Application of the N + 2 Transversal Network Method to the Study of a Coupled Resonator Filter

This paper presents a new approach to synthesize admittance function polynomials and coupling matrices for coupled resonator filters. The N + 2 transversal network method is applied to study a coupled resonator filter. This method allowed us to determine the polynomials of the reflection and transmission coefficients. A study is made for a 4 poles filter with 2 transmission zeros between the N + 2 transversal network method and the one found in the literature. A MATLAB code was designed for the numerical simulation of these coefficients for the 6, 8, and 10 pole filter with 4 transmission zeros.

Study on Robust H_∞ Filtering in Networked Environments

This paper is concerned with the robust H ∞ filter problem for networked environments, which are subject to both transmission delay and packet dropouts randomly. By employing random series which have Bernoulli distributions taking value of 0 or 1, the data transmission model is obtained. Based on state augmentation and stochastic theory, the sufficient condition for robust stability with H ∞ constraints is derived for the filtering error system. The robust filter is designed in terms of feasibility of one certain linear matrix inequality （LMI）, which is formed by adopting matrix congruence transformations. A numerical example is given to show the effectiveness of the proposed filtering method.

Design of twodimensional digital filters using neural networks

A new approach for the design of two-dimensional （2-D） linear phase FIR digital filters based on a new neural networks algorithm （NNA） is provided. A compact expression for the transfer function of a 2-D linear phase FIR filter is derived based on its frequency response characteristic, and the NNA, based on minimizing the square-error in the frequency-domain, is established according to the compact expression. To illustrate the stability of the NNA, the convergence theorem is presented and proved. Design examples are also given, and the results show that the ripple is considerably small in passband and stopband, and the NNA-based method is of powerful stability and requires quite little amount of computations.

A Filter-Based Uniform Algorithm for Optimizing Top-k Query in Distributed Networks

In this paper we propose a Filter-based Uniform Algorithm （FbUA） for optimizing top-κ query in distributed networks, which has been a topic of much recent interest. The basic idea of FhUA is to set a filter at each node to pre vent it from sending out the data with little chance to contrib ute to the top-κ result. FbUA can gain exact answers to top-κ query through two phrases of round trip communications between query station and participant nodes. The experiment results show that FbUA reduces network bandwidth consumption dramatically.

Fault Detection of Networked Control Systems Based on Optimal Robust Fault Detection Filter

有可能比一个采样时期大的随机、未知的导致网络的延期的联网的控制系统(NCS ) 的差错察觉在这份报纸被学习。首先，导致网络的延期引起的影响被转变成为错误建模组织，然后存在连续时间域基于引用，模型被扩大到分离时间领域并且适用指责联网的控制系统的察觉的柔韧的差错察觉方法。建议方法能被 Matlab LMI 工具箱容易实现，并且它的表演被一个模拟例子最后评估。

Filter Bank Networks for Few-Shot Class-Incremental Learning

Deep Convolution Neural Networks(DCNNs)can capture discriminative features from large datasets.However,how to incrementally learn new samples without forgetting old ones and recognize novel classes that arise in the dynamically changing world,e.g.,classifying newly discovered fish species,remains an open problem.We address an even more challenging and realistic setting of this problem where new class samples are insufficient,i.e.,Few-Shot Class-Incremental Learning(FSCIL).Current FSCIL methods augment the training data to alleviate the overfitting of novel classes.By contrast,we propose Filter Bank Networks(FBNs)that augment the learnable filters to capture fine-detailed features for adapting to future new classes.In the forward pass,FBNs augment each convolutional filter to a virtual filter bank containing the canonical one,i.e.,itself,and multiple transformed versions.During back-propagation,FBNs explicitly stimulate fine-detailed features to emerge and collectively align all gradients of each filter bank to learn the canonical one.FBNs capture pattern variants that do not yet exist in the pretraining session,thus making it easy to incorporate new classes in the incremental learning phase.Moreover,FBNs introduce model-level prior knowledge to efficiently utilize the limited few-shot data.Extensive experiments on MNIST,CIFAR100,CUB200,andMini-ImageNet datasets show that FBNs consistently outperformthe baseline by a significantmargin,reporting new state-of-the-art FSCIL results.In addition,we contribute a challenging FSCIL benchmark,Fishshot1K,which contains 8261 underwater images covering 1000 ocean fish species.The code is included in the supplementary materials.

Kalman filter based fault diagnosis of networked control system with white noise

The networked control system NCS is regarded as a sampled control system withoutput time-variant delay. White noise is considered in the model construction of NCS. By using theKalman filter theory to compute the filter parameters, a Kalman filter is constructed for this NCS.By comparing the output of the filter and the practical system, a residual is generated to diagnoseme sensor faults and the actuator faults. Finally, an example is given to show the feasibility ofthe approach.

Optimal design study of high order FIR digital filters based on neural network algorithm

An optimal design approach of high order FIR digital filter is developed based on the algorithm of neural networks with cosine basis function . The main idea is to minimize the sum of the square errors between the amplitude response of the desired FIR filter and that of the designed by training the weights of neural networks, then obtains the impulse response of FIR digital filter . The convergence theorem of the neural networks algorithm is presented and proved, and the optimal design method is introduced by designing four kinds of FIR digital filters , i.e., low-pass, high-pass, bandpass , and band-stop FIR digital filter. The results of the amplitude responses show that attenuation in stop-bands is more than 60 dB with no ripple and pulse existing in pass-bands, and cutoff frequency of passband and stop-band is easily controlled precisely .The presented optimal design approach of high order FIR digital filter is significantly effective.

ADAPTIVE RECURRENT NEURAL NETWORKS TRACKING-FILTER FOR MANEUVERING TARGET

It is a challenge to track the maneuvering targets with noise disturbance and unknown dynamics. In this paper, an adaptive recurrent neural network tracking filter (ARNNF) for use in maneuvering target tracking was provided. The scheme is based on recurrent neural networks of which the recurrence provides a potentially unlimited memory depth adjusted by the network adaptively ( i.e. , it finds the best duration to represent the input signals past), and thus can actually capture the dynamics of the system that produced a temporal signal. On the other hand, recurrent neural network can approximate arbitrary nonlinear functions in L 2 space. The theoretical analysis indicates that the ARNNF can track the maneuvering targets with optimal filtering performance. Comparisons with IMM and AIMM algorithm show that ARNNF has better performance, and furthermore the ARNNF does not rely on the assumption with the known maneuvering target models, measurement noise and system noise.

Fuzzy neural network image filter based on GA

A new nonlinear image filter using fuzzy neural network based on genetic algorithm is proposed. The learning of network parameters is performed by genetic algorithm with the efficient binary encoding scheme. In the following, fuzzy reasoning embedded in the network aims at restoring noisy pixels without degrading the quality of fine details. It is shown by experiments that the filter is very effective in removing impulse noise and significantly outperforms conventional filters.

Nonlinear Filtering With Sample-Based Approximation Under Constrained Communication:Progress, Insights and Trends

The nonlinear filtering problem has enduringly been an active research topic in both academia and industry due to its ever-growing theoretical importance and practical significance.The main objective of nonlinear filtering is to infer the states of a nonlinear dynamical system of interest based on the available noisy measurements. In recent years, the advance of network communication technology has not only popularized the networked systems with apparent advantages in terms of installation,cost and maintenance, but also brought about a series of challenges to the design of nonlinear filtering algorithms, among which the communication constraint has been recognized as a dominating concern. In this context, a great number of investigations have been launched towards the networked nonlinear filtering problem with communication constraints, and many samplebased nonlinear filters have been developed to deal with the highly nonlinear and/or non-Gaussian scenarios. The aim of this paper is to provide a timely survey about the recent advances on the sample-based networked nonlinear filtering problem from the perspective of communication constraints. More specifically, we first review three important families of sample-based filtering methods known as the unscented Kalman filter, particle filter,and maximum correntropy filter. Then, the latest developments are surveyed with stress on the topics regarding incomplete/imperfect information, limited resources and cyber security.Finally, several challenges and open problems are highlighted to shed some lights on the possible trends of future research in this realm.

Simulation of Cellular Neural Networks by Wave Digital Filter Principles

Based on wave digital filter(WDF) principles, this paper presents a digital model of cellular neural networks(CNNs). The model can precisely simulate the dynamic behavior of CNNs.

Multiple vehicle signals separation based on particle filtering in wireless sensor network

A novel statistical method based on particle filtering is presented for multiple vehicle acoustic signals separation problem in wireless sensor network. The particle filtering method is able to deal with non-Gaussian and nonlinear models and non-stationary sources. Using some instantaneously mixed observations of several real-world vehicle acoustic signals, the proposed statistical method is compared with a conventional non-stationary Blind Source Separation algorithm and attractive simulation results are achieved. Moreover, considering the natural convenience to transmit particles between sensor nodes, the algorithm based on particle filtering is believed to have potential to enable the task of multiple vehicles recognition collaboratively performed by sensor nodes in distributed wireless sensor network.

Filtering and Estimation of Vehicular Dead Reckoning System Based on Hopfield Neural Network

The algorithm of Hopfield neural network filtering and estimation is studied. The model of vehicular dead reckoning system fitting for the algorithm is constructed, and the design scheme of system filtering and estimation based on Hopfield network is proposed. Compared with Kalman filter, the algorithm does not require very precise system model and the prior knowledge of noise statistics and does not diverge easily. The simulation results show that the vehicular dead reckoning system based on Hopfield network filtering and estimation has the good position precision, and needn't require the inertial sensors with high precision. Therefore, the algorithm has the good practicability.

Kalman Filters versus Neural Networks in Battery State-of-Charge Estimation: A Comparative Study

Battery management systems (BMS) must estimate the state-of-charge (SOC) of the battery accurately to prolong its lifetime and ensure a reliable operation. Since batteries have a wide range of applications, the SOC estimation requirements and methods vary from an application to another. This paper compares two SOC estimation methods, namely extended Kalman filters (EKF) and artificial neural networks (ANN). EKF is a nonlinear optimal estimator that is used to estimate the inner state of a nonlinear dynamic system using a state-space model. On the other hand, ANN is a mathematical model that consists of interconnected artificial neurons inspired by biological neural networks and is used to predict the output of a dynamic system based on some historical data of that system. A pulse-discharge test was performed on a commercial lithium-ion (Li-ion) battery cell in order to collect data to evaluate those methods. Results are presented and compared.

Fuzzy Based Assignment Method of Filtering Nodes in Wireless Sensor Networks

Wireless sensor networks (WSNs) are networked systems that are able to sense various events and report the events to a user to enable appropriate responses. One of security threats to a WSN is false data injection attacks in which an attacker steals some sensor nodes in the network and injects forged event messages into the network through the captured nodes. As a result, the intermediate nodes on the forwarding paths of the false event messages waste their limited energy. Additionally, the network cannot provide the user with correct information. There have been many studies on en-route detection of false event messages for WSNs. Yang et al. proposed the commutative cipher-based en-route filtering scheme (CCEF) which establishes a secure session between a sink node and a cluster head (CH) based on the commutative cipher. In CCEF, each intermediate node on the path between the sink node and the CH receives an event message and verifies the authenticity of the session based on a probability. Due to the probabilistic approach, it is hard to adapt to the change of false traffic ratio in the network and energy inefficiency may occur. We propose a filtering scheme which applies a deterministic approach to assign filtering nodes to a given session. The proposed method consumes less energy than that of CCEF without sacrificing security.

Neural network-based H∞ filtering for nonlinear systems with time-delays

A novel H∞ design methodology for a neural network-based nonlinear filtering scheme is addressed. Firstly, neural networks are employed to approximate the nonlinearities. Next, the nonlinear dynamic system is represented by the mode-dependent linear difference inclusion （LDI）. Finally, based on the LDI model, a neural network-based nonlinear filter （NNBNF） is developed to minimize the upper bound of H∞ gain index of the estimation error under some linear matrix inequality （LMI） constraints. Compared with the existing nonlinear filters, NNBNF is time-invariant and numerically tractable. The validity and applicability of the proposed approach are successfully demonstrated in an illustrative example.