Indoor device-free passive localization with DCNN for location-based services

As an emerging technique with a promising application prospect, the device-free localization (DFL) technique has drawn considerable attention due to its ability of realizing wireless localization without the need of equipping the target with any device. The DFL technique detects the shadowed links and realizes localization with the received signal strength (RSS) measurements of these links. However, one major disadvantage of the DFL technique is that the RSS signal is too sensitive, and a slight variation of the environment will cause the variation of RSS measurements, which incurs the misjudgment of shadowed links and degradation of localization performance. To solve this problem, a robust DFL scheme based on differential RSS is proposed. The scheme utilizes the novel differential RSS to judge whether a link is shadowed, which not only eliminates the need of acquiring reference RSS measurements but also overcomes the negative effect incurred by the environment. Meanwhile, an outlier detection scheme is presented to filter out the outlier links that are far away from the target. We present the observation model of the shadowed links and incorporate it into the particle filter framework to realize location estimation robustly. Experimental results demonstrate the outstanding performance of the proposed scheme.

In this study, we present the performance of Support Vector Machine (SVM), Convolution Neural Network (CNN), and Artificial Neural Network (ANN) with Bag of Words (BoW), Histogram of Oriented Gradients (HOG), and Image Pixels (IP) for face recognition. SVM, CNN, and ANN are machine learning approaches and has been used for pattern recognition, especially in face recognition technology. BoW, HOG, and IP are being used for image feature extraction. The testing has been conducted from publicly available AT&T face database. Every individual subject consists of 10 images with different facial expression, different illumination and the dimensions of the images is unified as 92-by-112 pixels with PGM formats. SVM achieved recognition accuracy of 97.00%, 96.00%, and 98.00% with subsequently BoW, HOG, and IP. CNN achieved recognition accuracy of 94.00%, 99.00%, and 99.50% with subsequently BoW, HOG, and IP. ANN achieved recognition accuracy of 96.00%, 99.00%, and 99.50% with subsequently BoW, HOG, and IP. The experimental results show that the IP with ANN approach clearly outperformed the others approaches.

Low frequency sound %3C zo Hz, known as infrasound, is generated by a variety of natural and anthropogenic sources. Following an event, infrasonic waves travel through a dynamic atmosphere that can change on the order of minutes. This makes infrasound event classification a difficult problem as waveforms from the same source type can look drastically different. Event classification usually requires ground truth information from seismic or other methods. This is time consuming, inefficient, and does not allow for a classification if the event locates somewhere other than a known source, the location accuracy is poor, or ground truth from seismic data is lacking. Here we compare the performance of the state of the art for infrasound event classification, support vector machine (SVM), to the performance of a convolutional neural network (CNN), a method that has been proven in tangential fields such as seismology. For a 1-class catalog consisting of only volcanic activity and earthquake events, the 4-fold average SVM classification accuracy is 75%, while it is 74% when using a CNN. Classification accuracies from the 4-class catalog consisting of the most common infrasound events detected at the global scale are 55% and 56% for the SVM and CNN architectures, respectively. These results demonstrate that using a CNN does not increase performance for infrasound event classification. This suggests that SVM should be the preferred classification method as it is a simpler and more trustworthy architecture and can be tied to the physical properties of the waveforms. The SVM and CNN algorithms described in this paper are not yet generalizable to other infrasound event catalogs. We anticipate this study to be a starting point for the development of large and comprehensive, systematically labeled, infrasound event catalogs as such catalogs will be necessary to provide an increase in the value of deep learning on event classification.

With the rapid growth of the demand for indoor location-based services (LBS), Wi-Fi received signal strength (RSS) fingerprints database has attracted significant attention because it is easy to obtain. The fingerprints algorithm based on convolution neural network (CNN) is often used to improve indoor localization accuracy. However, the number of reference points used for position estimation has significant effects on the positioning accuracy. Meanwhile, it is always selected arbitraily without any guiding standards. As a result, a novel location estimation method based on Jenks natural breaks algorithm (JNBA), which can adaptively choose more reasonable reference points, is proposed in this paper. The output of CNN is processed by JNBA, which can select the number of reference points according to different environments. Then, the location is estimated by weighted K-nearest neighbors (WKNN). Experimental results show that the proposed method has higher positioning accuracy without sacrificing more time cost than the existing indoor localization methods based on CNN.

In this paper, we propose a surface correlation based fingerprinting method in sparse Wi-Fi environment. The proposed method can estimate location even with very few Wi-Fi APs by using accumulated spatial RSS pattern. We verify performance of the proposed method through several tests in a building with complex paths. Many studies have developed localization technologies that use measurements such as RSS (Received Signal Strength) or RTT (Round Trip Time) for Wi-Fi signal. All RF have signal noise and distortion problems as well as multipath problem in indoor environments. As the distance from the RF signal source increases, the discrimination of the RSS decreases due to RF signal noise and distortion. Therefore, ambiguity of RSS increases as the distance increases, the performance deteriorates. The conventional RF based localization method have tried to maintain dense RF environment to guarantee stable and accurate performance. However, in order to maintain the dense RF environment, it is becoming a barrier to commercialization of localization technology due to high cost of constructing RF infrastructure and maintenance difficulty. In this paper, we propose a Surface Correlation (SC) based fingerprinting technique to maintain high positioning performance even with few Wi-Fi APs. The surface is the accumulated RSS pattern data during the movement of a pedestrian, which means the spatial RSS pattern. Conventional fingerprinting methods using kNN (k-Nearest Neighbors) and WkNN (Weighted kNN) are very vulnerable to signal noise and distortion because they utilize instantaneous RSS data. The performance of Particle Filter (PF) based localization technology is also degraded in areas where Wi-Fi AP is sparsely distributed. However, the proposed method is very robust against signal noise and distortion because it compares the spatial RSS pattern with the reference RSS pattern on the fingerprinting database. This increases localization accuracy and availability for Wi-Fi coverage. For localization, the proposed method conduct correlation process for the spatial RSS pattern of a pedestrian with reference RSS pattern of database. So, the reference RSS pattern should be extracted from a reference trajectory that matches a pedestrian`s walking path. The proposed method utilize a database of ‘link-node’ structure to dynamically generate a reference trajectory in indoor environment with complicated path. It generate candidate reference trajectories using only MAC address and turn event of Pedestrian Dead-Reckoning (PDR). The proposed method fix a reference trajectory between these candidate trajectories through correlation process of measured RSS pattern with each reference RSS pattern. A reference point on the fixed reference having most similar RSS pattern to a pedestrian`s is considered as a current location. In order to improve performance at a fork where RSS ambiguity increases, the proposed method estimates preferentially for a reference trajectory matching with PDR turn event. The turn event can be wrong due to various motion of a pedestrian. The proposed method selects a correct reference trajectory by continuously performing correlation process on candidate reference trajectories. In order to verify the performance of the proposed surface correlation based fingerprinting method, we conducted several tests in a three-story building. Before the tests, we conducted a survey of Wi-Fi environment to construct fingerprinting database in the building. As a result of the survey, 170 APs are distributed in the building. We reduced the number of APs by 20% from 170 to 17, and analyzed performance of the proposed and conventional localization methods. The test results show that the performance of the conventional localization methods degrade as the number of applied APs decreases. The proposed fingerprinting method shows a 3 meter localization error even with very few Wi-Fi APs, and unlike the conventional fingerprinting method, the localization error is bounded to be within 5m due to the use of spatial RSS pattern. In this paper, we presented the fingerprinting method using the surface correlation in sparse Wi-Fi environment. The proposed fingerprinting method accumulates Wi-Fi RSS during movement of a pedestrian. The proposed method compares the spatial RSS pattern, which is accumulated RSS data, with reference RSS patterns of fingerprinting database. By comparing the RSS patterns, the proposed method greatly improves accuracy as well as stability. In addition, it can be effectively applied in a sparse Wi-Fi environment by improving the availability of ambiguous areas of Wi-Fi coverage. Through these results, we expect to be able to provide accurate and reliable location information even in indoor weak RF environment.

Indoor localization of mobile nodes is emerging as a promising field in the past decades due to the advances in mobile devices and the increasing number of wireless services and applications. Wi-Fi fingerprints which is a vector of Received Signal Strength (RSS) values in a specific location are considered a good feature for indoor localization due to their simplicity. However, the existing methods such as decision tree (DT) and Naïve Bayes (NB) techniques have some drawbacks. DT has low efficiency because it uses the Boolean function, and NB provides a low accuracy of prediction for the sparse dataset. In this research, fingerprint referencing with two important methods, Support Vector Machine (SVM) and Long-Short Term Memory (LSTM) machine learning algorithms, are used to enable accurate indoor localization. Furthermore, data normalization is implemented to reduce positioning error by increasing the inconsistency of the RSS values. Further, several experiments have been conducted by using three mobile devices in the area of 280 m <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> to create our dataset by taking 480 RSS records for both training and testing phases. The results show that the accuracy LSTM algorithm is 0.9 m, whereas the SVM algorithm accuracy is 1.1 m. Both experimental results show better accuracy for indoor environments can be achieved by the proposed approach than results recorded by several prior studies.

Device-free passive (DFP) localization systems are a key solution for location-based services because they do not require any wireless device on a human body. Most of the existing DFP localization systems are based on the received signal strength (RSS) measurement only. However, the localization accuracy of RSS only-based systems is easily affected by the spatial and temporal variances of RSS due to multipath fading and noise, even in a static environment. In this paper, we propose a novel localization system for DFP using signal subspace eigenvectors from an antenna array. We present a fingerprinting technique using multiclass support vector machines (SVMs) based on a combination of array signal features with spatial and temporal averaging. We then evaluate the localization accuracy of our proposed system in different propagation environments: line-of-sight (LOS) and non-line-of-sight (NLOS). In addition, we analyze two types of receive antenna placement: centralized and distributed antennas. The experimental results show that the localization accuracy can be improved by the proposed system, particularly in the centralized antenna case. Moreover, they show that the proposed system can improve localization accuracy compared to the conventional RSS-only based system.

Wi-Fi-based indoor positioning systems are becoming increasingly prevalent in digital transitions; therefore, ensuring accurate and robust positioning is essential to supporting the growth in demand for smartphones’ location-based services. The indoor positioning system on a smartphone, which is generally based on Wi-Fi received signal strength (RSS) measurements or the fingerprinting comparison technique, uses the K-NN algorithm to estimate the position due to its high accuracy. The fingerprinting algorithm is popular due to its ease of implementation and its ability to produce the desired accuracy. However, in a practical environment, the Wi-Fi signal strength-based positioning system is highly influenced by external factors such as changes in the environment, human interventions, obstacles in the signal path, signal inconsistency, signal loss due to the barriers, the non-line of sight (NLOS) during signal propagation, and the high level of fluctuations in the RSS, which affects location accuracy. In this paper, we propose a method that combines pedestrian dead reckoning (PDR) and Wi-Fi fingerprinting to select a k-node to participate in the K-NN algorithm for fingerprinting-based IPSs. The selected K-node is used for the K-NN algorithm to improve the robustness and overall accuracy. The proposed hybrid method can overcome practical environmental issues and reduces the KNN algorithm’s complexity by selecting the nearest neighbors’ search space for comparison using the PDR position estimate as the reference position. Our approach provides a sustainable solution for indoor positioning systems, reducing energy consumption and improving the overall environmental impact. The proposed method has potential applications in various domains, such as smart buildings, healthcare, and retail. The proposed method outperforms the traditional KNN algorithm in our experimental condition since its average position error is less than 1.2 m, and provides better accuracy.

In the field of medicine, with the introduction of computer systems that can collect and analyze massive amounts of data, many non-invasive diagnostic methods are being developed for a variety of conditions. In this study, our aim is to develop a non-invasive method of classifying respiratory sounds that are recorded by an electronic stethoscope and the audio recording software that uses various machine learning algorithms.In order to store respiratory sounds on a computer, we developed a cost-effective and easy-to-use electronic stethoscope that can be used with any device. Using this device, we recorded 17,930 lung sounds from 1630 subjects.We employed two types of machine learning algorithms; mel frequency cepstral coefficient (MFCC) features in a support vector machine (SVM) and spectrogram images in the convolutional neural network (CNN). Since using MFCC features with a SVM algorithm is a generally accepted classification method for audio, we utilized its results to benchmark the CNN algorithm. We prepared four data sets for each CNN and SVM algorithm to classify respiratory audio: (1) healthy versus pathological classification; (2) rale, rhonchus, and normal sound classification; (3) singular respiratory sound type classification; and (4) audio type classification with all sound types. Accuracy results of the experiments were; (1) CNN 86%, SVM 86%, (2) CNN 76%, SVM 75%, (3) CNN 80%, SVM 80%, and (4) CNN 62%, SVM 62%, respectively.As a result, we found out that spectrogram image classification with CNN algorithm works as well as the SVM algorithm, and given the large amount of data, CNN and SVM machine learning algorithms can accurately classify and pre-diagnose respiratory audio.

This study aims to evaluate the effectiveness of the Convolutional Neural Network ( CNN ) method combined with Support Vector Machine ( SVM ) in detecting URLs.Phishing.Phishing is one of the significant cyber threats, where attackers try to trick users into providing sensitive information through fake websites.With the increasing number of phishing attacks , there is a need for effective methods to detect and prevent this threat.In this study, a dataset containing URLs phishing and non-phishing data were used to train the CNN -SVM model .The training process involved feature extraction from URLs using CNN , which is capable of capturing complex patterns in the data, followed by classification using SVM , which is known for its ability to handle high-dimensional data.Testing was conducted across nine different scenarios to evaluate the performance of the model under various conditions.The test results showed that the hybrid CNN -SVM model achieved a precision of 95%, a recall of 92%, and an F1-Score of 93%, with an overall accuracy of 94%.These results indicate that the model is not only effective in detecting URLs phishing , but also has a good balance between precision and recall.This study indicates that the combination of CNN and SVM can be an effective solution for detecting URLs phishing , making a significant contribution to the development of better cyber security systems.

Convolutional neural networks (CNNs) are being used in this paper to investigate waste classification. The study compared the categorization of three waste categories (plastic, paper, and metal) using a CNN and a Support Vector Machine (SVM). The results showed that while the SVM had an accuracy of 94.8%, the CNN's accuracy was lower at 83%. Another study divided waste into six categories and showed that the classification accuracy was 22% with a CNN, while the SVM had a test accuracy of 63%. The authors noted challenges in optimizing the hyperparameters, which prevented the CNN from being trained to its full potential. However, they concluded that the CNN should attain higher accuracy levels than the SVM once these challenges are overcome. The authors also used two popular CNN architectures, VGG16 and FastNet-34, in their proposed model for waste classification. The dataset used for the model was provided by Kaggle and consisted of 22564 waste photos divided into recyclable (R) and organic (O) categories. The results of the proposed model showed improved accuracy in waste classification compared to previous studies. The study highlights the potential for the use of CNNs in the field of waste management but highlights the importance of overcoming challenges in hyperparameter optimization for higher computer vision accuracy.

A genetic algorithm (GA), is used to optimize the many parameters of a convolutional neural network (CNN) that control the structure of the network. CNNs are used in image classification problems where it is necessary to generate feature descriptors to discern between image classes. Because of the deep representation of image data that CNNs are capable of generating, they are increasingly popular in research and industry applications. With the increasing number of use cases for CNNs, more and more time is being spent to come up with optimal CNN structures for different applications. Where one CNN might succeed at classification, another can fail. As a result, it is desirable to more easily find an optimal CNN structure to increase classification accuracy. In the proposed method, a GA is used to evolve the parameters that influence the structure of a CNN. The GA compares CNNs by training them on images of concrete containing cracks. The best CNN after several generations of the GA is then compared to the state-of-the-art CNN for crack detection. This work shows that it is possible to generalize the process of optimizing a CNN for image classification through the use of a GA.

Emotion recognition plays a pivotal role in advancing Human-Computer Interaction (HCI) by enabling systems to understand and respond to human emotional states. Among various modalities, Speech Emotion Recognition (SER) stands out as a non-invasive, cost-effective, and temporally efficient approach for detecting emotions. This study leverages the RAVDESS dataset, which provides high-quality audio recordings for emotion classification. The proposed methodology involves preprocessing audio signals to remove noise, extracting temporal and spectral features in both time and frequency domains, and implementing machine learning models for multi-class emotion classification. The study evaluates and compares the performance of several classification models, including Random Forest (RF), Multilayer Perceptron (MLP), Support Vector Machine (SVM), Convolutional Neural Networks (CNN), and Decision Trees (DT). Experimental results demonstrate promising accuracy levels, highlighting the potential of these machine learning techniques in SER. This research contributes to the integration of SER within Brain-Computer Interfaces (BCI) and other emotion-aware applications, paving the way for enhanced interactive systems. Keywords – Speech Emotion Recognition -Machine Learning - Random Forest -Multilayer Perceptron - Support Vector Machine - Convolutional Neural Networks - Decision Tree - RAVDESS Dataset

Among the list of dreadful diseases, 'Brain Tumor and Brain Cancer' is one of them. According to survival rate, after the detection of cancerous Brain or Central Nervous System (CNS) tumor, 34% of men and 36% of women survive at least 5 years. Therefore, 'Early Detection of tumor' can be a savior for so many lives and artificial intelligence will make the way very accurate and appropriate. A new non-invasive approach to detect brain tumors in early-stage, cost-efficiently with the lowest risk factor is-'Liquid Biopsy'. It is a technique for analyzing non-solid biological tissue, i.e. - blood, plasma, cerebrospinal fluid (CSF). As human inspection maybe error full and inaccurate, in this paper, we have used 'k-means clustering' in 'Image Processing' as visual perception of artificial intelligence for detection of 'CTC' and we have used the technique 'Convolutional Neural Network' (CNN) for ctDNA detection and 'Support Vector Machine'(SVM) algorithm for classification of ctDNA and normal cell-free DNA (cfDNA). 'ctDNA' could provide clues for growth factor of cancerous cells, type of tumors, location, etc. Therefore, our work provides a new way to early detection and a new prospect for early cure.

Detection of brain tumor is very challenging task in today’s medical world. Nowadays, many doctors prefer ready-made methods to diagnose the tumor from given magnetic resonance imaging (MRI) images. Accurate and less time-consuming methods are used frequently to identify it. This paper introduces new techniques for finding and differentiating brain tumors. The new methods include support vector machine (SVM) and convolution neural network (CNN) with Softmax classifier. The SVM uses histogram of oriented gradients (HOGs) as a feature to label the given dataset. Further, the same feature of the test image is verified against labeled sets and classified. The SVM gives accuracy of 87.33%. In CNN, multilayer structure is used. It contains a convolution layer followed by the pooling layer and finally fully connected layer. In CNN, specifically, Alexnet is preferred for classification. In Alexnet, rectified linear unit (ReLU) is used as an activation function. At the output of it, Softmax is used as activation to classify the images. The CNN with Softmax classifier gives improved accuracy of 98.68%. In addition, the performance measures are also found out such as specificity, sensitivity, and precision. The sensitivity with SVM is 93.18%, and with CNN, it increases up to 100%. Comparing both methods with the obtained results, the CNN with Softmax classifier gives better results. Proposed method with CNN produces only one misclassified image as an advantage over the existing method, while SVM gives nine misclassified images. The methods are tested on 253 MRI images which are detected and classified correctly with Matlab.KeywordsSVMCNNHOGMRI