A Deep Learning-Based Compression Algorithm for 9-DOF Inertial Measurement Unit Signals Along With an Error Compensating Mechanism

Sepahvand, Majid; Abdali-Mohammadi, Fardin

doi:10.1109/jsen.2018.2877360

Cited by 16 publications

(4 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Generalized auto-encoders-based data compression and signal-compressed sensing have also achieved significant evolution [9,10,23]. These methods usually utilize multi-level auto-encoders to overcome the disadvantage that single-level auto-encoders have in being unable to achieve lossless data reconstruction; these methods use auto-encoders to replace one unit of the classical data compression and signal-compressed sensing model, such as the prediction, transformation, or quantization unit of data compression, as well as the measurement or recovery unit of signal-compressed sensing.…”

Section: Related Researchmentioning

confidence: 99%

See 1 more Smart Citation

Serial Decoders-Based Auto-Encoders for Image Reconstruction

Trocan

Galayko

2022

Applied Sciences

View full text Add to dashboard Cite

Auto-encoders are composed of coding and decoding units; hence, they hold an inherent potential of being used for high-performance data compression and signal-compressed sensing. The main disadvantages of current auto-encoders comprise the following aspects: the research objective is not to achieve lossless data reconstruction but efficient feature representation; the evaluation of data recovery performance is neglected; it is difficult to achieve lossless data reconstruction using pure auto-encoders, even with pure deep learning. This paper aims at performing image reconstruction using auto-encoders, employs cascade decoders-based auto-encoders, perfects the performance of image reconstruction, approaches gradually lossless image recovery, and provides a solid theoretical and applicational basis for auto-encoders-based image compression and compressed sensing. The proposed serial decoders-based auto-encoders include the architectures of multi-level decoders and their related progressive optimization sub-problems. The cascade decoders consist of general decoders, residual decoders, adversarial decoders, and their combinations. The effectiveness of residual cascade decoders for image reconstruction is proven in mathematics. Progressive training can efficiently enhance the quality, stability, and variation of image reconstruction. It has been shown by the experimental results that the proposed auto-encoders outperform classical auto-encoders in the performance of image reconstruction.

show abstract

Section: Related Researchmentioning

confidence: 99%

“…The first-level auto-encoders compress the prediction residuals, and the next-level auto-encoders compress the reconstruction residuals of the previous-level auto-encoders to a great extent [10]. Majid Sepahvand et al employed auto-encoders to implement the prediction coding unit of compressed sensing of sensor signals [23].…”

Section: Related Researchmentioning

confidence: 99%

Serial Decoders-Based Auto-Encoders for Image Reconstruction

Trocan

Galayko

2022

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Similarly, Sepahvand et al 12 developed an approach to achieve lossless multi‐axis inertial signal compression. In this approach, pre‐processing is done to do independent components extraction by using the principal component analysis approach.…”

Section: Related Workmentioning

confidence: 99%

An efficient priority‐based convolutional auto‐encoder approach for electrocardiogram signal compression in Internet of Things based healthcare system

Mahendran

Parthasarathy

Parthasarathy³

et al. 2020

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Due to advancements in healthcare monitoring systems, the Internet of Things concepts are proficiently utilized in the medical field to detect and diagnose the physical health problems. The compression of more substantial medical information is a significant issue that requires ample data storage space and takes longer transmission time. Though several compression algorithms are actualized in past cases, there is an absence of an upgraded approach to achieve improved signal compression without influencing signal quality. Hence a proficient signal compression algorithm is proposed in our work to provide an enhanced electrocardiogram (ECG) signal compression without any data loss and to acquire increased compression ratio (CR) and zero construction error. In this proposed approach, the input ECG signal dataset from the MIT‐BIH arrhythmia database gets influenced by noise because of the electrical measuring gadget. Hence, preprocessing is done by the proposed multi‐scoop notch filter (MSNF) to denoise this signal by removing the specified noise frequency range of around (1‐50) Hz. This proposed MSNF is designed with adaptiveness that has achieved the enhanced denoising by adjusting the notch frequency. In addition, to extricate the sophisticated ECG signal features, Fast Fourier Transform is being utilized, and that performs and decomposes the signal elegantly and obtains characteristics of the signal in the frequency domain. After feature extraction, optimal signal compression is performed by our proposed priority‐based convolutional auto‐encoder (PCAE) that provides better compression with almost zero reconstruction error by encoding the signals into lower‐dimensional vectors in convolutional layers which are again reconstructed using a decoding approach. The experimental results are then assessed using the performance metrics that include signal to noise ratio (SNR), CR, and percentage root‐mean‐square difference (PRD). The attained results are 1.83% as average PRD value, average SNR is about 33 dB, and average CR is about 35.2% whereas the traditional CAE approach has average values of 2.05% PRD, 23.45 dB SNR, and 32.2% CR.

show abstract

“…This integration enables precise monitoring of the health status of wind turbine blades, exemplifying an approach to applying IMU in the context of renewable energy technology. Nicola et al [14] have designed a structural monitoring system relying on micro inertial sensor network, which utilizes fusion algorithm based on complementary filters to extract tilt angles from triaxial acceleration and triaxial angular velocity. This study employs Micro Inertial Measurement Unit (MIMU) to precisely measure the triaxial acceleration and angular velocity of transmission towers, enabling continuous and detailed monitoring of their structural integrity.…”

Section: Introductionmentioning

confidence: 99%

Looseness Identification of Transmission Tower Based on Inertial Measurement

Shen,

Yang

2024

IEEE Access

View full text Add to dashboard Cite

Transmission towers play a critical role in power systems. In this study, we propose an innovative approach for the structural health monitoring of transmission towers with the aim of accurately identifying structural looseness and its location, facilitating timely warnings. By utilizing Micro Inertial Measurement Unit (MIMU) technology, we successfully capture the comprehensive dynamic behavior of transmission towers. Subsequently, we employ the extended Dynamic Mode Decomposition (DMD) to successfully capture the looseness features of transmission towers, leading to the development of looseness features: Dynamic Mode Participation Ratio and Degree of Freedom Mode Shape. This process, developed with signal window, delay embedding and cluster stability diagrams, addresses the shortcomings of the DMD algorithm when dealing with vibration data in complex engineering environments. To further identify the areas of looseness in transmission towers, we introduce the Dempster-Shafer (D-S) evidence theory, enabling the full utilization of information within the looseness indicators. Experimental results demonstrate that our proposed method consistently delivers precise and robust identification results for looseness and its location in various cases. This research not only provides an innovative solution for the detection of looseness in transmission towers but also offers valuable insights for the health monitoring and maintenance of similar structures, contributing to the reduction of potential accidents.

show abstract

A Deep Learning-Based Compression Algorithm for 9-DOF Inertial Measurement Unit Signals Along With an Error Compensating Mechanism

Cited by 16 publications

References 37 publications

Serial Decoders-Based Auto-Encoders for Image Reconstruction

Serial Decoders-Based Auto-Encoders for Image Reconstruction

An efficient priority‐based convolutional auto‐encoder approach for electrocardiogram signal compression in Internet of Things based healthcare system

Looseness Identification of Transmission Tower Based on Inertial Measurement

Contact Info

Product

Resources

About