International Journal of Computer and Applications

Artificial neural networks (ANNs) constitute a class of flexible nonlinear models designed to mimic biological neural systems. In this entry, we introduce ANN using familiar econometric terminology and provide an overview of ANN modeling approach and its implementation methods. † Correspondence: Chung-Ming Kuan, Institute of Economics, Academia Sinica, 128 Academia Road, Sec. 2, Taipei 115, Taiwan; ckuan@econ.sinica.edu.tw. †† I would like to express my sincere gratitude to the editor, Professor Steven Durlauf, for his patience and constructive comments on early drafts of this entry. I also thank Shih-Hsun Hsu and Yu-Lieh Huang for very helpful suggestions. The remaining errors are all mine.

Artificial neural networks

MANETs have unique characteristics like dynamic topology, wireless radio medium, limited resources and lack of centralized administration, as a result, they are vulnerable to different types of attacks in different layers of protocol stack. Each node in a MANET is capable of acting as a router. Routing is one of the aspects having various security concerns. In this paper, we will present survey of common Denial-of-Service (DoS) attacks on network layer namely Wormhole attack, Black hole attack and Gray hole attack which are serious threats for MANETs. We will also discuss some proposed solutions to detect and prevent these attacks. As MANETs are widely used in many vital applications, lots of research work has to be done to find efficient solutions against these DoS attacks that can work for different routing protocols.

DoS Attacks in Mobile Ad Hoc Networks: A Survey

A survey on physical unclonable function (PUF)-based security solutions for Internet of Things

Due to wireless communication, dynamic topology, limited resources and lack of centralized administration, MANETs are vulnerable to various types of DoS attacks on network layer. In Gray hole and Black hole attacks malicious nodes deliberately disrupt data transmission in the network by sending incorrect routing information. It is a challenge to keep the communication route free from such attackers. In this paper, we propose a scheme for Ad-hoc On-demand Distance Vector (AODV) protocol, in which an intermediate node detects the malicious node sending false routing information, routing packets are used not only to pass routing information, but also to pass information about malicious nodes. The proposed scheme not only detects but also removes malicious node by isolating it, to make safe and secure communication.

A Novel Approach for GrayHole and BlackHole Attacks in Mobile Ad Hoc Networks

Summary: In this era of wireless devices, Mobile Ad-hoc Network (MANET) has become an indivisible part for communication for mobile devices. Therefore, interest in research of Mobile Ad-hoc Network has been growing since last few years. In this paper we have discussed some basic routing protocols in MANET like Destination Sequenced Distance Vector, Dynamic Source Routing, Temporally-Ordered Routing Algorithm and Ad-hoc On Demand Distance Vector. Security is a big issue in MANETs as they are infrastructure-less and autonomous. Main objective of writing this paper is to address some basic security concerns in MANET, operation of wormhole attack and securing the wellknown routing protocol Ad-hoc On Demand Distance Vector. This article would be a great help for the people conducting research on real world problems in MANET security.

/pdf/manet-routing-protocols-and-wormhole-attack-against-aodv-4pu4t7stl6.pdf

MANET Routing Protocols and Wormhole Attack against AODV

Sensor-based activity recognition in the context of ambient assisted living systems: A review

The system envisioned is an automatic collision detection and warning system relying on GPS module and a GSM modem. The vehicle to be safeguarded is to be fitted with the system sturdily ensuring good mechanical coupling with the entire chassis. In the case of an accident the system detects it using the fact that the vehicle would be suddenly decelerated in such a condition. An accelerometer continuously monitors the acceleration of the vehicle and will detect decelerations greater than threshold value and send the data to the microcontroller via an ADC. The controller compares this with the threshold set value and immediately sends an SOS message to preset numbers. With this message the controller also transmits the GPS coordinates of the vehicle which it continuously obtains from the GPS module. This system will highly aid the search and rescue of vehicles that have met with an accident.

/pdf/automatic-accident-alert-and-safety-system-using-embedded-8di7x4qvxd.pdf

Automatic Accident Alert and Safety System using Embedded GSM Interface

A significant challenge to provide services to the inhabitant in a smart environment resides in the effective implementation of models. Most of the proposed models are conceptual and lacks practical consideration. Human activity recognition is one of the most challenging tasks to offer the solution for ambient assisted living. In this work, we explore a time series classification problem - human activity recognition. Total of nine machine learning and deep learning algorithms implemented and evaluated using the same dataset. The results are analyzed using different parameters. This paper aims at providing help to select a practical machine learning approach for activity recognition process in ambient assisted living systems. The comparative analysis shows that in deep learning, Long Short-Term Memory (LSTM) network performed best with a classification accuracy of 92%. In machine learning, Logistic Regression and Gradient Boosting came out with a classification accuracy of greater than 90%, and others came out the worst with classification accuracy less than 90%.

Performance Analysis of Supervised Machine Learning Algorithms to Recognize Human Activity in Ambient Assisted Living Environment

The primary objective of the ambient assisted living (AAL) is to enhance the living standards of the aged person by presenting various assistive solutions. The services presented to the occupants in AAL are localization, context awareness, activity and behaviour monitoring, personalization, overcoming loneliness, privacy, security, and many more. To endeavour the user with these services, human-activity recognition and human behaviour modeling is a fundamental necessity. Many researchers have assumed only the periodic and consistent behaviour of the users while designing their frameworks; in a real scenario, human behaviour is unpredictable. There arises a need for converting the proposal into the working prototype due to the enormous difference between the available solutions and actual needs. This paper introduces a novel hybrid framework for human behaviour modeling in AAL, which accommodates the varying nature of activity and behaviour of the inhabitant. The proposal uses feedback-based mechanism to recursively append new events and behaviour, such that they are classified into normal or abnormal human behaviour. The proposed solution uses the pattern mining and clustering mechanism to determine the anomalies in addition to the camera sensor, which detects human fall to deal with any critical situation. Once the action is recognized, it is more convenient to extend essential services like medical assistance or emergency response.

Ashish Patel

Papers

MANET Routing Protocols and Wormhole Attack against AODV

Sensor-based activity recognition in the context of ambient assisted living systems: A review

Automatic Accident Alert and Safety System using Embedded GSM Interface

Performance Analysis of Supervised Machine Learning Algorithms to Recognize Human Activity in Ambient Assisted Living Environment

Real-time human behaviour monitoring using hybrid ambient assisted living framework