The research on the power control strategy in Wireless LAN is a very important issue in the wireless communication field. In this paper, a Perfect Scheduling Power Control strategy PS_PC is presented for decreasing the power consumption of equipments in WLAN, and it schedules the traffics buffered in Access Point (AP) according to their arrival time, so that the QoS guarantee can be provided and the consumed system resources by collision or backoff procedure are decreased. The theoretical analysis and simulation results show that the PS_PC is an effective power saving method.

From the viewpoint of tracking strategies and link establishment, this paper studies the space geometrical relationship between a TDRS and user satellites in detail. Taking into account of the occultation caused by the Earth and the body of a satellite, the bidirectional tracking rules for a TDRS and user satellites are investigated based on classical orbital elements and the satellite's attitude. The coverage of a TDRS over a user satellite with different orbital altitudes is calculated and simulated. In order to overcome the occultation caused by the body of a user satellite, a way to fix the antenna to a mast which extrudes above the side towards space of a user satellite is presented. With an appropriate height of the mast, the occultation can be eliminated. All these formulas and strategies are the key elements for TT&C and applications of China's future TDRS.

According to the strong nonlinearity and big lag of the cable feed system, an adaptive sliding mode controller is proposed. The MIMO control law is derived with the linear model of the system. In addition, the model error and wind force are considered as the external disturbances, which are estimated on line and compensated. A series simulation is done with the parameters of a 5-meter scaled large spherical radio telescope model, and the results show that not only is the position error with the adaptive sliding mode controller reduced to 40% of that with the PID controller, but also the robustness of the system is improved.

The analytical model which comprises the temperature compensation for the SiC MOSFET is modified by considering the Gauss model of interface state density and the Poole-Frenkel effect. The simulation results for the 6H-SiC NMOS and PMOS show that as the non-uniformity distribution of interface state density, the percentage of current change due to the threshold voltage becomes smaller when the temperature is increased; the main reason for the large leakage current is the existence of surface defects. The larger the defect density, the more rapidly the current value increases with the temperature.

To study the problem of fast and accurate Doppler frequency estimation in the noisy environment, this paper presents the algorithm for high-accuracy frequency estimation based on the continuous zoom in the local spectrum. First, the algorithm cursorily estimates the signal spectrum with FFT; then, the narrowband spectrum of interest is obtained; finally, the narrowband spectrum is analyzed by the method of frequency subdivision. Due to adopting twice window functions in time domain and frequency domain during the subdivision process, straining noise and analyzing the spectrum are organically combined. Not only are the estimate precision and frequency stability greatly improved but also the computational quantity is largely reduced.

The generalized likelihood ratio test (GLRT) is proposed to achieve optimum detection for differential UWB signals. Under unknown channel conditions, maximum likehood estimation of the channel response is implemented, statistical detection of differential UWB signals is performed subsequently, and the receiver architecture based on the generalized likelihood ratio test is obtained. The analytical conclusion shows that the GLRT receiver of differential UWB is a correlation receiver in which the local template waveform is computed as the average of the channel responses to former symbol pulses. Based on the conclusion, the performance of bit error probability is analyzed by using a Gaussian approximation of the noise components at the output of the receiver. The GLRT receiver can serve as a performance benchmark for the conventional differential UWB receivers.

An internal locomotion mechanism of a spherical omnidirectional rolling robot is designed. Four linear motors move inward and outward respectively along four spokes with geometrical symmetry in three dimensional space, which will cause the robot to roll in all directions on a flat surface by displacing its gravity center. Based on the nonholonomic constraint arising from rolling contact and the discrete motion nature of stepper linear motors, the motion model of the robot is established. The second-order control differential equations are derived. Then an algorithm for trajectory planning is developed. A numerical example is given to illustrate the methods for motion analysis and trajectory planning. The virtual prototype technology is used to validate the effectiveness of the methods.

A particle swarm optimization algorithm is presented for the layout of IC design. Particle swarm optimization based on swarm intelligence is a new evolutionary computational tool and is successfully applied in function optimization, neural network design, classification, pattern recognition, signal processing, robot technology and so on. A modified algorithm is presented and applied to the layout of IC design. For a given layout plane, first of all, this algorithm generates the corresponding grid group by barriers and nets’ports with the thought of gridless net routing, establishes the initialization fuzzy matrix, then utilizes the global optimization character to find out the best layout route only if it exits. The results of model simulation indicate that the PSO algorithm is feasible and efficient in IC layout design.

Rotation, scaling and translation (RST) attacks can desynchronize watermark detection, thus making many watermarking systems fail. In this paper, an image adaptive RST invariant watermark (AWPZM) is proposed by using the rotation invariant property of pseudo-Zernike moments and odd-even quantization. Pseudo-Zernike moments of the original image are computed firstly, and those suitable for watermark generation are selected. Then, their magnitudes are odd-even quantized to generate the watermark. In detection, the normalized hamming function is employed to weigh the similarity of the watermark. Experimental results show its robustness to rotation and scaling. For traditional attacks such as JPEG compression, added noise and filtering, the similarities are all above 0．95.

A blind spot analysis is made in rectangular waveguide end-slot phased arrays of a triangular arrangement by using the method of moments (MoM). By increasing array elements of the E plane and H plane, the reflection coefficients versus scanning for the central elements in various sized arrays have been calculated. As the number of array elements of the E plane increases, the blind effect changes greatly, and as the number of array elements of the H plane increases, it will almost not change. By analyzing the coupling coefficient between array elements, an explanation of this phenomenon is given. In the analysis of the blind spot problem, the relationship between the radiation pattern of the center element and the reflection coefficient has been derived. We find their disunity in describing the blind effect. The sufficient condition for the unity that the array be equal scale excited and that the phase of the reflection coefficient be zero(resonance) is obtained.

A novel interactive volume clipping method based on region marking is proposed. The method uses an efficient dataset region marking algorithm to archive the interactive result, regardless of the shape of the clipping object to define which parts of the volume have to be clipped. The proposed technique is integrated into the general graphics hardware accelerated volume rendering. The sophisticated graphics hardware is utilized to make it possible to interactively select and explore regions of the volume datasets under the standard PC.

Concerning the problem of low precision and easy divergence in estimation of the moving parameters for objects in the Bistatic Forward Scattering Barrier Radar, a new tracking filter algorithm is presented based on the Extended Kalman Filter (EKF) and Unscented Kalman Filter (UKF).The new filter applies insensitiveness of UKF to initial estimation error to catch the object,which avoids the divergence due to the big error in initial estimation when using EKF,and then uses EKF to keep the tracking of objects,which avoids low precision and divergence due to sensitiveness of UKF to the system model error when the objects traverse the base line. Simulation results show the validity of the algorithm proposed in the paper.

By employing the topological adjacent relations among triangles, the highest resolution triangle mesh is parted into a set of generalized triangle strips in which the dual graph of every strip is the Hamiltonian path or cycle. The multiresolution representation of the triangle mesh is constructed successively by simplifying the generalized triangle strips while preserving the properties of the Hamiltonian triangulation. The single and multi resolution representations have same codec strategies in this method. The code rate is proportional to c directly when a triangle mesh is represented in c of resolutions.

A new Clos switching structure termed MmM is proposed. All the switching elements in the MmM use input memory. So MmM is adaptive to high speed switching fabrics. In MmM the middle stage provides the routing message for the first stage. The distributed routing and scheduling algorithm named Clos load balance routing(CLBR) for the MmM is proposed. CLBR can balance the traffic load and decease the collisions of output ports. Simulation results show that the MmM with CLBR performs better than the MSM with CRRD.

Ensemble Methods are learning algorithms that construct a collection of individual classifiers which are independent and yet accurate, and then classify a new data point by taking vote of their predictions. The support Vector Machine (SVM) presents excellent performance in solving the problems with a small number of simple, nonlinear and local minima. The combination of the Support Vector Machine with Ensemble methods has been done by Hyun-Chul Kim based on the bagging algorithm, yet it does not show high robustness for its randomicity. In this paper, by a deep investigation into the principle of the SVM and the Ensemble Method, we propose two possible ways, cross validated committees and manipulating of the input feature strategies, to construct the SVM ensemble, which provides strong robustness according to experimental results.

A precoding scheme for the MIMO-OFDMA downlink is discussed, which includes subcarrier assigning, power allocation and adaptive modulation. By using the beamforming codebook constructed according to the Grassmannian space packing, the optimal beamforming vector is determined at the receiver and its index is conveyed to the transmitter. Then power and bits are allocated to each subcarrier using ‘greedy algorithm’. The feedback data are greatly decreased compared with the existing schemes while maintaining a large capacity.

In the case of the non-saturation state, a simple and effective analysis model is proposed for the IEEE 802．11 DCF with multiple traffic supporting. In the model, the concepts of “Virtual Time Slot” and “Virtual Time Slot State” are introduced, which discretize the behavior of a wireless terminal. Based on it, a two dimensional Markov chain is used in the model. By using the proposed model, the approximated closed form solutions of the average packet delays are obtained. Moreover, the method for controlling the average packet delays of all the traffic flows is given. The proposed model and the obtained analytical results help one to further study how to predict and optimize the system performance.

According to the characteristic that the definition point of the Q matrix of wave front is always located on the extremum point on the axis of the beam, a three-dimensional formula for the rectificative factor of phase in complex ray paraxial approximation is proposed, which calculates the rectificative factor of phase directly without considering the calculation of the main directions and curvatures and the rotation of the coordinates, thus simplifying the paraxial approximation calculation by the complex ray expansion (CRE) method. A big cell parallel calculating method is proposed, by which the calculation tasks are distributed to the processes，evenly spaced，according to RCS calculating angles. Results show that the calculating speed of the Large Open-ended Cavities’ RCS is about 102 higher than that of the traditional shooting and bouncing ray (SBR).

We propose a phase unwrapping approach. With local frequencies estimated by the maximum likelihood (ML) estimator, the unwrapped phase at the central pixel of a local image is obtained by fitting unwrapped phases with an ideal plane assumed. We take a region-growing strategy in which a quality map guides the phase unwrapping from high-quality pixels to low-quality pixels, minimizing the propagation of errors to a maximum degree.

Based on a short signature scheme and discrete logarithm problem, a new fair exchange protocol with the off-line semi-Trusted Third Party (TTP) is proposed. The off-line semi-TTP intervenes into the protocol in case of trouble, which makes the fair exchange optimistic. The off-line TTP need not be completely trusted, since it can get neither signature exchanged when disputation is solved. Because this protocol is based on short signature, which needs low storage and little communication, it can be used in low-bandwidth communication and low-storage environments.

Based on the Lennard-Jones potential, the adhesive forces among the atoms of two spheres are considered as the distributing load acting on the surfaces of the two spheres. And according to the classically elastic theory, a novel model of elastic nanocontact of two spheres is established to solve the contact problems emerging in micro-mechanical systems. The model is capable of obtaining the variations with the distance of the adhesive force, the deformations and the contours of the two spheres at the same time. The results from the model established are consistent with the conclusions drawn from the scanning experiments available of atomic force microscopy, which confirms the validity of the model in investigating the nanocontact problems in micro-mechanical systems

In terms of the Karush-Kuhn-Tucker conditions of convex programming, a special nonlinear bilevel programming problem, whose follower-level problem is a convex programming, is transformed into an equivalent single-level programming problem. To solve the transformed problem effectively by using the genetic algorithm, firstly, a new constraint-handling scheme is proposed by adding slack terms to linear inequality constraints and by solving boundary points on nonlinear constraints; secondly, a linear model is constructed to decrease the dimensions of the search space; finally, a new crossover operator is designed, based on the best individuals generated by the algorithm, and it is helpful to improve the fitness values of crossover offspring.

A new efficient group signature scheme proposed by J.H.Zhang et al is analyzed. We point out its shortages in the module and prove that this scheme can be attacked completely in the same module algorithm. We present its improved scheme on RSA and discrete logarithm assumption, which contains the merits of the original scheme, unforgeability and anonymity, and gets rid of the original shortages. The total computation of signature and verification take only 6 exponents and the efficiency raised by nearly two times in the improved scheme.

Trust management based on CA can not suit the Peer-to-Peer network of dynamic, symmetry and decentralization. The CMPJV is proposed to determine whether or not to permit one applicant to join the peer group by joint votes of all the peers in the peer group, and to initialize the applicant’s key attributes in the decentralized environment. The final result of joint votes is calculated by WCVTS, which introduces two factors: original role trust that permits peer to vote more precisely instead of yes or no, and role delegation ability that is used to limit the authorization scope horizontally and vertically. In addition, because any one peer is his own CA to issue his vote for the applicant, and CMPJV is one pure peer-to-peer group admission protocol , it can better overcome the inherent drawbacks of Client/Server. Finally, the running model’s analysis is used to verify its formal security.

An efficient scheme is presented for protecting the ownership of the digital image content by applying iris feature watermarking. The main idea is as follows. Iris image preprocessing is conducted firstly, which includes iris image de-noising based on wavelet analysis, a coarse- to- precise iris localization, normalization and wavelet-based feature codes extraction, with the watermark generated by the operation between the iris codes and NVF. The second and third steps deal with adaptive iris-based watermark embedding and detecting using the human visual system (HVS) and independent component analysis (ICA). The final step is authentication between the iris codes transformed from the extracted watermark and the iris database. An adaptive PSNR metric is proposed for the more objective evaluation of the watermarking system. Experimental results show that the algorithm has good performance in dealing with various kinds of attack.

Problems on link failure recoveries in multicast networks are considered. From the idea of erasure codes, a way to construct a reliable multicast system based on network coding is presented. When limited link failures have occurred in the network, we can have a recover-based solution for all recoverable failures with a static network code, and do not need to change the coding of interior nodes. Based on the information flow decomposition, a way to construct the network codes in a smaller field is proposed. In this way, we can provide nearly instantaneous self-healing multicast communication networks to enhance the robustness of the multicast connections, and achieve a higher rate of multicasting with less required network capacity.

This paper studies the problem of integrating heterogeneous semi-structured Web objects into relational database. A generalized sequential learning model named the Combined Conditional Random Fields is presented for solving the problem of schema matching between pairs of heterogeneous Web data sources. The proposed model is able to learn on the manually labeled training data and unlabeled database records, thereby reducing the dependence on tediously labeled samples. It also provides a novel way to incorporate the two-dimensional neighborhood dependencies between Web data elements. Moreover, a constrained Viterbi algorithm is implemented to resolve the imposed labels inference for optimal data integration. Experimental results using a large number of Web pages from diverse domains show that the proposed method can improve the matching accuracy significantly.

The algorithm for direction-of-arrival of the wideband chirp signals based on ESPRIT using the modified spatial time-frequency matrix is presented. The modified STFD matrix which has the similar mathematical construction with the covariance matrix can be obtained with the cross Wigner-Ville distributions of the outputs of the array. Under the condition of uniform line array, the modified STFD matrix can be transformed into the matrix which has the property of rotational invariance. Then the ESPRIT can be applied to DOA estimation.

A low complexity group Gaussian approximation iterative detection for Turbo-BLAST is proposed. The proposed algorithm divides the transmitted signals into two groups. After assuming the superstition of the transmitted signals in one group in a Gaussian variable, the log-likelihood ratio of each bit in another group is calculated via maximum a posterior principle. The proposed algorithm can considerably decrease the receiver complexity. Furthermore, the performance of the new algorithm can also approach the optimal detection when the system has more receiver antennas than transmitter antennas.

A full-rate space-time coding based CDMA system model is firstly given. Considering that the existing space-time coded CDMA system has high decoding complexity, a low-complexity multiuser receiver scheme is proposed. In this scheme, each user employs the multiuser detection method to achieve the single user receiver signal form, which benefits the application of complex orthogonality of space-time coding. After maximum ratio combination, the decoding computation is only simple linear combination, and thus the exponential decoding complexity of the existing scheme is effectively decreased. On the condition of the same system throughput, the given system can adopt the low-order modulation scheme, and utilize more spatial redundant information. As a result, the concatenation of channel coding significantly strengthens the transmission reliability of the system in the fading channel. Simulation results also show that the system has a lower bit error rate than corresponding full-diversity space-time coded CDMA systems.

Aiming at the steering gear machinery in some missiles，and adopting the combined configuration of a small rare earth torque motor and a harmonic drive reducer, the increasing PID （Proportional-Integral-Derivative）control system based on the rate character of fractal geometry is designed. This control system has some advantages as follows: considering the change with velocity and small damp of harmonic drive, improving the ability of the control system, adopting the complement between linear model and fractal model, using Short-time Fractal Dimension of the feedback signal to attain the difference between the real angle and the ideal angle, and consulting the change above to modify the output of PID. Experimental shows that the 25°rank-jump response constant of this gear machine system is less than 100ms, and with 20Hz and amplitude of ±1.5°, the frequency response phase delay is less than 20°, and orientation precision is higher than 0.06°.

A location-based grid routing (LBGR) algorithm for mobile Ad Hoc networks is presented. To improve the capability of strong route maintenance in the presence of frequent unpredictable topology changes due to node mobility, the LBGR algorithm utilizes position information to employ a forwarding approach based on a grid-constructed trajectory. Based on the position relationship of nodes and downstream grids, each intermediate node infers the next hop to relay packets. Simulation results demonstrate that in the case of lower communication overhead the LBGR algorithm can reduce both the packet dropped rate and the average number of hops per route.

The focused beamformer is an efficient broadband beamforming method because of the advantages of the ability to operate in multipath environments and computational efficiency. However, the robustness of the focused beamformer is not good enough when the error of signal direction exists. To overcome the shortcoming, a novel robust broadband beamforming method based on the focusing approach is presented in this paper to achieve a constant beamwidth for acoustic imaging. The proposed method mainly applies the idea of Robust Capon Beamforming (RCB) to the focusing approach to get the more exact focusing transformation matrix and then reduce the focused data error. Thus it can greatly improve the robustness performance of the focused beamformer.

In order to establish the mathematical foundation of the intelligent information processing system and develop the logic algebraic theory itself, a new class of logic algebra——pseudo NM algebra has been proposed recently. This paper introduces and investigates the filter theory of pseudo NM algebra. The notions of filters，generated filters，normal filters and prime filters etc. are defined. Their basic properties are investigated, with the structures of generated filters obtained. We show that prime filters and irreducible filters are equivalent，and that maximum filters must be prime filters. The prime filter theorem of pseudo NM algebra and its some corollaries are established. All these results are the non-commutative generalizations of the corresponding ones of the filter theory of NM algebra.

A feasible algorithm for inverse synthetic aperture radar(ISAR) imaging of spatial diversity angles is proposed. For the interiors and exteriors of echo data sub-apertures of different visual angle, different alignment reference methods are used for envelope alignment, and then a multiple-spot integration method is used for autofocusing. For fusing received signals from various visual angles, an algorithm of using parameterization and non-parameterization combination for interpolation of the sparse aperture is given, and then the cross range FFT is used to obtain target ISAR imaging, which indicates that the algorithm can be applied in sparse aperture imaging of synthetic aperture radar(SAR). Finally the experimental data processing result is given, which demonstrates the effectiveness of the proposed algorithm.

Based on frequency-offset estimation for the OFDM system， a frequency-offset estimation technique for the Multi-symbol encapsulated OFDM system utilizing the repetitive signal structure to enlarge the estimation range is implemented. The channel estimation method based on the maximum likelihood principle is proposed， in which the training sequence has a relatively flat spectrum and constant peak to average ratio(PAR). Simulation results indicate that the proposed frequency offset and channel estimation methods are feasible for the MSE-OFDM system and improve its performance with a training sequence flat in the time and frequency domain.

A highly efficient Domain Decomposition Method (DDM) based on the auxiliary excitation source algorithm is presented for solving electrically large electromagnetics scattering problems. The original domain is partitioned into nonoverlapping subdomains to reduce the computational scale and complexity. To ensure the convergence of the algorithm, an auxiliary excitation source is used to exchange information between subdomains. The resulting linear system of equations for the auxiliary excitation source is established and solved with a fast algorithm. Once the auxiliary excitation source is obtained, the field on each subdomain can be computed independently. Compared with the traditional method, the proposed method can not only improve the computation efficiency greatly, but also decrease the memory requirement and the computational errors significantly. Moreover, it is especially appealing for problems with geometric repetitions, such as, gratings, photonic/electromagnetic band gap structures (PBG/EBG), and frequency selective surfaces (FSS).

We have designed and implemented an algorithm for discovering neighborhood and estimating the timeliness link quality between neighbors using active probing， passive overhearing and WMEWMA(Window Mean Exponentially Weighted Moving Average Estimator). The link layer relay service presented in this paper mitigates the effect of asymmetric links so as to discover more neighbors. We have evaluated the algorithm using both static analysis and simulation through the TOSSIM simulator. Results show that the algorithm is effective, that more than 40% of nodes identify more outbound neighbors and that the percentage of increased outbound neighbors is between 14% and 100%.

The problem of finding stabilizer quantum-error-correcting codes can be transformed into the problem of finding classical self-orthogonal linear codes over the Galois field GF(4) under a Hermitian inner product. A classical cyclic code over the field GF(4) is self-orthogonal with respect to the Hermitian inner product iff the generator polynomial of its dual code is a factor of its generator polynomial. Based on this connection, a class of quantum cyclic codes is constructed by finding corresponding classical cyclic codes. Some examples are also given in detail.

A novel side information transmission method for PAPR(peak-to-average power ratio) reduction in OFDM(orthogonal frequency division multiplexing) systems is presented, which can avoid the peak re-growth and the change of the structure of original OFDM signals which normally accompany the conventional methods. Simulation shows that the PAPR is reduced about 1.57dB with the new method for the CCDF(complementary cumulative density function) of 10^-3. At the same time, the accuracy of the side information can be ensured by repeating this certain OFDM symbol. Because the two same symbols have the maximal relativity, this characteristic can be used to achieve frame synchronization. Simulation result shows that the proposed scheme can prominently reduce the occurrence of peak re-growth brought by the conventional methods and help to achieve frame synchronization.