Search

Wireless sensor network (WSN) comprises several sensor nodes which are randomly deployed in the interesting region for data gathering operations. Since the sensor nodes are majorly deployed in harsh and accessible open regions, it is highly vulnerable to attacks and thereby necessitates an effective intrusion detection system (IDS). The rise of machine learning (ML) algorithms finds useful in the design of IDS, particularly for energy constrained WSN. In this view, this study introduces Optimal Multilayer Perceptron (OMLP) with Dragonfly Algorithm (DA) for intrusion detection in WSN. The major intention of OSVM is to determine the possible intrusions and identifies the class type effectually. The OMLP technique is mainly used to determine the weights and bias of the MLP by the DA. The utilization of DA results in the effective choice of weight and bias values and thereby led to improved detection performance. The experimental validation of the OMLP model takes place on the benchmark NSL KDDCup 99 dataset. The obtained experimental outcome reported the proficient performance of the OMLP model with higher accuracy and detection rate of 94.21% and 95.18% respectively.

Wireless sensor networks (WSN) hold numerous battery operated, compact sized, and inexpensive sensor nodes, which are commonly employed to observe the physical parameters in the target environment. As the sensor nodes undergo arbitrary placement in the open areas, there is a higher possibility of affected by distinct kinds of attacks. For resolving the issue, intrusion detection system (IDS) is developed. This paper presents a new optimal Support Vector Machine (OSVM) based IDS in WSN. The presented OSVM model involves the proficient selection of optimal kernels in the SVM model using whale optimization algorithm (WOA) for intrusion detection. Since the SVM kernel gets altered using WOA, the application of OSVM model can be used for the detection of intrusions with proficient results. The performance of the OSVM model has been investigated on the benchmark NSL KDDCup 99 dataset. The resultant simulation values portrayed the effectual results of the OSVM model by obtaining a superior accuracy of 94.09% and detection rate of 95.02%.

Any superficial skin growth that does not resemble the surrounding area is referred to as skin lesion. It can occur in the form of mole, bump, cyst, rash or other changes that can be classified either as primary or secondary lesion. While primary skin lesions correspond to those changes in color or texture, secondary lesions occur as a primary lesion progression. Skin lesion image segmentation and classification at the early stages can help the patients recover through proper medication and treatment. Many algorithms for segmentation and classification are available in the literature but they all fail to extract lesion boundaries perfectly and classify them with more accuracy. To improve the reliability of the skin image segmentation and classification, we propose to use decision trees and random forest algorithms in this works and compare them with different data sets. The proposed method can generate high-resolution feature maps that can help to preserve the spatial details of the image. While tested against the ISIC 2017 and HAM10000 dataset, we found that the proposed method is more accurate as compared to the existing algorithms in this domain and is also very robust to artifacts or hair fibers present in the skin images.

In the current scenario, monitoring and tracking of the Alzheimer’s patient who suffer from lapses of memory and concentration creates various problems to the caretakers. Many existing systems failed to produce a secure optimal solution for tracking the Alzheimer’s patient. The proposed approach is designed and implemented to track and locate the geographical region of the Alzheimer patient. To achieve an optimal solution, a secured transmission is required between the IOT device and cloud. In this regard, an enhanced RSA algorithm is designed to encrypt and transfer the sensitive data between the devices. Our experimental results show better performance when compared with the existing system.

Clustering is one important method for extending the network energy in WSNs. It mires grouping of sensor nodes into clusters and recruiting cluster Head (CHs) for all the clusters. CHs rally the data from relevant cluster’s nodes and forward the accumulated data to cluster Master. WSN are inclusive to various attacks in which Blackhole a kind of Denial of Service (DoS) attack is very crucial to encounter and defend the mechanism. The attackers elect a set of neighbor nodes in the network and reconstruct the route to downfall the received packets instead of forwarding them to the sink node, which will emerge in a situation where packets enter the blackhole area but never reaches the destination resulting in higher end-to-end delay and decline in the throughput. In this paper, We use SHA(Secure Hashing algorithm) Cluster based Futuristic Method for detecting black hole attack in WSNs. By using the proposed technique we adapts filtering technique to the attacker and prevent from the BlackHole attacker and improving the performance such as throughput, delay and packet delivery ratio.

The Vehicular Ad hoc Networks are a different type of ad hoc network, which contain communicating entities that are in motion with different velocity along with lacking in infrastructure. Hence, this work needs the establishment of more consistent end-to-end communicating pathways and efficiently transferring data. We have presented a new security-aware routing algorithm called Deep learning-based prediction. The presented scheme is more efficient and reliable against different kinds of attacks such as black hole and malicious node penetration attempts to the entire network. It basically depends upon route link error recover and shortest path by using Dijkstra’s algorithm. The aim of this scheme is to identify malicious data and black hole nodes. The simulation results of eDC-NC are compared with already existing techniques called COPE in terms of Energy consumption (EC), throughput and network lifetime (NLT). This helps to achieve better performance and filter unwanted data by applying deep learning filtering technique.

Any system or network is always susceptible to attacks. If an adversary has access to the network, they may disrupt the network in several ways. Sybil attack, selective forwarding, tampering, etc., are some of the attacks that disrupt the network. If the sensor network is employed in critical areas such as warzones or border protection systems, any attack or breach of security may prove to be fatal. Therefore, proper security mechanisms have to be integrated into the system to ensure that the data remains secure. In our project, we intend to identify and isolate the compromised nodes that selectively forwards data packets. The proposed protocol uses a reputation based system to determine the integrity of the neighboring nodes. Once a compromised node has been identified, that node is excluded from further communication after verification from base station.