提出了一种改进型自适应遗传算法,该算法用logistic函数拟合交叉概率和变异概率,以赌轮盘选择和精英保留相结合的方式,在全局寻找最优解.与经典遗传算法相比,改进型自适应遗传算法可以大大提高算法的求解质量.本文基于改进的自适应遗传算法研究设计了-3 d B范围为0—12,-10 d B波束宽度为65,波束覆盖为65,天线频带范围为8.5—9.8 GHz,中心频率为9.05 GHz的X波段微带余割平方扩展波束天线阵.研究结果表明改进型自适应遗传算法对方向图的拟合程度具有较大提高,适应度值可以从0.07以下提升到0.09以上.
Based on the transmission line theory of the buried metallic structures, the concerned harmful effects to the buried oil and natural gas pipelines caused by the power line short-circuit fault are further discussed. A closed-form expression of the induced voltage caused by the short-circuit fault of the ultra-high voltage power （UHVAC） transmission line is given. The transmission line model of the buried pipeline is set up and a set of formulas for calculating induced voltage on the pipeline and the parameters of the buried pipeline in actual environment condition are given. At last, the characteristic of the harm^hl effects on the buried pipeline from the power line short-circuit fault are discussed.
The display image of the video display unit （VDU） can be reconstructed by receiving and analyzing the unexpected electromagnetic noise emanated from a personal computer system. This fact is one of the information leakage threats in information security. Therefore, in order to choose the best frequency range for receiving the electromagnetic compromising emanations form a PC display effectively and quickly, a method based on the correlation power analysis （CPA） and the common equipments is proposed. Firstly, the frequency spectrum of 10 MHz to 1 GHz is measured about 10 times respectively when the PC is displaying the white-black vertical strips image and black image. The frequency range which contains the display information can be obtained through the CPA. And then, the special image signal is searched around the dot frequency and harmonics. If the special signal disappears when displaying the total black image, it can be judged to be the real special signal. Finally, the best receiving frequency range can be chosen by comprehensive considering the results of CPA and the signal to noise ratio （SNR） of the special image signal. In order to check the validity of the chosen frequency range, eavesdropping experiments are carried out. The results demonstrate that the selected frequency range can retrieve the higher quality image than other frequency bands.
A kind of dual-band-stop filter with an unsymmetrical T-shape defected ground structure （DGS） is proposed. The filter has two band gap centers at 4.9 GHz and 7.35 GHz, respectively, where the two centre frequencies are just the second and third harmonics 2.45 GHz. The filters with high band-stop performance and wideband spurious suppression can be achieved by cascading miniaturized unsymmetrical T-shape DGS. The attenuation of three-cascaded T-shape DGS filter is 30.77 dB at 4.9 GHz and 17.96 dB at 7.35 GHz respectively. The above experimental results are in good agreement with the simulation results, especially for the third harmonic.
The co-site interference exists in kinds of communication systems, especially in digital communication systems, also is one of the biggest threats ＂in electromagnetic compatibility （EMC） designation of the systems. Analyzing the co-site interference is an important process in achieving EMC performance of the system. Under the situation, the co-site interference simulation analysis of a vehicular digital communication system based on interference margin （IM） prediction method by bit error ratio （BER） was proposed in the paper. In the simulation analysis, some physical models of the vehicular communication system were established on SIMULINK environment, the upper limit of BER level for audio signals was defined as the IM, three different types of co-site interference was analyzed respectively, and the simulation results were obtained in the form of system interference bandwidth. From these results, the EMC performance evaluation of the vehicular communication system was predicted based on the IM prediction method. By comparing with the actual test results, the correctness of the simulation results was validated and the simulation analysis method can be used for reference by other communication systems was indicated.
Machine learning has a powerful potential for performing the template attack(TA) of cryptographic device. To improve the accuracy and time consuming of electromagnetic template attack(ETA), a multi-class directed acyclic graph support vector machine(DAGSVM) method is proposed to predict the Hamming weight of the key. The method needs to generate K(K ? 1)/2 binary support vector machine(SVM) classifiers and realizes the K-class prediction using a rooted binary directed acyclic graph(DAG) testing model. Further, particle swarm optimization(PSO) is used for optimal selection of DAGSVM model parameters to improve the performance of DAGSVM. By exploiting the electromagnetic emanations captured while a chip was implementing the RC4 algorithm in software, the computation complexity and performance of several multi-class machine learning methods, such as DAGSVM, one-versus-one(OVO)SVM, one-versus-all(OVA)SVM, Probabilistic neural networks(PNN), K-means clustering and fuzzy neural network(FNN) are investigated. In the same scenario, the highest classification accuracy of Hamming weight for the key reached 100%, 95.33%, 85%, 74%, 49.67% and 38% for DAGSVM, OVOSVM, OVASVM, PNN, K-means and FNN, respectively. The experiment results demonstrate the proposed model performs higher predictive accuracy and faster convergence speed.
