Search

This paper proposes a machine learning approach to detect threats using short-term PPG (photoplethysmogram) signals from a commercial smartwatch. In supervised learning, having accurately annotated training data is essential. However, a key challenge in the threat detection problem is the uncertainty regarding how accurately data labeled as ‘threat’ reflect actual threat responses since participants may react differently to the same experiments. In this paper, Gaussian Mixture Models are learned to remove ambiguously labeled training, and those models are also used to remove ambiguous test data. For the realistic test scenario, PPG measurements are collected from participants playing a horror VR (Virtual Reality) game, and the proposed method validates the superiority of our proposed approach in comparison with other methods. Also, the proposed filtering with GMM improves prediction accuracy by 23% compared to the method that does not incorporate the filtering.

Wi-Fi fingerprint indoor localization uses Wi-Fi signal strength measurements obtained from a number of access points. This method needs manual data collection across a positioning area and an annotation process to label locations to the measurement sets. To reduce the cost and effort, this paper proposes a Wi-Fi Semi-Supervised Generative Adversarial Network (SSGAN), which produces artificial but realistic trainable fingerprint data. The Wi-Fi SSGAN is based on a deep learning, which is extended from GAN in a semi-supervised learning manner. It is designed to create location-labeled Wi-Fi fingerprint data, which is different to unlabeled data generation by a normal GAN. Also, the proposed Wi-Fi SSGAN network includes a positioning model, so it does not need a external positioning method. When the Wi-Fi SSGAN is applied to a multi-story landmark localization, the experimental results demonstrate a 35% more accurate performance in comparison to a standard supervised deep neural network.

Fingerprinting localization is a promising indoor positioning methods thanks to its advantage of using preinstalled infrastructure. For example, WiFi signal strength can be measured by pre-existing WiFi routers. In the offline phase, the fingerprinting localization method first stores of position and RSSI measurement pairs in a dataset. Second, it predicts a target’s location by comparing the stored fingerprint database to the current measurement. The database size is normally huge, and data patterns are complicated; thus, an artificial neural network is used to model the relationship of fingerprints and locations. The existing fingerprinting locations, however, have been developed to predict only single locations. In practice, many users may require positioning services, and as such, the core algorithm should be capable of multiple localizations, which is the main contribution of this paper. In this paper, multiple fingerprinting localization is developed based on an artificial neural network and an analysis of the number of targets that can be estimated without loss of accuracy is conducted by experiments.

Background and Objectives: Muscle strengthening exercise is suggested to beneficial for patients with knee osteoarthritis (OA) and electrical muscular stimulation is reported to be effective in improvement of muscle strength. This study examined whether isometric exercise combined with whole body-electromyostimulation (WB-EMS) can improve serum cytokine levels, muscle strength, and knee function in elderly women with early knee OA.Materials and Methods: This randomized controlled study included 75 participants assigned into three groups: the control group (CON), isometric exercise group (ISOM), and isometric exercise and electromyostimulation group (ISOM + EMS). The two exercise groups performed their respective programs for 8 weeks, 3 days a week, 30 min a day. The main exercises for both groups were performed continuously during the 20 min in an alternation of a 6-s contraction with a 4-s break. At pre- and post-intervention, anthropometric variables, muscle strength, Knee Injury and Osteoarthritis Outcome Score (KOOS), and blood sampling for biomarkers including interleukin-6, tumor necrosis factor-α, C-reactive protein, and resistin were performed.Results: All variables at pre-intervention showed no significant differences among the three groups. However, there were significant differences between groups for body composition, muscle strength, KOOS subscale scores, and biomarkers. ISOM + EMS group resulted in a significant reduction in body weight, fat mass, fat percentage, inflammatory cytokine levels, and increased muscle strength. An ISOM + EMS group had the best KOOS score among all groups.Conclusion: Isometric exercise combined with WB-EMS resulted in the best overall improvements in knee function and alleviating the pain and symptoms of patients with early knee OA. Further, reduced levels of inflammatory cytokines were observed. These non-pharmacologic, non-invasive interventions should be considered by healthcare specialists for elderly patients with early knee OA.

With a steady increase in impervious surfaces due to urbanization in Korea, there is a growing burden and an urgent need to fund better management of nonpoint sources of pollution and stormwater. A prerequisite for securing the necessary financial resources is the determination of basic data for the accurate calculation of impervious surfaces as the basis for estimating the costs of nonpoint source pollution control and billing of stormwater utility fees. This requires the extraction of landcover information in a Geographic Information System (GIS) environment and development of a landcover map that accurately delineates pervious surfaces within impervious surface areas. However, since landcover maps in Korea are generated and updated at irregular intervals, it is difficult to properly track land use changes. To address this problem, this study has developed a new method for the detailed updating of landcover maps in developing urban areas to facilitate the calculation of stormwater utility fees. Sejong City was selected as the study site because it has experienced large-scale land use changes due to the recent relocation of the national administrative capital and continuous urbanization. The methodology proposed in this study is based on various spatial data such as aerial photographs and digital topographic maps and follows four process steps: preprocessing, first and second updates of landcover information, and quality assurance. In a test of this method, a total of 19,049 reclassified items were generated in the first and second updates, affecting a total area of 26.49 km2 within the original landcover map. The accuracy of these updates reached 99.78%, considering the changed areas and rate of change. This study provides fundamental data for further application of a stormwater utility fee policy in Korea. However, further research is required to automate the generation of accurate pervious/impervious maps and develop pertinent guidelines so that individual municipal and provincial governments can generate and update their own pervious/impervious maps as a basis for calculating the impervious surfaces in their regions.

This study introduces a semi-empirical algorithm to estimate the extent of the phycocyanin (PC) concentration in eutrophic freshwater bodies; this is achieved by studying the reflectance characteristics of the red and near-red spectral regions, especially the shifting of the peak near 700 nm towards longer wavelengths. Spectral measurements in a darkroom environment over the pure-cultured cyanobacteria Microcystis showed that the shift is proportional to the algal biomass. A similar proportional trend was found from extensive field measurement data. The data also showed that the correlation of the magnitude of the shift with the PC concentration was greater than that with chlorophyll-a. This indicates that the characteristic can be a useful index to quantify cyanobacterial biomass. Based on these observations, a new PC algorithm was proposed that uses the remote sensing reflectance of the peak band around 700 nm and the trough band around 620 nm, and the magnitude of the peak shift near 700 nm. The efficacy of the algorithm was tested with 300 sets of field data, and the results were compared to select algorithms for the PC concentration prediction. The new algorithm performed better than the other algorithms with respect to most error indices, especially the mean relative error, indicating that the algorithm can reduce errors when PC concentrations are low. The algorithm was also applied to a hyperspectral dataset obtained through aerial imaging, in order to predict the spatial distribution of the PC concentration in an approximately 86 km long reach of the Nakdong River.

Machine learning has been successfully used for target localization in wireless sensor networks (WSNs) due to its accurate and robust estimation against highly nonlinear and noisy sensor measurement. For efficient and adaptive learning, this paper introduces online semi-supervised support vector regression (OSS-SVR). The first advantage of the proposed algorithm is that, based on semi-supervised learning framework, it can reduce the requirement on the amount of the labeled training data, maintaining accurate estimation. Second, with an extension to online learning, the proposed OSS-SVR automatically tracks changes of the system to be learned, such as varied noise characteristics. We compare the proposed algorithm with semi-supervised manifold learning, an online Gaussian process and online semi-supervised colocalization. The algorithms are evaluated for estimating the unknown location of a mobile robot in a WSN. The experimental results show that the proposed algorithm is more accurate under the smaller amount of labeled training data and is robust to varying noise. Moreover, the suggested algorithm performs fast computation, maintaining the best localization performance in comparison with the other methods.

Tracking the locations and identities of moving targets in the surveillance area of wireless sensor networks is studied. In order to not rely on high-cost sensors that have been used in previous researches, we propose the integrated localization and classification based on semi-supervised learning, which uses both labeled and unlabeled data obtained from low-cost distributed sensor network. In our setting, labeled data are obtained by seismic and PIR sensors that contain information about the types of the targets. Unlabeled data are generated from the RF signal strength by applying Gaussian process, which represents the probability of predicted target locations. Finally, by using classified unlabeled data produced by semi-supervised learning, identities and locations of multiple targets are estimated. In addition, we consider a case when the labeled data are absent, which can happen due to fault or lack of the deployed sensor nodes and communication failure. We overcome this situation by defining artificial labeled data utilizing characteristics of support vector machine, which provides information on the importance of each training data point. Experimental results demonstrate the accuracy of the proposed tracking algorithm and its robustness to the absence of the labeled data thanks to the artificial labeled data.

A mapping guidance algorithm of a quadrotor for unknown indoor environments is proposed. A sensor with limited sensing range is assumed to be mounted on the quadrotor to obtain object data points. With obtained data, the quadrotor computes velocity vector and yaw commands to move around the object while maintaining a safe distance. The magnitude of the velocity vector is also controlled to prevent a collision. The distance transform method is applied to establish dead-end situation logic as well as exploration completion logic. When a dead-end situation occurs, the guidance algorithm of the quadrotor is switched to a particular maneuver. The proposed maneuver enables the quadrotor not only to escape from the dead-end situation, but also to find undiscovered area to continue mapping. Various numerical simulations are performed to verify the performance of the proposed mapping guidance algorithm.

This paper studies the indoor localization based on Wi-Fi received signal strength indicator (RSSI). In addition to position estimation, this study examines the expansion of applications using Wi-Fi RSSI data sets in three areas: (i) feature extraction, (ii) mobile fingerprinting, and (iii) mapless localization. First, the features of Wi-Fi RSSI observations are extracted with respect to different floor levels and designated landmarks. Second, the mobile fingerprinting method is proposed to allow a trainer to collect training data efficiently, which is faster and more efficient than the conventional static fingerprinting method. Third, in the case of the unknown-map situation, the trajectory learning method is suggested to learn map information using crowdsourced data. All of these parts are interconnected from the feature extraction and mobile fingerprinting to the map learning and the estimation. Based on the experimental results, we observed (i) clearly classified data points by the feature extraction method as regards the floors and landmarks, (ii) efficient mobile fingerprinting compared to conventional static fingerprinting, and (iii) improvement of the positioning accuracy owing to the trajectory learning.

Machine learning-based indoor localization used to suffer from the collection, construction, and maintenance of labeled training databases for practical implementation. Semi-supervised learning methods have been developed as efficient indoor localization methods to reduce use of labeled training data. To boost the efficiency and the accuracy of indoor localization, this paper proposes a new time-series semi-supervised learning algorithm. The key aspect of the developed method, which distinguishes it from conventional semi-supervised algorithms, is the use of unlabeled data. The learning algorithm finds spatio-temporal relationships in the unlabeled data, and pseudolabels are generated to compensate for the lack of labeled training data. In the next step, another balancing-optimization learning algorithm learns a positioning model. The proposed method is evaluated for estimating the location of a smartphone user by using a Wi-Fi received signal strength indicator (RSSI) measurement. The experimental results show that the developed learning algorithm outperforms some existing semi-supervised algorithms according to the variation of the number of training data and access points. Also, the proposed method is discussed in terms of why it gives better performance, by the analysis of the impact of the learning parameters. Moreover, the extended localization scheme in conjunction with a particle filter is executed to include additional information, such as a floor plan.

Supervised machine learning has been widely used in context-aware wireless sensor networks (WSNs) to discover context descriptions from sensor data. However, collecting a lot of labeled training data in order to guarantee good performance requires much cost and time. For this reason, the semisupervised learning has been recently developed due to its superior performance despite using only a small amount of the labeled data. In this paper, we extend the standard support vector regression (SVR) to the semisupervised SVR by employing manifold regularization, which we call Laplacian SVR (LapSVR). The LapSVR is compared with the standard SVR and the semisupervised least square algorithm that is another recently developed semisupervised regression algorithm. The algorithms are evaluated for location awareness of multiple mobile robots in a WSN. The experimental results show that the proposed algorithm yields more accurate location estimates than the other algorithms.

Abstract This study introduces a novel deep learning-based technique for predicting pressure distribution images, aimed at application in image-based approximate optimal design. The proposed approach integrates both unsupervised and supervised learning paradigms, employing autoencoders (AE) for the unsupervised component and fully connected neural networks (FNN) for the supervised component. A surrogate model based on 2D image data was developed, enabling a comparative analysis of three distinct methods: the conventional AE, the convolutional autoencoder (CAE), and a hybrid CAE, which combines the CAE with a conventional AE. Extensive experiments demonstrated that the CAE method achieved the highest learning capability and restoration rate for pressure distribution images of 2D airfoils. The compressed latent image data were utilized as inputs for the FNN, which was trained to predict latent features. These features were decoded to forecast the corresponding pressure distribution images. The results showed excellent concordance with those derived from computational fluid dynamics (CFD) simulations, achieving a match rate exceeding 99.99%. This methodology significantly simplifies and accelerates image prediction, rendering it feasible without requiring specialized CFD knowledge. Moreover, it enhances accuracy while streamlining the neural network structure. Consequently, it provides foundational technology for image data-based optimization, establishing a platform for future AI-driven design and optimization advancements.