Search

This is a preprint version of the paper entitled “EWOk: Towards Efficient Multidimensional Compression of Indoor Positioning Datasets”, published in IEEE Transactions on Mobile Computing. Please cite this paper as L. Klus, R. Klus, J. Torres-Sospedra, E. S. Lohan, C. Granell and J. Nurmi, "EWOk: Towards Efficient Multidimensional Compression of Indoor Positioning Datasets," in IEEE Transactions on Mobile Computing, doi: 10.1109/TMC.2023.3277333. DOI 10.1109/TMC.2023.3277333 Publisher Name IEEE License This publication is copyrighted. You may download, display and print it for your own personal use. Commercial use is prohibited. Link to publication https://ieeexplore.ieee.org/document/10128720 Take down policy If you believe that this document breaches copyright, please contact the authors, and we will investigate your claim

Emerging technologies enable massive deployment of wireless sensor networks across many industries. Internet of Things (IoT) devices are often deployed in critical infrastructure or health monitoring and require fast reaction time,reasonable accuracy, and high energy efficiency. In this letter, we introduce a lossy compression method for time-seriesdata, named direct lightweight temporal compression (DLTC), enabling energy-efficient data transfer for power-restricteddevices. Our method is based on the lightweight temporal compression method, targeting further reconstruction errorminimization and complexity reduction. This letter highlights the key advantages of the proposed method and evaluates themethod’s performance on several sensor-based, time-series data types. We prove that DLTC outperforms the consideredbenchmark methods in compression efficiency at the same reconstruction error level.

Ponencia presentada en la 2021 International Conference on Indoor Positioning and Indoor Navigation (IPIN), 29 Nov.-2 Dec. 2021, Lloret de Mar (Spain) Fingerprinting is a widely used technique in indoor positioning, mainly due to its simplicity. Usually, this technique is used with the deterministic k - Nearest Neighbors (k-NN) algorithm. Utilizing a neural network model for fingerprinting positioning purposes can greatly improve the prediction speed compared to the k-NN approach, but requires a voluminous training dataset to achieve comparable performance. In many indoor positioning datasets, the number of samples is only at a level of hundreds, which results in poor performance of the neural network solution. In this work, we develop a novel algorithm based on a transfer learning approach, which combines samples from 15 different Wi-Fi RSS indoor positioning datasets, to train a single convolutional neural network model, which learns the common patterns in the combined data. The proposed model is then fine-tuned to optimally fit the individual databases. We show that the proposed solution reduces the positioning error by up to 25% compared to the benchmark model while reducing the number of outlier predictions.

This paper gives an overview of crowdsourcing databases and crowdsourcing-related challenges and open research issues for data collected from wearable devices. It is shown that, with the advent of smarter wearable devices, the complexity of data gathering, storage, and processing in crowdsourced modes will increase exponentially and new solutions are needed in order to cope with larger data sets and low energy consumption in wearable devices, while ensuring the integrity and quality of the collected data. publishedVersion