Logarithmic Quantizer Research Articles

Lag quasi-synchronization for periodic neural networks with unreliable redundant communication channels

This work studies lag quasi-synchronization (LQS) for discrete-time master–slave (MS) periodic neural networks (NNs) with the communication channel (CC) constraint. A logarithmic quantizer is used to overcome the CC constraint, and a redundant CC is introduced to transmit more information to improve the system performance. Two independent bernoulli processes are considered to model the packet dropouts of the main and the redundant CCs, respectively. A sufficient condition, ensuring LQS of MS NNs, is achieved, which also gives the boundary of the LQS error. Finally, a controller is designed on the basis of the obtained sufficient condition and the derived result is proved by a numerical example.

Event-triggered scheduling control for linear systems with quantization and time-delay

Quantized feedback for event-triggered control systems is addressed in this paper. The state of the controlled linear plant is sampled and quantized by a logarithmic quantizer upon an event-triggered scheme, then, the quantized data is transmitted and arrives at the controllers’ side along the network after a short time delay. Since event-triggered control can reduce the need for feedback, an event-triggered scheduler is concerned, which decides when continuous measurement state is sampled and quantized. Moreover, based on the event-based quantized feedback, the stabilization of the controlled plant is ensured. It is shown that the proposed event-triggering scheme is robust to quantized error and time-delay. Meanwhile, the minimum inter-event time is established to avoid Zeno behavior. An illustrative example is given to show the effectiveness of our result.

Compression-Robust and Fuzzy-Based Feature-Fusion Model for Optimizing the Iris Recognition

Iris Recognition is gaining popularity in various online and offline authentication and multi-model biometric systems. The non-altering and non-obscuring nature of Iris have increased its reliability in authentication systems. The iris images captured in an uncontrolled environment and situation is the challenging issue of the iris recognition. In this paper, a compression robust and KPCA-Gabor fused model is presented to recognize the iris image accurately under these complexities. The illumination and noise robustness is included in this pre-processing stage for gaining the robustness and reliability against complex capturing. The effective compression features are generated as a phase pre-treatment vector using the Logarithmic quantization method. (Kernel Principal Component Analysis) KPCA and Gabor filters are applied to the rectified image for generating the textural features. The compression is also applied to Gabor and KPCA filtered images. The fuzzy adaptive content level fusion is applied to the compression image, KPCA-Compression, and Gabor-Compression iris-image. (K-Nearest Neighbors) KNN based mapping is used to this composite-fused and reduced feature set to recognize the individual. The proposed compression and fusion-feature based model is applied to CASIA-Iris, UBIRIS, and IITD datasets. The comparative evaluations against earlier approaches identify that the proposed model has improved the recognition accuracy and the reduction in error-rate is also achieved.

Robust $\mathscr{H}_{\infty }$ Consensus for Markov Jump Multiagent Systems Under Mode-Dependent Observer and Quantizer

This article investigates the robust \mathscr H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sub> consensus issue for Markov jump multiagent systems with external disturbances. The agent's information is quantized via a mode-dependent logarithmic quantizer before being transmitted to its neighbors. For each agent, a mode-dependent observer is designed for that only agent's measurement output can be directly transmitted to the neighbors. The original closed-loop system is converted to a reduced-order decomposed system. Then, a mode-dependent and decomposed-system-related Lyapunov function is constructed. According to such a Lyapunov function, some criteria in term of linear matrix inequality on stochastic stability and \mathscr H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sub> -based condition analysis for the reduced-order decomposed system are derived. The observer and consensus controller gains are explicitly deduced with the aid of the singular value decomposition method. In the end, a numerical example and a practical example of LC oscillator network show the feasibility of the developed consensus protocol.

Base-Reconfigurable Segmented Logarithmic Quantization and Hardware Design for Deep Neural Networks

The growth in the size of deep neural network (DNN) models poses both computational and memory challenges to the efficient and effective implementation of DNNs on platforms with limited hardware resources. Our work on segmented logarithmic (SegLog) quantization, adopting both base-2 and base- $\sqrt {2}$ logarithmic encoding, is able to reduce inference cost with a little accuracy penalty. However, weight distribution varies among layers in different DNN models, and requires different base-2 : base- $\sqrt {2}$ ratios to reach the best accuracy. This means different hardware designs for the decoding and computing parts are required. This paper extends the idea of SegLog quantization by using layer-wise base-2 : base- $\sqrt {2}$ ratio on weight quantization. The proposed base-reconfigurable segmented logarithmic (BRSLog) quantization is able to achieve 6.4x weight compression with 1.66% Top-5 accuracy drop on AlexNet at 5-bit resolution. An arithmetic element supporting BRSLog-quantified DNN inference is proposed to adapt to different base-2 : base- $\sqrt {2}$ ratios. With $\sqrt {2}$ approximation, the resource-consuming multipliers can be replaced by shifters and adders with only 0.54% accuracy penalty. The proposed arithmetic element is simulated in UMC 55nm Low Power Process, and it is 50.42% smaller in area and 55.60% lower in power consumption than the widely-used 16-bit fixed-point multiplier. Compared with equivalent SegLog arithmetic element designed for fixed base-2 : base- $\sqrt {2}$ ratio, the base-reconfigurable part only increases the area by 22.96 μm2 and energy cost by 2.6 μW.

Filtering for Discrete-Time Takagi–Sugeno Fuzzy Nonhomogeneous Markov Jump Systems With Quantization Effects

This article deals with the problem of H∞ and l2-l∞ filtering for discrete-time Takagi-Sugeno fuzzy nonhomogeneous Markov jump systems with quantization effects, respectively. The time-varying transition probabilities are in a polytope set. To reduce conservativeness, a mode-dependent logarithmic quantizer is considered in this article. Based on the fuzzy-rule-dependent Lyapunov function, sufficient conditions are given such that the filtering error system is stochastically stable and has a prescribed H∞ or l2-l∞ performance index, respectively. Finally, a practical example is provided to illustrate the effectiveness of the proposed fuzzy filter design methods.

Quantized Control for Synchronization of Delayed Fractional-Order Memristive Neural Networks

This research addresses the synchronization of delayed fractional-order memristive neural networks (DFMNNs) via quantized control. The motivations are twofold: (1) the transmitted information may be constrained by limited bandwidths; (2) the existing analysis techniques are difficult to establish LMI-based synchronization criteria for DFMNNs within a networked control environment. To overcome these difficulties, the logarithmic quantization is adopted to design two types of energy-saving and cost-effective quantized controllers. Then, under the framework of sector bound approach, the closed-loop drive-response DFMNNs can be represented as an interval system with uncertain feedback gains. By utilizing appropriate fractional-order Lyapunov functional and some inequality techniques, two LMI-based synchronization criteria for DFMNNs are derived to establish the relationship between the feedback gain and the quantization parameter. Finally, two illustrative examples are presented to validate the effectiveness of the proposed control schemes.

Fault-tolerant H∞ filtering for fuzzy networked control systems with quantisation effects

ABSTRACTThe issue of memory filter design is investigated for a class of discrete-time Takagi–Sugeno (T-S) fuzzy networked control systems with time-varying delay and quantisation effects. The main objective of this paper is to formulate an memory filter that can guarantee the resulting filtering error system to be finite-time bounded. Further, by considering the communication limitations, the state signals are quantised before transmission through a logarithmic quantiser and the quantisation error is considered in the filter design. In particular, the quantisation effects are dealt by employing the generalised sector bound approach. Based on Lyapunov stability theory, novel sufficient constraints are established in the form of linear matrix inequalities, to ensure the finite-time boundedness of the augmented filtering error system. Then, a new fault-tolerant memory filter is formulated and also the filter gain parameters are attained by solving the derived LMI based constraints. Finally, a numerical example is furnished to support the efficacy and applicability of the developed filter design.

Extended dissipative asynchronous filtering for T–S fuzzy switched systems with partial transition descriptions and incomplete measurements

This paper analyzes the problem of extended dissipative asynchronous filtering for Markov jump T–S fuzzy systems (MJTSFSs) with sensor failures and incomplete measurements. The highlight of this work lies in the fact that we introduce an asynchronous filter (AF) in which mode transition probability matrix (TPM) is non-homogeneous. “Asynchronous” means that the switching of the filters to be designed may be different from that of the systems. Thanks to this AF, partial information of system modes can be fully utilized to achieve the improved and extended dissipative performance including the dissipativity, passivity, H∞ performance and l2−l∞ performance. Specifically, we first attempt to show that the transition probability information (TPI) of the two different Markov chains is not fully known, which can be regarded as an extension of existing work. In the meantime, this is also an arduous problem to be solved in this article. Additionally, with respect to the asynchronous filtering of MJTSFSs, we not only consider that sensor failures occur randomly in the filter systems, but also that research that the measured output is assumed to be quantized by the logarithmic quantizer. Then, an AF is designed for MJTSFSs with sensor failures and incomplete transition probability (ITP) for the first time. Finally, through three examples, the effects of sensor failures, quantizers, and degrees of asynchronism on system performance are examined.

Exponential synchronization of memristive neural networks with time-varying delays via quantized sliding-mode control

In the paper, exponential synchronization issue is considered for memristive neural networks (MNNs) with time-varying delays via quantized sliding-mode algorithm. Quantized Sliding-mode controller is introduced to ensure the slave system can be exponentially synchronized with the host system via the super-twisting algorithm, which has been proved in the main results. Quantization function consists of uniform quantizer and logarithmic quantizer. Simulation results are given with comparisons between two quantizers in the end.

Analysis of stabilization for nonlinear control systems with logarithmic quantized signals

AbstractThe paper is concerned with the problem of input feedback stabilization for nonlinear control systems with logarithmic quantized signals. First of all, the above‐mentioned system is converted into a hybrid control system, and a relation is given between the solutions of the two systems. Then, by using a Lyapunov function approach, a sufficient condition is proposed to ensure that the hybrid system is uniformly globally pre‐asymptotically stable (UGpAS). A crucial step is to find an appropriate Lyapunov function for the system which verifies its stability in the sense of the hybrid framework. Furthermore, a bound of the quantization density is given to guarantee the stability for the nonlinear feedback control systems with the logarithmic quantization. Finally, we give some examples to illustrate the effectiveness of the proposed theorems and compare our results with the existing results.

Quantized passification of delayed memristor-based neural networks via sliding model control

In this paper, quantized passification is investigated for memristive neural networks (MNNs) with time-varying delays via sliding model control. The controller is designed with quantized schemes to reduce the computational complexity via uniform quantization and logarithmic quantizer. By choosing suitable Lyapunov functional and using LMI toolbox, some specific conditions are obtained to make MNN passive. At last, we give an illustrative example to ensure the correctness of the theorem.

Dissipative Fuzzy Filtering for Nonlinear Networked Systems With Limited Communication Links

This paper aims to design a dissipative fuzzy filter for a class of discrete-time nonlinear networked systems. In order to adopt the limited communication links, we employ an event-triggered scheme to reduce the number of transmitted data in the network by preventing the unnecessary ones from releasing. Due to the digital channel, the data to be transmitted should be quantized and a logarithmic quantizer is employed. Then, when the part of released data is transmitted in the network, data losses is captured by a Bernoulli process. Consequently, the uncomplete data sequence is compensated by the buffer and then send to the filter. During this process, a new random series is developed to help constructing the filtering systems. Thus, a novel method is presented to guarantee the filter error system to be dissipative based on the T–S fuzzy model approach. Finally, an example concerned with Henon mapping system is provided to verify the validity of the proposed design method.

Synchronization of discrete-time recurrent neural networks with time-varying delays via quantized sliding mode control

In this paper, we discuss synchronization of discrete-time recurrent neural networks (DRNNs) with time-varying delays via quantized sliding mode control. A feedback controller based on sliding mode control is firstly imported in the synchronization of DRNNs. The activation functional in our paper can be more relaxed than the other papers which should satisfy the Lipschitz conditions. For the sake of reducing the computational complexity and conservatism, we consider two quantized methods with uniform and logarithmic quantizer. We gain some specific conditions to ensure the synchronization of discrete-time system. Several examples are presented to support our theorem in the ending.

Distributed bipartite tracking consensus of nonlinear multi-agent systems with quantized communication

This paper mainly tends to address distributed bipartite tracking consensus problem for nonlinear multi-agent systems subject to logarithmic quantization. Based on appropriate quantized criterion, a distributed protocol with static coupling gain is developed. By employing Lyapunov technique and other mathematical analysis, the results show that distributed bipartite tracking consensus of multi-agent systems can be ensured with quantized relative state measurements, if some conditions are met. Additionally, compared with distributed protocol, the fully distributed protocol with dynamic coupling gain is further designed and analyzed via adaptive control, which is independent of any global information. Finally, simulations are exploited to support our theoretical analysis.

Non-fragile Suboptimal Set-membership Estimation for Delayed Memristive Neural Networks with Quantization via Maximum-error-first Protocol

This paper is concerned with the non-fragile protocol-based set-membership estimation problem for a class of discrete memristive neural networks (MNNs) with mixed time-delays, quantization and unknown but bounded noises. The nonlinear neural activation function satisfies the sector-bounded condition and the logarithmic quantization error is transformed to the norm-bounded uncertainty. In order to save the networks resources, the maximum-error-first (MEF) protocol is introduced to allocate the utilization order of the network channel. The focus is on the design of non-fragile state estimator to ensure such that, in the simultaneous presence of the mixed time-delays, quantization errors and estimator gain perturbations, real state is confined to the ellipsoid. In particular, a minimization problem is given to determine the radius of the designed ellipsoid and the estimator gain matrix by testifying the feasibility of some recursive matrix inequalities. Finally, some simulations are used to show the feasibility of the developed non-fragile suboptimal state estimation strategy.

Robust adaptive attitude control for non-rigid spacecraft with quantized control input

In this paper, an adaptive backstepping control scheme is proposed for attitude tracking of non-rigid spacecraft in the presence of input quantization, inertial uncertainty and external disturbance. The control signal for each actuator is quantized by sector-bounded quantizers, including the logarithmic quantizer and the hysteresis quantizer. By describing the impact of quantization in a new affine model and introducing a smooth function and a novel form of the control signal, the influence caused by input quantization and external disturbance is properly compensated for. Moreover, with the aid of the adaptive control technique, our approach can achieve attitude tracking without the explicit knowledge of inertial parameters. Unlike existing attitude control schemes for spacecraft, in this paper, the quantization parameters can be unknown, and the bounds of inertial parameters and disturbance are also not needed. In addition to proving the stability of the closed-loop system, the relationship between the control performance and design parameters is analyzed. Simulation results are presented to illustrate the effectiveness of the proposed scheme.

A multiple hierarchical structure strategy to quantized control of Markovian switching systems

This paper deals with the issue of nonstationary quantized control of Markovian switching systems. A more general networked Markovian switching system is exploited, where the randomly occurring PDs and a mode-dependent logarithmic quantizer are involved. Different from existing works, a novel nonstationary quantizer is introduced by utilizing a multiple hierarchical structure strategy. On the basis of mode-dependent Lyapunov functional, a sufficient condition is formulated and quantized controller gain is designed. In the end, the effectiveness of the theoretical results are proved by an automotive electronic throttle model.

Quantized $$H_{\infty }$$ Filtering for Continuous-Time Nonhomogeneous Markov Jump Systems

This paper investigates the problems of quantized $$H_{\infty }$$ filtering for continuous-time nonhomogeneous Markov jump systems. The transition probability matrix is assumed to be time-varying and lies in a convex bounded domain. Firstly, we design the $$H_{\infty }$$ filter with a mode-dependent logarithmic quantizer. Then based on the mode-dependent and parameter-dependent Lyapunov function, the stochastic stability with a prescribed $$H_{\infty }$$ performance index is guaranteed by fully considering the information of time-varying transition probability. Specifically, stability criteria are established to make system stochastically stable and a cost function is given to satisfy the $$H_{\infty }$$ performance. Finally, both numerical examples and practical example are given to illustrate the less conservatism and the feasibility of the proposed quantized filer design methods.

Quantized Fuzzy Finite-Time Control for Nonlinear Semi-Markov Switching Systems

This brief considers quantized control for finite-time synthesis of nonlinear semi-Markov switching systems (S-MSSs) via T-S fuzzy strategy. The stochastic phenomena of structural and parametric changes are modeled by the semi-Markov process, in which the sojourn-time (ST) is deemed to obey a non-exponential distribution. Compared with previous works, the input quantization is firstly investigated for studying the finite-time control via a logarithmic quantizer. A key issue under the consideration is how to design a fuzzy-model-based finite-time control law in the presence of quantized error. For this purpose, by using the key point of Lyapunov function, the finite-time boundedness (FTBs) performance is analyzed via establishing sojourn-time-dependent sufficient conditions within a given finite-time level. Then, the existence of a quantized controller is given in standard LMIs. Finally, an example for an electric circuit shows the effectiveness of the finite-time control scheme.