Biblio

Password-based authentication is the most common authentication method for either online or offline system. Password composition policies become too burdensome and put the user in a state of struggle to remember their password. Thus, most of the user save their password on the browser or even list it down in their personal gadgets. Therefore, a multi agent-based password management application have been developed to helps user in keeping their password safely. However, multi-agent system facing security issues such as man in the middle attack, data modification and eavesdropping. This paper proposed a security protocol for multi agent-based architecture in order to reduce potential threats. The security protocol focuess on the authentication of mobile agents, data transmission and the data local protection. The communication channels are secured using cryptography techniques.

Mobile systems are always growing, automatically they need enough resources to secure them. Indeed, traditional techniques for protecting the mobile environment are no longer effective. We need to look for new mechanisms to protect the mobile environment from malicious behavior. In this paper, we examine one of the most popular systems, Android OS. Next, we will propose a distributed architecture based on IDS-AM to detect intrusions by mobile agents (IDS-AM).

Mobile Ad-Hoc Networks (MANETs) are prone to many security attacks. One such attack is the blackhole attack. This work proposes a simple and effective application layer based intrusion detection scheme in a MANET to detect blackholes. The proposed algorithm utilizes mobile agents (MA) and wtracert (modified version of Traceroute for MANET) to detect multiple black holes in a DSR protocol based MANET. Use of MAs ensure that no modifications need to be carried out in the underlying routing algorithms or other lower layers. Simulation results show successful detection of single and multiple blackhole nodes, using the proposed detection mechanism, across varying mobility speeds of the nodes.

Software agents represent an assured computing paradigm that tends to emerge to be an elegant technology to solve present day problems. The eminent Scientific Community has proved us with the usage or implementation of software agent's usage approach that simplifies the proposed solution in various types to solve the traditional computing problems arise. The proof of the same is implemented in several applications that exist based on this area of technology where the software agents have maximum benefits but on the same hand absence of the suitable security mechanisms that endures for systems that are based on representation of barriers exists in the paradigm with respect to present day industry. As the application proposing present security mechanisms is not a trivial one as the agent based system builders or developers who are not often security experts as they subsequently do not count on the area of expertise. This paper presents a novel approach for protecting the infrastructure for solving the issues considered to be malicious host in mobile agent system by implementing a secure protocol to migrate agents from host to host relying in various elements based on the enhanced Trusted Platforms Modules (TPM) for processing data. We use enhanced extension to the Java Agent Development framework (JADE) in our proposed system and a migrating protocol is used to validate the proposed framework (AVASPA).

Technological advances in wearable and implanted medical devices are enabling wireless body area networks to alter the current landscape of medical and healthcare applications. These systems have the potential to significantly improve real time patient monitoring, provide accurate diagnosis and deliver faster treatment. In spite of their growth, securing the sensitive medical and patient data relayed in these networks to protect patients' privacy and safety still remains an open challenge. The resource constraints of wireless medical sensors limit the adoption of traditional security measures in this domain. In this work, we propose a distributed mobile agent based intrusion detection system to secure these networks. Specifically, our autonomous mobile agents use machine learning algorithms to perform local and network level anomaly detection to detect various security attacks targeted on healthcare systems. Simulation results show that our system performs efficiently with high detection accuracy and low energy consumption.

The rapid increase of connected devices and the major advances in information and communication technologies have led to great emergence in the Internet of Things (IoT). IoT devices require software adaptation as they are in continuous transition. Multi-agent based solutions offer adaptable composition for IoT systems. Mobile agents can also be used to enable interoperability and global intelligence with smart objects in the Internet of Things. The use of agents carrying personal data and the rapid increasing number of connected IoT devices require the use of security protocols to secure the user data. Elliptic Curve Cryptography (ECC) Algorithm has emerged as an attractive and efficient public-key cryptosystem. We recommend the use of ECC in the proposed Broadcast based Secure Mobile Agent Protocol (BROSMAP) which is one of the most secure protocols that provides confidentiality, authentication, authorization, accountability, integrity and non-repudiation. We provide a methodology to improve BROSMAP to fulfill the needs of Multi-agent based IoT Systems in general. The new BROSMAP performs better than its predecessor and provides the same security requirements. We have formally verified ECC-BROSMAP using Scyther and compared it with BROSMAP in terms of execution time and computational cost. The effect of varying the key size on BROSMAP is also presented. A new ECC-based BROSMAP takes half the time of Rivest-Shamir-Adleman (RSA) 2048 BROSMAP and 4 times better than its equivalent RSA 3072 version. The computational cost was found in favor of ECC-BROSMAP which is more efficient by a factor of 561 as compared to the RSA-BROSMAP.

Mobile Ad-Hoc Networks are dynamic and wireless self-organization networks that many mobile nodes connect to each other weakly. To compare with traditional networks, they suffer failures that prevent the system from working properly. Nevertheless, we have to cope with many security issues such as unauthorized attempts, security threats and reliability. Using mobile agents in having low level fault tolerance ad-hoc networks provides fault masking that the users never notice. Mobile agent migration among nodes, choosing an alternative paths autonomous and, having high level fault tolerance provide networks that have low bandwidth and high failure ratio, more reliable. In this paper we declare that mobile agents fault tolerance peculiarity and existing fault tolerance method based on mobile agents. Also in ad-hoc networks that need security precautions behind fault tolerance, we express the new model: Secure Mobil Agent Based Fault Tolerance Model.

Cover time measures the time (or number of steps) required for a mobile agent to visit each node in a network (graph) at least once. A short cover time is important for search or foraging applications that require mobile agents to quickly inspect or monitor nodes in a network, such as providing situational awareness or security. Speed can be achieved if details about the graph are known or if the agent maintains a history of visited nodes, however, these requirements may not be feasible for agents with limited resources, they are difficult in dynamic graph topologies, and they do not easily scale to large networks. This paper introduces a set-based form of heading (directional bias) that allows an agent to more efficiently explore any connected graph, static or dynamic. When deciding the next node to visit, agents are discouraged from visiting nodes that neighbor both their previous and current locations. Modifying a traditional movement method, e.g., random walk, with this concept encourages an agent to move toward nodes that are less likely to have been previously visited, reducing cover time. Simulation results with grid, scale-free, and minimum distance graphs demonstrate heading can consistently reduce cover time as compared to non-heading movement techniques.
 

Wireless Sensor Networking is one of the most promising technologies that have applications ranging from health care to tactical military. Although Wireless Sensor Networks (WSNs) have appealing features (e.g., low installation cost, unattended network operation), due to the lack of a physical line of defense (i.e., there are no gateways or switches to monitor the information flow), the security of such networks is a big concern, especially for the applications where confidentiality has prime importance. Therefore, in order to operate WSNs in a secure way, any kind of intrusions should be detected before attackers can harm the network (i.e., sensor nodes) and/or information destination (i.e., data sink or base station). In this article, a survey of the state-of-the-art in Intrusion Detection Systems (IDSs) that are proposed for WSNs is presented. Firstly, detailed information about IDSs is provided. Secondly, a brief survey of IDSs proposed for Mobile Ad-Hoc Networks (MANETs) is presented and applicability of those systems to WSNs are discussed. Thirdly, IDSs proposed for WSNs are presented. This is followed by the analysis and comparison of each scheme along with their advantages and disadvantages. Finally, guidelines on IDSs that are potentially applicable to WSNs are provided. Our survey is concluded by highlighting open research issues in the field.

An intrusion detection system (IDS) inspects all inbound and outbound network activity and identifies suspicious patterns that may indicate a network or system attack from someone attempting to break into or compromise a system. A networkbased system, or NIDS, the individual packets flowing through a network are analyzed. In a host-based system, the IDS examines at the activity on each individual computer or host. IDS techniques are divided into two categories including misuse detection and anomaly detection. In recently years, Mobile Agent based technology has been used for distributed systems with having characteristic of mobility and autonomy. In this working we aimed to combine IDS with Mobile Agent concept for more scale, effective, knowledgeable system.