1D-CLANet: A Novel Network for NLoS Classification in UWB Indoor Positioning System
Abstract
:1. Introduction
- (1)
- A novel network that integrates CNN, LSTM, and SE attention mechanisms is proposed. This architecture efficiently combines the deep feature extraction capability of 1D-CNN, the time series processing strength of LSTM, and the SE attention mechanism’s focus on critical features, significantly enhancing NLoS classification accuracy for UWB indoor positioning.
- (2)
- By leveraging 1D-CNN and other lightweight design considerations, the proposed method achieves a balance between high classification accuracy and low processing time. The integration of LSTM spatiotemporal features with the SE attention mechanism enables the model to effectively distinguish challenging features that other methods often confuse, such as LoS and human obstructions.
- (3)
- We validated the efficacy of our proposed method on both publicly available datasets and multiclass datasets collected from experiments. The experimental outcomes indicate that our approach not only significantly improves NLoS binary classification performance but also shows excellent performance in multiclass recognition of common NLoS obstructions in indoor environments.
2. Related Work
3. Theoretical Analysis
4. Proposed Method
4.1. Convolutional Neural Network
4.2. Long Short-Term Memory Cell
4.3. Squeeze-and-Excitation Block
4.4. Feature Fusion and Classification Module
5. Experiments and Analysis
5.1. Experimental Setting Up and Dataset Description
5.2. Results and Analysis
6. Conclusions
Author Contributions
Funding
Data Availability Statement
Acknowledgments
Conflicts of Interest
Abbreviations
Abbreviation | Full Name of Abbreviation |
UWB | Ultra-Wideband |
NLoS/LoS | Non-Line-of-Sight/Line-of-Sight |
1D-CLANet | 1D-ConvLSTM-Attention network |
CIR | Channel impulse response |
CNN | Convolutional neural network |
LSTM | Long Short-Term memory |
SE | Squeeze-and-Excitation |
IoT | Internet of things |
IPS | Indoor positioning systems |
TX | Transmitter |
RX | Receiver |
ToF | Time of flight |
ML | Machine learning |
DL | Deep learning |
SVMs | Support vector machines |
RNNs | Recurrent neural networks |
TOA | Time of arrival |
TDOA | Time difference in arrival |
AOA | Angle of arrival |
RSSI | Received signal strength indicator |
MKF | Map-corrected extended Kalman filter |
GD | Gaussian distribution |
GGD | Generalized Gaussian distribution |
MLP | Multilayer perceptron |
HQCNN | Hybrid quantum CNN |
BiLSTM | Bi-directional LSTM |
ECA | Efficient channel attention |
GRU | Gated recurrent units |
AWGN | Additive white Gaussian noise |
FPH | First path |
BN | Batch normalization |
CA | Coordinate attention |
GAP | Global average pooling |
FC | Fully connected |
TP/TN | True positive/True negative |
FP/FN | False positive/False negative |
CNSM | CNN-LSTM |
ROC | Receiver operating characteristic |
AUC | Area under the curve |
FPSD | First path strength difference |
FPDD | First path distance difference |
NPP | Number of pseudo peaks |
References
- Ngamakeur, K.; Yongchareon, S.; Yu, J.; Rehman, S.U. A survey on device-free indoor localization and tracking in the multi-resident environment. ACM Comput. Surv. CSUR 2020, 53, 71. [Google Scholar] [CrossRef]
- Nascita, A.; Montieri, A.; Aceto, G.; Ciuonzo, D.; Persico, V.; Pescapé, A. Xai meets mobile traffic classification: Understanding and improving multimodal deep learning architectures. IEEE Trans. Netw. Serv. 2021, 18, 4225–4246. [Google Scholar] [CrossRef]
- Xiao, Y.; Zhu, J.; Yan, S.; Song, H.; Zhang, S. PEiD: Precise and Real-Time LOS/NLOS Path Identification Based on Peak Energy Index Distribution. Appl. Sci. 2023, 13, 7458. [Google Scholar] [CrossRef]
- Martalo, M.; Perri, S.; Verdano, G.; De Mola, F.; Monica, F.; Ferrari, G. Improved UWB TDoA-based positioning using a single hotspot for industrial IoT applications. IEEE Trans. Industr. Inform. 2022, 18, 3915–3925. [Google Scholar] [CrossRef]
- Yu, K.; Wen, K.; Li, Y.; Zhang, S.; Zhang, K. A novel nlos mitigation algorithm for UWB localization in harsh indoor environments. IEEE Trans. Veh. Technol. 2019, 68, 686–699. [Google Scholar] [CrossRef]
- Zhao, Y.; Wang, M. The LOS/NLOS classification method based on deep learning for the UWB localization system in coal mines. Appl. Sci. 2022, 12, 6484. [Google Scholar] [CrossRef]
- Yang, X.; Wang, J.; Song, D.; Feng, B.; Ye, H. A novel NLOS error compensation method based IMU for UWB indoor positioning system. IEEE Sens. J. 2021, 21, 11203–11212. [Google Scholar] [CrossRef]
- Csík, D.; Sarcevic, P.; Pesti, R.; Odry, Á. Comparison of different radio communication-based technologies for indoor localization using trilateration. In Proceedings of the 2023 IEEE 17th International Symposium on Applied Computational Intelligence and Informatics (SACI), Timisoara, Romania, 23–26 May 2023; pp. 487–492. [Google Scholar]
- Yang, H.; Wang, Y.; Seow, C.K.; Sun, M.; Si, M.; Huang, L. UWB sensor-based indoor LoS/NLoS localization with support vector machine learning. IEEE Sens. J. 2023, 23, 2988–3004. [Google Scholar] [CrossRef]
- Djosic, S.; Stojanovic, I.; Jovanovic, M.; Djordjevic, G.L. Multi-algorithm UWB-based localization method for mixed LOS/NLOS environments. Comput. Commun. 2022, 181, 365–373. [Google Scholar] [CrossRef]
- Yao, L.; Yao, L.; Wu, Y. Analysis and improvement of indoor positioning accuracy for uwb sensors. Sensors 2021, 21, 5731. [Google Scholar] [CrossRef]
- Silva, B.; Hancke, G.P. Ir-uwb-based non-line-of-sight identification in harsh environments: Principles and challenges. IEEE Trans. Industr. Inform. 2016, 12, 1188–1195. [Google Scholar] [CrossRef]
- Sang, C.L.; Steinhagen, B.; Homburg, J.D.; Adams, M.; Hesse, M.; Rückert, U. Identification of NLOS and multi-path conditions in UWB localization using machine learning methods. Appl. Sci. 2020, 10, 3980. [Google Scholar] [CrossRef]
- Hajiakhondi-Meybodi, Z.; Mohammadi, A.; Hou, M.; Plataniotis, K.N. DQLEL: Deep Q-learning for energy-optimized LoS/NLoS UWB node selection. IEEE Trans. Signal Process. 2022, 70, 2532–2547. [Google Scholar] [CrossRef]
- Kim, D.H.; Farhad, A.; Pyun, J.Y. UWB positioning system based on LSTM classification with mitigated NLOS effects. IEEE Internet Things J. 2022, 10, 1822–1835. [Google Scholar] [CrossRef]
- Wang, Q.; Chen, M.; Wang, G.; Li, K.; Lin, Y.; Li, Z.; Zhang, C. A novel nlos identification and error mitigation method for uwb ranging and positioning. IEEE Commun. Lett. 2024, 28, 48–52. [Google Scholar] [CrossRef]
- Liu, Q.; Yin, Z.; Zhao, Y.; Wu, Z.; Wu, M. UWB LOS/NLOS identification in multiple indoor environments using deep learning methods. Phys. Commun. 2022, 52, 101695. [Google Scholar] [CrossRef]
- Vaghefi, R.M.; Gholami, M.R.; Buehrer, R.M.; Ström, E.G. Cooperative Received Signal Strength-Based Sensor Localization with Unknown Transmit Powers. IEEE Trans. Signal Process. 2013, 61, 1389–1403. [Google Scholar] [CrossRef]
- Zhang, Q.; Cheng, X.; Wang, K.; Cao, Z.; Hong, Y. NLOS Error Suppression Method based on UWB Indoor Positioning. In Proceedings of the 2023 IEEE International Conference on Mechatronics and Automation (ICMA), Harbin, China, 6–9 August 2023; pp. 1125–1130. [Google Scholar]
- Shi, Z.; Wang, J.; Zeng, X.; Yang, H. An improved positioning method based on compensation and optimization of ultra-wideband ranging results. Meas. Sci. Technol. 2024, 35, 086305. [Google Scholar] [CrossRef]
- Liu, J.; Zhang, L.; Xu, J.; Shi, J. Dynamic Feasible Region Based IMU/UWB Fusion Method for Indoor Positioning. IEEE Sens. J. 2024, 24, 21447–21457. [Google Scholar] [CrossRef]
- Zhu, X.; Yi, J.; Cheng, J.; He, L. Adapted error map based mobile robot UWB indoor positioning. IEEE Trans. Instrum. Meas. 2020, 69, 6336–6350. [Google Scholar] [CrossRef]
- Wang, Q.; Li, Z.; Zhang, H.; Yang, Y.; Meng, X. An Indoor UWB NLOS Correction Positioning Method Based on Anchor LOS/NLOS Map. IEEE Sens. J. 2023, 23, 30739–30750. [Google Scholar] [CrossRef]
- Ferreira, A.G.; Fernandes, D.; Branco, S.; Catarino, A.P.; Monteiro, J.L. Feature selection for real-time NLOS identification and mitigation for body-mounted UWB transceivers. IEEE Trans. Instrum. Meas. 2021, 70, 1–10. [Google Scholar] [CrossRef]
- Wang, F.; Tang, H.; Chen, J. Survey on NLOS identification and error mitigation for UWB indoor positioning. Electronics 2023, 12, 1678. [Google Scholar] [CrossRef]
- Wymeersch, H.; Maranò, S.; Gifford, W.M.; Win, M.Z. A machine learning approach to ranging error mitigation for UWB localization. IEEE Trans. Commun. 2012, 60, 1719–1728. [Google Scholar] [CrossRef]
- Che, F.; Ahmed, Q.Z.; Fontainc, J.; Van Herbruggen, B.; Shahid, A.; De Poorter, E.; Lazaridis, P.I. Feature-based generalized gaussian distribution method for nlos detection in ultra-wideband (uwb) indoor positioning system. IEEE Sens. J. 2022, 22, 18726–18739. [Google Scholar] [CrossRef]
- Yang, H.; Wang, Y.; Xu, S.; Bi, J.; Jia, H.; Seow, C.K. UWB ranging errors mitigation with novel CIR feature parameters and two-step NLOS identification. Sensors 2024, 24, 1703. [Google Scholar] [CrossRef] [PubMed]
- Si, M.; Wang, Y.; Siljak, H.; Seow, C.; Yang, H. A lightweight CIR-based CNN with MLP for nlos/los identification in a UWB positioning system. IEEE Commun. Lett. 2023, 27, 1332–1336. [Google Scholar] [CrossRef]
- Jiang, C.; Chen, S.; Chen, Y.; Liu, D.; Bo, Y. An UWB Channel Impulse Response De-Noising Method for NLOS/LOS Classification Boosting. IEEE Commun. Lett. 2020, 24, 2513–2517. [Google Scholar] [CrossRef]
- Miramá, V.; Bahillo, A.; Quintero, V.; Díez, L.E. NLOS detection generated by body shadowing in a 6.5 GHz UWB localization system using machine learning. IEEE Sens. J. 2023, 23, 20400–20411. [Google Scholar] [CrossRef]
- Jeong, S.-G.; Do, Q.-V.; Hwang, H.-J.; Hasegawa, M.; Sekiya, H.; Hwang, W.-J. UWB NLOS/LOS Classification Using Hybrid Quantum Convolutional Neural Networks. In Proceedings of the 2023 IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia), Busan, Republic of Korea, 23–25 October 2023; pp. 1–2. [Google Scholar]
- Yang, Y.; Ke, H.; Gan, W.; Deng, Z. CNN-BiLSTM-ATTENTION: A Novel Neural Network with Attention Mechanism for NLOS Identification of UWB Signal. In Proceedings of the 2023 3rd International Conference on Intelligent Communications and Computing (ICC), Nanchang, China, 24–26 November 2023; pp. 279–283. [Google Scholar]
- Niu, Z.; Yang, H.; Zhou, L.; Taha, M.F.; He, Y.; Qiu, Z. Deep learning-based ranging error mitigation method for UWB localization system in greenhouse. Comput. Electron. Agric. 2023, 205, 107573. [Google Scholar] [CrossRef]
- Tian, Y.; Lian, Z.; Amparo Núñez-Andrés, M.; Yue, Z.; Li, K.; Wang, P.; Wang, M. The application of gated recurrent unit algorithm with fused attention mechanism in UWB indoor localization. Measurement 2024, 234, 114835. [Google Scholar] [CrossRef]
- Wei, J.; Wang, H.; Su, S.; Tang, Y.; Guo, X.; Sun, X. NLOS identification using parallel deep learning model and time-frequency information in UWB-based positioning system. Measurement 2022, 195, 111191. [Google Scholar] [CrossRef]
- Hochreiter, S.; Schmidhuber, J. Long short-term memory. Neural Comput. 1997, 9, 1735–1780. [Google Scholar] [CrossRef] [PubMed]
- Smits, J.R.M.; Melssen, W.J.; Buydens, L.M.C.; Kateman, G. Using artificial neural networks for solving chemical problems: Part I. Multi-layer feed-forward networks. Chemometr. Intell. Lab. Syst. 1994, 22, 165–189. [Google Scholar] [CrossRef]
- Jin, X.; Xie, Y.; Wei, X.-S.; Zhao, B.-R.; Chen, Z.-M.; Tan, X. Delving deep into spatial pooling for squecze-and-excitation networks. Pattern Recognit. 2022, 121, 108159. [Google Scholar] [CrossRef]
- Zhao, H.; Jia, J.; Koltun, V. Exploring self-attention for image recognition. In Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 13–19 June 2020; pp. 10076–10085. [Google Scholar]
- Hou, Q.; Zhou, D.; Feng, J. Coordinate attention for efficient mobile network design. In Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA, 20–25 June 2021; pp. 13713–13722. [Google Scholar]
- Bregar, K.; Andrej, H.; Mohorcic, M. NLoS channel detection with multilayer perceptron in low-rate personal area networks for indoor localization accuracy improvement. In Proceedings of the 8th Jožef Stefan International Postgraduate School Student Conference, Ljubljana, Slovenia, 1 June 2016; Voume 31, pp. 130–139. [Google Scholar]
Notation | Definition |
---|---|
The distance from the tag to i-th anchor | |
The UWB measurements under LoS/NLoS | |
The true distance between tag and anchor | |
The UWB measurement noise | |
The transmitted symbol | |
The amplitude of the transmitted pulse | |
The single Gaussian pulse waveform | |
The number of pulses | |
The period of each pulse | |
The fading coefficient of the c-th path | |
The time delay of the c-th path | |
Additive white Gaussian noise with variance | |
Input CIR signal | |
The length of input CIR signal | |
The output feature map after convolution operation | |
The spatial dimensions of feature map | |
The set of learned filter kernels | |
The learnable parameters of BN operation | |
A small constant added to the variance | |
The outputs after the batch normalization operation | |
The forget gate’s activation vector | |
The input/update gate’s activation vector | |
Cell input activation vector | |
Current cell memory | |
Previous cell memory | |
The updated cell state | |
The output gate’s activation vector | |
Current cell output | |
tanh | The tanh activation function |
The previous cell output | |
The bias vector | |
The weight matrices | |
Weights of the FC layers | |
Biases of the FC layers | |
The ReLU function | |
The SE attention excitation output of the c-th channel | |
The per-channel multiplication output | |
The final output of the SE attention block | |
Deep spatiotemporal features extracted by LSTM | |
Fused feature vector | |
The is-th/js-th input elements for Softmax function | |
The sigmoid function | |
The balancing factor | |
The focusing parameter | |
The predicted probability for class iy | |
The LSTM input | |
The five types of CIR signal propagation channel | |
Energy of CIR | |
Maximum amplitude of CIR | |
Mean excess delay of CIR | |
The Root-mean-square delay spread of CIR | |
CIR kurtosis | |
UWB ToF estimated distance |
Category | Advantage | Shortcoming | Related Methods | |
---|---|---|---|---|
Traditional NLoS identification methods | Distance-based methods | The identification principle is simple, and the computational efficiency is high | The identification accuracy is easily affected by fluctuations in distance information | Kurtosis [20] SVM [21] |
Prior map-based methods | The NLoS identification performs well in static, well-mapped, known indoor environments | It requires extensive data and cannot identify dynamic obstacles causing NLoS | MKF [23] | |
Channel feature-based NLoS identification methods | Machine learning-based methods | It offers higher accuracy than traditional methods and greater efficiency than deep learning methods | Collecting a representative UWB channel feature dataset is challenging, and ML models often struggle with generalization in NLoS identification, leading to reduced accuracy | SVM [24,26] GD [25] GGD [27] Two-Step [28] |
Deep learning-based methods | NLoS deep feature extraction and scene recognition excel, with no need for manual feature extraction and strong robustness | Research on NLoS multi-class classification is limited, and complex models with attention mechanisms require costly hardware and may lack efficiency | MLP [29] CNN [30] LSTM [31] HQCNN [32] BiLSTM-CNN [33] ECA-ResNet [34] GRU [35] |
Module | Network Layer | Kernel Size | Output Size |
---|---|---|---|
Feature Extractor | Conv1 | 16 × 3 | 16 × 512 |
BN1 + ReLU + Pool1 | 1 × 2 | 16 × 256 | |
Conv2 | 32 × 3 | 32 × 256 | |
BN2 + ReLU + Pool2 | 1 × 2 | 32 × 128 | |
Conv3 | 64 × 3 | 64 × 256 | |
BN3 + ReLU + Pool3 | 1 × 2 | 64 × 64 | |
Flatten1 | / | 1 × 4096 | |
LSTM | 512 | Sequence | |
SE | 512, 4 | / | |
Concat | / | / | |
Label predictor | Flatten2 | / | / |
FC1 + ReLU | 128 | / | |
Dropout | 0.5 | / | |
FC2 | 5 | / | |
Output | 1 × 5 | / |
Hyper-Parameters | Set Value |
---|---|
Batch size | 32 |
Training times | 20 |
Optimizer | Adam |
Learning rate | 0.0001 |
Loss function | Focal loss |
Parameter | Configuration Value |
---|---|
Central frequency | 6489.6 MHz |
Bandwidth | 499.2 MHz |
Pulse repetition frequency | 64 MHz |
Channel | 5 |
Inter-ranging delay | 200 ms |
Method | Feature or Data Used |
---|---|
Kurtosis [20] | CIR feature |
SVM [26] | CIR feature , , , , , and |
CNN [30] | CIR signal |
ResNet (without ECA) [34] | CIR signal |
CNSM | CIR signal |
1D-CLANet | CIR signal |
Method | Accuracy | Precision | Recall | F1 |
---|---|---|---|---|
Kurtosis | 70.80% | 70.87% | 70.84% | 70.86% |
SVM | 79.66% | 81.15% | 79.54% | 80.28% |
CNN | 82.97% | 83.92% | 82.88% | 83.35% |
ResNet | 85.80% | 86.55% | 85.72% | 86.09% |
CNSM | 91.03% | 91.09% | 91.00% | 91.04% |
1D-CLANet | 96.89% | 96.92% | 96.88% | 96.89% |
Method | Feature or Data Used |
---|---|
LSTM | CIR signal |
SVM [26] | CIR feature , , , , , and |
HQCNN [32] | CIR signal |
MLP [29] | CIR signal andCIR feature FPSD, FPDD, NPP, , and |
CNSM | CIR signal |
1D-CLANet | CIR signal |
Model | 1D-CLANet without Attention | 1D-CLANet with Self | 1D-CLANet with CA | 1D-CLANet with SE |
---|---|---|---|---|
Structure |
Feature Extractor | Parameters/M | Accuracy/% | Time/s |
---|---|---|---|
1D-CLANet without | 1.2 | 89.82 | 85.43 |
1D-CLANet with Self | 2.3 | 93.37 | 143.79 |
1D-CLANet with CA | 1.5 | 94.91 | 120.82 |
1D-CLANet with SE | 1.4 | 95.58 | 100.31 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Wang, Q.; Chen, M.; Liu, J.; Lin, Y.; Li, K.; Yan, X.; Zhang, C. 1D-CLANet: A Novel Network for NLoS Classification in UWB Indoor Positioning System. Appl. Sci. 2024, 14, 7609. https://doi.org/10.3390/app14177609
Wang Q, Chen M, Liu J, Lin Y, Li K, Yan X, Zhang C. 1D-CLANet: A Novel Network for NLoS Classification in UWB Indoor Positioning System. Applied Sciences. 2024; 14(17):7609. https://doi.org/10.3390/app14177609
Chicago/Turabian StyleWang, Qiu, Mingsong Chen, Jiajie Liu, Yongcheng Lin, Kai Li, Xin Yan, and Chizhou Zhang. 2024. "1D-CLANet: A Novel Network for NLoS Classification in UWB Indoor Positioning System" Applied Sciences 14, no. 17: 7609. https://doi.org/10.3390/app14177609