摘要：This piece of work contributes to sensor-grid computing by attenuating latency in the computation of the data deployment from the decision fusion centre of the sensor network to the Grid computing architecture. With the application of the Bayesian network computational processes which can first be applied as a computing algorithm, the sensors and the Optimal Decision Fusion (ODF) centre can be used to obtain vital information on some hazardous environment such as in military zone or from unmanned orbit mission using the Grid computing facilities in some secured state. The information so obtained must first be filtered to remove the Gaussian noise in the disseminating information of the hazardous source using Shannon's entropy algorithm which then predict coherently some well refined unbiased information. The conversion of the decision fusion into binary expedites the processing of the data from the sensor network hazardous environment and made to assembly in the grid computing remote environment from the Internet network. Many exposures can be identified to improve the Sensor-Grid computing environment where vital conflicting scenarios based on the data can be mined using machining learning algorithms. This paper is concerned with making predictive suggestion on the need to harness the sensors with the grid computing environments if worthwhile algorithms can be developed to meet the targets.
关键词：Decision fusion; Data fusion; Maximum a Posteriori; Marginal likelihood; ; Prior; Entropy; Sensor-grid; Optimal Decision Fusion; Shannon's information; ; (SensorML) and Bayesian Network
