Low Power Unsupervised Anomaly Detection by Non-Parametric Modeling of\n Sensor Statistics Article Swipe

MAKE IMPACT

Ahish Shylendra , Priyesh Shukla , Saibal Mukhopadhyay , Swarup Bhunia , Amit Ranjan Trivedi · YOU? · · 2020 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2003.10088

This work presents AEGIS, a novel mixed-signal framework for real-time\nanomaly detection by examining sensor stream statistics. AEGIS utilizes Kernel\nDensity Estimation (KDE)-based non-parametric density estimation to generate a\nreal-time statistical model of the sensor data stream. The likelihood estimate\nof the sensor data point can be obtained based on the generated statistical\nmodel to detect outliers. We present CMOS Gilbert Gaussian cell-based design to\nrealize Gaussian kernels for KDE. For outlier detection, the decision boundary\nis defined in terms of kernel standard deviation ($\\sigma_{Kernel}$) and\nlikelihood threshold ($P_{Thres}$). We adopt a sliding window to update the\ndetection model in real-time. We use time-series dataset provided from Yahoo to\nbenchmark the performance of AEGIS. A f1-score higher than 0.87 is achieved by\noptimizing parameters such as length of the sliding window and decision\nthresholds which are programmable in AEGIS. Discussed architecture is designed\nusing 45nm technology node and our approach on average consumes $\\sim$75 $\\mu$W\npower at a sampling rate of 2 MHz while using ten recent inlier samples for\ndensity estimation. \\textcolor{red}{Full-version of this research has been\npublished at IEEE TVLSI}\n

Type

preprint

Landing Page

http://arxiv.org/abs/2003.10088

PDF

https://arxiv.org/pdf/2003.10088

OA Status

green

Related Works

10

OpenAlex ID

https://openalex.org/W4287824466

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4287824466

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2003.10088

Digital Object Identifier

Title

Low Power Unsupervised Anomaly Detection by Non-Parametric Modeling of\n Sensor Statistics

Work title

Type

preprint

OpenAlex work type

Publication year

2020

Year of publication

Publication date

2020-03-23

Full publication date if available

Authors

Ahish Shylendra, Priyesh Shukla, Saibal Mukhopadhyay, Swarup Bhunia, Amit Ranjan Trivedi

List of authors in order

Landing page

https://arxiv.org/abs/2003.10088

Publisher landing page

PDF URL

https://arxiv.org/pdf/2003.10088

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2003.10088

Direct OA link when available

Concepts

Anomaly detection, Kernel density estimation, Outlier, Computer science, Sliding window protocol, Kernel (algebra), Benchmark (surveying), Gaussian, Parametric statistics, Artificial intelligence, Algorithm, Statistics, Window (computing), Mathematics, Geodesy, Combinatorics, Physics, Operating system, Estimator, Geography, Quantum mechanics

Top concepts (fields/topics) attached by OpenAlex

Cited by

0

Total citation count in OpenAlex

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload