Search

Target detection is an important problem in multiple-input multiple-output (MIMO) radar. Many existing target detection algorithms were proposed without taking into consideration the quantization error caused by analog-to-digital converters (ADCs). This paper addresses the problem of target detection for MIMO radar with one-bit ADCs and derives a Rao’s test-based detector. The proposed method has several appealing features: 1) it is a closed-form detector; 2) it allows us to handle sign measurements straightforwardly; 3) there are closed-form approximations of the detector’s distributions, which allow us to theoretically evaluate its performance. Moreover, the closed-form distributions allow us to study the performance degradation due to the one-bit ADCs, yielding an approximate 2 dB loss in the low-signal-to-noise-ratio (SNR) regime compared to $\infty$ -bit ADCs. Simulation results are included to showcase the advantage of the proposed detector and validate the accuracy of the theoretical results. [ABSTRACT FROM AUTHOR]

Determining the quality of sensing devices exhibiting minimal digitization complexity is addressed. Measurements of such sensor systems are characterized by multivariate binary distributions and assessing sensitivity via the Cramér-Rao lower bound turns out to be intractable. In this context, the Fisher matrix of the exponential family and a lower bound for arbitrary probabilistic models are discussed. The conservative approximation for Fisher’s information matrix rests on a surrogate exponential family distribution connected to the actual data-generating system by two compact equivalences. Without characterizing the likelihood and its support, this probabilistic notion enables designing estimators that consistently achieve the sensitivity level defined by the inverse of the conservative information matrix. For hard-limited multivariate Gaussian signal models, a quadratic exponential surrogate distribution tames statistical complexity such that a quantitative and conservative assessment of Fisher information becomes possible. This result is exploited for the Fisherian quantization loss analysis of an array with low-complexity binary sensors in comparison to an ideal system featuring infinite amplitude resolution. Additionally, data-driven assessment by estimating a conservative approximation for the Fisher matrix under recursive binary sampling as implemented in ΣΔ-modulating analog-to-digital converters is demonstrated. [ABSTRACT FROM AUTHOR]

We consider the problem of estimating the mean of a symmetric log-concave distribution under the constraint that only a single bit per sample from this distribution is available to the estimator. We study the mean squared error as a function of the sample size (and hence the number of bits). We consider three settings: first, a centralized setting, where an encoder may release $n$ bits given a sample of size $n$ , and for which there is no asymptotic penalty for quantization; second, an adaptive setting in which each bit is a function of the current observation and previously recorded bits, where we show that the optimal relative efficiency compared to the sample mean is precisely the efficiency of the median; lastly, we show that in a distributed setting where each bit is only a function of a local sample, no estimator can achieve optimal efficiency uniformly over the parameter space. We additionally complement our results in the adaptive setting by showing that one round of adaptivity is sufficient to achieve optimal mean-square error. [ABSTRACT FROM AUTHOR]

Several information-theoretic studies on channels with output quantization have identified the capacity-achieving input distributions for different fading channels with 1-bit in-phase and quadrature (I/Q) output quantization. However, an exact characterization of the capacity-achieving input distribution for channels with multi-bit phase quantization has not been provided. In this paper, we consider four different channel models with multi-bit phase quantization at the output and identify the optimal input distribution for each channel model. We first consider a complex Gaussian channel with $b$ -bit phase-quantized output and prove that the capacity-achieving distribution is a rotated $2^{b}$ -phase shift keying (PSK). The analysis is then extended to multiple fading scenarios. We show that the optimality of rotated $2^{b}$ -PSK continues to hold under noncoherent fast fading Rician channels with $b$ -bit phase quantization when line-of-sight (LoS) is present. When channel state information (CSI) is available at the receiver, we identify $\frac {2\pi }{2^{b}}$ -symmetry and constant amplitude as the necessary and sufficient conditions for the ergodic capacity-achieving input distribution; which a $2^{b}$ -PSK satisfies. Finally, an optimum power control scheme is presented which achieves ergodic capacity when CSI is also available at the transmitter. [ABSTRACT FROM AUTHOR]

It is well known that for Gaussian channels, a nearest neighbor decoding rule, which seeks the minimum Euclidean distance between a codeword and the received channel output vector, is the maximum likelihood solution and hence capacity-achieving. Nearest neighbor decoding remains a convenient and yet mismatched solution for general channels, and the key message of this paper is that the performance of nearest neighbor decoding can be improved by generalizing its decoding metric to incorporate channel state dependent output processing and codeword scaling. Using generalized mutual information, which is a lower bound to the mismatched capacity under independent and identically distributed codebook ensemble, as the performance measure, this paper establishes the optimal generalized nearest neighbor decoding rule, under Gaussian channel input. Several restricted forms of the generalized nearest neighbor decoding rule are also derived and compared with existing solutions. The results are illustrated through several case studies for fading channels with imperfect receiver channel state information and for channels with quantization effects. [ABSTRACT FROM AUTHOR]

Consider the following distributed optimization scenario. A worker has access to training data that it uses to compute the gradients while a server decides when to stop iterative computation based on its target accuracy or delay constraints. The server receives all its information about the problem instance from the worker via a rate-limited noiseless communication channel. We introduce the principle we call differential quantization (DQ) that prescribes compensating the past quantization errors to direct the descent trajectory of a quantized algorithm towards that of its unquantized counterpart. Assuming that the objective function is smooth and strongly convex, we prove that differentially quantized gradient descent (DQ-GD) attains a linear contraction factor of $\max \{\sigma _{\mathrm {GD}}, \rho _{n} 2^{-R}\}$ , where $\sigma _{\mathrm {GD}}$ is the contraction factor of unquantized gradient descent (GD), $\rho _{n} \geq 1$ is the covering efficiency of the quantizer, and $R$ is the bitrate per problem dimension $n$. Thus at any $R\geq \log _{2} \rho _{n} /\sigma _{\mathrm {GD}}$ bits, the contraction factor of DQ-GD is the same as that of unquantized GD, i.e., there is no loss due to quantization. We show a converse demonstrating that no algorithm within a certain class can converge faster than $\max \{\sigma _{\mathrm {GD}}, 2^{-R}\}$. Since quantizers exist with $\rho _{n} \to 1$ as $n \to \infty $ (Rogers, 1963), this means that DQ-GD is asymptotically optimal. In contrast, naively quantized GD where the worker directly quantizes the gradient barely attains $\sigma _{\mathrm {GD}} + \rho _{n}2^{-R}$. The principle of differential quantization continues to apply to gradient methods with momentum such as Nesterov’s accelerated gradient descent, and Polyak’s heavy ball method. For these algorithms as well, if the rate is above a certain threshold, there is no loss in contraction factor obtained by the differentially quantized algorithm compared to its unquantized counterpart, and furthermore, the differentially quantized heavy ball method attains the optimal contraction achievable among all (even unquantized) gradient methods. Experimental results on least-squares problems validate our theoretical analysis. [ABSTRACT FROM AUTHOR]

This paper concerns the problem of defending against spoofing attacks without a secret key. We address the problem using the Physical-Layer-Authentication (PLA) because of its high security, low overhead, and high compatibility. However, many PLA schemes have the following limitations: quantization errors, local optimums, and performance loss due to the change in the communication environment. In this paper, two phase-noise-based PLA schemes are proposed to address the limitations of the prior schemes. We denote the first scheme as the Multiple Phase Noises PLA (MPP) scheme, which realizes the PLA by using multiple phase noise innovations. Note that since the MPP scheme avoids using any quantization algorithm, it outperforms the prior schemes on the authentication performance. We denote the second scheme as the Enhanced Multiple Phase Noises PLA (EMPP) scheme, which introduces an artificial random phase to the transmitted symbols at the transmitter to further improve the authentication performance. The theoretical analyses of the proposed schemes over fading channels are provided, where the closed-form expressions are derived. Theoretical comparisons between the proposed schemes and prior schemes are provided. The theoretical analyses and simulation results demonstrated the superiority of the proposed schemes. In comparison with the prior schemes, the MPP scheme achieves 13% authentication-performance gain without demodulation-performance loss, while the EMPP scheme achieves 42% authentication-performance gain with merely 16% demodulation-performance loss. [ABSTRACT FROM AUTHOR]

Real-time monitoring has received considerable attention recently due to its potential in automatic driving, tele-surgery, and factory automation in the 6G era. In remote estimation or reconstruction of stochastic processes, the statistical properties of stochastic processes to be monitored play a central role. Among the stochastic processes interested by real-time applications, the Brownian motion, also known as the Wiener processes is a typical one. In this paper, we are interested in how to monitor Brownian motions efficiently, timely, and reliably. To achieve this goal, we reveal that the real-time estimation error is jointly determined by the quantization error and freshness of data samples. Based on this observation, we present an optimal joint sampling and quantization scheme that efficiently balances the quantization distortion and the age-of-information (AoI). Furthermore, we find that the error accumulation will lead to infinite distortion as monitoring time increases. To overcome this, a multi-layer error correction method is presented for infinite-time monitoring, in which bounded distortion can be achieved with limited data rate. Finally, to conquer the accumulation of transmission errors in unreliable channels, we present an error correction mechanism based on periodic feedback. Diffusion approximation is then adopted to determine the optimal feedback rate and interval. [ABSTRACT FROM AUTHOR]

The polar receiver architecture is a receiver design that captures the envelope and phase information of the signal rather than its in-phase and quadrature components. Several studies have demonstrated the robustness of polar receivers to phase noise and other nonlinearities. Yet, the information-theoretic limits of polar receivers with finite-precision quantizers have not been investigated in the literature. The main contribution of this work is to identify the optimal signaling strategy for the additive white Gaussian noise (AWGN) channel with polar quantization at the output. More precisely, we show that the capacity-achieving modulation scheme has an amplitude phase shift keying (APSK) structure. Using this result, the capacity of the AWGN channel with polar quantization at the output is established by numerically optimizing the probability mass function of the amplitude. The capacity of the polar-quantized AWGN channel with $b_{1}$ -bit phase quantizer and optimized single-bit magnitude quantizer is also presented. Our numerical findings suggest the existence of signal-to-noise ratio (SNR) thresholds, above which the number of amplitude levels of the optimal APSK scheme and their respective probabilities change abruptly. Moreover, the manner in which the capacity-achieving input evolves with increasing SNR depends on the number of phase quantization bits. [ABSTRACT FROM AUTHOR]

Linear prediction and entropy coding are widely used in speech, audio, video, and other coding systems. Rate-distortion optimization (RDO) is another relevant coding technique, which is especially useful when the encoder performance depends on tunable parameters from distinct coding subsystems. This work presents a RDO strategy for variable-rate encoders that use linear prediction and entropy coding. The method is suitable when the signal to be compressed is short-term stationary and the prediction order varies over signal segments. The second contribution of this paper is to customize the method to the compression of time-domain OFDM-based cellular radio signals. The method enables a 4G / 5G packet-based fronthaul to flexibly trade-off rate and distortion depending on the amount of information to be transmitted and, consequently, benefit from multiplexing gain. Simulation results with LTE signals show that the proposed method achieves improved performance with a relatively low computational cost, while operating with time-domain signals under different traffic scenarios. [ABSTRACT FROM AUTHOR]

It is observed that if the premise variables are differentiable, the time derivatives of membership functions are positive or nonpositive. Based on the above fact, the switching observer gains and controller gains are designed to guarantee that the fuzzy system is globally exponentially asymptotically stable by applying the average dwell-time (ADT) technique. In addition, two methods are proposed to find the controller gains and observer gains by dividing the matrices into $3\times 1$ blocks or $2\times 2$ blocks. Finally, two examples are given to verify the conclusion. [ABSTRACT FROM AUTHOR]

In this paper, we compress convolutional neural network (CNN) weights post-training via transform quantization. Previous CNN quantization techniques tend to ignore the joint statistics of weights and activations, producing sub-optimal CNN performance at a given quantization bit-rate, or consider their joint statistics during training only and do not facilitate efficient compression of already trained CNN models. We optimally transform (decorrelate) and quantize the weights post-training using a rate–distortion framework to improve compression at any given quantization bit-rate. Transform quantization unifies quantization and dimensionality reduction (decorrelation) techniques in a single framework to facilitate low bit-rate compression of CNNs and efficient inference in the transform domain. We first introduce a theory of rate and distortion for CNN quantization and pose optimum quantization as a rate–distortion optimization problem. We then show that this problem can be solved using optimal bit-depth allocation following decorrelation by the optimal End-to-end Learned Transform (ELT) we derive in this paper. Experiments demonstrate that transform quantization advances the state of the art in CNN compression in both retrained and non-retrained quantization scenarios. In particular, we find that transform quantization with retraining is able to compress CNN models such as AlexNet, ResNet and DenseNet to very low bit-rates (1–2 bits). [ABSTRACT FROM AUTHOR]

Binary optimization problems (BOPs) arise naturally in many fields, such as information retrieval, computer vision, and machine learning. Most existing binary optimization methods either use continuous relaxation which can cause large quantization errors, or incorporate a highly specific algorithm that can only be used for particular loss functions. To overcome these difficulties, we propose a novel generalized optimization method, named Alternating Binary Matrix Optimization (ABMO), for solving BOPs. ABMO can handle BOPs with/without orthogonality or linear constraints for a large class of loss functions. ABMO involves rewriting the binary, orthogonality and linear constraints for BOPs as an intersection of two closed sets, then iteratively dividing the original problems into several small optimization problems that can be solved as closed forms. To provide a strict theoretical convergence analysis, we add a sufficiently small perturbation and translate the original problem to an approximated problem whose feasible set is continuous. We not only provide rigorous mathematical proof for the convergence to a stationary and feasible point, but also derive the convergence rate of the proposed algorithm. The promising results obtained from four binary optimization tasks validate the superiority and the generality of ABMO compared with the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The recent trend in neural image compression (NIC) research could be generally grounded into two categories: analysis-synthesis transform network improvements and entropy estimation optimization. They promote the compression efficiency of NIC by leveraging more expressive network structures and advanced entropy models respectively. From a different but more systematic viewpoint, we extend the horizon of NIC from software- to hardware-based lossy compression using more resource-constrained platforms, such as field programmable gate array (FPGA) or deep-learning processor unit (DPU). In this paper, we propose a novel hardware-oriented NIC system for real-time edge-computing video services. We for the first time present FPX-NIC, an FPGA-accelerated NIC framework designed for hardware encoding, which consists of a novel NIC scheme and an energy-efficient neural network (NN) deployment method. The former contribution is a block-based adaptive NIC approach based on local content characteristics. Essential side-information is also signalled to realize adaptive patch representation. The critical advantage of our latter contribution lies in the network-reconfigurable framework plus fixed-precision weights quantization method that takes advantage of quantization-aware post training procedure to compensate the performance degradation caused by quantization error. Therefore it is able to improve both processing speed and energy efficiency. We finally establish an intelligent video coding system using the proposed scheme, enabling visual capturing, neural encoding, decoding, and display, realizing 4K ultra-high-definition (UHD) all intra neural video coding on edge-computing devices. [ABSTRACT FROM AUTHOR]

JPEG steganalysis aims to detect stego JPEG images. For some robust steganography methods, in order to enhance stego images’ robustness of resisting JPEG recompression from lossy channel such SNS or photo sharing websites, steganographer may intentionally recompress cover image several times with quantization matrix of targeted channel, which thereby make it possible to transmit stego data in such channel for better disguise. In addition, there are huge number of cover JPEG images may be recompressed for various reasons, such as processing by some tools. Thus a better steganalysis method for such images is needed. In this paper, we investigate the steganalysis method for images recompressed with the same quantization matrix, namely, discriminate recompressed JPEG cover images and its stego images. We present some observed phenomenon on recompressed JPEG images, and design methods to enhance the sensitivity of feature based and deep model based steganalysis methods for this task. To verify their effectiveness with different acquisition of recompression prior-knowledge, we conduct experiments in various settings including conventional setting and mixing samples of different recompressing times in training. Their results demonstrate that the proposed method can notably improve detection accuracy on recompressed JPEG images. [ABSTRACT FROM AUTHOR]

This article is concerned with the issue of quantized sliding-mode control (SMC) design methodology for nonlinear stochastic switching systems subject to semi-Markovian switching parameters, T–S fuzzy strategy, uncertainty, signal quantization, and nonlinearity. Compared with the previous literature, the quantized control input is first considered in studying T–S fuzzy stochastic switching systems with a semi-Markovian process. A mode-independent sliding surface is adopted to avoid the potential repetitive jumping effects. Then, by means of the Lyapunov function, stochastic stability criteria are proposed to be dependent of sojourn time for the corresponding sliding-mode dynamics. Furthermore, the fuzzy-model-based SMC law is proposed to ensure the finite-time reachability of the sliding-mode dynamics. Finally, an application example of a modified series dc motor model is provided to demonstrate the effectiveness of the theoretical findings. [ABSTRACT FROM AUTHOR]

This work addresses quasisynchronization (QS) of the master–slave (MS) neural networks (NNs) with mismatched parameters. The logarithmic quantizer and the round-robin protocol (RRP) are used to deal with the limited communication channel (CC) capacity, then the intermittent control strategy is employed to improve the efficiency of CC and the controller. A transmission-dependent controller is designed, and the synchronization error system (SES) is established. The QS with a boundary is ensured for the MS NNs by a developed sufficient condition, and the controller design method is given. A numerical simulation is given to show the effectiveness of the obtained method. [ABSTRACT FROM AUTHOR]

This article is concerned with the quantized output-feedback control problem for unmanned marine vehicles (UMVs) with thruster faults and ocean environment disturbances via a sliding-mode technique. First, based on output information and compensator states, an augmented sliding surface is constructed and sliding-mode stability through linear matrix inequalities can be guaranteed. An improved quantization parameter dynamic adjustment scheme, with a larger quantization parameter adjustment range, is then given to compensate for quantization errors effectively. Combining the quantization parameter adjustment strategy and adaptive mechanism, a novel robust sliding-mode controller is designed to guarantee the asymptotic stability of a closed-loop UMV system. As a result, a smaller lower bound of the thruster fault factor than that of the existing result can be tolerated, which brings more practical applications. Finally, the comparison simulation results have illustrated the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Gradient compression (e.g., gradient quantization and gradient sparsification) is a core technique in reducing communication costs in distributed learning systems. The recent trend of gradient compression is to use a varying number of bits across iterations, however, relying on empirical observations or engineering heuristics without a systematic treatment and analysis. To the best of our knowledge, a general dynamic gradient compression that leverages both quantization and sparsification techniques is still far from understanding. This paper proposes a novel Adaptively-Compressed Stochastic Gradient Descent (AC-SGD) strategy to adjust the number of quantization bits and the sparsification size with respect to the norm of gradients, the communication budget, and the remaining number of iterations. In particular, we derive an upper bound, tight in some cases, of the convergence error for arbitrary dynamic compression strategy. Then we consider communication budget constraints and propose an optimization formulation - denoted as the Adaptive Compression Problem (ACP) - for minimizing the deep model’s convergence error under such constraints. By solving the ACP, we obtain an enhanced compression algorithm that significantly improves model accuracy under given communication budget constraints. Finally, through extensive experiments on computer vision and natural language processing tasks on MNIST, CIFAR-10, CIFAR-100 and AG-News datasets, respectively, we demonstrate that our compression scheme significantly outperforms the state-of-the-art gradient compression methods in terms of mitigating communication costs. [ABSTRACT FROM AUTHOR]

This article designs performance adjustable event- and self-triggered policies for nonlinear multiagent systems with controller output fluctuations. In the provided event-triggered mechanism, a performance adjustable function is introduced, which can freely adjust the system sampling frequencies and state convergence rates. In the proposed self-triggered policy, continuous measurement error monitoring and neighbors’ event listening are avoided, thereby further saving system sensor resources. Furthermore, under a unified framework, the controllers’ inherent nonlinearities and the influence of data quantization in digital communication networks are studied. In comparison with the previous event-triggered sampling schemes, the developed policies do not only study the multiagent coordinated control in terms of system sampling performance and state convergence performance but are also suitable for the scenarios with controller output fluctuations. Numerical simulation results verify the effectiveness and superiority of the proposed policies. [ABSTRACT FROM AUTHOR]

In this article, the effects of multiantenna users and low-resolution analog-to-digital converters/digital-to-analog converters (ADCs/DACs) on multigroup multicast downlink cell-free massive multi-input multi-output (MIMO) systems are explored. Leveraging on the distributed conjugate beamforming receiver and the generalized additive quantization noise model, we first derive a closed-form expression for the achievable sum spectral efficiency (SE). The resultant expression enables us to precisely quantify the effects of key parameters, such as the transmit power of access points (APs), the numbers of groups, users’ antennas, APs’ antennas, and quantization bits. Due to the joint influence of channel hardening property, multiantenna gain, and intergroup interference, the research becomes sophisticated particularly. The result shows that when the sum-SE is small, the multicast is better than the unicast and vice versa. Typically, it reveals that the full-resolution DAC can be replaced by 5-bit low-resolution DAC to achieve the same performance, and the additional antenna at users can effectively enhance the system performance. Then, according to a generated energy cost model, the formula of the energy efficiency (EE) is obtained. When the DAC quantization is equal to 2-bit, it can achieve the maximum sum EE. Finally, the simulation results are provided to verify the derived results. [ABSTRACT FROM AUTHOR]

Recently, analog in-memory computing (IMC) systems exhibit the considerable potential to break through the inherent high computational latency and energy cost of Von Neumann’s computer architecture. However, inefficient data convertor will inhibit the performance improvement of this system. The tradeoff between different data conversion circuit technologies has turned into one of the major driving forces for the analog IMC system-level improvements. The primary contribution is in two aspects. First, this article shows a digital-to-time-to-analog converter (DTAC) with the tradeoff of latency, area, and power consumption compared to a digital-to-time converter (DTC) and digital-to-analog converter (DAC). Second, we develop an innovative reconfigurable joint-quantization nonlinear analog-to-digital convertor (JQNL-ADC) architecture with lower quantization error by merging the two paradigms of uniform input quantization and uniform output quantization. Compared to conventional DAC, DTAC can reduce power and area by $50\times $ and $3\times $ , respectively. Compared to SAR-ADC, our JQNL-ADC can reduce area and power by $1.6\times $ and $2\times $ , respectively. In an example of ReRAM-based reconfigurable function-IMC (RFIMC) macro with 256-kb memory, our design can reach 112.9 TOPS/W@8bIN-8bW-8bO under the 28-nm process conditions. [ABSTRACT FROM AUTHOR]

Keyword spotting (KWS) has gained popularity as a natural way to interact with consumer devices in recent years. However, because of its always on nature and the variety of speech, it necessitates a low-power design as well as user customization. This article describes a low-power, energy-efficient KWS accelerator with static random access memory (SRAM)-based in-memory computing (IMC) and on-chip learning for user customization. However, IMC is constrained by macro size, limited precision, and nonideal effects. To address the issues mentioned above, this article proposes bias compensation and fine-tuning using an IMC-aware model design. Furthermore, because learning with low-precision edge devices results in zero error and gradient values due to quantization, this article proposes error scaling and small gradient accumulation to achieve the same accuracy as ideal model training. The simulation results show that with user customization, we can recover the accuracy loss from 51.08% to 89.76% with compensation and fine-tuning and further improve to 96.71% with customization. The chip implementation can successfully run the model with only 14 $\mu \text{J}$ per decision. When compared to the state-of-the-art works, the presented design has higher energy efficiency with additional on-chip model customization capabilities for higher accuracy. [ABSTRACT FROM AUTHOR]

The use of low-resolution digital-to-analog and analog-to-digital converters (DACs and ADCs) significantly benefits energy efficiency (EE) at the cost of high quantization noise for massive multiple-input multiple-output (MIMO) systems. This paper considers a precoding optimization problem for maximizing EE in quantized downlink massive MIMO systems. To this end, we jointly optimize an active antenna set, precoding vectors, and allocated power; yet acquiring such joint optimal solution is challenging. To resolve this challenge, we decompose the problem into precoding direction and power optimization problems. For precoding direction, we characterize the first-order optimality condition, which entails the effects of quantization distortion and antenna selection. We cast the derived condition as a functional eigenvalue problem, wherein finding the principal eigenvector attains the best local optimal point. To this end, we propose generalized power iteration based algorithm. To optimize precoding power for given precoding direction, we adopt a gradient descent algorithm for the EE maximization. Alternating these two methods, our algorithm identifies a joint solution of the active antenna set, the precoding direction, and allocated power. In simulations, the proposed methods provide considerable performance gains. Our results suggest that a few-bit DACs are sufficient for achieving high EE in massive MIMO systems. [ABSTRACT FROM AUTHOR]

This article proposes a robust $\mathcal {H}_\infty $ output feedback control strategy to improve vehicle lateral stability through active front wheel steering with quantization error, transmission delay and data dropouts. Since vehicle lateral dynamics presents inherently uncertain challenges, the polytopic technique is used and the variation of tire cornering stiffness is compensated via norm-bounded uncertainty. Meanwhile, considering that the measurement outputs are inevitably suffer from network-induced constrains, i.e. the quantization error, transmission delay and data dropouts, which degrade the stability and handling performance of vehicle, a robust gain-scheduling $\mathcal {H}_\infty $ output feedback controller design strategy is proposed and the stability conditions are presented in the form of linear matrix inequalities that are derived from Lyapunov asymptotic stability and quadratic $\mathcal {H}_\infty $ performance. Finally, two simulation cases are carried out with MATLAB/Simulink-Carsim to validate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

The issues of input–output finite-time stability and stabilization of Takagi–Sugeno (T–S) fuzzy systems with time-varying state delay and exogenous input signal are studied in this article. For the stabilization process, a controller that does not share the same membership functions as the system is considered and the state feedback is quantized by using a logarithmic quantizer. The desired stability criterion for the addressed fuzzy system is first derived without the control term. It is then extended to the case in which the proposed controller is present. The stability criteria are given in the form of matrix inequalities, which are supported by the augmented Lyapunov–Krasovskii functional, generalized free-weighting-matrix approach and Finsler’s lemma. Moreover, as a special case, an asymptotic stability criterion is obtained and used in a comparative study. As a result, the time-delay interval is much larger than some of the works published very recently, which is due to the augmented Lyapunov–Krasovskii functional construction. In addition, the effectiveness and practicability of the theoretical findings are validated with simulation examples. [ABSTRACT FROM AUTHOR]

Entropy is a powerful tool for nonlinear analysis of time series. We have recently introduced dispersion entropy (DispEn), which has widely drawn attention from researchers in a variety of settings in data analysis. However, DispEn is sensitive to its parameters, especially number of classes (quantization level). Additionally, in some cases, a small change in the signal amplitude due to noise changes the quantized series and so entropy value. To address these limitations, we develop fuzzy dispersion entropy (FuzzDispEn) on the basis of fuzzy membership functions for the quantization of a signal and Shannon entropy (ShEn). In FuzzDispEn, the concept of fuzzification in ShEn-based methods is introduced for the first time. Several straightforward concepts in signal processing using a set of time series were utilized to show the advantages of FuzzDispEn over DispEn in terms of the detection of the dynamical variability of signals. The results showed that FuzzDispEn has lower sensitivity to noise and signal length compared to DispEn. FuzzDispEn was also applied to five real world applications: mechanical fault detection in gearbox, fault diagnosis of rolling element bearings, damage level detection of bearings, detection of blood pressure in salt-sensitive rats, and detection of seizure and seizure-free electroencephalographics (EEGs). The results suggest that FuzzDispEn outperforms DispEn as well as popular sample entropy and permutation entropy in distinguishing different kinds of dynamical patterns in synthetic and real data. [ABSTRACT FROM AUTHOR]

A hybrid noise-coupling (HNC) technique is proposed to mitigate the issues with a digital noise-coupling (DNC) structure. In the proposed technique, the unwanted quantization noise that is generated from the noise-coupling analog-to-digital converter (ADC) is injected into the input of the quantizer with one-clock cycle delay, thus increasing the order of noise-shaping of this error. Behavioral simulation shows that the HNC is highly robust against gain variation. The proposed fourth-order continuous-time (CT) delta-sigma modulator (DSM) with HNC scheme was designed in a 28-nm CMOS technology with a core size of 0.3mm2. With an oversampling ratio (OSR) of 32, the proposed design achieves a signal-to-noise-distortion ratio (SNDR) of 97.20-dB at a signal bandwidth of 15.625-kHz, which has a 7-dB improvement in comparison to DNC only without introducing significant hardware complexity. [ABSTRACT FROM AUTHOR]

Resistive random access memory (RRAM)-based compute-in-memory (CIM) has shown great potential for accelerating deep neural network (DNN) inference. However, device characteristics, such as low-resistance values, susceptibility to drift, and single-level cells, may limit the capabilities of RRAM-based CIM. In addition, prior works generally used the off-chip write-verify scheme to tighten RRAM resistance distributions and used off-chip analog-to-digital converter (ADC) references for fine-tuning partial sum quantization. Although off-chip techniques are viable for testing purposes, they may be unsuitable for practical applications. In this work, we present an RRAM-CIM macro to accelerate DNN inference. The chip features: 1) multi-level cell (MLC) RRAM for improving compute performance and density; 2) sparsity-aware input control to leverage the high activation sparsity in DNN models; 3) on-chip write-verify to speed up initial weight programming and periodically refresh cells to compensate for resistance drift under stress; and 4) on-chip ADC reference generation that provides column-wise tunability and stability with varying temperatures to guarantee the CIFAR-10 accuracy of 85.8% at 120 °C. The design is fabricated in TSMC 40-nm process with embedded RRAM technology and achieves a macro-level peak performance of 97.8 GOPS/mm2 and 44.5 TOPS/W for multiply-and-accumulate (MAC) operations on VGG-8 network with ternary weights. [ABSTRACT FROM AUTHOR]

Resistive random-access memory (ReRAM) is an emerging non-volatile memory technology for high-density and high-speed data storage. However, the sneak path interference (SPI) occurred in the ReRAM crossbar array seriously affects its data recovery performance. In this letter, we first propose a quantized channel model of ReRAM, based on which we design both the one-bit and multi-bit channel quantizers by maximizing the mutual information of the channel. A key channel parameter that affects the quantizer design is the sneak path occurrence probability (SPOP) of the memory cell. We first use the average SPOP calculated statistically to design the quantizer, which leads to the same channel detector for different memory arrays. We then adopt the SPOP estimated separately for each memory array for the quantizer design, which is generated by an effective channel estimator and through an iterative detection and decoding scheme for the ReRAM channel. This results in an array-level SPI-aware adaptive detection and decoding approach. Moreover, since there is a strong correlation of the SPI that affects memory cells in the same rows/columns than that affecting cells in different rows/columns, we further derive a column-level scheme which outperforms the array-level scheme. We also propose a channel decomposition method that enables effective ways for theoretically analyzing the ReRAM channel. Simulation results show that the proposed SPI-aware adaptive detection and decoding schemes can approach the ideal performance with three quantization bits, with only one decoding iteration. [ABSTRACT FROM AUTHOR]

Very recently, convolution neural network (CNN) based deep-learning (DL) models have been used in automatic modulation classification (AMC) systems and achieved superior performance. However, the huge CNN model size and floating-point weights and activation make deployment of such systems very complex with limited hardware resources such as FPGAs. In this work, we have designed CNN-based AMC schemes for the complex-valued temporal radio signal domain and implemented them on an FPGA platform. This work focuses on quantized CNN and low precision mathematics to solve the problem of bigger model size and floating-point weights and activations. However, the low precision weights, activation, and quantized CNN significantly impact the model’s accuracy. To overcome this, we proposed an RU-based AMC scheme and used an iterative pruning-based training mechanism to maintain the overall accuracy above a certain threshold while decreasing the model size on hardware. The proposed scheme achieves at least 1.4% higher accuracy and takes only 40% hardware resources than the baseline. The RFSoc implemented model achieves a real-time throughput of $527k$ classification per second with a latency of $7.5\mu\text{s}$. [ABSTRACT FROM AUTHOR]

Signal-to-noise ratio (SNR) estimation for the one-bit analog-to-digital converter (ADC) is a challenging task due to the nonlinearity, lots of information loss, and complex probability distribution. And classical SNR estimators are unavailable in 1-bit sampling. Given this issue, we propose novel one-bit SNR estimators with the closed-form solution based on harmonic analysis to estimate the SNR from 1-bit measurements. We first derive the statistical characteristic of each-order harmonic caused by 1-bit sampling and then give the mathematical relationship between the expectation of harmonic amplitude (EoHA) and the SNR. The ${m}$ th-order harmonic-based estimator (mHE) with high precision is devised by this relationship. Further, an approximate alternative, the ${m}$ th- and ${n}$ th-order harmonics-based ratio estimator (mnHRE), is presented to reduce the computational complexity. Moreover, two schemes for calculating the EoHA are provided to assist the proposed one-bit SNR estimators. Simulation results verify the effectiveness and superiority of our proposed estimators. [ABSTRACT FROM AUTHOR]

This paper is concerned with security control of nonlinear unmanned marine vehicle (UMV) systems under a networked environment. The UMV system and land-based control station are connected by a communication network. Considering the limited communication resource in the marine environment, the dynamic event-triggering mechanisms are proposed in the sensor to controller and controller to actuator sides simultaneously. Meanwhile, the triggered output data is then quantized by a logarithmic quantizer before being sent to the remote control station. First, based on the Takagi-Sugeno (T-S) fuzzy theory, the nonlinear UMV system is molded as a T-S fuzzy model. Then a hybrid switched fuzzy system is established by taking the DoS attack and quantization effect into account. An observer-based sliding mode control (SMC) scheme is proposed to stabilize the system under DoS attack, and the observer gains and controller gains can be obtained by solving a set of matrix inequalities. Finally, a benchmark UMV system is used to show the effectiveness of control scheme. [ABSTRACT FROM AUTHOR]

In contrast to their seemingly simple and shared structure of independence and stationarity, Lévy processes exhibit a wide variety of behaviors, from the self-similar Wiener process to piecewise-constant compound Poisson processes. Inspired by the recent paper of Ghourchian, Amini, and Gohari, we characterize their compressibility by studying the entropy of their double discretization (both in time and amplitude) in the regime of vanishing discretization steps. For a Lévy process with absolutely continuous marginals, this reduces to understanding the asymptotics of the differential entropy of its marginals at small times, for which we obtain a new local central limit theorem. We generalize known results for stable processes to the non-stable case, with a special focus on Lévy processes that are locally self-similar, and conceptualize a new compressibility hierarchy of Lévy processes, captured by their Blumenthal–Getoor index. [ABSTRACT FROM AUTHOR]

This article proposes a novel control design method for high-precision positioning systems. The method aims to eliminate the tracking error caused by measurement quantization present in positioning systems with incremental encoders. By employing a combined internal model based feedback and quantized feedforward design, we are able to make the output of the positioning system asymptotically track any input signal with one or more sinusoidal components of known frequencies and a possible constant component. When combined with a micro-actuator, the resulting dual-stage positioning system is able to track any continuous periodic signal with a known period. Besides theoretical guarantees, the proposed design is validated experimentally and proved able to achieve asymptotic tracking error below $\pm \text{1}\ \mu$ m when subject to a sensor quantization level of 5 $\mu$ m. [ABSTRACT FROM AUTHOR]

In this article, the adaptive attitude-tracking problem of a rigid body is investigated, where the input and output are transmitted via a network. To reduce the communication burden in a network, a quantizer is introduced in both uplink and downlink communication channels. An adaptive backstepping-based control scheme is developed for a class of multiple-input and multiple-output (MIMO) rigid body systems. The proposed control algorithm can overcome the difficulty to proceed with the recursive design of virtual controls with quantized output vector, and a new approach to stability analysis is developed by constructing a new compensation scheme for the effects of the vector output quantization and input quantization. It is shown that all closed-loop signals are ensured uniformly bounded and the tracking errors converge to a compact set containing the origin. Experiments on a 2 degrees-of-freedom helicopter system illustrate the effectiveness of the proposed control scheme. [ABSTRACT FROM AUTHOR]

In this paper, we study a quantized feedback scheme to maximize the goodput of a finite blocklength communication scenario over a quasi-static fading channel. It is assumed that the receiver has perfect channel state information (CSI) and sends back the CSI to the transmitter over a resolution-limited error-free feedback channel. With this partial CSI, the transmitter is supposed to select the optimum transmission rate, such that it maximizes the overall goodput of the communication system. This problem has been studied for the asymptotic blocklength regime, however, no solution has so far been presented for finite blocklength. Here, we study this problem in two cases: with and without constraint on reliability. We first formulate the optimization problems and analytically solve them. Iterative algorithms that successfully exploit the system parameters for both cases are presented. It is shown that although the achievable maximum goodput decreases with shorter blocklengths and higher reliability requirements, significant improvement can be achieved even with coarsely quantized feedback schemes. [ABSTRACT FROM AUTHOR]

The nearest lattice point problem in $\mathbb {R}^{n}$ is formulated in a distributed network with $n$ nodes. The objective is to minimize the probability that an incorrect lattice point is found, subject to a constraint on inter-node communication. Algorithms with a single as well as an unbounded number of rounds of communication are considered for the case $n=2$. For the algorithm with a single round, expressions are derived for the error probability as a function of the total number of communicated bits. We observe that the error exponent depends on the lattice structure and that zero error requires an infinite number of communicated bits. In contrast, with an infinite number of allowed communication rounds, the nearest lattice point can be determined without error with a finite average number of communicated bits and a finite average number of rounds of communication. In two dimensions, the hexagonal lattice, which is most efficient for communication and compression, is found to be the most expensive in terms of communication cost. [ABSTRACT FROM AUTHOR]

This article presents two novel digital predistortion (DPD) based architectures that jointly mitigate the inphase/quadrature (IQ) modulator impairments and the power amplifier (PA) nonlinear distortion in wireless transmitters. The proposed architectures are multibit cartesian and complex delta-sigma modulator-based joint DPDs, called CDSM-JDPD and CXDSM-JDPD, respectively, which enable using low-cost digital-to-analog converters (DACs) while offering versatile linearization capabilities to combat the coexisting distortions of the PA and the IQ modulator. The proposed approach alleviates the need for reverse modeling and implementation of extra hardware to separately deal with frequency-dependent IQ impairments. Moreover, the CXDSM-JDPD enhances the linearization performance and relaxes the high oversampling ratio (OSR) requirement by quantizing the signal more efficiently. Furthermore, the presented concepts inherently support the use of low-resolution DACs, which offers a tremendous advantage in designing and implementing low-cost and energy-efficient radio transmitters. Extensive set of hardware-in-the-loop RF verification measurements with a commercial PA are provided, including two timely 5G New Radio (NR) scenarios at NR bands n3 and n78, while covering channel bandwidths up to 100 MHz and varying the OSR and the DAC bit resolution. The obtained results demonstrate the excellent linearization capabilities of the proposed solutions and their superiority compared to other DSM-based DPD approaches. [ABSTRACT FROM AUTHOR]

Low-rank tensor recovery (LRTR) is a natural extension of low-rank matrix recovery (LRMR) to high-dimensional arrays, which aims to reconstruct an underlying tensor $\boldsymbol{\mathcal {X}}$ X from incomplete linear measurements $\mathfrak {M}(\boldsymbol{\mathcal {X}})$ M (X) . However, LRTR ignores the error caused by quantization, limiting its application when the quantization is low-level. In this work, we take into account the impact of extreme quantization and suppose the quantizer degrades into a comparator that only acquires the signs of $\mathfrak {M}(\boldsymbol{\mathcal {X}})$ M (X) . We still hope to recover $\boldsymbol{\mathcal {X}}$ X from these binary measurements. Under the tensor Singular Value Decomposition (t-SVD) framework, two recovery methods are proposed—the first is a tensor hard singular tube thresholding method; the second is a constrained tensor nuclear norm minimization method. These methods can recover a real $n_1\times n_2\times n_3$ n 1 × n 2 × n 3 tensor $\boldsymbol{\mathcal {X}}$ X with tubal rank $r$ r from $m$ m random Gaussian binary measurements with errors decaying at a polynomial speed of the oversampling factor $\lambda :=m/((n_1+n_2)n_3r)$ λ : = m / ((n 1 + n 2) n 3 r) . To improve the convergence rate, we develop a new quantization scheme under which the convergence rate can be accelerated to an exponential function of $\lambda$ λ . Numerical experiments verify our results, and the applications to real-world data demonstrate the promising performance of the proposed methods. [ABSTRACT FROM AUTHOR]

We study network pruning which aims to remove redundant channels/kernels and hence speed up the inference of deep networks. Existing pruning methods either train from scratch with sparsity constraints or minimize the reconstruction error between the feature maps of the pre-trained models and the compressed ones. Both strategies suffer from some limitations: the former kind is computationally expensive and difficult to converge, while the latter kind optimizes the reconstruction error but ignores the discriminative power of channels. In this paper, we propose a simple-yet-effective method called discrimination-aware channel pruning (DCP) to choose the channels that actually contribute to the discriminative power. To this end, we first introduce additional discrimination-aware losses into the network to increase the discriminative power of the intermediate layers. Next, we select the most discriminative channels for each layer by considering the discrimination-aware loss and the reconstruction error, simultaneously. We then formulate channel pruning as a sparsity-inducing optimization problem with a convex objective and propose a greedy algorithm to solve the resultant problem. Note that a channel (3D tensor) often consists of a set of kernels (each with a 2D matrix). Besides the redundancy in channels, some kernels in a channel may also be redundant and fail to contribute to the discriminative power of the network, resulting in kernel level redundancy. To solve this issue, we propose a discrimination-aware kernel pruning (DKP) method to further compress deep networks by removing redundant kernels. To avoid manually determining the pruning rate for each layer, we propose two adaptive stopping conditions to automatically determine the number of selected channels/kernels. The proposed adaptive stopping conditions tend to yield more efficient models with better performance in practice. Extensive experiments on both image classification and face recognition demonstrate the effectiveness of our methods. For example, on ILSVRC-12, the resultant ResNet-50 model with 30 percent reduction of channels even outperforms the baseline model by 0.36 percent in terms of Top-1 accuracy. We also deploy the pruned models on a smartphone (equipped with a Qualcomm Snapdragon 845 processor). The pruned MobileNetV1 and MobileNetV2 achieve 1.93× and 1.42× inference acceleration on the mobile device, respectively, with negligible performance degradation. The source code and the pre-trained models are available at https://github.com/SCUT-AILab/DCP. [ABSTRACT FROM AUTHOR]

Approximate nearest neighbor search (ANNS) in high-dimensional space is essential in database and information retrieval. Recently, there has been a surge of interest in exploring efficient graph-based indices for the ANNS problem. Among them, navigating spreading-out graph (NSG) provides fine theoretical analysis and achieves state-of-the-art performance. However, we find there are several limitations with NSG: 1) NSG has no theoretical guarantee on nearest neighbor search when the query is not indexed in the database; and 2) NSG is too sparse which harms the search performance. In addition, NSG suffers from high indexing complexity. To address above problems, we propose the satellite system graphs (SSG) and a practical variant NSSG. Specifically, we propose a novel pruning strategy to produce SSGs from the complete graph. SSGs define a new family of MSNETs in which the out-edges of each node are distributed evenly in all directions. Each node in the graph builds effective connections to its neighborhood omnidirectionally, whereupon we derive SSG's excellent theoretical properties for both indexed and unindexed queries. We can adaptively adjust the sparsity of an SSG with a hyper-parameter to optimize the search performance. Further, NSSG is proposed to reduce the indexing complexity of the SSG for large-scale applications. Both theoretical and extensive experimental analysis are provided to demonstrate the strengths of the proposed approach over the existing representative algorithms. Our code has been released at https://github.com/ZJULearning/SSG. [ABSTRACT FROM AUTHOR]

We present a new numerical scheme to solve efficiently scattering problems involving an elongated and flat heterogeneous dielectric material assumed to be invariant along a direction of space. The technique consists of compressing the integral operators of an integro-differential formulation with a so-called quantized tensor train (QTT) algorithm whose use is rather original in this context. We show that it allows to compute and store operators with a notably small memory footprint while having at the same time a fast matrix-vector product (MVP) leading to a competitive method compared with the more classical $\mathcal {H}$ -matrix approach. [ABSTRACT FROM AUTHOR]

Quantization of phase and/or amplitude has far-reaching effects on the radiation characteristics of the phased array (PA), including gain, minor lobe level, and point deviation. Traditionally, one common method to address such a nonlinear distortion is using random-phasing to interrupt the error periodicity. Here, we show that the distortion due to the quantization can be better remediated by spectrally shaping the error compared to the random-phasing (dithering) approaches. We adapted the method for phase-only and amplitude-phase synthesis of planar array designed based on analog beamforming (ABF). To do that, for the first time, 2-D real- and complex-coefficient minimum-phase digital finite impulse response (FIR) filters are designed based on the discrete Hilbert transform (DHT) method. In particular, the digital filter design for phase-only synthesis is comprehensively investigated, respecting the error spectra in the beamspace domain. It is shown that by pushing the error out of the so-called visible region, the decrease of antenna directivity due to the quantization can be compensated to some extent, which provides a quite advantage over the uniform distribution of error. For some cases, pushing the error out of the visible region might be impossible. For such cases, we proposed using the spaced-notches filter. It is also shown that the method is on maximum efficacy when both phase and amplitude of the excitation signal are controllable. Thus, complex-valued noise shaping (CV-NS) can be exploited for the phase-amplitude synthesis of the PA, showing quite promising performance. [ABSTRACT FROM AUTHOR]

Quantization is an important technique to transform the input sample values from a large set (or a continuous range) into the output sample values in a small set (or a finite set). It has been applied broadly for lossy-data compression, pattern recognition, probability density estimation, and clustering. Vector quantization (VQ) is a prevalent image-compression technique, which treats image matrices as stretched vectors and then finds the representative stretched vectors accordingly for a given image data set. One can use tensor data representation to directly characterize the original two-dimensional image data rather than stretch the image matrix into a long vector so as to destroy the original two-dimensional data structure. In this work, we propose a new tensor quantization (TQ) framework which does not need to reduce the dimensionality of the original image data and destroy the original two-dimensional spatial relationship among data; these two serious drawbacks of vector quantization are well known. We first present tensor calculus and then propose a new parallel tensor-inversion algorithm for TQ thereupon. We also establish the pertinent theoretical proof to justify that our proposed new TQ approach is superior to the existing VQ approach especially as the image dimension becomes large. Finally, numerical experiments to evaluate the image-compression performances of VQ and TQ are demonstrated and their corresponding computational-complexities are also compared. [ABSTRACT FROM AUTHOR]

Even if Application-Specific Integrated Circuits (ASIC) have proven to be a relevant choice for integrating inference at the edge, they are often limited in terms of applicability. In this paper, we demonstrate that an ASIC neural network accelerator dedicated to image processing can be applied to multiple tasks of different levels: image classification and compression, while requiring a very limited hardware. The key component is a reconfigurable, mixed-precision (3b/2b/1b) encoder that takes advantage of proper weight and activation quantizations combined with convolutional layer structural pruning to lower hardware-related constraints (memory and computing). We introduce an automatic adaptation of linear symmetric quantizer scaling factors to perform quantized levels equalization, aiming at stabilizing quinary and ternary weights training. In addition, a proposed layer-shared Bit-Shift Normalization significantly simplifies the implementation of the hardware-expensive Batch Normalization. For a specific configuration in which the encoder design only requires 1Mb, the classification accuracy reaches 87.5% on CIFAR-10. Besides, we also show that this quantized encoder can be used to compress image patch-by-patch while the reconstruction can performed remotely, by a dedicated full-frame decoder. This solution typically enables an end-to-end compression almost without any block artifacts, outperforming patch-based state-of-the-art techniques employing a patch-constant bitrate. [ABSTRACT FROM AUTHOR]

This article studies the $H_{\infty }$ exponential synchronization problem for complex networks with quantized control input. An aperiodic sampled-data-based event-triggered scheme is introduced to reduce the network workload. Based on the discrete-time Lyapunov theorem, a new method is adopted to solve the sampled-data problem. In view of the aforementioned method, several sufficient conditions to ensure the $H_{\infty }$ exponential synchronization are acquired. Numerical simulations show that the proposed control schemes can significantly reduce the amount of transmitted signals while preserving the desired system performance. [ABSTRACT FROM AUTHOR]

In this article, a pinning control strategy is developed for the finite-horizon $H_{\infty }$ synchronization problem for a kind of discrete time-varying nonlinear complex dynamical network in a digital communication circumstance. For the sake of complying with the digitized data exchange, a feedback-type dynamic quantizer is introduced to reflect the transformation from the raw signals into the discrete-valued ones. Then, a quantized pinning control scheme takes place on a small fraction of the network nodes with the hope of cutting down the control expenses while achieving the expected global synchronization objective. Subsequently, by resorting to the completing-the-square technique, a sufficient condition is established to ensure the finite-horizon $H_{\infty }$ index of the synchronization error dynamics against both quantization errors and external noises. Moreover, a controller design algorithm is put forward via an auxiliary $H_{2}$ -type criterion, and the desired controller gains are acquired in terms of two coupled backward Riccati equations. Finally, the validity of the presented results is verified via a simulation example. [ABSTRACT FROM AUTHOR]