When the tracks of the multi-target get approached or crossed, it is easy to lead to combining or even to get wrong tracks for the traditional tracking methods, since the traditional methods only utilize the information on the target position to finish the data association. Aiming at this problem, a multi-target tracking algorithm aided by the high resolution range profile (HRRP) is proposed in this paper. Firstly, the target attitude angle is estimated in real time on the principle that the HRRP is sensitive to the attitude angle. And then the attitude angle is added to the target measurement state to construct a multi-dimension correlating gate. The data association is accomplished with the multi-dimension information. So the problem of multi-target data association is simplified to multiple sub-problems of data association for a single target. Finally, each target motion state is estimated by the probabilistic data association-unscented Kalman filter (PDA-UKF). Simulation results reveal that the computing complexity is reduced, and that the correct probability of data association is improved by using the target HRRP on the one hand. On the other hand, the tracking accuracy is improved with the aid of the target attitude angle.

We propose a network selection algorithm based on the residual service time for the network selection problem in heterogeneous networks. There have been already many research works and achievements in this area, but most of the existing works just consider the optimal user or network revenue which does not consider the impact of new users. This paper presents the concept of the residual service time, and uses it to model the impact of the new users, in order to get a better network option on long time scales. In this paper, we use the non-cooperative game to model the network, and prove that the Nash equilibrium of the model is also the global optimal solution. Finally, we use the Lyapunov stability theory to show that the proposed algorithm is stable. Simulation results show that the introduction of the residual service time can improve the network performance, reduce the blocking rate, and increase the total network revenue.

In view of the limitation of fixed complete orthogonal transformation, represented by two-dimensional wavelet transform and discrete cosine transform in compressed sensing high-resolution image reconstruction, this paper proposes a new method for high-resolution image reconstruction based on adaptive redundant dictionary sparse representation with the total variation constraint.The algorithm takes the intermediate image in the process of iteration as the training sample to get a redundant dictionary suitable for sample characteristics by adaptive learning. It makes full use of the correlation between dictionary atoms and the image to get an ideal complete sparse representation, thus reducing the sampling rate and improving the quality of image reconstruction. Finally, the algorithm takes the total variation as a constraint and uses the split Bregman iterative method to solve the sparse optimization problem. Simulation shows that the proposed method can reconstruct high quality images under a low sampling rate.

We consider the phase noise filtering problem for interferometric synthetic aperture radar (InSAR) based on the dictionary learning technique. Due to the non-convexity of the optimization problem is difficult to solve. By using the splitting technique and employing the augmented Lagrangian framework, we obtain a relaxed nonlinear constraint optimization problem with l₁-norm regularization which can be solved efficiently by the alternating direction method of multipliers (ADMM). Specifically, we firstly train dictionaries from the InSAR complex phase data, and then reconstruct the desired complex phase image from the sparse representation. Simulation results based on simulated and measured data show that this new InSAR phase noise reduction method has a much better performance than several classical phase filtering methods in terms of residual count, mean square error (MSE) and preservation of the fringe completeness.

In view of the problem of poor carrier frequency offset estimation performance over multipath channels for orthogonal frequency division multiplexing(OFDM) systems in non-cooperative communication, a new blind carrier frequency offset(CFO) estimation method is proposed. Firstly, the cost function based on the difference in amplitude is proposed. Secondly, the terminal cost function is derived by the product of the adjacent subcarrier's amplitude. Finally, the CFO estimation based on the polynomial interpolation is realized by the transformation cost function. Simulations show that the proposed CFO estimation method has a better performance and lower computational complexity over the multipath channels.

Aiming at the transmitted and received steering vectors mismatch problem, an iterative dimension-reducing robust adaptive beamformer for MIMO radar is presented. The General Linear Combined(GLC) method is applied in MIMO radar to obtain the enhanced covariance matrix estimation, and the transmitted and received steering vectors mismatch model is established. The cost function is established based on the desired signal output power maximum principle to estimate the transmitted and received steering vectors. The bi-iteration method is proposed to solve the cost function and it is merely necessary to find out two low-dimensional convex quadratically constrained quadratic programming(QCQP) problems in per iteration. Simulation results show that the proposed method can obtain the higher output signal-to-noise-plus-interference(SINR) under the condition of severe steering vector mismatch than the conventional robust beamformers, and that the proposed method can converge fast so that it has the lower computational complexity.

Multiple kernel learning (MKL) combines multiple kernels in a convex optimization framework and seeks the best line combination of them. Generally, MKL can get better results than single kernel learning, but heavy computational burden makes MKL impractical. Inspired by the extreme learning machine (ELM), a novel fast MKL method based on the random kernel is proposed. When the framework of ELM is satisfied, the kernel parameters can be given randomly, which produces the random kernel. Thus, the sub-kernel scale is reduced largely, which accelerates the training time and saves the memory. Furthermore, the reduced kernel scale can reduce the error bound of MKL by analyzing the empirical Rademacher complexity of MKL. It gives a theoretical guarantee that the proposed method gets a higher classification accuracy than traditional MKL methods. Experiments indicate that the proposed method uses a faster speed, more small memory and gets better results than several classical fast MKL methods.

The resonant properties of the interdigital capacitance loading loop resonator (IDCLLR) and its dual counterpart complimentary interdigital capacitance loading loop resonator (CIDCLLR) are studied. Due to the high capacitance/inductance of the IDCLLR/CIDCLLR, the resonators can generate lower resonant frequencies and narrower resonant bandwidths than many other structures. A compact ultra-wideband (UWB) antenna with quadruple sharp band-notching characteristics is presented by utilizing IDCLLRs and CIDCLLRs. Measured results show that the proposed antenna has a wide bandwidth from 3 to 11GHz(VSWR<2), with four sharp notched bands centered on 3.5, 5.2, 5.8 and 7.5GHz, respectively, avoiding the elimination of many useful frequencies. Moreover, the presence of IDCLLRs and CIDCLLRs scarcely affects the performance of the UWB antenna at its operating frequencies.

Divisible-load scheduling has become an increasingly hot subject in the research on information technologies in recent years. Most existing divisible-load scheduling models assume that all processors are idle at the beginning of workload assignment. In fact, many processors may still in the busy state when a new workload arrives. Processors may have different waiting times from the busy state to the idle, that is, processors have different release times. This paper proposes a new release time aware divisible-load scheduling model with hybrid time constraints and designs an effective global optimization genetic algorithm to solve it. Finally, experimental results show the effectiveness of the proposed model and the efficiency of the proposed algorithm.

The performance of the selected bands in stealth aircraft target multi-spectral detection is hard to analyze and evaluate. A unit energy spectrum detection system is designed to analyze the selected detection band, demonstrating the availability of the selected detection bands. The selected narrow band receives less energy than the traditional wide band, while the signal itself can not represent the performance of target detection. By analyzing the unit energy spectrum detection principle and considering the alternative model of the system signal, the model of the wave quality factor is established to objectively analyze the performance of the selected band in different visibilities, cloud backgrounds and detection ranges. The result shows that the unit energy detection system can analyze each band detection objectively and accurately. At the same time, it also gives a clear and intuitive display of the jitter property while using it in different cloud backgrounds, proving that the detection effect of the narrow band is superior to the wide band and consistent with the theoretical study.

The dimensions reducing technique combined with the Maehly approximation is applied to the Method of Moment (MoM) for analysis of the two-dimensional radar cross section (RCS) of the target. The Gauss-Green-Ostrogradsky (GGO) algorithm is utilized to transform the two-dimensional expression for the surface currents in both spatial and frequency domains to one dimension and its derivatives. This procedure avoids the solution of expansion coefficients in the two-dimensional expression and makes the accuracy adjustable by the order of derivatives. Compared with the two-dimensional Maehly approximation method, the proposed scheme can acquire RCS data efficiently with good accuracy and reduce procedural complexity. Finally, numerical results show that memory requirement and calculation time are about 1/6 and 1/3 of what are needed in the original method.

Based on the three-layer altitude spectrum, this paper analytically investigates the bit error rate (BER) performance of the slanted laser communication system under the weak turbulence condition. The expressions for BER using on-off keying (OOK) and pulse position modulation (PPM) are derived. An analysis of impacts of the zenith angle, wavelength, power-law index and modulations on BER is made. Theoretical results show that the descriptions of turbulence given by the three-layer altitude spectrum is consistent with the actual situation. With the wavelength decreasing and zenith angle increasing, BER increases. Furthermore, PPM requires less average transmitted power than OOK, which may achieve the aim of saving cost without increasing the error rate in system design.

If the sum capacity of a channel can be achieved by treating interference as noise, we say that this channel is under noisy interference. By analyzing the sum capacity of Gaussian X channels, sufficient conditions for such channels under noisy interference are proposed. Providing vector genies and auxiliary sets as side information to the receivers, a sum-rate upper bound is obtained. Under the noisy interference sufficient conditions, this sum-rate upper bound can be achieved by operating the original channel as the underlying Gaussian interference channel and treating interference as noise. Therefore, the sum capacity of the Gaussian X channel is determined for noisy interference.

The MSVL is a temporal logic programming language. It can be used to verify C, Verilog/VHDL programs. To do so, a program written in C or Verilog/VHDL is translated to an MSVL program, and then the task is changed to verify MSVL programs. However, at present, the correctness of MSVL programs can only be proved by hand with deductive approaches. This is tedious and error-prone. To handle this problem, an automatic theorem proving technique for the MSVL based on the interactive theorem prover PVS is proposed. To this end, first the syntax and semantics of the MSVL are described in the specification language of PVS, which enables MSVL programs to be correctly recognized by PVS. Further, an axiomatic system of the MSVL and some theorems are specified. Then the proof commands of PVS are input for invoking the PVS prover to deduce MSVL programs. During verification, simple details can be proved by PVS automatically while complex steps are controlled by human. In this way, MSVL programs can be verified semi-automatically, which facilitates the deduction of MSVL programs. An instance of the bakery algorithm is given to show that our method is feasible.

In ordor to investigate the dynamic rain attenuation forecast in countermeasure rain fade techniques on the microwave link, in typical temperate and subtropical rain areas in China, the power spectra for the rain attenuation time series simulated by the EMB (Enhanced Maseng and Bakken) model are in agreement with the measured results at 12.5GHz, but there are differences between the fade slopes statistics based on simulated series and measured data. The rain attenuation time series by the EMB model are improved by using a circular inserting data method. By the improved method, the statistical probabilities of the fade slopes are very close approximation to the measured results. The probabilities of the fade duration based on the simulated time series by the improved processes are compared with the results by unimproved processes, the results based on the measured data and by the ITU-R predicted results. It is shown that the statistical results for the fade slopes are improved, and that the differences between the fade duration probabilities by the improved series and measured data are smaller than that between the results by unimproved and measured. The results show that the improved process to the EMB model is feasible in typical rain areas in China.

Firstly, in order to assess the stability of the camera robots, a stability performance index with combination of force and position is proposed based on the determinations of the cable tensions for a camera robot. Furthermore, the stability performance index is described using the weighted average method, and meanwhile，the stability workspace is designed with the stability performance index. Secondly, a robust workspace with the external wrench is selected to compare with the stable workspace above for the camera robots. Finally, simulation results show that it is suitable to employ the stable performance to evaluate the stability of the camera robots.

In order to implement quick and effective search, save the storage space and improve the poor performance of affinity relationshaps between high dimensional data and its codes in image retrieval, a new linear embedding hashing is proposed by introducing the preserving similarity. First, the whole data set is clustered into several classes, and then the similarity predicted function is used to maintain affinity relationships between high dimensional data and its codes so as to establish the objective function. By minimizing the margin loss function, the optimal embedded matrix can be obtained. Compared with the existing classic hashing algorithm, experimental results show that the performance of the linear embedding hash algorithm is superior to the other binary encoding strategy on precision and recall.

When focusing beyond diffraction-limit through the scattering medium with amplitude modulation by the feedback control algorithm, the grid-by-grid modulation scheme of the continuous sequential algorithm will severely degrade anti-noise performance, which is detrimental to focusing of light under low signal-to-noise ratio environment. To overcome the drawback of the continuous sequential algorithm, the genetic algorithm is introduced to shape the wavefront amplitude of incident light. With the help of its characteristic global optimization strategy, the amplitude of all the grids can be simultaneously modulated and optimized so as to reduce the sensitivity to noise. Simulation results show that the target focus intensity of the genetic algorithm can be hardly affected by noise level, and that the resultant focusing intensity acquired at different noise levels are very close. The focusing intensity value up to 80.67％ that of SNR is ∞ can be achieved even if the noise level rises to SNR is 10dB.

Aquatic Microbial Fuel Cell(AMFC) must be inoculated and work in water environment. Terrestrial Microbial Fuel Cell(TMFC) can overcome the shortcoming. In order to realize the practical application of TMFC, a single-hop Wireless Sensor Network(WSN) powered by a TMFC experimental setup is designed and established. Power generation performance of the TMFC and sensor data acquisition, wireless transmission and processing of the WSN are tested by experiments. Experimental results show that the proposed TMFC can drive the single-hop WSN periodically, which validates the feasibility of TMFC powered WSN.

As the off-fed truss reflector is connected to the platform by a single edge and deployed rapidly through being driven by the spring, the impact influence acts obviously on both the reflector and the platform. Due to limitations of the ground test, the in-orbit deployment impact characteristics of truss reflector are difficult to obtain. Then, by taking a certain reflector as the research object, a dynamic model is established with multibody dynamic software. Based on the modified model with ground deployable tests, in-orbit deployment impact characteristics are obtained by simulation analysis. Therefore, the method for multibody dynamic modeling is suitable for truss reflector dynamic research, and the result is useful for the deployable reliability optimization.

To recover the 3D deformable structure from an un-calibration image sequence, a factorization reconstruction method for 3D deformable reconstruction is presented. Based on the low rank constraint on the image matrix, the projective reconstruction can be obtained by single value decomposition. By using the orthogonality of the projective matrix, the Euclidean reconstruction is upgraded from projective reconstruction. The innovation of the method is that the solving of the 3D non-rigid reconstruction is linear and that all the images and the image points are treated uniformly. The experiments with both simulated and real data show that the method presented in the paper is efficient.

Rate adaptation, which is one indispensable mechanism for the multi-rate wireless local area network (WLAN), is critical to the system performance. In this paper, we propose a highly efficient rate adaptation algorithm for IEEE 802.11ac based on the receiver's information feedback without modifying the MAC frame format. The basic idea is that the RTS frame is used to estimate the channel condition accurately, and then the CTS frame is sent back with the information of the chosen rate in a scrambling sequence. Moreover, according to the probability of the error received, the threshold of rate selection can be adaptively adjusted, thus further enhancing the stability of system performance. Simulation and verification by NS-3 show that the throughput performance of this algorithm outperforms that of three well-known rate adaptation solutions (AARF, ONOE, and Minstrel) in different channel environments.

In millimeter wave high resolution SAR imaging, the existing squint SAR cross-track motion compensation algorithm brings into a phase error of more than π/4 during the imaging process. The phase error would worsen the compensation effect. This paper transforms the squint slant range with the motion error into the side-looking slant range. We adopt the side-looking cross-track motion compensation algorithm to cross-track motion compensation. The proposed algorithm works well on the millimeter wave real data in the 10° squint angle and the results are better than those obtained by existing algorithms in the same squint angle.

To improve the fairness performance of the downlink traffic scheduling algorithm, a network flow based downlink traffic scheduling algorithm is proposed for the roadside unit (RSU) in vehicular networks. In the proposed algorithm, a bipartite graph is constructed firstly, where the node set is composed by the vehicle set and the timeslot set. At any given timeslot if a vehicle can communicate with the RSU, then an edge between the given timeslot and that vehicle is added into the edge set. Next, a flow network graph is constructed based on the bipartite graph by adding a virtual source node and a virtual sink node. By applying the conventional minimum cost maximum flow algorithms, a minimum cost maximum flow can be computed, which is converted to the fair traffic scheduling strategy. Simulation results show that, when the total vehicle requirements are maximized, compared with the existing algorithms, the fairness performance of the proposed algorithm is improved by 116.4％ in the offline case, and by 25.9％ in the online case.

As the ternary comparator faces problems of slow speed and high power consumption, this paper proposes a scheme for the magnitude ternary comparator of CNFET, which is researching on theory of multi-valued logic circuit and the structure of comparator circuit. First, the ternary decode signal is propagated to the comparator, then the encoder circuit encodes the results of comparison, and finally the magnitude ternary comparator is made up of all kinds of modules. The magnitude comparator is simulated by software, which demonstrates that it has a correct logic function, fast speed and low power consumption characteristics.

Nowadays, the optimal defense strategies selection based on the incomplete information game model has many disadvantages, such as ignoring the type of the defender, using the simple cost quantitative method, and choosing defense strategies improperly. To solve the problem, this paper proposes an active defense strategy selection based on the static Bayesian game theory, and constructs the static Bayesian game model. The model considers the types of the attacker and the defender, and improves the classical strategies taxonomy and cost quantitative method by considering the strike back act of the defender and the success attack rate. Then, this paper calculates and comprehensively analyzes the Bayesian equilibrium of the game. Taking mixed strategies Bayesian equilibrium of the attacker as the defender's prediction of the attacker's action, this paper calculates the defense effectiveness of defense strategies and performs a defense strategies selection algorithm. Finally, an example is provided to analyze and demonstrate the effectiveness of the model and algorithm.

A new image interpolation algorithm based on the rational function model with both smoothing and surface constraints is proposed. The rational function has a simple and explicit expression with two parameters. The image edge regions can be detected adaptively by using contour analysis. Basically, the detective threshold is selected based on the rational function construction. Selecting different parameters in the rational function model, the edge regions and smooth regions are interpolated by bicubic interpolation and rational interpolation, respectively. The optimal rational interpolation parameters can be obtained with a set of exact solutions by solving a maximization problem based on the contrast sensitivity enhancement of human eyes. Experimental results demonstrate that the proposed method achieves competitive performance with the state-of-the-art interpolation algorithms, especially in image details features.

Selecting features with high sensor interoperability is of great importance but it is not been investigated enough. Based on the application of classifying lung diseases on CT (Computed Tomography) images, the sensor interoperability of 4 features is studied. An evaluation criterion is proposed to select features by considering interoperability and discrimination ability of features. After doing experiments on 3 different datasets, it is shown that sensor interoperability affects the disease recognition or information retrieval methods. Moreover, the rationality and effectiveness of the proposed feature evaluation criterion is verified.

A high speed and medium accuracy multiplying digital-to-analog converter (MDAC) circuit optimization design is presented for meeting the requirements of the 8bit, 80MS/s pipelined analog-to-digital (A/D) converter. An optimized transmission gate is adopted to improve the linearity of the MDAC circuit. In view of the high gain two-stage operational amplifier, design method in wideband operational amplifier design optimization is proposed and the settling time and power consumption of operational amplifier can be effectively decreased In addition, an improved high speed dynamic comparator is used in this design Fabricated in a 1.8V 0.18μm CMOS process, this A/D converter with the proposed MDAC circuit achieves a signal to noise and distortion ratio (SNDR) of 54.6dB and an effective number of bits (ENOB) of 7.83bit with a 35MHz input signal at the 80MHz sample rate.

To overcome the biased estimation of the pseudo-linear algorithm in bearings-only target tracking, a modified instrumental variable algorithm is proposed. In the new algorithm, the current bearings-only estimation angle is acquired by polyfitting several previous bearings angles. Then the estimated angle is used as the instrumental variable to get the motion parameters by applying the least-square method. The algorithm can achieve the theoretical unbiased estimation. Simulation results illustrate that the new modified instrumental variable algorithm has a better convergence rate and estimation accuracy than the existing research and that it is more suitable for engineering practice.

Aimed to improve the time-validity and accuracy of radar under the information guiding, the cueing and handoff model is proposed and the division method of the searching airspace is analyzed according to the elements of confidence coefficient, timeliness and false handoff probability. Based on dynamic detection theory, the dynamic beam position arrangement is proposed by predicting the global information gain. Simulation results show that the proposed method can capture the target quickly and accurately on condition of a large guiding error and improve the cooperative detection capability of passive sensor guiding of radar.

Aiming at the flaw of the PMJ cognitive computation model and discovering the law of cognition, the single fuzzy cognitive unit event PMJ cognitive computation model is proposed. The correlation selection device is added in perception representation, the long-term memory bank and short-term memory bank are designed in memory representation, the uncompleted work is mapped as the goal in judgment representation, the input data of perception representation is the single fuzzy cognitive unit event, the single fuzzy cognitive unit event PMJ cognitive computation model fulfills the cognitive mechanisms such as selected attention, long-term memory,short-term memory,recalling and forgetting. The schema memory experimental results demonstrate that the rise in long-term memory bank capacity enlarges the completion rate of the task when forgetting occurs, the variation in long-term memory bank capacity does not enlarge the completion rate of the task when no forgetting does not occur, and the variation in short-term memory bank length does not change the completion rate of the task whether forgetting occurs or not. So the single fuzzy cognitive unit event PMJ cognitive computation model corrects the flaw of PMJ cognitive computation model and discovers the cognitive law that the long-term memory bank capacity and short-term memory bank length influence the completion rate of the task, so that it is an optimal cognitive computation model.

A novel method is presented for the purpose of recovering sparse high dimensional signals from few linear measurements, especially in the noisy case. The proposed method works in the following two steps: ①The support of signal is approximately identified via Thresholded Basis Pursuit(TBP), the weighting matrix and parameters needed for the next step are also computed; ②The Iteratively Reweighted Lp Minimization(IRLp) procedure is used to solve the non-convex objective function. As theoretic interpretation and simulation results show, lower computational complexity is required for the proposed Support Driven IRLp(SD_IRLp) algorithm for high probability recovery, in comparison to 7 analogous methods(including an oracle estimator).

Considering SUs with multiple radios, this paper discusses the cross-layer resource allocation problem for video transmissions in cognitive radio networks. Because of the coupling relationship between radio and channel, this paper proposes a new conflict graph based on the link-radio-channel, and models the resource allocation problem into an optimization problem. The constraints ensure the conflict-free and interference-free video transmissions, and provide flow routing and rate allocation for video services based on the feature of scalable video coding streams. The objective function achieves the network-level throughput maximization with fairness consideration. Simulation results show that the multi-radio transmission could gain a higher throughput. Also, the proposed scheme could provide fairness resource allocation, and use resource efficiently based on the feature of video services.

Considering the two limitations of scenario assumption and performance optimization in distributed opportunistic channel access under the continuous transmission rate, a distributed opportunistic channel access strategy is proposed in DF relay networks under the discrete transmission rate. The proposed strategy maximizes the system average throughput with the instruction of optimal stopping theory. By optimizing the second-hop transmission time in DF relay networks completely, the total transmission time is reduced, and the system performance is improved. Compared with the current strategy, simulation result demonstrates the validity of the proposed strategy in system average throughput.

Device-to-Device (D2D) communication underlaying LTE-A cellular networks is effective to improving spectral efficiency and offload traffic of the base station by reusing cellular resources. However, the mutual interference between D2D and cellular communications can degrade the performance of both D2D and cellular users. In this paper, a resource reusing selection scheme based on minimizing power increase is proposed, which enables selective compensation for signal to interference plus noise ratio (SINR) diminution caused by interference. Simulation results demonstrate that the proposed scheme notably improves the access grant probability of D2D users and increases the link spectral efficiency of D2D and cellular networks, without sacrificing the quality of cellular and D2D communication.

A statistical modeling method based on multitask sparse learning is proposed to realize the recognition of the high resolution range profile (HRRP) with a small training data size. The statistical modeling of each training aspect-frame is considered as a single task in our method. Since the training aspect-frames are not independent but inter-related, they can share a compact dictionary to make full use of the information. However, with the different targets and the aspect sensitivity of the same target, it is usually hard to assess the task relatedness, and joint learning with unrelated tasks may degrade the recognition performance. Therefore, we adopt the Bernoulli-Beta prior to learn the needed atoms of each aspect-frame automatically with the given training data. Then the relatedness between frames is determined by the number of shared atoms, and multitask learning can be realized adaptively. The recognition experiments of the measured HRRP data demonstrate the performance of the proposed method.

The acquisition of high image quality and high resolution spectral data is limited by light flux. A push-broom spectral imaging should reduce the spatial resolution if it amplified its light flux to increase its signal noise ratio (SNR). According to this problem, the theory of compressive sensing (CS) is introduced for modeling the push-broom spectral imaging system from the signal processing analysis, so that the number of slits of the imaging system can be increased to amplify its light flux. Under the guidance of the theory of compressive sensing, the light flux can increase without reducing the spatial resolution. In the simulation, if its exposure frequency dropped to 1/4  the original, and its light flux increased to 128 times the original, the spectral image with the resolution of 512×512 could be well obtained. This method is suitable for remote sensing by using a smaller number of times for imaging and less memory for storage and transmission compared with the traditional one.

With the gradual increase in image resolution of the spacecraft camera, it is highly required to figure out the problem how to process a huge amount of image data on board at a high speed. As a solution, the CCSDS proposes a space-oriented image-coding standard. For the sake of high image-coding performance, it adopts wavelet transformation as a method of image data transformation. However, wavelet transformation contains multi-level data processing, which causes more computational time consumption and more memory utilization. In order to solve this problem, we propose a highly efficient VLSI architecture for DWT with low-storage. By revising the traditional lifting structure and employing time-multiplex data processing strategy to perform the second and third level of wavelet transformation by the same logic module, the usage of logic resource is reduced with no sacrifice on speed.Using a small amount of on-chip memory instead of off-chip memory to save certain parts of DWT coefficients and sending the coefficients in a specific sequence to entropy coder timely, the off-chip memory for storage of DWT coefficients is no longer required. The proposed VLSI architecture of DWT is already implemented on the Xilinx FPGA XC4VSX55, which can achieve a high performance, in terms of data throughput, reaching 95.91MPixels/s.

Aiming at the probably existing performance loss and high computational complexity of the robust beamforming based on steering vector estimation with as little prior information as possible which is solved by the semi-definite relaxation (SDR) approach, a novel robust beamforming algorithm using sequential quadratic programming (SQP) is proposed. The original non-convex problem is linearly approximated to a convex subproblem using the first order Taylor's series, and the optimal solution is found out by solving the convex subproblem iteratively. Moreover, considering the mismatch of the sample covariance matrix, the SQP-WC method based on worst-case performance optimization is presented to improve the performance of the proposed SQP method. Theoretical analysis and simulation results show that the proposed SQP algorithm can converge fast and its convergence point approximates the optimal solution to the original problem, which indicates that the SQP method can effectively reduce the computational complexity compared with the SDR method, and furthermore, the SQP-WC method can effectively improve the performance of the SQP method with a small parameter.