Lightweight Lossy Compression of Biometric Patterns via Denoising Autoencoders

Abstract

Wearable Internet of Things (IoT) devices permit the massive collection of biosignals (e.g., heart-rate, oxygen level, respiration, blood pressure, photo-plethysmographic signal, etc.) at low cost. These, can be used to help address the individual fitness needs of the users and could be exploited within personalized healthcare plans. In this letter, we are concerned with the design of lightweight and efficient algorithms for the lossy compression of these signals. In fact, we underline that compression is a key functionality to improve the lifetime of IoT devices, which are often energy constrained, allowing the optimization of their internal memory space and the efficient transmission of data over their wireless interface. To this end, we advocate the use of autoencoders as an efficient and computationally lightweight means to compress biometric signals. While the presented techniques can be used with any signal showing a certain degree of periodicity, in this letter we apply them to ECG traces, showing quantitative results in terms of compression ratio, reconstruction error and computational complexity. State of the art solutions are also compared with our approach

Study on data compression algorithm and its implementation in portable electronic device for Internet of Things applications

“…Regarding its method, MAS project concentrated on the implementation of the data compression algorithm in WSN. This approach is similar to projects [5], [6], [7], [8], and [10]. The last similarity of the project [9] was that it focused on improving the performance of the data compression.…”

“…According to various researches, data compression is divided into two categories: lossless and lossy compressions. The Autoencoders (AE) compression algorithm in [5] is known as a lossy compression. The improvement of the IoT device's lifetime such as energy constrained, optimization of the internal memory space, and efficient data transmission via wireless connection was brought into focus in this experiment, while biomedical signals such as ECG, photoplethysmography, and respiratory traces were used as the dataset.…”

“…The performance of the AE compression algorithm was then compared with Discrete Cosine Transforms (DCT), Wavelet transforms, Principal Component Analysis (PCA), and linear approximations. Based on the results produced in the paper [5], the EPJ Web of Conferences 162, 01073 (2017) DOI: 10.1051/epjconf/201716201073 InCAPE2017 energy consumption required for the online compression of the signal is quite small when compared with a scheme using linear approximations. According to researchers, this project also posed a weakness.…”

“…There are different types of compression algorithms in this world, some of which can support image files, while some can only support text files. Some data compressions cannot preserve the originality of the data after decoding, for example, the Autoencoder algorithm [5]. In Section 3, the data compression algorithms that were previously discussed in Section 2 are analyzed and displayed in a table for comparison.…”

“…The last similarity of the project [9] was that it focused on improving the performance of the data compression. Project [9] was focused on reducing the usage of power consumption during data transmission, which was similar to projects [5], [6], and [7].…”