Quantization Mechanism Research Articles

Event-triggered nonlinear state estimation with quantized innovations

This paper proposes an event-triggered (ET) estimator with quantization mechanism to relieve the bandwidth and energy limitations of wireless sensors, where the ET strategy reduces the overall communication rate of the system, and the quantization mechanism mitigates the instantaneous bit rate constraint. Then, Gaussian approximation is applied to derive the analytical form of the minimum mean squared error (MMSE) estimator under the event-triggered strategy and quantization mechanism. Furthermore, the relationship between the event-triggered parameter and the communication rate is explicitly established. Finally, theoretical results are validated through some simulations on a target tracking system.

Trainable pruned ternary quantization for medical signal classification models

The field of deep learning is renowned for its resource-intensive nature, hence improving its environmental impact is crucial. In this paper, we propose a novel model compression method to mitigate the energy demands of deep learning for a greener, and more sustainable AI landscape. Our approach relies on an asymmetric weakly-differentiable pruning function that leverages weight statistics to directly incorporate adaptable pruning into the quantization mechanism. This enables us to achieve higher compression rates globally while simultaneously reducing energy consumption and minimizing classification performance degradation. The efficacy of our approach was evaluated using three distinct models on three distinct datasets: cerebral emboli (HITS), epileptic seizure recognition (ESR), and MNIST. Our method demonstrated a superior balance between compression, energy consumption, and classification performance compared to other state-of-the-art extreme quantization methods, across all models and datasets. In fact, on the HITS dataset with a two-dimensional convolutional neural network, we achieved strong gains of 50.6%, 54.9%, 52.1% in compression rates (of the global model and the quantized layers only, respectively) and energy consumption, respectively, while improving the Matthews correlation coefficient by 2.5% compared to other approaches. The code is available at: https://github.com/yamilvindas/pTTQ.

Multi-scale image compression and reconstruction algorithm for structural health monitoring system

Image data is increasingly common in structural health monitoring systems. However, due to the limitation of hardware resources such as network bandwidth, the data collected by sensors cannot be transmitted efficiently. In order to solve the problem of inefficient data transmission and ensure data recoverability, an end-to-end compression and reconstruction algorithm with an attention-enhanced residual module and a quantization mechanism is proposed. In order to meet the needs of SHM systems for image diversity, multi-scale images of local and global structural damage are used for research. The performance of the proposed algorithm is compared with traditional image compression standards and baseline models. Experimental results show that the peak signal-to-noise ratio of the reconstructed image is greater than 30 dB at different bit rates, effectively reconstructing the original image information; for images with complex textures, the residual block attention module can improve the rate-distortion performance of the proposed algorithm; when the bit rate is less than 0.44 bit per pixel, the peak signal-to-noise ratio and structural similarity of the reconstructed image of the proposed algorithm are better than those of the autoencoder baseline model and Joint Photographic Experts Group compression method. Structural damage detection examples verify the usability of compressed-reconstructed images in engineering applications.

[formula omitted] filtering for uncertain discrete-time singular switched positive systems with time delay and output quantization

This paper focuses on the design problem of an l1 filter for singular switched positive systems with time delay and uncertainty. In order to enhance resource efficiency and information transmission security, the quantization mechanism is introduced when designing the filter. Furthermore, sufficient conditions for causality, regularity, positivity, and exponential stability of the filtering error system are provided under the constraint of mode-dependent minimum dwell time (MDMDT). These conditions are obtained by selecting an appropriate co-positive Lyapunov function and are presented in the form of linear programming (LP). Subsequently, the disturbance attenuation performance is further analyzed, and a design methodology for the corresponding filter is outlined. Finally, a numerical example is presented to verify the validity of the theoretical results.

Observer-based fuzzy control for fractional order PMSG wind turbine systems with adaptive quantized-mechanism

This study aims to present an observer-based fuzzy control approach for a wind turbine system (WTS) equipped with a fractional-order permanent magnet synchronous generator (PMSG) by utilizing an adaptive quantized-mechanism. To do this, firstly, the fractional order control is introduced into the nonlinear PMSG model to improve the convergence rate beyond the existing integer order control techniques. In the next step, the nonlinear PMSG model is converted into linear sub-models using Takagi–Sugeno (T–S) fuzzy models. Using a fractional order Lyapunov function, an adaptive quantization mechanism, and a linear matrix inequality (LMI), we then obtain the necessary conditions for global stabilization of the PMSG-based WTS. Finally, the method proposed in this study demonstrates its superiority and efficiency through comparison with existing methods.

Numerical simulation of Rashba spin–orbit coupling effect on quantum Hall resistivity within tight-binding approximation

A numerical simulation of the effect of Rashba spin–orbit coupling (RSO) on quantum transport properties of electrons confined to a two-dimensional electron gas (2DEG) in the integer quantum Hall regime is presented. The mechanism of quantization and the dependence of the Hall resistivity (HR) on the system built-in and external parameters are studied using the non-equilibrium Green’s function (NEGF) technique. Relevant Green’s functions are computed numerically using spectral decomposition method in combination with recursive techniques. The tight-binding-like representation of the system Hamiltonian including magnetic field is obtained within the finite difference method, using Dirichlet boundary conditions. Corresponding formula linking the Hall resistivity with the transmission function of the system has been derived analytically within the NEGF formalism in a general form, in the sense that the formula can also be used beyond the linear response limit and for nonzero temperatures. In particular, it was shown explicitly that in the presence of the RSO coupling the even- and odd plateaus emerging in the magnetic-field dependence of the quantum Hall resistivity are related to different physical mechanisms.

Minimal Quantization Model in Pilot-Wave Hydrodynamics.

Investigating how classical systems may manifest dynamics analogous to those of quantum systems is a broad subject of fundamental interest. Walking droplets, which self-propel through a resonant interaction with their own wave field, provide a unique macroscopic realization of wave-particle duality that exhibits behaviors previously thought exclusive to quantum particles. Despite significant efforts, elucidating the precise origin and form of the wave-mediated forces responsible for the walker's quantumlike behavior remained elusive. Here, we demonstrate that, owing to wave interference, the force responsible for orbital quantization originates from waves excited near stationary points on the walker's past trajectory. Moreover, we derive a minimal model with the essential ingredients to capture quantized orbital dynamics, including quasiperiodic and chaotic orbits. Notably, this minimal model provides an explicit distinction between local forces, which account for the walker's preferred speed and wave-induced added mass, and spatiotemporal nonlocal forces responsible for quantization. The quantization mechanism revealed here is generic, and will thus play a role in other hydrodynamic quantum analogs.

Dynamic Quantized Consensus Under DoS Attacks: Towards a Tight Zooming-Out Factor

This paper deals with dynamic quantized consensus of dynamical agents in a general form under packet losses induced by Denial-of-Service (DoS) attacks. The communication channel has limited bandwidth and hence the transmitted signals over the network are subject to quantization. To deal with agent's output, an observer is implemented at each node. The state of the observer is quantized by a finite-level quantizer and then transmitted over the network. To solve the problem of quantizer overflow under malicious packet losses, a zooming-in and out dynamic quantization mechanism is designed. By the new quantized controller proposed in the paper, the zooming-out factor is lower bounded by the spectral radius of the agent's dynamic matrix. A sufficient condition of quantization range is provided under which the finite-level quantizer is free of overflow. A sufficient condition of tolerable DoS attacks for achieving consensus is also provided. At last, we study scalar dynamical agents as a special case and further tighten the zooming-out factor to a value smaller than the agent's dynamic parameter. Under such a zooming-out factor, it is possible to recover the level of tolerable DoS attacks to that of unquantized consensus, and the quantizer is free of overflow.

Privacy preserving unique identity generation from multimodal biometric data for privacy and security applications

AbstractThis study presents a novel approach for generating unique identities from multi‐modal biometric data using ensemble feature descriptors extracted from the consistent regions of fingerprint and iris images. The method employs prominent feature selection and discriminant vector generation to enhance intra‐class similarity and inter‐class separability. Finally, a novel quantization mechanism is used to generate a unique identity. This identity might be vulnerable to many attacks. A shielding mechanism is proposed to address this issue. Experimental results substantiate the method's efficacy, satisfying criteria for distinctiveness, randomness, revocability, and irreversibility. Security analyses showcase resilience against diverse attacks, establishing its applicability in forensic investigations, digital wallets, remote authentication, and other privacy‐focused applications. The confidential UID generation scheme ensures privacy and security without involving biometric data or UID enrollment, enhancing its suitability across various applications.

To the theory of dimensional quantization in crystals in the Kane approximation

The microscopic mechanism of size quantization of the conduction band and valence band of semiconductors of cubic and tetrahedral symmetry has been theoretically studied in the Kane approximation. It is shown that the matrix Schrödinger equation obtained on the basis of the Kane model cannot be solved analytically for a potential well of an arbitrary profile. Therefore, expressions for the energy spectrum are obtained depending on the two-dimensional wave vector directed along the interface of the heterostructure for various cases that differ in the range of current carrier energy values.

VaBTFER: An Effective Variant Binary Transformer for Facial Expression Recognition.

Existing Transformer-based models have achieved impressive success in facial expression recognition (FER) by modeling the long-range relationships among facial muscle movements. However, the size of pure Transformer-based models tends to be in the million-parameter level, which poses a challenge for deploying these models. Moreover, the lack of inductive bias in Transformer usually leads to the difficulty of training from scratch on limited FER datasets. To address these problems, we propose an effective and lightweight variant Transformer for FER called VaTFER. In VaTFER, we firstly construct action unit (AU) tokens by utilizing action unit-based regions and their histogram of oriented gradient (HOG) features. Then, we present a novel spatial-channel feature relevance Transformer (SCFRT) module, which incorporates multilayer channel reduction self-attention (MLCRSA) and a dynamic learnable information extraction (DLIE) mechanism. MLCRSA is utilized to model long-range dependencies among all tokens and decrease the number of parameters. DLIE's goal is to alleviate the lack of inductive bias and improve the learning ability of the model. Furthermore, we use an excitation module to replace the vanilla multilayer perception (MLP) for accurate prediction. To further reduce computing and memory resources, we introduce a binary quantization mechanism, formulating a novel lightweight Transformer model called variant binary Transformer for FER (VaBTFER). We conduct extensive experiments on several commonly used facial expression datasets, and the results attest to the effectiveness of our methods.

Disturbance suppression based quantized tracking control for periodic piecewise polynomial systems

The investigation presented here deals with the issue of disturbance suppression and the design of robust tracking control with quantization mechanism for periodic piecewise polynomial systems prone to parameter uncertainties, time delays and external disturbances by means of a proportional integral observer (PIO)-based parallel equivalent input disturbance (PEID) approach. Primarily, in the PEID technique, in order to reduce estimate errors, the PEID notion has been put forward, wherein these errors are perceived as artificial disturbances and a chain of EID compensators has been employed to make amends for them. Meanwhile, PIO’s integrating part aids in blending a relaxing variable into the system’s layout, which allows for greater flexibility in the system’s framework while rendering the system more robust. Subsequently, with the information of estimates from PEID and PIO, disturbance suppression-based quantized tracking control is designed, which simultaneously makes the system states follow the reference states and mitigates the disturbances from the system. Further, the input signals are quantized before being sent as a result of the limited capacity of the channel via which the data is transmitted. Subsequently, through configuring a periodic piecewise polynomial matrix and blending Lyapunov stability theory with the matrix polynomial lemma, adequate requirements are derived in the framework of linear matrix inequalities which ensure the desired outcomes. After which, the requisite controller and observer gain matrices are generated by solving the stated linear matrix inequality-based relations. Ultimately, the simulation portion offers a numerical example that verifies the potential of the acquired findings.

Quantized Berry winding from an emergent $\mathcal{PT}$ symmetry

Linear crossings of energy bands occur in a wide variety of materials. In this paper we address the question of the quantization of the Berry winding characterizing the topology of these crossings in dimension D=2D=2. Based on the historical example of 22-bands crossing occuring in graphene, we propose to relate these Berry windings to the topological Chern number within a D=3D=3 dimensional extension of these crossings. This dimensional embedding is obtained through a choice of a gap-opening potential. We show that the presence of an (emergent) \mathcal{PT}𝒫𝒯 symmetry, local in momentum and antiunitary, allows the quantization of the Berry windings as multiples of \piπ. We illustrate this quantization mechanism on a variety of three-band crossings.

Adaptive event-triggered consensus of multi-agent systems with spherical polar coordinate quantization mechanism

Adaptive event-triggered consensus of multi-agent systems with spherical polar coordinate quantization mechanism

Dynamic quantization‐driven stability of networked control systems under DoS attacks

AbstractThis paper focuses on the stability problem of networked control systems (NCSs) under denial‐of‐service (DoS) attacks and bandwidth limitation. Firstly, the periodic dynamic quantization mechanism is extended to insecure systems, and a periodic‐like quantization strategy is proposed. Due to the designable period, such mechanism allows a trade‐off between better system performance and less energy consumption. With finite quantization levels, a reasonable parametric design is given to avoid quantizer saturation under general energy‐constrained DoS attacks. Secondly, a suitable quantized controller is designed for the system affected by DoS attacks. The relationship between the quantized controller and the DoS attacks is analyzed, and the criterion for NCSs achieving global asymptotic stability is derived. Finally, two numerical examples are given to verify the correctness of the theoretical result.

Adaptive Quantization Mechanism for Federated Learning Models Based on DAG Blockchain

With the development of the power internet of things, the traditional centralized computing pattern has been difficult to apply to many power business scenarios, including power load forecasting, substation defect detection, and demand-side response. How to perform efficient and reliable machine learning tasks while ensuring that user data privacy is not violated has attracted the attention of the industry. Blockchain-based federated learning (FL), proposed as a new decentralized and distributed learning framework for building privacy-enhanced IoT systems, is receiving more and more attention from scholars. The framework provides some advantages, including decentralization, scalability, and data privacy, but at the same time its consensus mechanism consumes a significant amount of computational resources. Moreover, the number of model parameters has increased dramatically, leading to an increasing amount of transmitted data and a vast communication overhead. To reduce the communication overhead, we propose an FL framework in the directed acyclic graph (DAG)-based blockchain system, which achieves efficient and trusted sharing of FL networks. We design an adaptive model compression method based on k-means to compress the FL model and reduce the communication overhead of each round in FL training. Meanwhile, the original accuracy-based tips selection algorithm is optimized, and a tips selection algorithm considering multi-factor evaluation is proposed. Simulation experimental results show that the method proposed in this paper reduces the total bytes of communication of the blockchain-based federated learning system while balancing the accuracy of the FL model compared to previous work.

Set-membership filtering for two-dimensional shift-varying systems with stochastic communication protocol and uniform quantization

This paper is concerned with the set-membership filtering problem for the two-dimensional shift-varying systems subject to the stochastic communication protocol and uniform quantization effects. To prevent the data transmission from collisions, only one sensor can transmit the measured information to the filter at each sampling shift instant, and the selected sensor is determined by the scheduling strategy of the stochastic communication protocol. On the contrary, the uniform quantization mechanism has been employed to mitigate the influence of quantization error on filtering performance. Incorporating the stochastic communication protocol and uniform quantization mechanism, this paper proposes a set-membership filter design framework for the two-dimensional shift-varying systems with unknown-but-bounded noises. Sufficient conditions are derived for the existence of desired set-membership filter by utilizing double mathematical induction, such that the estimation error resides within the ellipsoidal set. Moreover, the optimal filtering algorithm is given by minimizing the ellipsoidal constraints. Finally, several examples are provided to illustrate the effectiveness of the proposed filter design algorithm.

Radiative stability of tiny cosmological constant from the swampland and quantized compactification

We address a quantization mechanism that can allow us to understand why the cosmological constant is not large under the quantum corrections from studying the circle compactification solution of the Standard Model coupled to Einstein gravity which is subject to the constraint of the Swampland conjectures. A novel result in the present work compared to the previous investigations in the literature is that the radius of the compactified dimension and the 4D cosmological constant $\Lambda_4$ must in fact be quantized. The quantization rule of the cosmological constant is given by $\Lambda_4\propto n^2$ with $n=1,2,3,4,\ldots$, which means that the values of $\Lambda_4$ are not arbitrary but only its specific values are allowed. In general, the quantum corrections as well as other effects would break this quantization rule. Hence, it could prevent the quantum fluctuations from generating zero-point energy contributions to the cosmological constant.

State Quantized Output Feedback Control for Nonlinear Systems via Event-Triggered Sampling

The current article proposes a novel output feedback event-triggered control scheme for nonlinear systems considering state quantization. First, we construct a state observer to track the unavailable system states. Second, a dynamic event-triggered and quantized mechanism is constructed to handle the state sampling and quantization. By designing the event-triggered mechanism to sample the system state, and using the quantizer to filtrate the sampled signals, the wasted resources in the data channels can be effectively reduced, and the communication efficiency can be improved. Then, based on the input-to-state stability method, the asymptotic stability is ensured for the closed-loop systems. Moreover, we propose an event-triggered and quantized control scheme for the nonlinear systems with input delay, and the asymptotic stability is ensured using the Lyapunov–Krasovskii functional method. The proposed output feedback event-triggered and quantized mechanisms can guarantee a positive lower bound of the transmission interval times of communication channels, which means that the Zeno behavior is avoided. Finally, a simulation example is presented to verify the feasibility of the proposed control schemes.

Nonuniform Sampling Control for Multibody High-Speed Train Systems With Quantization Mechanisms via Stochastic Faded Channels

This paper studies the dissipative control problem of the multibody high-speed train (HST) systems with nonuniform sampling mechanisms and logarithmic quantizers, in which the transmitted signals are subject to random fading phenomenon. The tracking error dynamic model of HST is firstly established and the logarithmic quantizers for both the input and output (I/O) signals are designed. The faded I/O signals are described by the Rice fading model, in which the mathematical expectation and variance are given in advance. Then, based on the Lyapunov-Krasovskii functional approach with the consideration of time-varying delay, sufficient conditions are derived to ensure the convergence of tracking error and that HST is strictly dissipative. Further, the design method of the gain matrix is obtained by employing the linear matrix inequalities (LMI) techniques and a compensation algorithm is designed to offset the adverse effect brought by the fading measurements. Finally, the effectiveness of the proposed controller is verified by a numerical example from Japan Shinkansen HST.