Machine Learning for Antenna Array Failure Analysis Lydia de Lange - - PowerPoint PPT Presentation

β–Ά
machine learning
SMART_READER_LITE
LIVE PREVIEW

Machine Learning for Antenna Array Failure Analysis Lydia de Lange - - PowerPoint PPT Presentation

Sep 30, 2023 387 likes β€’649 views

Machine Learning for Antenna Array Failure Analysis Lydia de Lange Under Dr DJ Ludick and Dr TL Grobler Dept. Electrical and Electronic Engineering, Stellenbosch University MML 2019 Outline 15/03/2019 3 Introduction 15/03/2019 4

slide-1
SLIDE 1

Machine Learning for Antenna Array Failure Analysis

Lydia de Lange Under Dr DJ Ludick and Dr TL Grobler

  • Dept. Electrical and Electronic Engineering,

Stellenbosch University MML 2019

slide-2
SLIDE 2

Outline

15/03/2019 3

slide-3
SLIDE 3

4 15/03/2019

Introduction

slide-4
SLIDE 4

Antenna Arrays

5 15/03/2019

slide-5
SLIDE 5

Reconstructed Sky Im Image

6 15/03/2019

slide-6
SLIDE 6

Square Kilometer Array (S (SKA)

7 15/03/2019

slide-7
SLIDE 7

Lydia’s Arrays (LA) and Far-Field Patterns

8 15/03/2019

slide-8
SLIDE 8

Problem Statement

Element failure Inaccurate far-field patterns (beam patterns) Distorted results (e.g. in reconstructed sky image)

Important applications:

  • Array failure correction
  • System health management of large antenna arrays
slide-9
SLIDE 9

Previous work

11 15/03/2019

Failed antenna element detection and location possible with machine learning techniques e.g.:

  • Feedforward neural

networks

  • Support vector

models

slide-10
SLIDE 10

Methodology

12 15/03/2019

slide-11
SLIDE 11

Methodology

Simulate scenarios for input data Sampling methods Train FNN

slide-12
SLIDE 12

Sampling Methods

Name Number of Samples πœ’ (Β°) πœ„ (Β°)

Single cut (πœ’ = 0) 181 πœ„ ∈ [βˆ’90, 90] Single cut (πœ’ = 45) 45 Single cut (πœ’ = 90) 90 Single cut (πœ’ = 135) 135 Principle cuts 362 0, 90 Diagonal cuts 45, 135 All cuts 724 0, 45, 90, 135 3-D pattern (182 samples) 182 3-D far-field pattern sampled in a (πœ„, πœ’) grid. 3-D pattern (361 samples) 361 3-D pattern (725 samples) 725

slide-13
SLIDE 13

Training of FNN

Sampled far-field observation of 1 failure scenario

𝑦

y = ON or OFF state of each antenna in the array β€œmulti-label” – 1 label for each antenna (25)

𝑧

Multi-label feedforward neural network

slide-14
SLIDE 14

Training of FNN

Adapt parameters 𝛾 with each pass until f is as similar as possible to true relationship.

slide-15
SLIDE 15

22 15/03/2019

Results

slide-16
SLIDE 16

FNN Results

23

15/03/2019

Nature of FNN*:

  • ↑ Number of samples (S)
  • ↑ Number of parameters (𝛾)

to be estimated

  • ↓ Accuracy
  • If accuracy ↑: sampling

pattern found a useful region in the 3-D far-field pattern to accurately identify failure scenarios

* # training iterations = const.

slide-17
SLIDE 17

FNN Results

Dataset Samples Training Time (sec) Accuracy (%)

90α΅’ cut 181 31.98 69.70 3-D pattern 182 32.17 87.88 Diagonal cuts 362 35.48 90.91 All cuts 724 40.73 75.76

slide-18
SLIDE 18

Additional experiments

25

slide-19
SLIDE 19

Additional Experiments

  • Compared 14 other classification algorithms1 according to accuracy

using the 10 sampling method datasets.

  • Best 4:
  • FNN
  • One vs Rest Classifier + Linear SVC
  • One vs Rest Classifier + Logistic Regression
  • One vs Rest Classifier + Logistic Regression CV

x1 Scikit-learn algorithms

slide-20
SLIDE 20

Additional Experiments

27 15/03/2019

  • Best: One vs Rest +

Logistic Regression CV

  • 100% accuracy

achieved

  • Number of parameters

vs accuracy relationship is different

  • 3-D sampling method

contains more information than combined single cuts

10 20 30 40 50 60 70 80 90 100

ACCURACY (%) SAMPLING METHOD DATASETS

Classification Algorithm Comparison

FNN OvR+LinearSVC OvR+LogisticRegression OvR+LogisticRegressionCV

slide-21
SLIDE 21

Conclusion

28 15/03/2019

slide-22
SLIDE 22

Conclusion

  • FNN used to detect and locate failed antenna elements in a bow-tie

antenna array

  • Investigated choice of training data on FNN accuracy and training time
  • Diagonal cuts – 90.91% accuracy, 35.48 secs
  • 3-D pattern (182 samples) – 87.88% accuracy, 32.17 secs
  • On larger datasets with more scenarios, the difference in training

time may become more significant.

  • Additional work:
  • Best algorithm: One vs Rest + Logistic Regression CV
  • Best sampling method: 3-D pattern
slide-23
SLIDE 23

Future work

  • Manufacturing and measuring

an antenna array with a spherical nearfield scanner!

  • Look at SVMs
  • Looking at other places in

pipeline to do ML on: Power Spectral Density and Correlations

slide-24
SLIDE 24

Acknowledgement

The financial assistance of the South African SKA project (SKA SA) towards this research is hereby acknowledged (www.ska.ac.za).

15/03/2019 31

slide-25
SLIDE 25

References

15/03/2019

  • [1] R. J. Mailloux, β€œArray Failure Correction With

A Digitally Beamformed Array,” IEEE Trans. Antennas Propag., vol. 44, no. 12, pp. 1543– 1550, 1996.

  • [2] P. Hall, β€œThe Square Kilometre Array: An

Engineering Perspective,” Springer, 2010.

  • [3] J. A. RodrΓ¬guez, et al., β€œA Comparison Among

Several Techniques For Finding Defective Elements In Antenna Arrays,” 2nd European Conference on Antennas and Propagation (EUCAP), pp. 1–8, 2007.

  • [4] I. Goodfellow, Y. Bengio, and A. Courville,

β€œDeep Learning,” MIT Press, pp. 164–167, 2016.