Anomaly and Degradation Detection Using Subspace Tracking in Streaming Data

Cha, K; Sadek, Carol; Asgharzadeh, Zohreh

doi:10.1109/bigdata50022.2020.9377804

Cited by 1 publication

(3 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2) SVD operation: SVD is used in the ISVD algorithm in (14). The Golub-Reinsch technique (GR-SVD) [60] is adopted for this purpose, which has computational complexity similar to the Chol-QR method.…”

Section: B Low-end Devicesmentioning

confidence: 99%

“…GR-SVD is implemented on the MCU using the CControl library 4 . To optimize the use of memory on the MCU, the CControl library is modified to eliminate the unnecessary Q matrix in (14) and to overwrite the input matrix with the P matrix.…”

Section: B Low-end Devicesmentioning

confidence: 99%

“…For instance, representing the signal information in a lower-dimensional space hinges on the possibility of signal compression [8]. Monitoring how the signal either distributes its energy over the principal components or the unexpected presence of energy in the minor subspace may reveal the presence of anomalies [11]- [14]. In the case of Code Division Multiple Access systems, it has been proven that the identification of the principal subspace of the received signal allows the design of a simpler and better performing receiver compared to classical detector [5].…”

mentioning

confidence: 99%

See 2 more Smart Citations

Streaming Algorithms for Subspace Analysis: Comparative Review and Implementation on IoT Devices

Marchioni

Prono

Mangia

et al. 2023

IEEE Internet Things J.

View full text Add to dashboard Cite

Subspace analysis is a widely used technique for coping with high-dimensional data and is becoming a fundamental step in the early treatment of many signal processing tasks. However, traditional subspace analysis often requires a large amount of memory and computational resources, as it is equivalent to eigenspace determination. To address this issue, specialized streaming algorithms have been developed, allowing subspace analysis to be run on low-power devices such as sensors or edge devices.Here, we present a classification and a comparison of these methods by providing a consistent description and highlighting their features and similarities. We also evaluate their performance in the task of subspace identification with a focus on computational complexity and memory footprint for different signal dimensions. Additionally, we test the implementation of these algorithms on common hardware platforms typically employed for sensors and edge devices.

show abstract

Section: B Low-end Devicesmentioning

confidence: 99%

Section: B Low-end Devicesmentioning

confidence: 99%