In the end they give two examples to demonstrate how to construct the relation matrix and how the ideas and methods of set covering can be applied into the problem of reduction of attributes. And at the same time we can infer that complexity and the precision of the reduction of attributes is o(r 2 m 3 +m 2 )and78percent respectively from the heuristic algorithm appearing in reference5.
When using improved Heuristic Algorithm, Close the switches of LAN to form circularity configuration and then seek for open cycle switches under conditions of guaranteeing the safety flow, bring about several failure recovery plans and at last determine the best recovery plan in terms of evaluation index.
In some wireless sensor network, usually, there is a sink node to achieve the global information, such as topology control or the node location, and the size of the wireless sensor network is large, so this thesis proposes a centralized heuristic algorithm to construct 1-connected 1-dominating set using the global information from the sink node.
This paper is concerned to the structure analyses of logic networks. A kind of independent variable ( abbreviated to INDV ) in the networks is defined. Using INDV, test generation can be simplified in two aspects, one is the dimension-reduced mithod, another is using a known test to generate a new test for an uncovered fault. The proposed method can be implemented in any heuristic algorithm.
This paper extends the heuristic algorithm for network design from one dimensional space into two dimensional space, taking minimal spanning tree as initial network, thus suggesting a heuristic algorithm which can be used efficiently for centralized network design on time delay constraint.
The general task scheduling problem is NP hard and usually we adopt a heuristic algorithm of polynomial time compexity. In this paper we present an optimal-scheduling problem and then propose an effective distr-ibuuted algorithm to solve it. Finaly, based on simulation and analysis, we give an informal discussion of the algorithm's behavior.