He You

Discretization of Continuous Interval-Valued Attributes in Rough Set Theory and its Application

Rough set theory is a relatively new soft computing tool to deal with vagueness and uncertainty, and is regarded as a field of leading edge. But it cannot deal with continuous attributes, thus discretization is a problem which we cannot neglect. Discretization based on rough set has some particular characteristics, and consistency must be satisfied for discretization of decision systems. Existing discretization methods cannot well process continuous interval-valued attributes in rough set theory. A new approach is proposed to discretize continuous interval-valued attributes in this paper, which enhances the precision of classification and accurate recognition rate in pattern recognition. In the simulation experiment, the decision table was composed of 3 features and 17 radar emitter signals, and the recognition results obtained from this discretization algorithm show that the proposed approach is valid and feasible. The approach expands the application scope of rough set theory.

Attribute measure recognition approach and its applications to emitter recognition

This paper studies the emitter recognition problem. A new recognition method based on attribute measure for emitter recognition is put forward. The steps of the method are presented. The approach to determining the weight coefficient is also discussed. Moreover, considering the temporal redundancy of emitter information detected by multi-sensor system, this new recognition method is generalized to multi-sensor system. A method based on the combination of attribute measure and D-S evidence theory is proposed. The implementation of D-S reasoning is always restricted by basic probability assignment function. Constructing basic probability assignment function based on attribute measure is presented in multi-sensor recognition system. Examples of recognizing the emitter purpose and system are selected to demonstrate the method proposed. Experimental results show that the performance of this new method is accurate and effective.This paper studies the emitter recognition problem. A new recognition method based on attribute measure for emitter recognition is put forward. The steps of the method are presented. The approach to determining the weight coefficient is also discussed. Moreover, considering the temporal redundancy of emitter information detected by multi-sensor system, this new recognition method is generalized to multi-sensor system. A method based on the combination of attribute measure and D-S evidence theory is proposed. The implementation of D-S reasoning is always restricted by basic probability assignment function. Constructing basic probability assignment function based on attribute measure is presented in multi-sensor recognition system. Examples of recognizing the emitter purpose and system are selected to demonstrate the method proposed. Experimental results show that the performance of this new method is accurate and effective.

Research on unobservability problem for two-dimensional bearings-only target motion analysis

The observability for bearings-only target motion analysis is a very important problem. The bearings-only system is to be said observable if and only if the target motion parameters can be uniquely determined by noise-free bearing measurements. The problem of unobservability for hearing-only target tracking system is discussed in this paper based on the target that travels in the 2-dimentional space with a constant acceleration. Utilizing Gramm rule and rank of matrix, a new proposition that the bearing-only locating and tracking system remains unsolvable prior to an observer maneuver is presented and proved. By proving the proposition, it is shown that for certain type of observer movement the estimation process remains unobservable, even if the target moves with a constant accelerate. The work done in this paper is a valuable study in solving the bearings-only target motion analysis.

A New Radar Emitter Recognition Method Based on Variable Precision Rough Set Model

Radar emitter information detected by multisensor system takes on uncertainty, illegibility and contradiction. In real reconnaissance environment, the patterns of radar classes often overlap, the accurate classification of the Pawlak rough set model restricts its application in the real world. In order to solve emitter recognition problem, a new method of finding decision rules is presented to classify radar emitter from the new point of view of variable precision rough set. This method is according to dependent degree of decision attributes on condition attributes. The decision rules proposed are more straightforward. At last, example of recognizing the radar emitter purposes is selected. During the experiment, discretization is conducted on extracted index data of radar emitter and metrical radar characteristic parameter firstly. Then, positive region of each condition attribute are calculated under the given error parameter, which is the basis of decision rules. Experimental results demonstrate this new radar emitter recognition method by finding decision rules based on variable precision rough set model is effective and feasibility

Cooperative location model under the Nearest Neighbor criterion

This paper presents current work on decentralized data fusion applied to the relative localization among multiple platform, which is one of the key formation control techniques for mobile robots or unmanned aerial vehicles. A novel Nearest Neighbor-based scheme is proposed to estimate the navigational states own from the range measurement to other platform. The model to calculate the pseudomeasurement and the concomitant error covariance matrix is deduced for the planar circumstance.

A Novel Emitter Signal Recognition Model Based on Rough Set

On the basis of classification, rough set theory regards knowledge as partition over data using equivalence relation. Rough set theory is deeply studied in this paper and introduced into the problem of emitter recognition, based on which a new emitter signal recognition model is presented. At the same time, a new method of determining weight coefficients is proposed, which is independent of a prior knowledge. And a new classification rule is also presented in this paper. At last, application example is given, which demonstrates this new method is accurate and effective. Moreover, computer simulation of recognizing radar emitter purpose is selected, and compared with fuzzy pattern recognition and classical statistical recognition algorithm through simulation. Experiments results demonstrate the excellent performance of this new recognition method as compared to existing two pattern recognition techniques. A brand-new method is provided for researching on emitter recognition.

A novel gray model for radar emitter recognition

Based on radar practical reconnaissance environment, the application of gray correlation analysis method in emitter recognition is deeply studied in this paper. Firstly, the detailed steps of the method are put forward. Secondly, two approaches to determining the weighed coefficients are also proposed, which overcome the subjectivity in traditional ones. Thirdly, the recognition criterion is discussed. On the base of which, the radar emitter recognition of model and mode are conducted. It can be used for radar emitter whose measurement parameters are incomplete. The simulation results show the feasibility and validity of the novel approach.

A Novel Target Recognition Method Based on Neural Network and Gray Correlation

In order to solve target recognition problems, D-S reasoning method based on information fusion is applied. The key problem to D-S reasoning is basic probability assignment function, so the algorithm implementation of D-S reasoning is a serious problem. For the special traits of target recognition, a new method of constructing basic probability assignment function based on neural network and gray correlation analysis is presented. Examples of recognizing the radar emitter purpose have been selected to demonstrate the new method. Experimental results show that this information fusion method is accurate and effective

Dynamic Multidimensional Assignment Algorithm for Multisensor Information Fusion System

Data association, one of the key and difficult problems for multitarget tracking, is the decision process of linking the measurements or the tracks deemed to be of common origin under certain criteria. All typical data association algorithms can be deducted into special Multidimensional Assignment Problem. However, present S-D assignment algorithm only associates the synchronous measurements from different sensors, which only generate the static result. In this paper, the static assignment algorithm(S-D) has been generalized to the dynamic Multidimensional Assignment Problem by means of combining the measurements set and the tracks set. The main challenge in the assignment problem is that it is NP-hard. The solution using Hopfield neural network (HNN) is presented in this paper. The simulation results illustrate that this method can decrease the computing burden greatly.

Research on Radar Emitter Attribute Recognition Method

In order to solve emitter recognition problems in a practical reconnaissance environment, attribute mathematics is introduced. The basic concepts and theory of attribute set and attribute measure are described in detail. A new attribute recognition method based on attribute measure is presented in this paper. Application example is given, which demonstrates this new method is accurate and effective. Moreover, computer simulation for recognizing the emitter purpose is selected, and compared with classical statistical pattern recognition through simulation. The excellent experimental results demonstrate that this is a brand-new attribute recognition method as compared to existing statistical pattern recognition techniques.