Localized Filter Bank Research Articles

An efficient detection of congestive heart failure using frequency localized filter banks for the diagnosis with ECG signals

Congestive heart failure (CHF) refers to the condition in which the heart is unable to maintain the required blood flow under normal heart pressure. CHF is one of the major causes of death worldwide and is commonly caused by coronary artery disease, diabetes and high blood pressure. It typically affects the elderly population. The diagnosis of CHF is mostly based on clinical assessment of symptoms, signs, imaging findings, and invasive intracardiac pressure measurement. The electrocardiogram (ECG) is neither sensitive nor specific for diagnosis of CHF, and the analysis of the ECG signal for the possible presence of CHF is manually intensive and requires adequate skills and expertise for discerning subtle abnormalities in the electrical activity of the heart that may be associated with CHF. We hypothesized that this task of recognizing the multiparametric patterns of ECG signal aberrations that might occur in CHF could be expedited and optimized using machine learning. In this paper, we present an automated approach for the diagnosis of CHF using ECG signals. The proposed approach was tested on four different sets of normal and CHF ECG signals obtained from established public databases. The experiments were performed using short (2 second (s)) ECG segments. Five different features (fuzzy entropy, Renyi entropy, Higuchi’s fractal dimension, Kraskov entropy and energy) were extracted from the wavelet decomposition of ECG segments using frequency localized filter banks. For training and classification, we employed quadratic support vector machine (QSVM). A 10-fold cross-validation technique was used for evaluation. Accuracy ⩾99.66%, sensitivity ⩾99.82%, and specificity ⩾99.28%, were obtained across all four data sets used. The system can be deployed in hospitals to facilitate the diagnosis of CHF. The proposed system can reduce the time requirement and error rate associated with manual reading of large ECG signals.

A Hybrid Kalman Filtering Approach Based on Federated Framework for Gas Turbine Engine Health Monitoring

The Kalman filter (KF) is the most common state estimation method for gas turbine health monitoring, and it runs in the centralized architecture. However, health estimation cannot be achieved by the KF-based method as sensor fault occurs, and malfunction of the central monitoring unit will unavoidably result to the termination of the diagnosis task. For these purposes, this paper develops a novel hybrid federated KF approach from the previous achievements. The hybrid KF consists of a bank of local filters and one master filter, and the federated filtering structure and asynchronous fusion mechanism are designed. Both the linearized KF and extended KF are employed as the local filters based on the linear correlation of thermodynamic parameters. The local state estimates and covariance are yielded in parallel, and then integrated in a master filter to produce global state estimate. The proposed methodology is evaluated and compared with the general federated KFs in terms of estimation accuracy, computational efforts, and robustness to sensor fault in the application of gas turbine health monitoring. The result shows that the hybrid KF is the best balance off the involved performance, and confirms our viewpoints in this paper.

Time–frequency localized three-band biorthogonal wavelet filter bank using semidefinite relaxation and nonlinear least squares with epileptic seizure EEG signal classification

In this paper, we design time–frequency localized three-band biorthogonal linear phase wavelet filter bank for epileptic seizure electroencephalograph (EEG) signal classification. Time–frequency localized analysis and synthesis low-pass filters (LPF) are designed using convex semidefinite programming (SDP) by transforming a nonconvex problem into a convex SDP using semidefinite relaxation technique. Three-band parameterized lattice biorthogonal linear phase perfect reconstruction filter bank (BOLPPRFB) is chosen and nonlinear least squares algorithm is used to determine its parameters values that generate the designed analysis and synthesis LPF such that the band-pass and high-pass filters are also well localized in time and frequency domain. The designed analysis and synthesis three-band wavelet filter banks are compared with the standard two-band filter banks like Daubechies maximally regular filter banks, Cohen–Daubechies–Feauveau (CDF) biorthogonal filter banks and orthogonal time–frequency localized filter banks. Kruskal–Wallis statistical test is employed to measure the statistical significance of the subband features obtained from the various two and three-band filter banks for epileptic seizure EEG signal classification. The results show that the designed three-band analysis and synthesis filter banks both outperform two-band filter banks in the classification of seizure and seizure-free EEG signals. The designed three-band filter banks and multi-layer perceptron neural network (MLPNN) are further used together to implement a signal classifier that provides classification accuracy better than the recently reported results for epileptic seizure EEG signal classification.

Design of Time–Frequency Localized Filter Banks: Transforming Non-convex Problem into Convex Via Semidefinite Relaxation Technique

We present a method for designing optimal biorthogonal wavelet filter banks (FBs). Joint time---frequency localization of the filters has been chosen as the optimality criterion. The design of filter banks has been cast as a constrained optimization problem. We design the filter either with the objective of minimizing its frequency spread (variance) subject to the constraint of prescribed time spread or with the objective of minimizing the time spread subject to the fixed frequency spread. The optimization problems considered are inherently non-convex quadratic constrained optimization problems. The non-convex optimization problems have been transformed into convex semidefinite programs (SDPs) employing the semidefinite relaxation technique. The regularity constraints have also been incorporated along with perfect reconstruction constraints in the optimization problem. In certain cases, the relaxed SDPs are found to be tight. The zero duality gap leads to the global optimal solutions. The design examples demonstrate that reasonably smooth wavelets can be designed from the proposed filter banks. The optimal filter banks have been compared with popular filter banks such as Cohen---Daubechies---Feauveau biorthogonal wavelet FBs, time---frequency optimized half-band pair FBs and maximally flat half-band pair FBs. The performance of optimal filter banks has been found better in terms of joint time---frequency localization.

Subband image coding using jointly localized filter banks and entropy coding based on vector quantization

A method for image compression based on subband decomposition is presented. We describe a new filter bank design method for image coding applications and a new entropy coding algorithm for the compression of subband images. A set of relevant optimization criteria is defined for the filter bank design. For the compression, a composite source model is defined by combining vector quantization (VQ) and scalar quantization (SQ) with entropy coding. In the proposed scheme, VQ exploits the remaining statistical dependencies among the subband samples, while SQ allows an optimal control on local distortions. The system is based on a statistical model that uses VQ information to generate low entropy probability tables for an arithmetic coder. The bit rate can be shared between the VQ rate and the SQ rate, allowing many possible configurations in terms of performance and implementation complexity. The proposed system shows improved performance when compared with other existing methods.