications are also expected to run for longer periods of time and at the same time battery recharging and replacement are costly, energy consumption in wireless sensor networks should be minimized while achieving the application goals. In this article, we propose a novel framework for application-specific data gathering which exploits a trade-off between coverage and latency, thereby minimizing energy consumption and extending the network lifetime. The proposed energy-efficient, constant-time, randomized scheme, called Coverage-Adaptive raNdom SEnsor sElection (CANSEE), selects a subset of k sensors to report at each round so as to fulfill the application-specific requirement of desired sensing coverage and bounded latency, instead of always guaranteeing full coverage and minimum latency. We present a probabilistic model to estimate: (i) the connectivity of those selected k sensors and the number of additional sensors needed to guarantee connectivity; (ii) a lower bound on k in each round; and (iii) the probability of almost surely having k data reporters using the Chernoff bound. The immediate event detection capability achieved by the proposed CANSEE scheme is also analyzed to compare the performance of our framework with other data gathering schemes that allow 100% coverage. Simulation results demonstrate that our framework leads to a significant conservation of energy (and thus extended network lifetime) with a small trade-off between coverage and data reporting latency, yet providing the required data reporting capability.