Yiping TANG Kohei HATANO Eiji TAKIMOTO
We introduce the Hexagonal Convolutional Neural Network (HCNN), a modified version of CNN that is robust against rotation. HCNN utilizes a hexagonal kernel and a multi-block structure that enjoys more degrees of rotation information sharing than standard convolution layers. Our structure is easy to use and does not affect the original tissue structure of the network. We achieve the complete rotational invariance on the recognition task of simple pattern images and demonstrate better performance on the recognition task of the rotated MNIST images, synthetic biomarker images and microscopic cell images than past methods, where the robustness to rotation matters.
Representation learning is a crucial and complex task for multivariate time series data analysis, with a wide range of applications including trend analysis, time series data search, and forecasting. In practice, unsupervised learning is strongly preferred owing to sparse labeling. However, most existing studies focus on the representation of individual subseries without considering relationships between different subseries. In certain scenarios, this may lead to downstream task failures. Here, an unsupervised representation learning model is proposed for multivariate time series that considers the semantic relationship among subseries of time series. Specifically, the covariance calculated by the Gaussian process (GP) is introduced to the self-attention mechanism, capturing relationship features of the subseries. Additionally, a novel unsupervised method is designed to learn the representation of multivariate time series. To address the challenges of variable lengths of input subseries, a temporal pyramid pooling (TPP) method is applied to construct input vectors with equal length. The experimental results show that our model has substantial advantages compared with other representation learning models. We conducted experiments on the proposed algorithm and baseline algorithms in two downstream tasks: classification and retrieval. In classification task, the proposed model demonstrated the best performance on seven of ten datasets, achieving an average accuracy of 76%. In retrieval task, the proposed algorithm achieved the best performance under different datasets and hidden sizes. The result of ablation study also demonstrates significance of semantic relationship in multivariate time series representation learning.
Quanxin MA Xiaolin DU Jianbo LI Yang JING Yuqing CHANG
The estimation problem of structured clutter covariance matrix (CCM) in space-time adaptive processing (STAP) for airborne radar systems is studied in this letter. By employing the prior knowledge and the persymmetric covariance structure, a new estimation algorithm is proposed based on the whitening ability of the covariance matrix. The proposed algorithm is robust to prior knowledge of different accuracy, and can whiten the observed interference data to obtain the optimal solution. In addition, the extended factored approach (EFA) is used in the optimization for dimensionality reduction, which reduces the computational burden. Simulation results show that the proposed algorithm can effectively improve STAP performance even under the condition of some errors in prior knowledge.
Jingke ZHANG Huina SONG Mengyuan WANG Zhaoyang QIU Xuyang TENG Qi ZHANG
Adaptive multilooking is a critical processing step in multi-temporal interferometric synthetic aperture radar (InSAR) measurement, especially in small temporal baseline subsets. Various amplitude-based adaptive multilook approaches have been proposed for the improvement of interferometric processing. However, the phase signal, which is fundamental in interferometric systems, is typically ignored in these methods. To fully exploit the information in complex SAR images, a nonlocal adaptive multilooking is proposed based on complex covariance matrix in this work. The complex signal is here exploited for the similiarity measurement between two pixels. Given the complexity of objects in SAR images, structure feature detection is introduced to adaptively estimate covariance matrix. The effectiveness and reliability of the proposed approach are demonstrated with experiments both on simulated and real data.
Shiwen LIN Yawen ZHOU Weiqin ZOU Huaguo ZHANG Lin GAO Hongshu LIAO Wanchun LI
Estimating the spatial parameters of the signals by using the effective data of a single snapshot is essential in the field of reconnaissance and confrontation. Major drawback of existing algorithms is that its constructed covariance matrix has a great degree of rank loss. The performance of existing algorithms gets degraded with low signal-to-noise ratio. In this paper, a three-parallel linear array based algorithm is proposed to achieve two-dimensional direction of arrival estimates in a single snapshot scenario. The key points of the proposed algorithm are: 1) construct three pseudo matrices with full rank and no rank loss by using the single snapshot data from the received signal model; 2) by using the rotation relation between pseudo matrices, the matched 2D-DOA is obtained with an efficient parameter matching method. Main objective of this work is on improving the angle estimation accuracy and reducing the loss of degree of freedom in single snapshot 2D-DOA estimation.
Expectation propagation (EP) is a powerful algorithm for signal recovery in compressed sensing. This letter proposes correction of a variance message before denoising to improve the performance of EP in the high signal-to-noise ratio (SNR) regime for finite-sized systems. The variance massage is replaced by an observation-dependent consistent estimator of the mean-square error in estimation before denoising. Massive multiple-input multiple-output (MIMO) is considered to verify the effectiveness of the proposed correction. Numerical simulations show that the proposed variance correction improves the high SNR performance of EP for massive MIMO with a few hundred transmit and receive antennas.
Longjiao ZHAO Yu WANG Jien KATO
Recently, local features computed using convolutional neural networks (CNNs) show good performance to image retrieval. The local convolutional features obtained by the CNNs (LC features) are designed to be translation invariant, however, they are inherently sensitive to rotation perturbations. This leads to miss-judgements in retrieval tasks. In this work, our objective is to enhance the robustness of LC features against image rotation. To do this, we conduct a thorough experimental evaluation of three candidate anti-rotation strategies (in-model data augmentation, in-model feature augmentation, and post-model feature augmentation), over two kinds of rotation attack (dataset attack and query attack). In the training procedure, we implement a data augmentation protocol and network augmentation method. In the test procedure, we develop a local transformed convolutional (LTC) feature extraction method, and evaluate it over different network configurations. We end up a series of good practices with steady quantitative supports, which lead to the best strategy for computing LC features with high rotation invariance in image retrieval.
This paper proposes a salient chromagram by removing local trend to improve cover song identification accuracy. The proposed salient chromagram emphasizes tonal contents of music, which are well-preserved between an original song and its cover version, while reducing the effects of timber difference. We apply the proposed salient chromagram to the sequence-alignment based cover song identification. Experiments on two cover song datasets confirm that the proposed salient chromagram improves the cover song identification accuracy.
Di YAO Xin ZHANG Bin HU Xiaochuan WU
A robust adaptive beamforming algorithm is proposed based on the precise interference-plus-noise covariance matrix reconstruction and steering vector estimation of the desired signal, even existing large gain-phase errors. Firstly, the model of array mismatches is proposed with the first-order Taylor series expansion. Then, an iterative method is designed to jointly estimate calibration coefficients and steering vectors of the desired signal and interferences. Next, the powers of interferences and noise are estimated by solving a quadratic optimization question with the derived closed-form solution. At last, the actual interference-plus-noise covariance matrix can be reconstructed as a weighted sum of the steering vectors and the corresponding powers. Simulation results demonstrate the effectiveness and advancement of the proposed method.
Yuta NAKAHARA Toshiyasu MATSUSHIMA
A spatially “Mt. Fuji” coupled (SFC) low-density parity-check (LDPC) ensemble is a modified version of the spatially coupled (SC) LDPC ensemble. Its decoding error probability in the waterfall region has been studied only in an experimental manner. In this paper, we theoretically analyze it over the binary erasure channel by modifying the expected graph evolution (EGE) and covariance evolution (CE) that have been used to analyze the original SC-LDPC ensemble. In particular, we derive the initial condition modified for the SFC-LDPC ensemble. Then, unlike the SC-LDPC ensemble, the SFC-LDPC ensemble has a local minimum on the solution of the EGE and CE. Considering the property of it, we theoretically expect the waterfall curve of the SFC-LDPC ensemble is steeper than that of the SC-LDPC ensemble. In addition, we also confirm it by numerical experiments.
Jun MUNEMORI Kohei KOMORI Junko ITOU
We propose an idea generation support system known as the “GUNGEN-Heartbeat” that uses heartbeat variations for creating high quality ideas during brainstorming. This system shows “An indication of a check list” or “An indication to promote deep breathing” at time beyond a value with variance of heart rates. We also carried out comparison experiments to evaluate the usefulness of the system.
Rachelle RIVERO Yuya ONUMA Tsuyoshi KATO
It has been reported repeatedly that discriminative learning of distance metric boosts the pattern recognition performance. Although the ITML (Information Theoretic Metric Learning)-based methods enjoy an advantage that the Bregman projection framework can be applied for optimization of distance metric, a weak point of ITML-based methods is that the distance threshold for similarity/dissimilarity constraints must be determined manually, onto which the generalization performance is sensitive. In this paper, we present a new formulation of metric learning algorithm in which the distance threshold is optimized together. Since the optimization is still in the Bregman projection framework, the Dykstra algorithm can be applied for optimization. A nonlinear equation has to be solved to project the solution onto a half-space in each iteration. We have developed an efficient technique for projection onto a half-space. We empirically show that although the distance threshold is automatically tuned for the proposed metric learning algorithm, the accuracy of pattern recognition for the proposed algorithm is comparable, if not better, to the existing metric learning methods.
Kaibo CUI Qingping WANG Quan WANG Jingjian HUANG Naichang YUAN
A novel algorithm is proposed for estimating the direction of arrival (DOA) of linear frequency modulated (LFM) signals for the uniform circular array (UCA). Firstly, the UCA is transformed into an equivalent virtual uniform linear array (ULA) using the mode-space algorithm. Then, the short time Fourier transform (STFT) of each element's output is worked out. We can obtain the spatial time-frequency distribution matrix of the virtual ULA by selecting the single-source time-frequency (t-f) points in the t-f plane and then get the signal subspace of the array. The characteristics nature of the Bessel function allow us to obtain the multiple invariance (MI) of the virtual ULA. So the multiple rotational invariant equation of the array can be obtained and its closed-form solution can be worked out using the multi-least-squares (MLS) criterion. Finally, the two dimensional (2-D) DOA estimation of LFM signals for UCA can be obtained. Numerical simulation results illustrate that the UCA-STFT-MI-ESPRIT algorithm proposed in this paper can improve the estimation precision greatly compared with the traditional ESPRIT-like algorithms and has much lower computational complexity than the MUSIC-like algorithms.
Zhe LI Yili XIA Qian WANG Wenjiang PEI Jinguang HAO
A novel time-series relationship among four consecutive real-valued single-tone sinusoid samples is proposed based on their linear prediction property. In order to achieve unbiased frequency estimates for a real sinusoid in white noise, based on the proposed four-point time-series relationship, a constrained least squares cost function is minimized based on the unit-norm principle. Closed-form expressions for the variance and the asymptotic expression for the variance of the proposed frequency estimator are derived, facilitating a theoretical performance comparison with the existing three-point counterpart, called as the reformed Pisarenko harmonic decomposer (RPHD). The region of performance advantage of the proposed four-point based constrained least squares frequency estimator over the RPHD is also discussed. Computer simulations are conducted to support our theoretical development and to compare the proposed estimator performance with the RPHD as well as the Cramer-Rao lower bound (CRLB).
Fengde JIA Zishu HE Yikai WANG Ruiyang LI
In this paper, we propose an online antenna-pulse selection method in space time adaptive processing, while maintaining considerable performance and low computational complexity. The proposed method considers the antenna-pulse selection and covariance matrix estimation at the same time by exploiting the structured clutter covariance matrix. Such prior knowledge can enhance the covariance matrix estimation accuracy and thus can provide a better objective function for antenna-pulse selection. Simulations also validate the effectiveness of the proposed method.
Yutian CHEN Wenyan GAN Shanshan JIAO Youwei XU Yuntian FENG
Recent researches on mobile robots show that convolutional neural network (CNN) has achieved impressive performance in visual place recognition especially for large-scale dynamic environment. However, CNN leads to the large space of image representation that cannot meet the real-time demand for robot navigation. Aiming at this problem, we evaluate the feature effectiveness of feature maps obtained from the layer of CNN by variance and propose a novel method that reserve salient feature maps and make adaptive binarization for them. Experimental results demonstrate the effectiveness and efficiency of our method. Compared with state of the art methods for visual place recognition, our method not only has no significant loss in precision, but also greatly reduces the space of image representation.
Mengbo ZHANG Lunwen WANG Yanqing FENG Haibo YIN
Spectrum sensing is the first task performed by cognitive radio (CR) networks. In this paper we propose a spectrum sensing algorithm for orthogonal frequency division multiplex (OFDM) signal based on deep learning and covariance matrix graph. The advantage of deep learning in image processing is applied to the spectrum sensing of OFDM signals. We start by building the spectrum sensing model of OFDM signal, and then analyze structural characteristics of covariance matrix (CM). Once CM has been normalized and transformed into a gray level representation, the gray scale map of covariance matrix (GSM-CM) is established. Then, the convolutional neural network (CNN) is designed based on the LeNet-5 network, which is used to learn the training data to obtain more abstract features hierarchically. Finally, the test data is input into the trained spectrum sensing network model, based on which spectrum sensing of OFDM signals is completed. Simulation results show that this method can complete the spectrum sensing task by taking advantage of the GSM-CM model, which has better spectrum sensing performance for OFDM signals under low SNR than existing methods.
Shi BAO Zhiqiang LIU Go TANAKA
A new projection-based color-to-gray conversion method is proposed in this letter. In the proposed method, an objective function which considers color contrasts in an input image is defined. Projection coefficients are determined by minimizing the objective function. Experimental results show the validity of the proposed method.
Tarek Hasan AL MAHMUD Zhongfu YE Kashif SHABIR Yawar Ali SHEIKH
Using local time frames to treat non-stationary real world signals as stationary yields Quasi-Stationary Signals (QSS). In this paper, direction of arrival (DOA) estimation of uncorrelated non-circular QSS is analyzed by applying a novel technique to achieve larger consecutive lags using coprime array. A scheme of virtual extension of coprime array is proposed that exploits the difference and sum co-array which can increase consecutive co-array lags in remarkable number by using less number of sensors. In the proposed method, cross lags as well as self lags are exploited for virtual extension of co-arrays both for differences and sums. The method offers higher degrees of freedom (DOF) with a larger number of non-negative consecutive lags equal to MN+2M+1 by using only M+N-1 number of sensors where M and N are coprime with congenial interelement spacings. A larger covariance matrix can be achieved by performing covariance like computations with the Khatri-Rao (KR) subspace based approach which can operate in undetermined cases and even can deal with unknown noise covariances. This paper concentrates on only non-negative consecutive lags and subspace based method like Multiple Signal Classification (MUSIC) based approach has been executed for DOA estimation. Hence, the proposed method, named Virtual Extension of Coprime Array imbibing Difference and Sum (VECADS), in this work is promising to create larger covariance matrix with higher DOF for high resolution DOA estimation. The coprime distribution yielded by the proposed approach can yield higher resolution DOA estimation while avoiding the mutual coupling effect. Simulation results demonstrate its effectiveness in terms of the accuracy of DOA estimation even with tightly aligned sources using fewer sensors compared with other techniques like prototype coprime, conventional coprime, Coprime Array with Displaced Subarrays (CADiS), CADiS after Coprime Array with Compressed Inter-element Spacing (CACIS) and nested array seizing only difference co-array.
In statistical approaches such as statistical static timing analysis, the distribution of the maximum of plural distributions is computed by repeating a maximum operation of two distributions. Moreover, since each distribution is represented by a linear combination of several explanatory random variables so as to handle correlations efficiently, sensitivity of the maximum of two distributions to each explanatory random variable, that is, covariance between the maximum and an explanatory random variable, must be calculated in every maximum operation. Since distribution of the maximum of two Gaussian distributions is not a Gaussian, Gaussian mixture model is used for representing a distribution. However, if Gaussian mixture models are used, then it is not always possible to make both variance and covariance of the maximum correct simultaneously. We propose a new algorithm to determine covariance without deteriorating the accuracy of variance of the maximum, and show experimental results to evaluate its performance.