Geomagnetic-based indoor positioning has drawn a great attention from academia and industry due to its advantage of being operable without infrastructure support and its reliable signal characteristics. However, it must overcome the problems of ambiguity that originate with the nature of geomagnetic data. Most studies manage this problem by incorporating particle filters along with inertial sensors. However, they cannot yield reliable positioning results because the inertial sensors in smartphones cannot precisely predict the movement of users. There have been attempts to recognize the magnetic sequence pattern, but these attempts are proven only in a one-dimensional space, because magnetic intensity fluctuates severely with even a slight change of locations. This paper proposes accurate magnetic indoor localization using deep learning (AMID), an indoor positioning system that recognizes magnetic sequence patterns using a deep neural network. Features are extracted from magnetic sequences, and then the deep neural network is used for classifying the sequences by patterns that are generated by nearby magnetic landmarks. Locations are estimated by detecting the landmarks. AMID manifested the proposed features and deep learning as an outstanding classifier, revealing the potential of accurate magnetic positioning with smartphone sensors alone. The landmark detection accuracy was over 80% in a two-dimensional environment.

Motivation: The increase in the amount of available protein–protein interaction (PPI) data enables us to develop computational methods for protein complex predictions. A protein complex is a group of proteins that interact with each other at the same time and place. The protein complex generally corresponds to a cluster in PPI network (PPIN). However, clusters correspond not only to protein complexes but also to sets of proteins that interact dynamically with each other. As a result, conventional graph-theoretic clustering methods that disregard interaction dynamics show high false positive rates in protein complex predictions. Results: In this article, a method of refining PPIN is proposed that uses the structural interface data of protein pairs for protein complex predictions. A simultaneous protein interaction network (SPIN) is introduced to specify mutually exclusive interactions (MEIs) as indicated from the overlapping interfaces and to exclude competition from MEIs that arise during the detection of protein complexes. After constructing SPINs, naive clustering algorithms are applied to the SPINs for protein complex predictions. The evaluation results show that the proposed method outperforms the simple PPIN-based method in terms of removing false positive proteins in the formation of complexes. This shows that excluding competition between MEIs can be effective for improving prediction accuracy in general computational approaches involving protein interactions. Availability:http://code.google.com/p/simultaneous-pin/ Contact:dshan@kaist.ac.kr Supplementary information: Supplementary data are available at Bioinformatics online.

If the construction and maintenance of Wi-Fi radio maps were fully automated, the implementation of a global-scale Wi-Fi indoor positioning system would be possible. This paper proposes a Wi-Fi radio map calibration system that automates the initial construction and maintenance of radio maps using crowdsourced fingerprints collected from numerous smartphones without location information. The system incorporates an unsupervised learning algorithm into an incremental and adaptive calibration process. The unsupervised learning algorithm constructs an initial radio map, using fingerprints collected from unknown locations, by finding a hidden structure among them. Once a positioning service is available based on the initial radio map, the radio map continues to adapt to signal changes in the environment through the incremental and adaptive calibration process using the fingerprints that are continuously collected from the service users. Experiments carried out in an office building have shown that the proposed system could successfully construct and maintain a precise radio map without requiring any location information. A longterm experiment that lasted for five months, the proposed system was able to not just maintain but also improve the quality of the radio map. These results indicate that Wi-Fi indoor positioning systems can be automatically constructed and maintained in continuously changing Wi-Fi environments without manual calibration efforts.

Radio channel state information (CSI) measured with many receivers is a good resource for localizing a transmitter device with machine learning with a discriminative model. However, CSI localization is nontrivial when the radio map is complicated, such as in building corridors. This article introduces a view-selective deep learning (VSDL) system for indoor localization using CSI of WiFi. The multiview training with CSI obtained from multiple groups of access points (APs) generates latent features on a supervised variational deep network. This information is then applied to an additional network for dominant view classification to minimize the regression loss of localization. As noninformative latent features from multiple views are rejected, we can achieve a localization accuracy of 1.28 m, which outperforms by 30% the best known accuracy in practical applications in a complex building environment. To the best of our knowledge, this is the first approach to apply variational inference and to construct a practical system for radio localization. Furthermore, our work investigates a methodology for supervised learning with multiview data where informative and noninformative views coexist.

In this paper, we propose a new geomagnetic localization scheme, named ILoA, to address error accumulation and global localization. Global localization is a fundamental problem that determines the initial pose under global uncertainty. Moreover, error accumulation using inertial navigation systems (INS) impacts robustness and drift error, making it challenging to achieve reliable estimation. The magnetic field in indoor space generates a unique signature/anomaly, which can be used as a local feature. Earth’s magnetic field can be easily influenced by ferromagnetic material from the indoor environment due to its weak intensity. The magnetic field vector measured by a magnetometer depends on the orientation of the sensor, which we term a direction variant. We devise a novel approach to identify location and heading through the direction-variant augmented vector. Since a magnetic field vector under varying poses can produce many different vectors, the geomagnetic map is trained with the transformation. We present experiments in two testbeds, covering open space, showing that the proposed method using the magnetic field vector is efficient for global localization and accuracy compared with a state-of-the-art approach.

In this study, polyvinyl alcohol (PVA)-mixed DL-alanine (PVA/DL-alanine) polycrystals are fabricated, and their piezoelectric characteristics in the d₃₃ mode are investigated. The d₃₃ piezoelectric coefficients of the PVA/DL-alanine polycrystals are found to increase with an increase in the weight ratio of DL-alanine, and the PVA/DL-alanine polycrystal composed of PVA and DL-alanine in a weight ratio of 1:3 exhibits a d₃₃ of ∼5 pC/N. The piezoelectric characteristics of the PVA/DL-alanine polycrystals are discussed in terms of the crystal structure by employing scanning electron microscopy and X-ray diffraction analyses. To confirm the piezoelectric performance of the polycrystals, the piezoelectric voltages of a piezoelectric device composed of a single layer of ZnO thin film and heterostructured devices consisting of a layer of PVA/DL-alanine polycrystal and a ZnO thin film layer are measured and compared. This study presents PVA/DL-alanine polycrystals as a potential piezoelectric material for bio-friendly piezoelectric-device applications.