Search results

Security is one of the major challenges in the design and implementation of protocols for mobile ad hoc networks (MANETs). In such systems, the cooperation between nodes is one of the important principles being followed in the current research works to formulate various security protocols. Many existing works assume that mobile nodes will follow prescribed protocols without deviation. However, this is not always the case, because these networks are subjected to a variety of malicious attacks. Since there are various models of attack, trust routing scheme can guarantee security and trust of the network. The purpose of this paper is to propose a novel trusted routing model for mitigating attacks in MANETs.

The proposed model incorporates the concept of trust into the MANETs and applies grey relational analysis theory combined with fuzzy sets to calculate a node’s trust level based on observations from neighbour nodes’ trust level, these trust levels are then used in the routing decision-making process.

In order to prove the applicability of the proposed solution, extensive experiments were conducted to evaluate the efficiency of the proposed model, aiming at improving the network interaction quality, malicious node mitigation and enhancements of the system’s security.

The proposed solution in this paper is a new approach combining the fundamental basics of fuzzy sets with the grey theory, where establishment of trust relationships among participating nodes is critical in order to enable collaborative optimisation of system metrics. Experimental results indicate that the proposed method is useful for reducing the effects of malicious nodes and for the enhancements of system’s security.

The purpose of this paper is to ensure power efficiency in wireless sensor networks (WSNs) through a new framework-oriented middleware, based on a biologically inspired mechanism that uses an evolutionary multi-objective optimization algorithm. The authors call this middleware framework multi-objective optimization for wireless sensor networks (MONet).

In MONet, the middleware level of each network node autonomously adjusts its routing parameters according to dynamic network conditions and seeks optimal trade-offs among performance objectives for a balance of its global performance. MONet controls the cooperation between agents (network nodes) while varying transmission paths to reduce and distribute power consumption equitably on all the sensor nodes of network. MONet-runtime uses a modified TinyDDS middleware platform.

Simulation results confirm that MONet allows power efficiency to WSN nodes while adapting their sleep periods and self-heal false-positive sensor data.

The framework implementation is lightweight and efficient enough to run on resource-limited nodes such as sensor nodes.

The collective intelligence emerging from behaviors of social insects has become an inspiration source that is impossible to avoid; guiding researchers in various domains to solutions of insolvent problems by traditional approaches. These behaviors are made possible because of the interactions individual‐individual and individual‐environment, representing support on which cooperative work within the same group is based and allowing emergence at macroscopic level of sophisticated achievements. Many models were inspired by this new and very promising vision, to find simple rules, leading mobile, autonomous robots with limited capacities in their environment to realize tasks, like those of: browsing, collecting or self‐assembly. In this context, the purpose of this paper is to suggest a method, making global behavior evolve within an homogeneous agent‐robots community to accomplish heap‐formation task based on appointment principle in changing environment which can be very difficult. Control device, comparable to the functioning of cellular automaton containing sensory‐motor rules, is then used to arbitrate between some given elementary attitudes with which each agent‐robot initially is equipped.

Evolutionary approach using genetic algorithm based on reverse emergence principle seeks, then, for cellular automaton whose arbitration succeeds to realize this adaptive oriented grouping task.

Rules as simulation results obtained according to reactive model of multi‐agent systems are provided, compared with those found at the ants and commented.

Discovered rules are adaptive; it means when training ground becomes more difficult, agent‐robots become more flexible by decreasing thresholds conditioning rules application. If environment state continues to turn into harsh, robots are able to seek for another direction to start new heap formation somewhere else. Such zones are like Saharan region, airports or supermarkets.