Search

A novel sorting algorithm based on clustering is proposed to resolve problems such as low sorting precision and even failure which are caused by conventional radar signal sorting algorithm applied in present high-density signal environment. According to the characters of electromagnetic parameters, the advantages and disadvantages of the traditional clustering algorithms in data mining have been discussed in this paper. By using traditional clustering algorithm, users have to specify in advance how many clusters are being sought and it cannot be processed well if signal samples are not in spherical cluster distribution. An improved clustering algorithm based on self-organizing iterative analytic is presented in this paper, which avoid relying on prior knowledge or depending on signals' distribution. Simulation result proves the validity and practicability of this algorithm. It provides a new way for object clustering problem in emitter identification.

A new possibility degree for comparing two projects is defined. A new method is proposed for interval multi-attribute decision-making, in which the attribute values and weights are in the form of interval numbers. The method can compare and rank interval numbers, and avoid computing the probability distribution of integral evaluation values, if all attribute values’ distributions are known. The proposed method is straightforward and can be performed on computer easily. Finally, an illustrative example, which assumes the weights and attribute values are uniform distributions, is given to demonstrate the feasibility and practicability of the proposed method.