Loading...
Office
Journal

Table of Content

    20 April 2004 Volume 31 Issue 2
      
    Original Articles

    Study of DC characteristics models of SiGe HBT
    DAI Xian-ying1;ZHANG He-ming1;HU Hui-yong1;Lü Yi1;WANG Wei1;LI Kai-cheng2
    J4. 2004, 31(2):  165-169. 
    Abstract ( 1522 )   Save
    Related Articles | Metrics
    The SiGe HBT DC models of and the analytic expressions for various currents with definite physical meanings are established, based on the solution to equations of drift and diffusion. The established current models are suitable for both large and small current injections. MEDICI is used to simulate the effects of Ge content and doping concentration on current density. The simulating results show that the decrease of the current gain is due to the decrease of the collector current and the increase of the neutral base recombination current caused by the increase of base doping cocentration. The effects of the collector current, backward hole injection current, neutral base recombination current and space-charge region recombination current on the current gain are also analyzed and discussed.

    A novel optimal model of and the solution to the layer yield of IC's
    JING Ming-e;HAO Yue;ZHAO Tian-xu;MA Pei-jun
    J4. 2004, 31(2):  170-173. 
    Abstract ( 1449 )   Save
    Related Articles | Metrics
    Based on the need for the method to improve the IC's integrated profit, a novel layer yield optimal model is presented. Firstly, an integrated performance function is formulated. Secondly, accoring to the integrated performance function, all factors, such as designable parameters, the number of layers and layered parameters, are considered together to establish the optimal model. Thirdly a special algorithm is designed for this model with efficient sampling technology--Uniform Design. Finally, encouraging results have been obtained thus far, with numerical and circuit examples given to demonstrate the model.

    The reliability study of EEPROM with the FLOTOX structure
    LUO Hong-wei1;2;YANG Yin-tang1;ZHU Zhang-ming1;XIE Bin3;WANG Jin-yan3
    J4. 2004, 31(2):  174-176. 
    Abstract ( 1559 )   Save
    Related Articles | Metrics
    A type of EEPROM with the FLOTOX structure is intoduced. The main reliability issues of the FLOTOX EEPROM are studied, i.e., the degradation of the programming windows, the degradation of the retention characteristics and the time dependent dielectric breakdown(TDDB). The reliability of FLOTOX is correlative to the quality of the tunnel oxide. The degradation of FLOTOX depends on the trap charges which come from the tunnel oxide, which is proved by experimental results.

    Fluid neural network based routing algorithm
    BO Hua1;MA Fu-long2
    J4. 2004, 31(2):  177-181. 
    Abstract ( 1529 )   Save
    Related Articles | Metrics
    From the heuristic viewpoing, the maximum probability criterion for the choice of the routing of a communication network is proposed by comparing the communication network with the fluid neural network. The routing algorithm based on the fluid neural network is developed. Experimental results indicate that the proposed algorithm is much faster than any existing routing algorithm.

    An implementation method for self-organized network management
    LIU Fu-jie;CHANG Yi-lin;SHEN Zhong;ZHANG Xin;LI Song-quan
    J4. 2004, 31(2):  182-185. 
    Abstract ( 1700 )   Save
    Related Articles | Metrics
    This paper presents a framework for managing the self-organized network. The framework uses hierarchical clustering of nodes to reduce the number of messages exchanged between the manager and the agents. We give a way for the manager to appoint a node as a new cluster head to manage its responsive domain in the Ad Hoc network and to assign their management tasks dynamically. Active clustering enables the network to keep track of varied network topology. These technologies are compatible with SNMP and are of real usable value.

    An SNR blind estimation technique based on virtual carriers in OFDM systems
    REN Guang-liang;ZHANG Hui;CHANG Yi-lin
    J4. 2004, 31(2):  186-189. 
    Abstract ( 1973 )   Save
    Related Articles | Metrics
    According to the requirements for the real-time control in an adaptive OFDM system, a novel SNR blind estimation technique based on virtual carriers is proposed. By statistical processing of the virtual sub-carrier signal and the modulated sub-carrier signal, the average power of noise and that of signal plus noise can be achieved respectively and are further processed to achieve the SNR. The proposed technique is of low complexity. The technique outperforms the M2M4 blind estimation technique and is close to the pilot aided SNR estimation technique as the number of estimation samples increases. he result of simlation shows its validity.

    Standardization status of the intelligent optical network
    WEN Ai-jun;XU Zhan-qi;ZHANG Bing
    J4. 2004, 31(2):  190-193. 
    Abstract ( 1367 )   Save
    Related Articles | Metrics
    The key technologies of ION(intelligent optical networks) include signaling, routing and standardized optical interfaces. Standardization organizations involved in ION are ITU-T, IETF and OIF, which emphasize architecture, protocol and internetworking respectively. The ION based on SONET/SDH is more viable at the present stage.

    A modeling and verification method for ingtelligent networks services with Petri nets
    MA Yu-xiang;LI Yuan-jun;LIU Yan-ming
    J4. 2004, 31(2):  194-198. 
    Abstract ( 1783 )   Save
    Related Articles | Metrics
    This paper introduces a specification of service logic based on the analysis of IN services and presents a model for the components of the specification that are related to concurrency and synchronization with Petri nets. The verification method for the model is also discussed. An example is given to illustrate the modeling and verification.

    Dual-satellite's groudn moving target detection at multi-operating frequencies
    NING Wei;LIAO Gui-sheng
    J4. 2004, 31(2):  199-204. 
    Abstract ( 1512 )   Save
    Related Articles | Metrics

    A study has been made of detecting the ground slow moving target with the dual-satellite mode. Utilizing the method of SAR-GMTI we adopt dual satellites to realize the detection of very slow ground targets, but the problem of blind speed arises. We design the satellites to work at several different operating-frequencies, thus overcoming the problem of blind speed. Meantime the targets can be orientated easily. Analyse or simulations of this method under different conditions are also made.

    ROI coding based on rate-distortion slope lifting
    DENG Jia-xian;WU Cheng-ke;LI Yun-song;ZHUANG Huai-yu
    J4. 2004, 31(2):  205-208. 
    Abstract ( 2173 )   Save
    Related Articles | Metrics
    A method for ROI encoding based on rate-distortion slope lifting which does not lift the coefficients is proposed. The decoder does not need to lift reversely the coefficients and works normally. The requirement for ROI can be attained. So it is suitable for image compression with the given ROI PSNR and for interactive client/server applications.

    An immune RBF neural network MUD method
    YANG Shu-yuan;JIAO Li-cheng;LIU Fang
    J4. 2004, 31(2):  209-213. 
    Abstract ( 1879 )   Save
    Related Articles | Metrics
    The Radical asis Function Neural Network(RBFNN) is a feed-forward NN model with the capacity of local approximation, so it can be used as a multiple-user detector(MUD) in the CDMA. Althouth it proves to be able to present a good performance in the detection, it has some disadvantages which lie in the determination of its structure and the optimization of its training algorithm. In this paper, a novel immune RBFNN MUD is proposed to balance computational complexity and performance of RBFNN. It is characteristic of more rapid convergence, better generalization and greater robustness, so its has better pracicability and real-time performance. Simulations also prove its effectiveness.

    Fuzzy entropy-based method for multilevel image restoration
    WANG Bao-ping1;FAN Jiu-lun2;XIE Wei-xin3
    J4. 2004, 31(2):  214-217. 
    Abstract ( 1364 )   Save
    Related Articles | Metrics
    A method for multi-level image restoration based on fuzzy entropy is presented. The method can be used to extend the entropy's expression in terms of entropy's limitation on application, and construct a kind of fuzzy entropy that can reflect characteristics of the multi-level image. The simulation results show that the performance of image restoration by using fuzzy entropy is much bettern than those by using one of two usual image noising-reducing methods, the median filter and the mean filter. In addition, noise-reduction in the multi-level images plays an improtant role in medical image processing and military image processing. Therefore, this paper contributes to both the theoretical researches on and the applications of fuzzy entropy.

    Pose-varied face recognition based on teh 3D model
    LIU Zhi-jing;LI Xia-zhong;WU Mang
    J4. 2004, 31(2):  218-222. 
    Abstract ( 1870 )   Save
    Related Articles | Metrics
    By estimating the rotating angle of the unknown photo using the front and side face photos to build the 3D virtual person model and then making the projection of the 3D model in the angle, we can resolve the Pose-Varied Face Recognition by matching the unknown photo with the projection of the 3D model. A way to resolve the problem of the Pose-Varied Face Recognition and a new algorithm for calculating the rotating angle of the unknown photo are proposed in the paper.

    Engropy-based image retrieval
    SUN Jun-ding1;WU Xiao-sheng2;ZHOU Li-hua1
    J4. 2004, 31(2):  223-228. 
    Abstract ( 1853 )   Save
    Related Articles | Metrics
    The content based image retrieval(CBIR) on colour, shape and texture is analyzed and an entropy-based image retrieval(EBIR) algorithm is proposed. Entropy is introduced to extract the features of the image. With the combination of colour property and grid descriptor(GD), and entropy eigenvalue vector descriptor is brought forward for the description of the image. Using teh entropy eigenvalue vector, we can simply retrieve the relevant images from the database. It is shown through performance experiments that this algorithm is of scale invariance and rotation invariance to some extent. Satisfactory retrieval results can also be observed for the image with analogical vision but different colour histogram.

    A no-reference video quality assessment method based on digital watermark
    WANG Xin-dai;YANG Fu-zheng;CHANG Yi-lin
    J4. 2004, 31(2):  229-233. 
    Abstract ( 1696 )   Save
    Related Articles | Metrics

    This paper proposes a novel no-reference video quality assessment method. By comparison of the extracted watermark embedded in transmitted video with the copy of original watermark in the destination, we can assess all restored video quality without any original reference video sequences. This method does not require any information about the test video sequence to be transmitted, so it can improve the utilization efficiency of the channel as well as keep the assessmeht in real time. Experiment shows that the assessed PSNR of reconstructed video quality is approximately equal to its true measured value, so this technique can be applied to the video quality assessment of wireless and IP video servies.

    Study of the feed of teh UHF radar antenna
    LI Ping1;2;LIANG Chang-hong1;ZHANG Dian-fu2
    J4. 2004, 31(2):  234-237. 
    Abstract ( 1586 )   Save
    Related Articles | Metrics
    aiming at spreading bandwidth, ministurization and high-gain of the air-borne SAR antenna, this paper puts forward a new type of prism radar antenna, whose feed based on the technices of CPW is a two-component H-shaped antenna array printed by dubble plane overlap. The frequency bandwidth is the 2.2 times the frequency, the VSWR within the bandwidth is less than 1.85 and the gain at the centural frequency is more than 9 dBi, which are found by means of the High-Frequency Structure Simulator and the larruping designed. The VSWR and the radiate direction plot by simulation and test are given. The results show that it can be the new feed for UHF radar antennas.

    Study of the algorithm for power density measurement of millimeter-wave irradiators
    ZHAO Jian-xun;LI Ji-xi
    J4. 2004, 31(2):  238-242. 
    Abstract ( 1669 )   Save
    Related Articles | Metrics
    An analysis is made of the evaluation of the millimeter-wave(MMW) power density(PD) in the area of the biological sample in the vicinity of MMW irradiators used in experiments on MMW biological effects. An idea is presented to calculate the MMW PD in the area of the biological sample using the electric field measured in the scanning plane based on the plane wave expansion technique. A method to calculate the MMW electric field via open-ended waveguide measurement is also given. With the application of the sampling theorem and the Fast Fourier Transform(FFT), two groups of numerical forms of the algorithm equations suitable for two kinds of FFT subroutines are derived. Finally, the measurement algorithm is fully tested by the measurement and calculation carrier out for an MMW irradiator.

    The eigen-driven analysis method in antenna pattern synthesis
    BU An-tao;SHI Xiao-wei
    J4. 2004, 31(2):  243-247. 
    Abstract ( 1808 )   Save
    Related Articles | Metrics
    The novel definition of the eigen-driven mode is given in antenna array analysis by the Moment Method. The practical array with mutual coupling between elements is approximated with the ideal one without mutual coupling. The excitation coefficients derived from the current distribution of ideal array elements are used to excite the array with mutual coupling, to produce the same radiation pattern as that of the ideal array. The process of pattern synthesis of a practical array is shortened because of the reduced electromagnetic analysis. Some examples are illustrated to prove the conclusion.

    Image segmentation algorithm based on teh morphological pyramid
    REN Huo-rong;WANG Jia-li;ZHANG Ping
    J4. 2004, 31(2):  248-251. 
    Abstract ( 2306 )   Save
    Related Articles | Metrics
    A new image segmentation algorithm based on the mathematical morphological pyramid is proposed, which is used to decompose an image level-wise, to process the detail information in different resolution spaces, and then segment different feature regions. The experimental result of the SAR image shows this algorithm can be used as a class of band-pass filters to extract different feature regions, thus implementing the image multi-resolution segmentation.

    Colour secret sharing scheme for general access structures
    MA Wen-ping;REN Ya-an
    J4. 2004, 31(2):  252-254. 
    Abstract ( 1355 )   Save
    Related Articles | Metrics
    An extended coloured k out of n secret sharing scheme for general access structures is proposed, and a new method to construct coloured k out of k secret sharing schemes is given, with a class of coloured k out of n secret sharing schemes constructed by block design. It is shown that there exists a coloured k out of n secret sharing scheme for any parameter k and n such than k≤n.

    Analysis of the cryptographic properties of the AES S-box
    LIU Jing-wei;WEI Bao-dian;Lü Ji-qiang;WANG Xin-mei
    J4. 2004, 31(2):  255-259. 
    Abstract ( 1936 )   Save
    Related Articles | Metrics
    As the only nonlinear structures in most block ciphers, S-boxes account for the block ciphers' secruity. A detail analysis of the cryptographic properties of teh AES S-box is made in this paper. A new simple and universal method, the Lagrange Interpolating Fundamental Polynomial is used to get the algebraic expression for teh S-box. Eight algebraic properties of S-boxes such as balanceness, strict avalanche criterion, differential uniformance, algebraic order, and so on are calculated and proved.

    An intrusion-tolerant conference key distribution scheme
    GUO Yuan-bo1;2;MA Jian-feng1
    J4. 2004, 31(2):  260-263. 
    Abstract ( 1518 )   Save
    Related Articles | Metrics
    This paper proposes a computationally secure and fault-tolerant conference key distribution scheme, which only requires the authenticated and encrypted point-to-point channels between each server and each user. By the combined use of knowledge proof and verifiable secret sharing, the scheme has the properties of simple structure and high security. Analysis shows that every honest user of a conference can get a common key after the running of the protocol, even if a minority of the servers malfunction or misbehave. We also show that on the assumption of a Diffie-Hellman decisional problem, a passive adversary gets zero knowledge about the conference key, and the active adversary cannot impersonate someone successfuly. Because the knowledge proof method we adopt is non-interactive, both operation and communication overheads are reduced, thus making the schemes very efficient.

    The self-editing generator
    BAI En-jian;DONG Qing-kuan;XIAO Guo-zhen
    J4. 2004, 31(2):  264-268. 
    Abstract ( 1629 )   Save
    Related Articles | Metrics
    Based on a single ternary linear feedback shift register(LFSR) which is a combined model of the clock-controlled generator and the self-shrinking generator, a construction of a pseudo-random generator, called a self-editing generator is presented. The period, linear complexity and symbol distribution of thw self-edited sequence are discussed and several cryptology indexes are compared with those of the self-shrinking sequence. The results of theoretic analysis show that the period and the linear complexity of the self-editing sequence are superior to those of the self-shrinking sequence. Furthermore, the sequence can resis the attacks from the application of the Berlekamp-Massey algorithm when the series of LFSR satisfies n>60. The construction is suitable for practical implementation of efficient stream cipher cryptosystems.

    Research on the relationship between the formats of computer information storage and the leakage of electromagnetic information
    QIU Yang;ZHANG Kun;LIU Hao;TIAN Jin
    J4. 2004, 31(2):  269-271. 
    Abstract ( 1566 )   Save
    Related Articles | Metrics
    Different storage and compression formats of computer information are discussed. After the relationship between the formats of computer information storage and the leakage of electromagnetic information has been tested in three aspects; time-domain, frequency-domain and radiation field, a conclusion is reached that in the same display mode, the leakage of electromagnetic information is independent of the storage formats of computer information. As the theoretical foundation, this useful conclusion is of value for the future system design of anti-leakage of computer electromagnetic information.

    A fair-scheduling algorithm for apordic and period tasks on multiprocessor systems
    ZHANG Hui-juan;ZHOU Shui-sheng;ZHOU Li-hua
    J4. 2004, 31(2):  272-275. 
    Abstract ( 1405 )   Save
    Related Articles | Metrics
    The fair scheduling theory is an important scheduling theory for period tasks on multiprocessor systems. PFair scheduling, proposed by Baruah, is presently the only known optimal scheduling method for period real-time tasks on multiprocessor systems. PD2 has been developed recently and is the most efficient. But in practice, the real-time tasks are made up of period and apordic taks. In this paper, based on the fair scheduling theory and algorithms, a fair scheduling algorithm based on the server is proposed for apordic and period tasks on real-time multiprocessor systems. Three scheduling ways for apordic tasks are presented. The worst-case response time for the soft apordic real-time task is also computed, with the computed expressions proved.

    On the tradeoff between time to market and performance under cannibalization strategy
    LI Hua1;DU Zhen-wu2;XIAO Yu1
    J4. 2004, 31(2):  276-280. 
    Abstract ( 1348 )   Save
    Related Articles | Metrics
    Reduction of the new product development(NPD) cycle time and improvements in product performance rate become strategic objectives for many technology-driven firms. In this paper, a model is developed to study the trandeoff between them under cannibalization strategy. Optimal time-to-market and duration of the overlapping stage conncerned with cannibalization rate are obtained. The analysis of our model shows that the overlapping stage reduces the time-to-market to some extent. The existing product with high performance contributes to the rduction of time to market, but not obviously; cannibalization rate wakens the ralation between the productivity of the overlapping stage and the optimal time-to-market, and the optimal duration of the overlapping stage respectively. These results modify the three conclusions drawn by Calantone. Two new conclusions are derived from our model. They are (1) the bigger the cannibalization rate, the more profit the firm obtains. (2) the less the duration of the overlapping stage, the longer the time-to-market, and the shorter the windows of opportunity becomes, so the cannibalization should be increased to bring more profit for the firm.

    Modal parameters identification using the wavelet transform
    XU Ya-lan;CHEN Jian-jun;HU Tai-bin
    J4. 2004, 31(2):  281-285. 
    Abstract ( 1486 )   Save
    Related Articles | Metrics
    The paper introduces a method for modal analysis using the wavelet transform, and discusses the possibilities of teh uncoupling of the system modal as well as the identification of time-varying parameters using the wavelet transform. In addition, the reasons why the Morlet wavelet is attractive in analyzing signals generated by dynamical systems is given, although there are a multitude of wavelets that could be used. Finally, based on the characteristics of transient responses and the Morlet wavelet, a family of wavelet functions is introduced and used in the simulation of modal analysis. The results are straightforward and satisfactory.

    2-D analysis of an iterative algorithm for the optimal control of the discrete nonlinear system
    MA Hao;LI Jun-min
    J4. 2004, 31(2):  286-290. 
    Abstract ( 1648 )   Save
    Related Articles | Metrics
    By means of the two-dimensional system theory, an optimal control algorithm for dynamic integrated system optimization and parameter estimation is analyzed for the discrete-time nonlinear system with terminal constraints. The optimlity and convergence of the algorithm are invertigated from a new viewpoint. An easily verified necessary and sufficient condition for the local convergence of the algorithm is derived, with a sufficient condition for the global convergence also given.

    Fuzzy c-mean clustering method for analyzing microarray gene expression data
    GONG Gai-yun1;MAO Yong-cai2;GAO Xin-bo2;LIU San-yang1
    J4. 2004, 31(2):  291-295. 
    Abstract ( 1455 )   Save
    Related Articles | Metrics
    Microarray technologies are emerging as a promising tool for genomic studies. Today the challenge is how to analyze the resulting amounts of data. For this purpose clustering technologies have been applied to this field, but fuzy clustering technology analysis has not been used for microarray gene expression data. in this paper the fuzzy c-mean(FCM) clustering method is used to analyze such data in order to detect differentially expressed genes. Our results indicate that fuzzy clustering can be a useful tool to exploit the differential gene expression for microarray data.

    Signal de-noising in wavelet domain based on a new kind of thresholding function
    ZHANG Wei-qiang;SONG Guo-xiang
    J4. 2004, 31(2):  296-299. 
    Abstract ( 2185 )   Save
    Related Articles | Metrics
    A novel thresholding function is presented based on the wavelet shrinkage put forward by D.L.Donoho and I.M.Johnstone. This new thresholding function has many advantages over DJ's soft- and hard-thresholding function. It is simple in expression and as continuous as the soft-thresholding function, and has a high order derivative which makes convenient some kinds of mathematical disposals. It also overcomes the shortcoming that there is an invariable dispersion between the estimated wavelet coefficients and the decomposed wavelet coefficients of the soft-thresholding function. All these advantages make it possible to construct an adaptive denoising algorithm. Simulation results indicate that the de-noising method adopting the new thresholding functionsuppresses the Pseudo-Gibbs phenomena near the singularities of the signal effectively, adn the numerical results also show the new method gives better MSE performance and SNR gains than DJ's hard- and soft-thresholding methods.

    On the solvability of finite groups based on normal indexes
    WEI Xian-biao1;2;YU Li1
    J4. 2004, 31(2):  300-303. 
    Abstract ( 1408 )   Save
    Related Articles | Metrics
    For a maximal subgroup M of a finite group G, the normal index of M is defined as the order of a chief factor H/K, where H is minimal normal supplements of M in G. In this paper some sufficient conditions for solvability of finite groups are obtained by using the normal indexes of maximal subgroups.

    Stochastic LQ control framework and its application in finance as the stock price follows the jump-diffusion process
    LIU Xuan-hui1;HU En-jian2;HOU Jian-rong3
    J4. 2004, 31(2):  304-309. 
    Abstract ( 1342 )   Save
    Related Articles | Metrics
    In the continuous time finance model the stock price valiatility is deemed the Brownian motion. However in teh real world as the significant information occurs, a discontinuous jump will occur in the stock price. This paper extends the classical stochastic LQ control to the jump-diffusion model. With the jump-diffusion stochastic Riccati equation introduced, the optimal feedback control can be obtained. With its application in hedging strategy and the Mean-Variance model, we obtain the optimal bedging strategy and the optimal portfolio strategy.

    The self-intersection local time of the generalized α-stable process
    CHEN Zhen-long1;2;LIU San-yang1
    J4. 2004, 31(2):  310-316. 
    Abstract ( 1304 )   Save
    Related Articles | Metrics
    We study the existence and the joint continuity of the self-intersection local time of N-parameter d-dimension generalized α-stable processes, and present a sufficient condition for the existence of square integrable and the joint continuity of this process. These results contain and extend the results of N-parameter d-dimension α-stable processes and N-parameter d-dimension generalized Brownian Sheet.

    Existence of solutions of a class of system of nonlinear elliptic equations
    ZHANG Guo-qing1;2;LIU San-yang1
    J4. 2004, 31(2):  317-320. 
    Abstract ( 1248 )   Save
    Related Articles | Metrics
    In this paper, the solution of a class of system of nonlinear elliptic equations is transformed into a problem of finding critical points of the given functional. Under various conditions, the existence of critical points and nontrival critical points is proved by using variational methods, the extreme theory and the Mountain Pass lemma, so that the solution and nontrivial solution of the nonlinear elliptic system is obtained.

    A fast 2-D virtual ESPRIT algorithm
    LI Guo-min1;XUE Qian2;WANG An-yi1
    J4. 2004, 31(2):  321-324. 
    Abstract ( 1780 )   Save
    Related Articles | Metrics
    This paper presents a new fast method based on teh 2-D virtual ESPRIT signal parameters estimation algorithm. It proves to be of the same estimation results but lower complication than that of the original algorithm in theory. Its effectiveness can be proved by computer simulations.

    Analysis and control of the backlash of the drive system for limited motion antennas
    SHI Zuo-ming
    J4. 2004, 31(2):  325-328. 
    Abstract ( 1341 )   Save
    Related Articles | Metrics
    This paper analyses the backlash of the drive system of limited motion antennas, and derives the functional relations between the backlash and the drive ratio, as well as the backlash and the antenna rotating angle. Finally an effective method for controlling the backlash is presented.
Copyright © 2018 Journal of Xidian University, All Rights Reserved.
Powered by Beijing Magtech Co. Ltd