Abstract:In this paper, the allocation approaches of auxiliary antenna based on the independent array elements、sub-array region and array element combination are designed for studying the impact of allocation of auxiliary antenna on performance of adaptive sidelobe canceller in air-borne phased array radar. The results showed that the performances of the cancellers are good and there are no obvious differences for most of proposed allocation approaches. The performance of canceller seriously deteriorate when the auxiliary antenna are allocated in the same a row or a column. The auxiliary antenna based on the single array elements have better interference suppression effectiveness than that based on multiple array element combination. Thus, this paper suggested that the auxiliary antenna can not be allocated in a row or a column and are constructed using single array elements as many as possible.

Abstract:The modified Gabor wavelet transform (MGWT) algorithm gave the best prediction results among the independent protein coding regions prediction algorithms. In this paper, a FIR (Finite Impulse Response) windowed narrow pass-band filter (WNPBF) based protein coding regions prediction algorithm is proposed. The algorithm is consisted of the following parts mainly：designing a WNPBF with appropriate length, which taking the DNA sequence F56F11.4 as an example; extending the DNA sequences using the boundary symmetric padding method and cutting off the beginning part of the outputs of the WNPBF to eliminate the side effects of the group delay of the filter upon prediction results; designing a moving average filter to smooth the power spectral density curve to get better prediction results. Experiments carried out on DNA data sets ALLSEQ and HMR195 respectively gave the prediction results which are close or equal to the best ones. The proposed algorithm is much more efficient than MGWT algorithm and can be used for comparing the prediction results of different filters directly and objectively.

Abstract:Because of the limited of experimental condition, there are missing values in gene expression data which make against to following use. Estimating missing values without destroying other data and information lost has becoming an important work of bio-information . By weighted kernel function，similar coefficient between those rows and columns the miss values in and other rows and columns is presented. In this paper an estimation method based on double weighted regression using weighted kernel function is introduced. It makes the information data more abundant by considering gene space correlation and time correlation in regression. Comparing with other methods, weighted double regression method can get better performance.

Abstract:In order to fully consider the inter-frame correlation in voice conversion, a convolutive nonnegative matrix factorization based voice conversion method is proposed. The personal characteristics and inter-frame correlation in voice can be well preserved in the time-frequency bases obtained from convolutive nonnegative matrix factorization. With this feature, during the training phase of voice conversion, the matching time-frequency bases of source and target speakers can be extracted from training data through convolutive nonnegative matrix factorization. Then in the conversion phase, the voice of source speaker is converted through time-frequency bases substitution. Compared to traditional methods, the inter-frame correlation in voice can be better preserved and converted in the proposed method. Experimental results using objective and subjective evaluations show that the proposed method outperforms the Gaussian Mixture Model and State Space Model based methods in the view of both speech quality and conversion similarity to the target speech.

Abstract:Copy-Move forgery , in which a part of a digital image is copied and then pasted to another portion of the same image in order to conceal an important object in the scene, is one of the common image forgery techniques. This paper describes an efficient and robust algorithm for detecting and localizing this type of malicious tampering. Firstly, Discrete Wavelets Transform (DWT) is applied to a tampered image, then the low-frequency component in wavelet sub-band is divided into overlapped blocks. Seven statistics of each block are calculated which are lexicographically sorted so that duplicated blocks are identified by searching for rows that get closer to other rows after sorting. Through finding out the duplicated region in the wavelet sub-band , we can find out the duplicated region in the tampered image. Simulation results show the efficiency of proposed algorithm.

Abstract:This paper proposed a modified scheme of embedding watermark in JPEG bitstream based on sorting, the motivation for this method is that it ensure the cover image is lossless after embedding the watermark and improve the embedding capacity. According to the statistical results of VLC usage in an image, we embed watermark by mapping an unused VLC to a used VLC. Sort the VLC(s) according to their appear rate in the bitstream before mapping, it will improve the VLC utilization rate and increase the embedding capacity after sorting. During the data embedding, we replace the codes appearing in the bitstream by the mapped codes according to the watermark. The results can proved that the proposed method has high embedding capacity and can preserve the image with no quality distortion.

Abstract:Phase measurement is widely used in all areas, and all phase FFT(apFFT) has outstanding accuracy in phase measuring among existing methods. To disclose the underlying reasons of the superiority of apFFT in measuring phases compared to other existing methods, based on [1], this paper combines the distribution characteristics of spectral leakages and theory of parameter estimation ,thus successfully leading to one Cramer-Rao lower bound (CRLB) involving 2 unknown parameters estimation suitable to apFFT measuring phases. Based on this, the relationship of apFFT phase error and the frequency offset distribution is also deduced. Moreover, it is also pointed that the novel CRLB is 12dB lower than the traditional CRLB involving three parameters. Consequently, this paper draws several conclusions of apFFT measuring phases under different signal to noise ratios and different frequency offsets. Simulation indicates that the phase measuring variance of apFFT can be defined by CRLB2 with different SNRs and different frequency offsets, verifying these conclusions and correctness of the novel CRLB.

Abstract:The angle estimation algorithms of Multiple Input and Multiple Out (MIMO) radar have been studied. A new algorithm based on bistatic MIMO radar signal model is proposed for coherent angle estimation for MIMO radar in the presence of the spatially Gaussian noise. Through the numeration of fourth order Cumulants by using the data vectors of bistatic MIMO radar system, the influence of spatial colored noise can be eliminated and the vector which obtains the target information can be got and proved. Then a block Hankel matrix can be reconstructed by the fourth order Cumulants vector and it is proved that the rank of this block Hankel matrix is equal to the number of targets and has no relations with the coherency of the targets. Through the eigenvalue decomposition of the matrix, the receiver and transmitter angle estimation of coherent targets can be estimated combined with ESPRIT algorithm. This algorithm, combining the capacity of parameter estimation of MIMO radar and fourth order Cumulants, restrains automatically the influence of the additive Gaussian white noise and colored noise and realizes angle estimation of coherent targets with automatic pairing, and which will be more available in practice. Simulation results verify that the proposed algorithm is effective.

Abstract:In this paper, on the basis of the traditional interpolation algorithm using FFT, an algorithm which can improve the accuracy of interpolation is proposed. The algorithm here overlap and cut the subsequences, then discard the samples with great error at both ends of interpolated sequence and reconstruct the relatively accurate samples. In this way, the accuracy of interpolation can be significantly improved. Experimental results show that, compared with the interpolation algorithm proposed by Prasad, in case of computational amount increase by 3.1%, the normalized mean square error corresponding to different subsequence length decrease to 1/19 of its original in average; in case of computational amount increase by twice, the normalized mean square error corresponding to different subsequence length decrease to 1/75 of its original in average.

Abstract:This paper propose a phone boundary detection method based on the phonological attributes posterior probability, and taking these features and boundary information in Conditional Random Fields based phone recognition system. Firstly, calculate the angles between posterior probability vectors of adjacent frames, and then mark the frames with maximum angle as the boundary candidates. Secondly, remove the false phone boundaries through several restrictions in these boundary candidates. Finally, the combination of phonological attributes and phone boundaries are presented as the observation vectors of Conditional Random Fields, experimental results show that the accuracy rate of phoneme recognition is superior to the base system which use phonological attributes features only.

Abstract:A SNR weighted-consensus based cooperative spectrum sensing algorithm was proposed on the basis of the distributed average consensus-based cooperative spectrum sensing algorithm. The algorithm incorporates weights reflecting users’ SNR into the average consensus and the weights are obtained based on the converged average consensus. Without any SNR knowledge, the algorithm enables effective weighted-consensus of the spectrum sensing information in distributed . Both the analysis and simulation indicate that the proposed algorithm can over the inability to distinguish the diversity on users’ SNR of the average consensus-based algorithm, as well as improve the performance of spectrum sensing.

Abstract:Considering that the conventional beam forming based on an acoustic vector sensor array (AVSA) cannot demonstrate the anti-interference capability of AVSA, a direction estimation algorithm based on the signal subspace projection is put forward. In this paper, the 3rd order tensor of the received signals from AVSA is modeled and the signal subspace is derived from higher order singular value decomposition(HOSVD) and the calculation process of spatial spectrum is given by the projection of array manifold vector onto it ,exploiting multidimensional data structure inherently, with lower side lobe which can suppress noise sufficiently. The simulation result shows that the proposed algorithm is feasible which has important engineering significance.

Abstract:Inorder to compare of the synchronization error affect on location accuracy between TOA(Time Of Arrival) and TDOA(Time Difference Of Arrival) schemes ,The synchronization error’s components were analyzed and the clock offset contained measurements model is established, the TOA and TOA/RSS(Received Signal Strength) schemes’ modified Cramer-Rao bound (MCRB) are derived, compare with the TDOA and TDOA/RSS schemes’s Cramer-Rao bound respectively. It’s shown that the transmitter’s clock offset are also the unknown variable in the new observation equation, it can be solved in the TOA scheme to eliminate part of synchronization error; the TDOA scheme sacrifice one measurement to subtract the transmitter’s clock offset from unknown variables lead synchronize requirements can be relaxed. The theoretical derivation and the computer simulation both show that the MCRB are the same for TOA and TDOA schemes and the MCRB are the same for TOA/RSS and TDOA/RSS hybrid schemes when using the same sensor array and the same amount synchronization error were introduced. Especially,the hybrid schemes improve the short distance location accuracy.

Abstract:In cellular network, Non-Line-of-Sight(NLOS) error is the key factor of affecting positioning accuracy. Based on kalman filter and Greenstein model, this paper identifies NLOS error by judgment and directly mitigates it with its mean modifying new abnormal information caused during kalman prediction process, then eliminates measurement error with polynomial smoothing, thus the Time-of-Arrival(TOA) measurements are reconstructed; after that, it modifies the measuring matrix and gets the locating results using the reconstructed measurements. The simulation results show that the proposed method can mitigate NLOS error effectively and improve the location accuracy in the condition of NLOS propagation, meets the demands of E-911 to some extent.

Abstract:The observability and measurement accuracy are low in single observer passive location, so the initial error is usually large. As the unscented kalman filter(ukf) in single observer passive location is sensitive to the initial value and will divergent because of the numerical calculation error, an improved forward-backward smoothing algorithm based on square-root unscented kalman filter(srukf) is presented. To guarantee the stability of filter, the algorithm used the covariance square root matrix instead of covariance matrix in the process of estimation. And the algorithm utilized backward smoothing to get a more accurate state estimate as an initial condition to improve the robustness of the initial value. Simulation results show that the algorithm has better performance to ukf and srukf in the filter’s stability, convergence velocity, positioning precision and the robustness of the initial value.

Abstract:GPS positioning data processing algorithm based on the Markov Chain and Monte Carlo(MCMC) particle filter was proposed to solve the problem that GPS positioning data processing based on the simple particle filter suffers from severe sample degeneracy. The standard MCMC method, Metropolis Hsstings (MH) sampling, was incorporated into the filtering framework , and was applied to the GPS positioning data processing problem. It is combined the particle filter with the GPS system nonlinear dynamic state-space model. The MCMC method is adapted to solve the degeneracy phenomenon of particle filter (PF). It is an effective algorithm to the nonlinear and non-Gaussian state estimation problem of GPS positioning data processing. Experimental results based on the real GPS data showed that the MCMC particle filter can increase the sample variety and reduce sample degeneracy. GPS positioning data processing based on the MCMC particle filter is much more accurate, compared with GPS positioning data processing based on the simple particle filter. Moreover, the MCMC particle filter can provide a high accurate positioning value as an aided plan when the GPS signal quality is poor.

Abstract:Space-time adaptive processing (STAP) is an effective method for moving target detection in airborne radar. However, fast velocity of air target induces serious range walk and Doppler ambiguity. To make things worse, acceleration of the target will result in Doppler spread. Due to the range walk and Doppler spread of maneuvering air target, the performance of the conventional STAP degrades dramatically. To detect the above mentioned fast air maneuvering targets, a new method is proposed in this paper, which removes the clutter before keystone formatting, hence the effects of keystone formatting for the ambiguity targets’ range walk compensation on the clutter distributions and further on the performance of STAP are avoided. Doppler spread of the target is compensated by the estimated acceleration by a modified optimum adaptive processor. Therefore, good detection performance of fast air maneuvering targets can be achieved. Effectiveness of the new method is verified via simulation examples.

Abstract:The threshold-array model is presented to study Suprathreshold Stochastic Resonance(SSR) phenomenon . Analysis of the model proves that that the threshold-array system can be decomposed into a cascade of a single threshold system and an ensemble averager. In order to study the Suprathreshold Stochastic Resonance with periodic input, the statistical properties of the output process of the threshold-array model is evaluated. With a fixed input signal-to-noise ratio, the output SNR Gain of the model varies in a non-monotonic way when injecting independent threshold noises into the array. If the input noise is Gaussian, when adding independent Gaussian white threshold noises into the array, a SNR Gain larger than unity can be obtained. Moreover, when the input noise is non-Gaussian, there will be a better SNR Gain.

Abstract:In this paper, a new method is proposed to detect tampering of intelligent video surveillance and to recognize types of tampering. In order to detect tampering and recognize types of tampering, the method gives every type of tampering a feature of detection such as covered camera, defocused, intensity error, color error, noise of video and so on. At the same time, adaptive threshold is applied to decrease complexity. The experimental results show that the proposed method not only satisfies the demand for real-time of video surveillance system but also has excellent performance of detection and recognition. At last, the accuracy rate is 92.2%.

Abstract:To test the performance of vowel cepstrum in forensic speaker recognition, a forensic speaker identification method based on likelihood ratio and MFCC features is presented in this paper. This method is tested in vowel /a/ of 45 people’s telephone dialog recordings and shows high identification ratio. Experiment results show that not only can the method identify the speaker, but also quantify the evidence strength according the acoustic difference between the questioned recording and the suspect’s recording, and provide the scientific and reasonable evaluation results to court. Compare to the manual method in formants extraction and pitch extraction, the usage of auto extraction of features increases efficiency and performance of forensic speaker identification system.

Abstract:The problem of direction-of-arrival (DOA) tracking for multi-signals is researched based on simplified polarization vector-sensor array, and a novel quaternion data projection method (QDPM) is proposed. The proposed algorithm has high robustness to the undulate phenomenon arisen from the initialization, and converges faster than the conventional DPM method, especially when the DOA is changing drastically. In addition, the QDPM algorithm shows a higher tracking precision than DPM algorithm at low signal-to-noise ration (SNR). Simulation results are finally shown to verify the efficacy of the proposed algorithm.

Abstract:A reliable high-precision and high-performance streamer cable digitizer for marine seismic exploration is proposed. In this precise digitizer, there are total 16 data acquisition channels, each of which can be connected to hydrophones directly. Analog signals are digitized by 24-bit high precise Δ-Σ ADCs controlled by Field Programmable Gate Array (FPGA) in real time. Schemes of real time system control and remote synchronization are also well designed and realized in the digitizer based on the architecture of serial concurrent bus which is implemented by FPGA. The results indicate that this digitizer can meet the requirements of marine seismic exploration.

Abstract:Test sieves are usually used to particle size analysis. Since limit of test facility, complex testing and low accuracy are problems regularly in calibration of test sieves. A automatic calibration method for test sieves using video measuring machine is discussed in this paper in order to increase calibration precision. Firstly, some key technologies are introduced, such as edge detection, mesh edge line fitting, mesh width calculation. Secondly, a sieves calibraion software , which is developed on video measuring machine to prove effectiveness of methods. Thirdly, through comparing with traditional methods, the advantage of video measuring machine on test sieves calibration is discussed, and it maybe get the result of full inspection on every mesh of test sieves.