Towards application of neural networks for optimal structural synthesis of distributed database systems

Although Distributed Database Management Systems increase usefulness of sensors networks, proper database design requires solution of various complex optimization problems. Synthesis of optimal logical structure belongs to the class of the most important database design problems, and practice recommends using heuristic optimization methods, such as utilization of the paradigm of artificial neural networks. This article describes a new approach to representation of a procedure of logical database structure synthesis in terms of Hopfield neural networks. Development of the energy function, modification of the stable state analysis, and some details of software implementation are illustrated by a simple example, but the same approach can be generalized to more complex situations.