To quickly and efficiently retrieve the target component from a large software component library, a component retrieval method based on identification of faceted classification and cluster tree was proposed. The component with facet classification identification was described by using the set of component identification, which overcomes the impact of subjective factors when only using facets classification to describe and retrieve components. By introducing cluster tree, the component cluster tree was established by analysis clustering of components based on semantic similarity, thus narrowing the retrieval area, reducing the number of comparisons with component libary, and improving the search efficiency. Finally, the proposed method was experimented and compared with other common retrieval methods. The results show that the precision of the proposed method is 88.3% and the recall ratio is 93.1%; moreover, the proposed method also has a good retrieval effect when searching in a large-scale component library.

In the large-scale Radio Frequency Identification (RFID) system, the estimated time of current tag estimation algorithms increase linearly with the increase of tags, and the deviation is large. Regarding these problems, a new estimation algorithm based on 0-1 distribution was proposed. By using the feature of 0-1 distribution, the algorithm set the specific frame length and selected flag to choose the collection of tags which responded to the command of query. In this way, estimation time was reduced to the logarithmic level of tag number and the deviation was reduced through picking numerous average values randomly. Compared with other algorithms, the simulation results show that the proposed algorithm drops deviation at least by 0.9%, and has less fluctuation.