Conference Paper
Gridbased SensorDCSP.
Conference: IJCAI03, Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence, Acapulco, Mexico, August 915, 2003
Source: DBLP
 Citations (4)
 Cited In (0)

Article: Distributed Constraint Satisfaction : Foundations of Cooperation in MultiAgent Systems / M. Yokoo.
[Show abstract] [Hide abstract]
ABSTRACT: Constraint satisfaction is a wellknown technique in artificial intelligence which can be applied to a large variety of problems. Many algorithms have been developed, and many implementations of these algorithms are available and have been studied. Hence, it is not surprising that the recent developments of dcstributed systems are also successfully extended to this area. Chapter 1 defines constraint satisfaction problems in general and presents the most important algorithms (e.g. backtracking, weakcommitment search, hillclimbing et al.). This introduction is used as the background for distributed constraint satisfaction problems. The given problem is assigned to agents, one agent is responsible for one variable of the problem, constraints are to be considered between different agents. One agent knows all constraints relevant to its variable.Based on these assumptions, the next chapters discuss different methods and algorithms to solve this problem: – asynchronous backtracking (chapter 3), – asynchronous weakcommitment search (chapter 4), – distributed breakout (chapter 5), – distributed consistency (chapter 6). In chapter 7, the assumption that one agent deals with one variable is deleted, multiple local variables are considered. Chapter 8 shows some overconstrained problems – the constraints do not allow any solution – the problem consists now in finding values for the variables so that only a “small” number of constraints remains unsatisfied. The presentation in this book is excellent. Each algorithm is presented by a sequence “basic ideas – details of the algorithm – example of algorithm execution – evaluation”. This allows a very economic representation. The examples are rather small, they allow, however, the understanding of the ideas, and the reader is not overloaded with too many (unimportant) details. By means of this book, an understanding of the problems, solutions and further research fields and topics can be achieved in a rather short time. It is pleasure to read this book.  [Show abstract] [Hide abstract]
ABSTRACT: this paper looks at optimization versions of random k+pSat06/2002; 
Conference Paper: Communication and Computation in Distributed CSP Algorithms.
[Show abstract] [Hide abstract]
ABSTRACT: We introduce SensorDCSP, a naturally distributed benchmark based on a realworld application that arises in the context of networked distributed systems. In order to study the performance of Distributed CSP (DisCSP) algorithms in a truly distributed setting, we use a discreteevent network simulator, which allows us to model the impact of different network traffic conditions on the performance of the algorithms. We consider two complete DisCSP algorithms: asynchronous backtracking (ABT) and asynchronous weakcommitment search (AWC). In our study of different network traffic distributions, we found that random delays, in some cases combined with a dynamic decentralized restart strategy, can improve the performance of DisCSP algorithms. More interestingly, we also found that the active introduction of message delays by agents can improve performance and robustness while reducing the overall network load. Finally, our work confirms that AWC performs better than ABT on satisfiable instances. On unsatisfiable instances, however, the performance of AWC is considerably worse than ABT.Principles and Practice of Constraint Programming  CP 2002, 8th International Conference, CP 2002, Ithaca, NY, USA, September 913, 2002, Proceedings; 01/2002
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.