Search

We provide a rigorous account of the equivalence between the complex-valued widely linear estimation method and the quaternion involution widely linear estimation method with their vector-valued real linear estimation counterparts. This is achieved by an account of degrees of freedom and by providing matrix mappings between a complex variable and an isomorphic bivariate real vector, and a quaternion variable versus a quadri-variate real vector. Furthermore, we show that the parameters in the complex-valued linear estimation method, the complex-valued widely linear estimation method, the quaternion linear estimation method, the quaternion semi-widely linear estimation method, and the quaternion involution widely linear estimation method include distinct geometric structures imposed on complex numbers and quaternions, respectively, whereas the real-valued linear estimation methods do not exhibit any structure. This key difference explains, both in theoretical and practical terms, the advantage of estimation in division algebras (complex, quaternion) over their multivariate real vector counterparts. In addition, we discuss the computational complexities of the estimators of the hypercomplex widely linear estimation methods. [ABSTRACT FROM AUTHOR]

Reduced biquaternion is a four dimensional hyper–complex number which is commutative algebra and an extension of complex. Due to this property, the corresponding fast algorithm is designed in time frequency analysis which can better fit the convolution theorem than the non-commutative quaternion. In addition, the reduced biquaternion can be interpreted as a single point in 2-dimensional hyperbolic space with its algebraic structure which has more capacity and variable ability than the Euclidean space. However, the algebra properties of the reduced biquaternion have not yet been well explored in the deep convolutional neural network. In this paper, we develop a new deep network structure, namely reduced biquaternion valued convolutional neural network (RQV-CNN). The proposed RQV-CNN includes basic modules of reduced biquaternion convolution layer and fully connection layer. Extensive experiments on color image classification and color image denoising are conducted to evaluate the promising performance of the RQV-CNN framework. The results show that RQV-CNN outperforms the real-valued CNN (RV-CNN), complex-valued CNN (CV-CNN), and quaternion-valued CNN (QV-CNN) with same structures. The source code can be found at https://github.com/tasteimage/RQVCNN. [ABSTRACT FROM AUTHOR]

To solve practical challenges for implementing deep learning-based algorithm in MIMO signal detector such as handling complex number or designing proper neural network for a specific communication system, (Baek et al., 2019) has proposed candidate implementation methodologies with simple verification experiments. According to (Baek et al., 2019), it was shown that the proposed algorithms can achieve the optimal ML performance. However, due to the lack of explanation on the experimental environment, it is difficult for readers to reproduce the presented experiments and obtain the same results. This document precisely explains on the experimental environments of (Baek et al., 2019) including the exact channel profiles. [ABSTRACT FROM AUTHOR]

Ship target recognition is a vital task in synthetic aperture radar (SAR) imaging applications. Although convolutional neural networks have been successfully employed for SAR image target recognition, surpassing traditional algorithms, most existing research concentrates on the amplitude domain and neglects the essential phase information. Furthermore, several complex-valued neural networks utilize average pooling to achieve full complex values, resulting in suboptimal performance. To address these concerns, this letter introduces a complex-valued convolutional neural network (CVGG-Net) specifically designed for SAR image ship recognition. CVGG-Net effectively leverages both the amplitude and phase information in complex-valued SAR data. Additionally, this study examines the impact of various widely used complex activation functions on network performance and presents a novel complex max-pooling (CMaxPool) method, called Complex Area Max-Pooling. Experimental results from two measured SAR datasets demonstrate that the proposed algorithm outperforms conventional real-valued convolutional neural networks. The proposed framework is validated on several SAR datasets. [ABSTRACT FROM AUTHOR]

We present two model-based neural network architectures purposed for sporadic user detection and channel estimation in massive machine-type communications. In the scenario under consideration, a base station assigns the users a set of pilot sequences that is linearly dependent, but because user activity is sporadic the detection/estimation problem is amenable to sparse recovery algorithms. Further, we consider a millimeter-wave wireless channel, so that the channel vectors are sparse in a known dictionary. We apply the deep unfolding framework to design custom neural network layers by unrolling two iterative optimization algorithms: (1) linearized alternating direction method of multipliers, which we apply to a constrained convex problem, and (2) vector approximate message passing featuring a novel denoiser based on the iterative shrinkage thresholding algorithm. The networks thus inherit domain knowledge as encapsulated by the signal model, and suitable operations as informed by the algorithms—in the same spirit as convolutional networks that exploit structure inherent in images and audio, except grounded in optimization and statistics. The networks, trained on synthetic data generated from the block-fading millimeter-wave multiple access channel model, offer improved complexity and accuracy relative to their iterative counterparts, and are potentially a boon to cell-free MIMO systems. [ABSTRACT FROM AUTHOR]

Digital watermarking technology and watermark attack methods are mutually reinforcing and complementary. Currently, traditional watermark attack methods are relatively mature, but these traditional attack methods will inevitably damage the visual quality of original images (OIs). Therefore, this paper proposes a covert attack method called residual dense based imperceptible watermark attack network (RD-IWAN). First, this paper designs a watermark attack residual dense network (WARDN) based on the residual dense network (RDN), which can effectively remove the watermark information in the middle and low frequency features of the watermarked image (WMI). Second, to improve the attack ability of the network, this paper innovatively proposes a progressive preprocessing method based on the information enhancement preprocessing method. Concurrently, to ensure the imperceptibility of this watermark attack method, a comprehensive loss function that combines the perceptual loss and mean square error loss (MSE) of OI and attacked watermarked image (AWMI) is designed in this study. Finally, attack experiments are designed and performed on watermarks with different embedding strengths and sizes. Experimental results show that, compared to traditional attack methods, the watermark attack method proposed in this paper exhibits stronger attack ability and higher imperceptibility. [ABSTRACT FROM AUTHOR]

In frequency division duplexing systems, the user equipment frequently reports the channel state information (CSI) to the base station. This information is crucial for link adaptation and beamforming tasks in massive MIMO systems. However, it consumes a large amount of bandwidth when the number of antennas and subcarriers increases. Currently, compression techniques have been used to reduce this bandwidth overhead. In this letter, we propose a novel learning-based technique for CSI compression. Our method exploits the bias/variance tradeoff for CSI compression based on a shallow neural network to approximate a sufficient statistic function for each sampled channel. Unlike prior work that adopts projection-based techniques for compression, our method interprets the model weights as the compressed representation. The experimental results confirm the advantages of the proposed method compared with other state-of-the-art models. [ABSTRACT FROM AUTHOR]

In this paper, we introduce a structure-based neural network architecture, namely RC-Struct, for MIMO-OFDM symbol detection. The RC-Struct exploits the temporal structure of the MIMO-OFDM signals through reservoir computing (RC). A binary classifier leverages the repetitive constellation structure in the system to perform multi-class detection. The incorporation of RC allows the RC-Struct to be learned in a purely online fashion with extremely limited pilot symbols in each OFDM subframe. The binary classifier enables the efficient utilization of the precious online training symbols and allows an easy extension to high-order modulations without a substantial increase in complexity. Experiments show that the introduced RC-Struct outperforms both the conventional model-based symbol detection approaches and the state-of-the-art learning-based strategies in terms of bit error rate (BER). The advantages of RC-Struct over existing methods become more significant when rank and link adaptation are adopted. The introduced RC-Struct sheds light on combining communication domain knowledge and learning-based receive processing for 5G/5G-Advanced and Beyond. [ABSTRACT FROM AUTHOR]

This article presents a tailored artificial neural network (ANN) that is designed for electromagnetic (EM) full-wave inversion (FWI). The ANN includes two subnets. The first subnet is a physics-driven fully connected neural network in which the known sensitivity matrix of EM FWI is assigned to the network model parameters. The second subnet is the traditional data-driven convolutional neural network (CNN) U-Net. The role of the first subnet is to convert the scattered EM fields recorded at the receiver array into the preliminary dielectric images of the scatterers with the aid of the fast calculation of the graphics processing unit. Then, the second subnet is used to further refine the scatterer dielectric parameters. In the off-line training, the network parameters in the first subnet are frozen, and only the parameters in the second subnet are optimized. In the online prediction, the whole ANN is used to directly invert the dielectric parameters of the scatterers. The proposed semiphysics-driven ANN is compared with the purely data-driven ANN for the inversion accuracy, training cost, generalization ability, antinoise ability, and so on. [ABSTRACT FROM AUTHOR]

The constraints of communication channel resources and control energies always lie in real networks, which bring large error and cause adverse affects on system performance, ultimately. To address these limitations, in this article, a distributed event-triggered impulsive control strategy for the consensus problem of fractional-order multiagent networks (FOMANs) is investigated. By collecting the local instantaneous access information, the distributed tracking algorithm with a novel triggering condition is proposed, which is available for both linear and nonlinear systems. Combining with general quadratic Lyapunov stability method and discontinuous control theory, the sufficient conditions of consensus in terms of gain parameter and impulsive interval are derived. Furthermore, the minimal length of interevent is proven to be strictly greater than zero, which guarantees the nonexistence of Zeno-behavior in the closed-loop system. In addition, the implementation of the fractional order algorithm, the quantitative relationship between fractional order $\alpha $ or parameter $a$ in the triggering function, and the number of event-triggered impulsive instants are discussed. At the end of this article, two illustrative examples with numerical and circuit simulations are provided to validate the effectiveness of above results. [ABSTRACT FROM AUTHOR]

Millimeter wave (mmWave) and massive MIMO systems are intrinsic components of 5G and beyond. These systems rely on using beamforming codebooks for both initial access and data transmission. Current beam codebooks, however, generally consist of a large number of narrow beams that scan all possible directions, leading to large training overhead. Further, these codebooks do not normally account for hardware impairments or possible non-uniform array geometries, and their calibration process is expensive. To overcome these limitations, this paper develops an efficient online machine learning framework that learns how to adapt the codebook beam patterns to the specific deployment, surrounding environment, user distribution, and hardware characteristics. This is done by designing a novel complex-valued neural network architecture in which the neuron weights directly model the beamforming weights of the analog phase shifters, accounting for the key hardware constraints. This model learns the codebook beams through online and self-supervised training avoiding the need for explicit channel state information. This respects the practical situations where the channel is imperfect or hard to obtain. Simulation results highlight the capability of the proposed solution in learning environment and hardware aware beam codebooks, which reduce the training overhead and improve the robustness against possible hardware impairments. [ABSTRACT FROM AUTHOR]

To reduce the computational complexity and improve the achievable rate of hybrid precoding, we propose a deep constrained and synchronous training framework, which enables effective learning from the fully digital precoders and approaches the upper bound performance. The key innovation focuses on integrating a constrained precoding network with synchronous loss to predict hybrid precoders that approximate the corresponding fully digital precoders, where the constrained neural network guarantees the unit module and total transmission power while the synchronous loss maximizes the achievable rate. Experimental results show that the hybrid precoder generated from the proposed framework outperforms the benchmark hybrid precoders and could save more than half the inference time. [ABSTRACT FROM AUTHOR]

A deep learning (DL)-based detector utilizing the Transformer framework is proposed for orthogonal frequency-division multiplexing with index modulation (OFDM-IM) systems, termed as TransIM. Concretely, TransIM adopts a two-step detection method. First, the neural networks with the Transformer block as the core provide soft probabilities of different transmitted symbols. Then, conventional signal detection methods are performed based on those probabilities to make final decisions. This method is verified to improve system error performance significantly, albeit at the cost of slightly increased complexity. Simulation results indicate that the proposed TransIM detector fares better than existing DL-based ones regarding bit error rate (BER) performance. [ABSTRACT FROM AUTHOR]

Atrial fibrillation (AF) is the most common arrhythmia, but an estimated 30% of patients with AF are unaware of their conditions. The purpose of this work is to design a model for AF screening from facial videos, with a focus on addressing typical motion disturbances in our real life, such as head movements and expression changes. This model detects a pulse signal from the skin color changes in a facial video by a convolution neural network, incorporating a phase-driven attention mechanism to suppress motion signals in the space domain. It then encodes the pulse signal into discriminative features for AF classification by a coding neural network, using a de-noise coding strategy to improve the robustness of the features to motion signals in the time domain. The proposed model was tested on a dataset containing 1200 samples of 100 AF patients and 100 non-AF subjects. Experimental results demonstrated that VidAF had significant robustness to facial motions, predicting clean pulse signals with the mean absolute error of inter-pulse intervals less than 100 milliseconds. Besides, the model achieved promising performance in AF identification, showing an accuracy of more than 90% in multiple challenging scenarios. VidAF provides a more convenient and cost-effective approach for opportunistic AF screening in the community. [ABSTRACT FROM AUTHOR]

This article, for the first time, extends the zeroing neural network (ZNN) method to address the problem of dynamic quaternion-valued matrix inversion. Due to the noncommutative property of quaternion multiplication, the complex representation method is first adopted to transform quaternion-valued matrices into the corresponding complex-valued matrices. Then, based on two kinds of ways to deal with nonlinear activation functions in the complex-valued domain, this article proposes two quaternion-valued ZNN (QVZNN) models for dynamic quaternion-valued matrix inversion. In addition, a novel nonlinear activation function is given to accelerate the convergence rate of the models to reach the predefined-time convergence. The detailed theoretical analysis, together with four theorems, are given to show the excellent properties of the QVZNN models. Furthermore, the upper bound of the convergence time is derived analytically with the residual error being zero theoretically. Finally, two numerical examples are provided to verify the theoretical results and the effectiveness of the QVZNN models for the dynamic quaternion-valued matrix inversion, and an application to mobile manipulator control is provided to indicate the practical application value of the QVZNN models. [ABSTRACT FROM AUTHOR]

Considering that synthetic aperture radar (SAR) images obtained directly after signal processing are in the form of complex matrices, we propose a complex convolutional network for SAR target recognition. In this article, we give a brief introduction to complex convolutional networks and compare them with the real counterpart. A complex activation function is applied to analyze the influence of phase information in complex neural networks. Inspired by the theory of network visualization, a special kind of transfer learning based on the electromagnetic property from the attributed scattering center model is applied in our networks to modulate the first convolutional layer. The experiment shows a better performance in terms of classification accuracy compared to random weight initialization. [ABSTRACT FROM AUTHOR]

The great success of deep learning (DL) has inspired researchers to develop more accurate and efficient symbol detectors for multi-input multi-output (MIMO) systems. Existing DL-based MIMO detectors, however, suffer several drawbacks. To address these issues, in this paper, we develop a model-driven DL detector based on variational Bayesian inference. Specifically, the proposed unrolled DL architecture is inspired by an inverse-free variational Bayesian learning framework which circumvents matrix inversion via maximizing a relaxed evidence lower bound. Two networks are respectively developed for independent and identically distributed (i.i.d.) Gaussian channels and arbitrarily correlated channels. The proposed networks, referred to as VBINet, have only a few learnable parameters and thus can be efficiently trained with a moderate amount of training samples. The proposed VBINet-based detectors can work in both offline and online training modes. An important advantage of our proposed networks over state-of-the-art MIMO detection networks such as OAMPNet and MMNet is that the VBINet can automatically learn the noise variance from data, thus yielding a significant performance improvement over the OAMPNet and MMNet in the presence of noise variance uncertainty. Simulation results show that the proposed VBINet-based detectors achieve competitive performance for both i.i.d. Gaussian and realistic 3GPP MIMO channels. [ABSTRACT FROM AUTHOR]

For solving complex-valued linear matrix equations with time-varying coefficients (CV-LME-TVC) in the complex field, this article proposes a parameter-changing and complex-valued zeroing neural network (PC-CVZNN) model through integrating a new parameter-changing function. As compared to previous complex-valued zeroing neural networks (CVZNNs) with fixed parameters and existing parameter-changing functions, the PC-CVZNN model can achieve superior performance due to the accelerated role of the new parameter-changing function. In parts of theoretical analysis, we take advantage of Lyapunov methodology to prove that the proposed PC-CVZNN model can acquire the global and super-exponential convergence when the linear activation function is adopted, and even acquire super finite-time convergence when the new sign-bi-power activation function and its modified one are used. In parts of numerical comparison experiments, it is shown that the PC-CVZNN model possesses faster convergence rate than fixed-parameter CVZNN models and other analogy neural networks with parameter-changing function, when applied to finding the solution of CV-LME-TVC. Importantly, an application of the proposed method to the mobile manipulator control provides the potential practical value of the PC-CVZNN model in the industrial field. [ABSTRACT FROM AUTHOR]

Thermostatically controlled load (TCL, such as heating, ventilation, and air conditioning system) is a desirable demand-side flexibility source in distribution networks. It can participate in regulation services and mitigate power imbalances from fluctuating distributed renewable generation. To effectively utilize the load flexibility from spatially and temporally distributed TCLs in a distribution network, it is necessary to consider power flow constraints to avoid possible voltage or current violations. Published works usually adopt optimal power flow models (OPF) to describe these constraints. However, these models require accurate topology of the distribution network that is often unobservable in practice. To bypass this challenge, this paper proposes a novel learning-based OPF to optimize TCLs for regulation services. This method trains three regression multi-layer perceptrons (MLPs) based on the distribution network's historical operation data to replicate its power flow constraints. The trained MLPs are further equivalently reformulated into linear constraints with binary variables so that the optimization problem becomes a mixed-integer linear program that can be effectively solved. Numerical experiments based on the IEEE 123-bus system validate that the proposed method can achieve better TCL power scheduling performance with guaranteed feasibility and optimality than other state-of-art models. [ABSTRACT FROM AUTHOR]

In this paper, a novel end-to-end learning approach, namely JTRD-Net, is proposed for uplink multiuser single-input multiple-output (MU-SIMO) joint transmitter and non-coherent receiver design (JTRD) in fading channels. The basic idea lies in the use of artificial neural networks (ANNs) to replace traditional communication modules at both transmitter and receiver sides. More specifically, the transmitter side is modeled as a group of parallel linear layers, which are responsible for multiuser waveform design; and the non-coherent receiver is formed by a deep feed-forward neural network (DFNN) so as to provide multiuser detection (MUD) capabilities. The entire JTRD-Net can be trained from end to end to adapt to channel statistics through deep learning. After training, JTRD-Net can work efficiently in a non-coherent manner without requiring any levels of channel state information (CSI). In addition to the network architecture, a novel weight-initialization method, namely symmetrical-interval initialization, is proposed for JTRD-Net. It is shown that the symmetrical-interval initialization outperforms the conventional method (e.g. Xavier initialization) in terms of well-balanced convergence-rate among users. Simulation results show that the proposed JTRD-Net approach takes significant advantages in terms of reliability and scalability over baseline schemes on both i.i.d. complex Gaussian channels and spatially-correlated channels. [ABSTRACT FROM AUTHOR]

As a promising neuromorphic framework, the optical neural network (ONN) demonstrates ultrahigh inference speed with low energy consumption. However, the previous ONN architectures have high area overhead which limits their practicality. In this article, we propose an area-efficient ONN architecture based on structured neural networks, leveraging optical fast Fourier transform for efficient computation. A two-phase software training flow with structured pruning is proposed to further reduce the optical component utilization. Experimental results demonstrate that the proposed architecture can achieve 2.2– $3.7\times $ area cost improvement compared with the previous singular value decomposition-based architecture with comparable inference accuracy. A novel optical microdisk-based convolutional neural network architecture with joint learnability is proposed as an extension to move beyond Fourier transform and multilayer perception, enabling hardware-aware ONN design space exploration with lower area cost, higher power efficiency, and better noise-robustness. [ABSTRACT FROM AUTHOR]

The key role of spectrum sensing in cognitive radios attracted substantial research attention to improve the performance of detectors. We consider a general model for the receiver noise with the potential of generalization towards modeling noise in various environments. This general noise model describes accurately noise characteristics ranging from the Gaussian noise to the severe impulsive noise. However, many previous studies are based on the ideal Gaussian noise, and they cannot capture the non-Gaussian models. To provide a robust detector against different behaviors of the noise in various environments, we employ a convolutional neural network (CNN) compatible with different noise models. The proposed CNN detector is data-driven, and due to its single-dimensional input layer, it is consistent with the received signal and requires no pre-processing. The likelihood ratio test (LRT), the Wald, and the Rao tests for this problem are derived to enrich the paper with comparative evaluations of the proposed CNN and conventional model-based approaches and other neural networks. Although various simulated scenarios substantiate the general superiority and robustness of the CNN-based against impulsive noise and mismatch of parameters, it requires higher computational complexity than other discussed detectors. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel class of networks named as quaternion-valued inertial memristor-based neural networks (QVIMNNs) by introducing inertial term and memristor into traditional quaternion-valued neural networks (QVNNs). The problem of fixed-time synchronization of the QVIMNNs is investigated based on the variable transformation and Lyapunov functional method. It is shown that two types of activation functions are considered and a novel criteria guaranteeing fixed-time synchronization for each cases are then achieved by designing different types of controllers. This paper attempts to pave a new way to investigate neural networks classes with numerical simulations support to demonstrate the correctness of the obtained results. [ABSTRACT FROM AUTHOR]

A prototype compressed sensing radar processor boosts the accuracy of target range and velocity estimations by over 6 $\times $ compared with conventional processing techniques. The prototype numerically solves basis pursuit denoising with a biologically plausible spiking neural network. A unique form of weight compression allows on-chip storage of all weights for the large fully connected network. Capable of producing over 200000 range-velocity scene reconstructions per second, the prototype improves throughput by 8 $\times $ and efficiency by 18 $\times $ over the state of the art. [ABSTRACT FROM AUTHOR]

In this paper, we propose a phase reconstruction framework, named Deep Griffin–Lim Iteration (DeGLI). Phase reconstruction is a fundamental technique for improving the quality of sound obtained through some process in the time-frequency domain. It has been shown that the recent methods using deep neural networks (DNN) outperformed the conventional iterative phase reconstruction methods such as the Griffin–Lim algorithm (GLA). However, the computational cost of DNN-based methods is not adjustable at the time of inference, which may limit the range of applications. To address this problem, we combine the iterative structure of GLA with a DNN so that the computational cost becomes adjustable by changing the number of iterations of the proposed DNN-based component. A training method that is independent of the number of iterations for inference is also proposed to minimize the computational cost of the training. This training method, named sub-block training by denoising (SBTD), avoids recursive use of the DNN and enables training of DeGLI with a single sub-block (corresponding to one GLA iteration). Furthermore, we propose a complex DNN based on complex convolution layers with gated mechanisms and investigated its performance in terms of the proposed framework. Through several experiments, we found that DeGLI significantly improved both objective and subjective measures from GLA by incorporating the DNN, and its sound quality was comparable to those of neural vocoders. [ABSTRACT FROM AUTHOR]

Domain adaptation has great values in unpaired cross-modality image segmentation, where the training images with gold standard segmentation are not available from the target image domain. The aim is to reduce the distribution discrepancy between the source and target domains. Hence, an effective measurement for this discrepancy is critical. In this work, we propose a new metric based on characteristic functions of distributions. This metric, referred to as CF distance, enables explicit domain adaptation, in contrast to the implicit manners minimizing domain discrepancy via adversarial training. Based on this CF distance, we propose an unsupervised domain adaptation framework for cross-modality cardiac segmentation, which consists of image reconstruction and prior distribution matching. We validated the method on two tasks, i.e., the CT-MR cross-modality segmentation and the multi-sequence cardiac MR segmentation. Results showed that the proposed explicit metric was effective in domain adaptation, and the segmentation method delivered promising and superior performance, compared to other state-of-the-art techniques. The data and source code of this work has been released via https://zmiclab.github.io/projects.html. [ABSTRACT FROM AUTHOR]

This paper proposes a novel learning to learn method, called learning to learn iterative search algorithm (LISA), for signal detection in a multi-input multi-output (MIMO) system. The idea is to regard the signal detection problem as a decision making problem over tree. The goal is to learn the optimal decision policy. In LISA, deep neural networks are used as parameterized policy function. Through training, optimal parameters of the neural networks are learned and thus optimal policy can be approximated. Different neural network-based architectures are used for fixed and varying channel models, respectively. LISA provides soft decisions and does not require any information about the additive white Gaussian noise. Simulation results show that LISA 1) obtains near maximum likelihood detection performance in both fixed and varying channel models under QPSK modulation; 2) achieves significantly better bit error rate (BER) performance than classical detectors and recently proposed deep/machine learning based detectors at various modulations and signal to noise (SNR) ratios both under i.i.d and correlated Rayleigh fading channels in the simulation experiments; 3) is robust to MIMO detection problems with imperfect channel state information; and 4) generalizes very well against channel correlation and SNRs. [ABSTRACT FROM AUTHOR]

This paper introduces a framework for systematic complexity scaling of deep neural network (DNN) based MIMO detectors. The model uses a fraction of the DNN inputs by scaling their values through weights that follow monotonically non-increasing functions. This allows for weight scaling across and within the different DNN layers in order to achieve accuracy-vs.-complexity scalability during inference. In order to further improve the performance of our proposal, we introduce a sparsity-inducing regularization constraint in conjunction with trainable weight-scaling functions. In this way, the network learns to balance detection accuracy versus complexity while also increasing robustness to changes in the activation patterns, leading to further improvement in the detection accuracy and BER performance at the same inference complexity. Numerical results show that our approach is 10-fold and 100-fold less complex than classical approaches based on semi-definite relaxation and ML detection, respectively. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) have shown good performance in polarimetric synthetic aperture radar (PolSAR) image classification. Excellent hand-crafted CNN architectures incorporated the wisdom of human experts, which is an important reason for CNNs success. However, the design of architectures is a difficult problem, which needs a lot of professional knowledge as well as computational resources. Moreover, the manually designed architecture might be suboptimal, because it is only one of the thousands of unobserved but objective existed paths. Considering that the success of deep learning is largely due to its automation of the feature engineering process, how to design automatic architecture search methods to replace the hand-crafted ones is an interesting topic. In this article, the application of neural architecture search (NAS) in the PolSAR area is explored for the first time. Different from the utilization of existing methods, a PolSAR-tailored Differentiable Architecture Search (DARTS) method, called PDAS, is proposed in order to adapt NAS to the PolSAR classification. A PolSAR-tailored search space and an improved one-shot search strategy are equipped with the proposed method. By PDAS, the architecture (corresponds to the hyperparameter but not the topology) parameters can be optimized with high efficiency by a stochastic gradient descent (SGD) method. The optimized architecture parameters should be transformed into corresponding architecture and retrained to achieve classification. In addition, a complex-valued PDAS (CVPDAS) is developed to fit the data form of PolSAR images so as to improve the performance. Experiments on three benchmark data sets show that the architectures obtained by searching have better classification performance than hand-crafted ones. [ABSTRACT FROM AUTHOR]

In this article, we cover the fundamentals of neural networks and Bayesian learning with a focus on signal and power integrity problems arising in packaging. Rather than only focus on mathematical formulations, we explain the important concepts and the intuition behind them, thereby demystifying the use of machine learning for these problems. We also share some of the recent developments in this area along with future research directions in the context of packaging. Links to open-source downloadable software for some of the methods discussed are also provided. [ABSTRACT FROM AUTHOR]

Index Modulations, in the form of Spatial Modulation or Polarized Modulation, are gaining traction for both satellite and terrestrial next generation communication systems. Adaptive Spatial Modulation based links are needed to fully exploit the transmission capacity of time-variant channels. The adaptation of code and/or modulation requires a real-time evaluation of the channel achievable rates. Some existing results in the literature present a computational complexity which scales quadratically with the number of transmit antennas and the constellation order. Moreover, the accuracy of these approximations is low and it can lead to wrong Modulation and Coding Scheme selection. In this work we apply a Multilayer Feedforward Neural Network to compute the achievable rate of a generic Index Modulation link. The case of two antennas/polarizations is analyzed in depth, showing not only a one-hundred fold decrement of the Mean Square Error in the estimation of the capacity as compared with existing analytical approximations, but also a fifty times reduction of the computational complexity. Moreover, the extension to an arbitrary number of antennas is explained and supported with simulations. More generally, neural networks can be considered as promising candidates for the practical estimation of complex metrics in communication related settings. [ABSTRACT FROM AUTHOR]

Domain-specific hardware is becoming a promising topic in the backdrop of improvement slow down for general-purpose processors due to the foreseeable end of Moore’s Law. Machine learning, especially deep neural networks (DNNs), has become the most dazzling domain witnessing successful applications in a wide spectrum of artificial intelligence (AI) tasks. The incomparable accuracy of DNNs is achieved by paying the cost of hungry memory consumption and high computational complexity, which greatly impedes their deployment in embedded systems. Therefore, the DNN compression concept was naturally proposed and widely used for memory saving and compute acceleration. In the past few years, a tremendous number of compression techniques have sprung up to pursue a satisfactory tradeoff between processing efficiency and application accuracy. Recently, this wave has spread to the design of neural network accelerators for gaining extremely high performance. However, the amount of related works is incredibly huge and the reported approaches are quite divergent. This research chaos motivates us to provide a comprehensive survey on the recent advances toward the goal of efficient compression and execution of DNNs without significantly compromising accuracy, involving both the high-level algorithms and their applications in hardware design. In this article, we review the mainstream compression approaches such as compact model, tensor decomposition, data quantization, and network sparsification. We explain their compression principles, evaluation metrics, sensitivity analysis, and joint-way use. Then, we answer the question of how to leverage these methods in the design of neural network accelerators and present the state-of-the-art hardware architectures. In the end, we discuss several existing issues such as fair comparison, testing workloads, automatic compression, influence on security, and framework/hardware-level support, and give promising topics in this field and the possible challenges as well. This article attempts to enable readers to quickly build up a big picture of neural network compression and acceleration, clearly evaluate various methods, and confidently get started in the right way. [ABSTRACT FROM AUTHOR]

The classification of urban areas in polarimetric synthetic aperture radar (PolSAR) data is a challenging task. Moreover, urban structures oriented away from the radar line of sight pose an additional complexity in the classification process. The characterization of such areas is important for disaster relief and urban sprawl monitoring applications. In this paper, a novel technique based on deep learning is proposed, which leverages a synthetic target database for data augmentation. The PolSAR dataset is rotated by uniform steps and collated to form a reference database. A stacked autoencoder network is used to transform the information in the augmented dataset into a compact representation. This significantly improves the generalization capabilities of the network. Finally, the classification is performed by a multilayer perceptron network. The modular architecture allows for easy optimization of the hyperparameters. The synthetic target database is created and the classification performance is evaluated on an L-band airborne UAVSAR dataset and L-band space-borne ALOS-2 dataset acquired over San Francisco, USA. The proposed technique shows an overall accuracy of 91.3\% . An improvement over state-of-the-art techniques is achieved, especially in urban areas rotated away from the radar line of sight. [ABSTRACT FROM PUBLISHER]