Vector Quantization Technique Research Articles

Event-triggered and quantised feedback stabilisation of switched networked control systems

In order to conserve network resources, this paper investigates the event-triggered control and quantised feedback stabilisation of switched networked control systems by introducing the vector quantizer technique. The state information undergoes quantisation and is transmitted to the controller after being sampled by the event-triggering technology. The designed event-triggered mechanism relies on quantised state values, and the generation of each quantised value also depends on the event-triggered condition. Addressing the asynchronous issue between the controller and the subsystem, we provide a sufficient condition for ensuring exponential stability of the closed-loop system and outline the controller gain design method by employing the multiple Lyapunov function and average dwell time methods. Besides, we demonstrate the explicit exclusion of the Zeno phenomenon in the event-triggered mechanism. Finally, the viability of our approach is demonstrated through simulation examples.

Comparison of support vector machines (SVMs) and the learning vector quantization (LVQ) techniques for geological domaining: a case study from Darehzar porphyry copper deposit, SE Iran

Comparison of support vector machines (SVMs) and the learning vector quantization (LVQ) techniques for geological domaining: a case study from Darehzar porphyry copper deposit, SE Iran

Fast Linde-Buzo-Gray (FLBG) Algorithm for Image Compression through Rescaling Using Bilinear Interpolation.

Vector quantization (VQ) is a block coding method that is famous for its high compression ratio and simple encoder and decoder implementation. Linde-Buzo-Gray (LBG) is a renowned technique for VQ that uses a clustering-based approach for finding the optimum codebook. Numerous algorithms, such as Particle Swarm Optimization (PSO), the Cuckoo search algorithm (CS), bat algorithm, and firefly algorithm (FA), are used for codebook design. These algorithms are primarily focused on improving the image quality in terms of the PSNR and SSIM but use exhaustive searching to find the optimum codebook, which causes the computational time to be very high. In our study, our algorithm enhances LBG by minimizing the computational complexity by reducing the total number of comparisons among the codebook and training vectors using a match function. The input image is taken as a training vector at the encoder side, which is initialized with the random selection of the vectors from the input image. Rescaling using bilinear interpolation through the nearest neighborhood method is performed to reduce the comparison of the codebook with the training vector. The compressed image is first downsized by the encoder, which is then upscaled at the decoder side during decompression. Based on the results, it is demonstrated that the proposed method reduces the computational complexity by 50.2% compared to LBG and above 97% compared to the other LBG-based algorithms. Moreover, a 20% reduction in the memory size is also obtained, with no significant loss in the image quality compared to the LBG algorithm.

Estimating Kolmogorov–Sinai entropy from time series of high-dimensional complex systems

A method is presented to estimate KS entropy from time series of a high-dimensional complex system. We focus on partitioned entropy, which measures the complexity of data points localized in a delay-coordinate space. By quantifying the time evolution of the partitioned entropy, we show that its growth rate gives a good estimate of the KS entropy. Our method is computationally efficient, since a time series is discretized into a coarse-grained symbolic sequence by a vector quantization technique. Using numerical data sets generated from prototypical models of chaotic systems, we showed that our approach works quite well for high-dimensional complex systems. Application to experimental data measured from physical models of the vocal membranes confirmed that our method is effective for characterizing the complexity of real-world systems.

Abstract and Explore: A Novel Behavioral Metric with Cyclic Dynamics in Reinforcement Learning

Intrinsic motivation lies at the heart of the exploration of reinforcement learning, which is primarily driven by the agent's inherent satisfaction rather than external feedback from the environment. However, in recent more challenging procedurally-generated environments with high stochasticity and uninformative extrinsic rewards, we identify two significant issues of applying intrinsic motivation. (1) State representation collapse: In existing methods, the learned representations within intrinsic motivation have a high probability to neglect the distinction among different states and be distracted by the task-irrelevant information brought by the stochasticity. (2) Insufficient interrelation among dynamics: Unsuccessful guidance provided by the uninformative extrinsic reward makes the dynamics learning in intrinsic motivation less effective. In light of the above observations, a novel Behavioral metric with Cyclic Dynamics (BCD) is proposed, which considers both cumulative and immediate effects and facilitates the abstraction and exploration of the agent. For the behavioral metric, the successor feature is utilized to reveal the expected future rewards and alleviate the heavy reliance of previous methods on extrinsic rewards. Moreover, the latent variable and vector quantization techniques are employed to enable an accurate measurement of the transition function in a discrete and interpretable manner. In addition, cyclic dynamics is established to capture the interrelations between state and action, thereby providing a thorough awareness of environmental dynamics. Extensive experiments conducted on procedurally-generated environments demonstrate the state-of-the-art performance of our proposed BCD.

A combined method of optimized learning vector quantization and neuro-fuzzy techniques for predicting unified Parkinson's disease rating scale using vocal features

Parkinson's Disease (PD) is a common disorder of the central nervous system. The Unified Parkinson's Disease Rating Scale or UPDRS is commonly used to track PD symptom progression because it displays the presence and severity of symptoms. To model the relationship between speech signal properties and UPDRS scores, this study develops a new method using Neuro-Fuzzy (ANFIS) and Optimized Learning Rate Learning Vector Quantization (OLVQ1). ANFIS is developed for different Membership Functions (MFs). The method is evaluated using Parkinson's telemonitoring dataset which includes a total of 5875 voice recordings from 42 individuals in the early stages of PD which comprises 28 men and 14 women. The dataset is comprised of 16 vocal features and Motor-UPDRS, and Total-UPDRS. The method is compared with other learning techniques. The results show that OLVQ1 combined with the ANFIS has provided the best results in predicting Motor-UPDRS and Total-UPDRS. The lowest Root Mean Square Error (RMSE) values (UPDRS (Total)=0.5732; UPDRS (Motor)=0.5645) and highest R-squared values (UPDRS (Total)=0.9876; UPDRS (Motor)=0.9911) are obtained by this method. The results are discussed and directions for future studies are presented.i.ANFIS and OLVQ1 are combined to predict UPDRS.ii.OLVQ1 is used for PD data segmentation.iii.ANFIS is developed for different MFs to predict Motor-UPDRS and Total-UPDRS.

Compression of an ECG Signal Using Mixed Transforms

Electrocardiogram (ECG) is an important physiological signal for cardiac disease diagnosis. With the increasing use of modern electrocardiogram monitoring devices that generate vast amount of data requiring huge storage capacity. In order to decrease storage costs or make ECG signals suitable and ready for transmission through common communication channels, the ECG datavolume must be reduced. So an effective data compression method is required. This paper presents an efficient technique for the compression of ECG signals. In this technique, different transforms have been used to compress the ECG signals. At first, a 1-D ECG data was segmented and aligned to a 2-D data array, then 2-D mixed transform was implemented to compress the ECG data in the 2-D form. The compression algorithms were implemented and tested using multiwavelet, wavelet and slantlet transforms to form the proposed method based on mixed transforms. Then vector quantization technique was employed to extract the mixed transform coefficients. Some selected records from MIT/BIH arrhythmia database were tested contrastively and the performance of theproposed methods was analyzed and evaluated using MATLAB package. Simulation results showed that the proposed methods gave a high compression ratio (CR) for the ECG signals comparing with other available methods. For example, the compression of one record (record 100) yielded CR of 24.4 associated with percent root mean square difference (PRD) of 2.56% was achieved.

A Physical Layer Key Generation Scheme Based on Deep Learning Compensation and Balanced Vector Quantization

Channel reciprocity is the foundation for physical layer key generation, which is influenced by noise, hardware impairments, and synchronization offsets. Weak channel reciprocity will result in a high key disagreement rate (KDR). The existing solutions for improving channel reciprocity cannot achieve satisfactory performance improvements. Furthermore, the existing quantization algorithms generally use one-dimensional channel features to quantize and generate secret keys, which cannot fully utilize channel information. The multidimensional vector quantization technique also needs to improve in terms of randomness and time complexity. This paper proposes a physical layer key generation scheme based on deep learning and balanced vector quantization. Specifically, we build a channel reciprocity compensation network (CRCNet) to learn the mapping relationship between Alice and Bob’s channel measurements. Alice compensates for channel measurements via a trained CRCNet to reduce channel measurement errors between legitimate users and enhance channel reciprocity. We also propose a balanced vector quantization algorithm based on integer linear programming (ILP-BVQ). ILP-BVQ reduces the time complexity of quantization on the basis of ensuring key randomness and a low KDR. Simulation results showed that the proposed CRCNet performs better in terms of channel reciprocity and KDR, while the proposed ILP-BVQ algorithm improves time consumption and key randomness.

Electroencephalography (EEG) eye state classification using learning vector quantization and bagged trees

The analysis of Electroencephalography (EEG) signals has been an effective way of eye state identification. Its significance is highlighted by studies that examined the classification of eye states using machine learning techniques. In previous studies, supervised learning techniques have been widely used in EEG signals analysis for eye state classification. Their main goal has been the improvement of classification accuracy through the use of novel algorithms. The trade-off between classification accuracy and computation complexity is an important task in EEG signals analysis. In this paper, a hybrid method that can handle multivariate signals and non-linear is proposed with supervised and un-supervised learning to achieve a fast EEG eye state classification with high prediction accuracy to provide real-time decision-making applicability. We use the Learning Vector Quantization (LVQ) technique and bagged tree techniques. The method was evaluated on a real-world EEG dataset which included 14976 instances after the removal of outlier instances. Using LVQ, 8 clusters were generated from the data. The bagged tree was applied on 8 clusters and compared with other classifiers. Our experiments revealed that LVQ combined with the bagged tree provides the best results (Accuracy = 0.9431) compared with the bagged tree, CART (Classification And Regression Tree) (Accuracy = 0.8200), LDA (Linear Discriminant Analysis) (Accuracy = 0.7931), Random Trees (Accuracy = 0.8311), Naïve Bayes (Accuracy = 0.8331) and Multilayer Perceptron (Accuracy = 0.7718), which demonstrates the effectiveness of incorporating ensemble learning and clustering approaches in the analysis of EEG signals. We also provided the time complexity of the methods for prediction speed (Observation/Second). The result showed that LVQ + Bagged Tree provides the best result for prediction speed (58942 Obs/Sec) in relation to Bagged Tree (28453 Obs/Sec), CART (27784 Obs/Sec), LDA (26435 Obs/Sec), Random Trees (27921), Naïve Bayes (27217) and Multilayer Perceptron (24163).

Cryptanalysis of Reversible Data Hiding in Encrypted Images Based on the VQ Attack

Reversible data hiding in encrypted images (RDHEI) is commonly used for privacy protection in images stored on cloud storage. Currently, block permutation and co-modulation (BPCM) encryption is commonly utilized in most existing RDHEI schemes to generate encrypted images. In this paper, we analyze the vulnerabilities of RDHEI based on BPCM encryption and then propose a cryptanalysis method based on the vector quantization (VQ) attack. Unlike the existing cryptanalysis method, our method does not require the help of a plaintext image instead of adopting the symmetric property between the original cover image and the encrypted cover image. To obtain the pixel-changing pattern of a block before and after co-modulation, the concept of a pixel difference block (PDB) is first defined. Then, the VQ technique is used to estimate the content of the ciphertext block. Finally, we propose a sequence recovery method to help obtain the final recovered image based on the premise that the generator is compromised. The experimental results demonstrate that when the block size is 4 × 4, our proposed cryptanalysis method can decrypt the contents of the ciphertext image well. The average similarity can exceed 75% when comparing the edge information of the estimated image and the original image. It is concluded from our study that the BPCM encryption algorithm is not robust enough.

DNA Computing with Water Strider Based Vector Quantization for Data Storage Systems

The exponential growth of data necessitates an effective data storage scheme, which helps to effectively manage the large quantity of data. To accomplish this, Deoxyribonucleic Acid (DNA) digital data storage process can be employed, which encodes and decodes binary data to and from synthesized strands of DNA. Vector quantization (VQ) is a commonly employed scheme for image compression and the optimal codebook generation is an effective process to reach maximum compression efficiency. This article introduces a new DNA Computing with Water Strider Algorithm based Vector Quantization (DNAC-WSAVQ) technique for Data Storage Systems. The proposed DNAC-WSAVQ technique enables encoding data using DNA computing and then compresses it for effective data storage. Besides, the DNAC-WSAVQ model initially performs DNA encoding on the input images to generate a binary encoded form. In addition, a Water Strider algorithm with Linde-Buzo-Gray (WSA-LBG) model is applied for the compression process and thereby storage area can be considerably minimized. In order to generate optimal codebook for LBG, the WSA is applied to it. The performance validation of the DNAC-WSAVQ model is carried out and the results are inspected under several measures. The comparative study highlighted the improved outcomes of the DNAC-WSAVQ model over the existing methods.

Accurate Reader Identification for the Arabic Holy Quran Recitations Based on an Enhanced VQ Algorithm

The Speaker identification process is not a new trend; however, for the Arabic Holy Quran recitation, there are still quite improvements that can make this process more accurate and reliable. This paper collected the input data from 14 native Arabic reciters, consisting of “Surah Al-Kawthar” speech signals from the Holy Quran. Moreover, this paper discusses the accuracy rates for 8 and 16 features. Indeed, a modified Vector Quantization (VQ) technique will be presented, in addition to realistically matching the centroids of the various codebooks and measuring systems’ effectiveness. Note that the VQ technique will be utilized to generate the codebooks by clustering these features into a finite number of centroids. The proposed system’s software was built and executed using MATLAB®. The proposed system’s total accuracy rate was 97.92% and 98.51% for 8 and 16 centroids codebooks, respectively. However, this study discussed two validation tactics to ensure that the outcomes are reliable and can be reproduced. Hence, the K-mean clustering algorithm has been used to validate the obtained results and discuss the outcomes of this study. Finally, it has been found that the improved VQ method gives a better result than the K-means method.

Voice Cloning in Real Time

Abstract: Deep learning models for natural voice cloning methods were first developed in 2016, and since then, the researchers' main attention has been on making the voice more realistic and obtaining the output voice in real time. Previously it used to take many hours of voice samples to clone a few seconds. It was decreased to a few seconds after utilizing deep learning models. We shall look at various voice cloning techniques in this paper. Multi-speaker generative models, speaker adoption, speaker encoding, vector quantization, and other techniques are among them

Context-Tree-Based Lossy Compression and Its Application to CSI Representation

We propose novel compression algorithms for time-varying channel state information (CSI) in wireless communications. The proposed scheme combines (lossy) vector quantisation and (lossless) compression. First, the new vector quantisation technique is based on a class of parametrised companders applied on each component of the normalised CSI vector. Our algorithm chooses a suitable compander in an intuitively simple way whenever empirical data are available. Then, the sequences of quantisation indices are compressed using a context-tree-based approach. Essentially, we update the estimate of the conditional distribution of the source at each instant and encode the current symbol with the estimated distribution. The algorithms have low complexity, are linear-time in both the spatial dimension and time duration, and can be implemented in an online fashion. We run simulations to demonstrate the effectiveness of the proposed algorithms in such scenarios.

Hybrid artificial neural network algorithm for air pollution estimation

In recent years, airborne broadcasting has grown more prevalent in cities. Air quality degradation is a severe air pollution issue that exists daily. To forecast the amount of pollutants, Artificial Neural Network (ANN) and Linear Vector Quantization (LVQ) techniques were utilized. The data set dimensions are defined by the pre-processing procedure and the feature extraction mechanism. The ANN model predicts categorization concentration, allowing the LVQ model to classify direct situations with greater accuracy using explanatory factors. The ANN+LVQ model outperformed other technologies in terms of classification accuracy. The raw data was cleaned to improve the accuracy of the prediction algorithms. The pollutants discovered in the collection are NO2, NOx, O3, Benzene, Xylene, NH3, CO, SO2, PM10, NO, and Toluene. The performance of the recommendation and forecast models were tested in this study using two datasets in two distinct experiments. In urban, rural, and industrial settings, the proposed ANN model is successful in detecting air quality and predicting pollution levels. The ANN-LVQ model obtained 90% percent sensitivity, 97.59% accuracy, and 99.46% specificity with 2.43% error rate. The suggested model's accuracy is much greater than that of other current research models.

Histogram Shifting-Based Quick Response Steganography Method for Secure Communication

Steganography is a tool which allows the data for transmission by concealing secret information in a tremendously growing network. In this paper, a novel technique quick response method (QRM) is proposed for the purpose of encryption and decryption. Existing system uses side match vector quantization (SMVQ) technique which has some challenges such as security issues and performance issues. To handle the security and performance issues, the proposed system uses two methods, namely, quick response method and shifting method. In the proposed system, encoding part calculates the performance for capacity, PSNR (peak signal-to-noise ratio), MSE (mean square error), and SSIM (structural similarity index method), and the decoding part calculates the performance of MSE (mean square error) and PSNR (peak signal-to-noise ratio). The shifting method is used to increase the data hiding capacity. In this system, the encryption part embeds the secret image using steganography and the decryption part extracts the original image. By analyzing and comparing the proposed system with the existing system, it is proved that the system proposed was much better than the existing systems.

NSVQ: Noise Substitution in Vector Quantization for Machine Learning

Machine learning algorithms have been shown to be highly effective in solving optimization problems in a wide range of applications. Such algorithms typically use gradient descent with backpropagation and the chain rule. Hence, the backpropagation fails if intermediate gradients are zero for some functions in the computational graph, because it causes the gradients to collapse when multiplying with zero. Vector quantization is one of those challenging functions for machine learning algorithms, since it is a piece-wise constant function and its gradient is zero almost everywhere. A typical solution is to apply the straight through estimator which simply copies the gradients over the vector quantization function in the backpropagation. Other solutions are based on smooth or stochastic approximation. This study proposes a vector quantization technique called NSVQ, which approximates the vector quantization behavior by substituting a multiplicative noise so that it can be used for machine learning problems. Specifically, the vector quantization error is replaced by product of the original error and a normalized noise vector, the samples of which are drawn from a zero-mean, unit-variance normal distribution. We test our proposed NSVQ in three scenarios with various types of applications. Based on the experiments, the proposed NSVQ achieves more accuracy and faster convergence in comparison to the straight through estimator, exponential moving averages, and the MiniBatchKmeans approaches.

Optimization Based Vector Quantization for Data Reduction in Multimedia Applications

Data reduction and image compression techniques in the present Internet and multi-media age are essential to increase image and video capacity in relation to memory, network bandwidth use and safe data transmission. There have been a different variety of image compression models with varying compression efficiency and visual image quality in the literature. Vector Quantization (VQ) is a widely used image coding scheme that is designed to generate an efficient coding book that includes a list of codewords that assign the input image vector to a minimum distance of Euclidea. The Linde–Buzo–Gray (LBG) historically widely used model produces the local optimal codebook. The LBG model’s codebook architecture is seen as an optimization challenge that can be resolved using metaheuristic algorithms. In this perspective, this paper introduces a new DPIO algorithm for codebook generation in VQ. The model presented uses LBG to initialise the DPIO algorithm to construct the VQ technique and is called the DPIO-LBG process. The performance of the DPIO-LBG model is validated with benchmark data and the effects of different performance aspects are investigated. The simulation values showed the DPIO-LBG model to be efficient interfaces of compression efficiency and image quality reconstructed. The presented model produces an efficient codebook with minimum computational time and a better signal to noise ratio (PSNR).

HHO-Based Vector Quantization Technique for Biomedical Image Compression in Cloud Computing

In the present digital era, the exploitation of medical technologies and massive generation of medical data using different imaging modalities, adequate storage, management, and transmission of biomedical images necessitate image compression techniques. Vector quantization (VQ) is an effective image compression approach, and the widely employed VQ technique is Linde–Buzo–Gray (LBG), which generates local optimum codebooks for image compression. The codebook construction is treated as an optimization issue solved with utilization of metaheuristic optimization techniques. In this view, this paper designs an effective biomedical image compression technique in the cloud computing (CC) environment using Harris Hawks Optimization (HHO)-based LBG techniques. The HHO-LBG algorithm achieves a smooth transition among exploration as well as exploitation. To investigate the better performance of the HHO-LBG technique, an extensive set of simulations was carried out on benchmark biomedical images. The proposed HHO-LBG technique has accomplished promising results in terms of compression performance and reconstructed image quality.

Solar activity classification based on Mg II spectra: Towards classification on compressed data

Although large volumes of solar data are available for study, the vast majority of these data remain unlabeled and are therefore not amenable to supervised machine learning methods. Having a way to accurately and automatically classify spectra into categories related to solar activity is highly desirable and will assist and speed up future research efforts in solar physics. At the same time, the large volume of raw observational data is a serious bottleneck for machine learning, requiring powerful computational means that are not at the disposal of many laboratories. Besides, the raw data communication imposes restrictions on real time data observations and requires considerable bandwidth and energy for the onboard solar observation systems. To solve these issues, we propose a framework to classify solar activity on compressed data. For this, we used a labeling scheme from a pre-existing vector quantization technique in conjunction with different machine learning algorithms to categorize spectra of singly-ionized magnesium Mg II measured by NASA's Interface Region Imaging Spectrograph satellite (IRIS) into five types of solar activity. Our training dataset is a human annotated list of 85 IRIS observations containing 29097 frames. The annotated types of Solar activities are active region, pre-flare activity, Solar flare, Sunspot, and quiet Sun. We compress these data and reduce its complexity before training classifiers. We found that the XGBoost classifier produces the most accurate results on the compressed data, yielding over a 95\% prediction rate, and outperforming other ML methods like convolution neural networks, K-nearest neighbors, naive Bayes classifiers, and SVM. We find that the classification performance on compressed and uncompressed data is comparable, implying the possibility of large compression rates for relatively low degrees of information loss.