Laser & Optoelectronics Progress, Volume. 58, Issue 8, 0812001(2021)

Research on Fingerprint Location Algorithm Based on OCAE-SOM

Xinchun Li1, Xiaolu Ji2、*, Wu Wei2, Liyan Wang2, Yongyan Gu2, and Dayan Cao1
Author Affiliations
  • 1School of Electronic and Information Engineering, Liaoning Technical University, Huludao, Liaoning 125105, China
  • 2Graduate School, Liaoning Technical University, Huludao, Liaoning 125105, China
  • show less
    Figures & Tables(11)
    Basic model of convolutional autoencoders
    Flow chart of overall algorithm
    Different environment layouts. (a) LOS layout; (b) NLOS layout
    Loss comparison curves of different sizes
    Contrastive curves of accuracy in different modes
    Comparison of positioning accuracy of four indoor positioning algorithms in different environments. (a) NLOS layout; (b) LOS layout
    • Table 1. Network structure of CAE

      View table

      Table 1. Network structure of CAE

      LayerParameterActivation functionOutput size
      Convolutional layer3×3 filter kernel16 feature imagesReLU64×64×16
      Max pooling layer2×2--32×32×16
      Convolutional layer3×3 filter kernel8 feature imagesReLU32×32×8
      Max pooling layer2×2--16×16×8
      Convolutional layer3×3 filter kernel8 feature imagesReLU16×16×8
      Max pooling layer2×2--8×8×8
    • Table 2. Parameter setting of Adam algorithm

      View table

      Table 2. Parameter setting of Adam algorithm

      ParameterMeaningValue
      αLearning rate0.01
      β1Exponential decay rate of 1stmoment estimate0.9
      β2Exponential decay rate of 2ndmoment estimate0.999
      εConstant1×10-8
    • Table 3. Comparison of location accuracy of feature extraction methods

      View table

      Table 3. Comparison of location accuracy of feature extraction methods

      MethodEnvironmentMaximum positioning error /mAverage positioning error /mTraining time /s
      CAELOS4.53821.27253.41
      NLOS7.96031.93416.32
      PCALOS6.03621.72607.29
      NLOS9.53742.307112.53
      Statistical averageLOS6.56071.95649.72
      NLOS10.62433.047216.74
    • Table 4. Performance comparison of proposed algorithm before and after improvement

      View table

      Table 4. Performance comparison of proposed algorithm before and after improvement

      AlgorithmProbability distributions /%Training time /s
      Accumulate within 1 mAccumulate within 2 m
      CAE-SOM62.7185.3610.42
      OCAE-SOM70.5496.111.86
    • Table 5. Average location response time of different algorithms

      View table

      Table 5. Average location response time of different algorithms

      AlgorithmCSI-improved WKNNsmoothing+SVMCSI-XGBoostOCAE-SOM
      Average location response time /ms10.2498.9358.0276.914
    Tools

    Get Citation

    Copy Citation Text

    Xinchun Li, Xiaolu Ji, Wu Wei, Liyan Wang, Yongyan Gu, Dayan Cao. Research on Fingerprint Location Algorithm Based on OCAE-SOM[J]. Laser & Optoelectronics Progress, 2021, 58(8): 0812001

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Instrumentation, Measurement and Metrology

    Received: Aug. 4, 2020

    Accepted: Sep. 9, 2020

    Published Online: Apr. 16, 2021

    The Author Email: Xiaolu Ji (3078929795@qq.com)

    DOI:10.3788/LOP202158.0812001

    Topics