Beijing, China
Beijing, China

Time filter

Source Type

Li P.,University of Science and Technology Beijing | Li P.,Beijing Key Laboratory of Knowledge Engineering for Materials Science | Wang C.,University of Science and Technology Beijing | Wang C.,Beijing Key Laboratory of Knowledge Engineering for Materials Science | Sheng S.,Unit 95899 of PLA
Journal of Computational Information Systems | Year: 2015

In order to overcome premature convergence and improve convergent speed, a new adaptive genetic algorithm based on normal distribution function (NDAGA) is proposed in this paper. Based on normal distribution function, we present two adaptive adjustment measures. One is about the range of fitness value, which is mainly used for maintaining population diversity to provide adequate power for evolution. The other is mainly reflected in adjusting the probabilities of crossover and mutation, which are the two crucial factors of the whole algorithm. Compared with SGA, AGA and IAGA through a series of classic test functions, the presented algorithm has a significant improvement on the convergent rate and can jump out of the local optimum easily. Moreover, the novel algorithm is efficient in solving deployment problem about radio station. ©, 2015, Binary Information Press. All right reserved.

Cui M.,University of Science and Technology Beijing | Cui M.,Beijing Key Laboratory of Knowledge Engineering for Materials Science | Wang C.,University of Science and Technology Beijing | Wang C.,Beijing Key Laboratory of Knowledge Engineering for Materials Science | Sheng S.,Unit 95899 of PLA
Journal of Information and Computational Science | Year: 2015

Fidelity and real-time performance are two key factors in simulation training system. The accurate jamming intensity and smooth picture display can improve simulation fidelity. In order to calculate jamming intensity, we propose a jamming intensity model which takes major influence into account. In order to improve jamming pictures fidelity, we propose fractal interpolation to interpolate in detection distance and point-wise blanking method to simulate the jamming blanking process. In order to meet real-time requirements, we propose a radar antenna pattern model based on sinc function instead of array synthesis. CUDA technology is proposed to solve the problem of huge computation. Experiment result shows that the model achieves both real-time requirements and system fidelity. The evaluation result based on Fuzzy Analytic Hierarchy Process (FAHP) shows that the fidelity of jamming pictures based on fractal interpolation is 0.8901. This model has been applied to simulation system of military equipment. © 2015 by Binary Information Press

Zou Z.-B.,The First Aeronautic Institute of the Air Force | Wu J.-B.,Unit 95899 of PLA | Ma M.,The First Aeronautic Institute of the Air Force
Applied Mechanics and Materials | Year: 2013

Maintenance training of airborne electronic equipment is the guarantee of working orderly for the airplane. The best choice of maintenance training is using real electronic equipment so that high training cost and potential damage to normally working equipment exist. To reduce the training cost and risk, a virtual maintenance training system of airborne electronic equipment is designed and implemented using virtual reality technology. The architecture of this virtual training system is given. In this architecture, three-dimensional model of electronic equipment is implemented by 3DMax software, and GL Studio software is used to create the operation response of all manipulations. Signal models based on multi signals flaw graph is used to simulate various faults. Finally, the application of this virtual maintenance training system in daily aircraft maintenance training show good training efficiency. © (2013) Trans Tech Publications, Switzerland.

Zhang S.,The First Aeronautic Institute of the Air Force | Zhu Z.-G.,Unit 95899 of PLA | Wu J.-B.,Unit 94620 of PLA
Applied Mechanics and Materials | Year: 2013

When several ultra-wide band (UWB) communication systems coexist, multiuser interferences (MUI) are unavoidable. To cope with the MUI of UWB systems, the independent of UWB signals are investigated and a novel receiving method is proposed. At first several antennas are utilized to receive the transmitted signals of different UWB users, and the mixture signals are gained. Then the mixed signals are separated by independent component analysis algorithm to recover the transmitted signals of each UWB users. Finally, the information of each UWB users transmitted is demodulated. Computer simulations show that the proposed method can efficiently resist the interference of different UWB users. © (2013) Trans Tech Publications, Switzerland.

Meng F.,National University of Defense Technology | Shi P.,Unit 95899 of PLA | Ou G.,National University of Defense Technology
Beijing Hangkong Hangtian Daxue Xuebao/Journal of Beijing University of Aeronautics and Astronautics | Year: 2014

The rotation of the autonomous navigation constellation is unobservable while using inter-satellite measurements only. Once the rotation occurs, errors will be introduced to the positioning results of ground users. The positioning error model of ground users was established to solve this problem. According to the positioning principle and spherical coordinates transformation, the relationship was mathematically proved between the rotation error of the constellation and the positioning error of ground users. Based on this relationship, a differential correction technique to the rotation error was proposed, including the structure and correction algorithm of the technique. Simulations of the Walker 12/3/1 constellation indicate that the constellation's rotation error will cause the longitude of the ground user to have the same error. After applying the differential correction technique, the ground users will effectively correct this error. While the rotation error is less than 1' (equivalent to a 31 m horizontal error in the equator), the user's horizontal error is less than 1.5 m and the vertical error is less than 0.003 m after differential correcting.

Fei A.-G.,Unit. 95899 of PLA | Zhang L.-Y.,Unit. 95899 of PLA | Liu G.,Unit. 95899 of PLA | Wang Y.,Unit. 95899 of PLA
Yuhang Xuebao/Journal of Astronautics | Year: 2013

In order to solve the problem of Air-to-Air Missile(AAM) guidance power handover during the Beyond-Visual-Rang (BVR) cooperative air combat, the mathematic models composed of the AAM guidance power handover mission assignment model and the guidance superiority models for our aircrafts to target or AAM are firstly built. Secondly, a Particle Swarm Auction Hybrid Algorithm (PSAHA) is proposed for the optimization of large scale AAM guidance superiority handover, this alogrithm has the both good performances of improcved particle swarm algorithm and auction algorithm at the same time, it can get a reasonable guidance superiority handover scheme within short time, and is high dynamics, good global optimization and easy for computer program realization. At last, simulation experiment results show that this method is reasonable and valid.

Loading Unit. 95899 of PLA collaborators
Loading Unit. 95899 of PLA collaborators