Abstract not found

Sensor networks arise as one of the most promising technologies for the next decades. The recent emergence of small and inexpensive sensors based upon microelectromechanical system (MEMS) ease the development and proliferation of this kind of networks in a wide range of real-world applications. Multi-Agent systems (MAS) have been identified as one of the most suitable technologies to contribute to this domain due to their appropriateness for modeling autonomous self-aware sensors in a flexible way. Firstly, this survey summarizes the actual challenges and research areas concerning sensor networks while identifying the most relevant MAS contributions. Secondly, we propose a taxonomy for sensor networks that classifies them depending on their features (and the research problems they pose). Finally, we identify some open future research directions and opportunities for MAS research. 1.

Market-based approaches proposed recently proved to be promising for competitive resource sharing in peer-to-peer and grid computing. Many approaches leverage on the Vickrey-Clarke-Groves (VCG) mechanism to achieve incentive compatibility which embraces truthful bidding of participating agents. This paper addresses a deficiency of VCG that to the best of our knowledge has not been studied. When one or more agents possess a large portion of the market share of resource, a monopoly situation arises. Applying VCG mechanism does not lead to an allocation because the second price cannot be mathematically determined. Using both theoretical and simulation analysis, we show the importance of addressing this problem. Our results show that monopoly situation arises in many types of market settings, from auction to exchange, and with a relatively high occurrence rate. To address this, we propose a new pricing method suitable for many market settings that achieve budget balanced and economic efficiency but relax the strategy proof property.

We consider a tactical data network with limited bandwidth, in which each agent is tracking objects and may have value for receiving data from other agents. The agents are selfinterested and would prefer to receive data than share data. Each agent has private information about the quality of its data and can misreport this quality and degrade or otherwise decline to share its data. The problem is one of interdependent value mechanism design because the value to one agent for the broadcast of data on an object depends on the quality of the data, which is privately known to the sender. A recent two-stage mechanism due to Mezzetti (2004) can be modified to our setting. Our mechanism achieves efficient bandwidth allocation and provides incentive compatibility by conditioning payments on the realized value for data shared between agents.

In this paper we use a reinforcement learning algorithm with the aim to increase the autonomous lifetime of a Wireless Sensor Network (WSN) and decrease latency in a decentralized manner. WSNs are collections of sensor nodes that gather environmental data, where the main challenges are the limited power supply of nodes and the need for decentralized control. To overcome these challenges, we make each sensor node adopt an algorithm to optimize the efficiency of a small group of surrounding nodes, so that in the end the performance of the whole system is improved. We compare our approach to conventional ad-hoc networks of different sizes and show that nodes in WSNs are able to develop an energy saving behaviour on their own and significantly reduce network latency, when using our reinforcement learning algorithm.