Limits...
Entropy-Based TOA Estimation and SVM-Based Ranging Error Mitigation in UWB Ranging Systems.

Yin Z, Cui K, Wu Z, Yin L - Sensors (Basel) (2015)

Bottom Line: The major challenges for Ultra-wide Band (UWB) indoor ranging systems are the dense multipath and non-line-of-sight (NLOS) problems of the indoor environment.The entropy is used to measure the randomness of the received signals and the FP can be determined by the decision of the sample which is followed by a great entropy decrease.The presented numerical simulation results show that the proposed approach achieves significant performance improvements in the CM1 to CM4 channels of the IEEE 802.15.4a standard, as compared to conventional approaches.

View Article: PubMed Central - PubMed

Affiliation: School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin 150001, China. yinzhendong@hit.edu.cn.

ABSTRACT
The major challenges for Ultra-wide Band (UWB) indoor ranging systems are the dense multipath and non-line-of-sight (NLOS) problems of the indoor environment. To precisely estimate the time of arrival (TOA) of the first path (FP) in such a poor environment, a novel approach of entropy-based TOA estimation and support vector machine (SVM) regression-based ranging error mitigation is proposed in this paper. The proposed method can estimate the TOA precisely by measuring the randomness of the received signals and mitigate the ranging error without the recognition of the channel conditions. The entropy is used to measure the randomness of the received signals and the FP can be determined by the decision of the sample which is followed by a great entropy decrease. The SVM regression is employed to perform the ranging-error mitigation by the modeling of the regressor between the characteristics of received signals and the ranging error. The presented numerical simulation results show that the proposed approach achieves significant performance improvements in the CM1 to CM4 channels of the IEEE 802.15.4a standard, as compared to conventional approaches.

No MeSH data available.


Related in: MedlinePlus

Illustration of conventional TOA approaches.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4481902&req=5

sensors-15-11701-f001: Illustration of conventional TOA approaches.

Mentions: Many approaches have been proposed to address these problems. For accuracy improvements, most of the approaches focus on modifications of the FP detection algorithms, such as maximum energy selection (MES), MES searching-back (MES-SB), coherent detection and some threshold-crossing-based methods [20,21,22,23,24], as shown in Figure 1. Other approaches try to build mathematical models for range information [25], which, however, turns out to be complicated and relies too much on prior information. As for the MES methods, the energy block with the maximal energy is considered as the one containing the FP, but the position of the FP in the energy block cannot be determined, and under NLOS conditions, there is a delay between the energy block that contains the FP and the maximal energy block, which results in a biased estimation. There have been some achievements in the MES-SB method, but the in-block problem remains. In the case of threshold-crossing-based methods, the sample or the energy block which first exceeds the threshold is taken as the FP, so it is very important to select the appropriate threshold. However, in NLOS conditions, there is a delay between the FP and the strongest path (SP), and sometimes the FP is even lower than the noise, so it is hard to detect the FP with conventional threshold-crossing. A number of modified algorithms are proposed to determine the threshold, but many of them are parametric and complex [26,27], which requires prior information or complicated calculations, and the estimation results are still far from optimal. Therefore, a method of which the threshold is not so strictly required and the accuracy is acceptable is worth consideration.


Entropy-Based TOA Estimation and SVM-Based Ranging Error Mitigation in UWB Ranging Systems.

Yin Z, Cui K, Wu Z, Yin L - Sensors (Basel) (2015)

Illustration of conventional TOA approaches.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4481902&req=5

sensors-15-11701-f001: Illustration of conventional TOA approaches.
Mentions: Many approaches have been proposed to address these problems. For accuracy improvements, most of the approaches focus on modifications of the FP detection algorithms, such as maximum energy selection (MES), MES searching-back (MES-SB), coherent detection and some threshold-crossing-based methods [20,21,22,23,24], as shown in Figure 1. Other approaches try to build mathematical models for range information [25], which, however, turns out to be complicated and relies too much on prior information. As for the MES methods, the energy block with the maximal energy is considered as the one containing the FP, but the position of the FP in the energy block cannot be determined, and under NLOS conditions, there is a delay between the energy block that contains the FP and the maximal energy block, which results in a biased estimation. There have been some achievements in the MES-SB method, but the in-block problem remains. In the case of threshold-crossing-based methods, the sample or the energy block which first exceeds the threshold is taken as the FP, so it is very important to select the appropriate threshold. However, in NLOS conditions, there is a delay between the FP and the strongest path (SP), and sometimes the FP is even lower than the noise, so it is hard to detect the FP with conventional threshold-crossing. A number of modified algorithms are proposed to determine the threshold, but many of them are parametric and complex [26,27], which requires prior information or complicated calculations, and the estimation results are still far from optimal. Therefore, a method of which the threshold is not so strictly required and the accuracy is acceptable is worth consideration.

Bottom Line: The major challenges for Ultra-wide Band (UWB) indoor ranging systems are the dense multipath and non-line-of-sight (NLOS) problems of the indoor environment.The entropy is used to measure the randomness of the received signals and the FP can be determined by the decision of the sample which is followed by a great entropy decrease.The presented numerical simulation results show that the proposed approach achieves significant performance improvements in the CM1 to CM4 channels of the IEEE 802.15.4a standard, as compared to conventional approaches.

View Article: PubMed Central - PubMed

Affiliation: School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin 150001, China. yinzhendong@hit.edu.cn.

ABSTRACT
The major challenges for Ultra-wide Band (UWB) indoor ranging systems are the dense multipath and non-line-of-sight (NLOS) problems of the indoor environment. To precisely estimate the time of arrival (TOA) of the first path (FP) in such a poor environment, a novel approach of entropy-based TOA estimation and support vector machine (SVM) regression-based ranging error mitigation is proposed in this paper. The proposed method can estimate the TOA precisely by measuring the randomness of the received signals and mitigate the ranging error without the recognition of the channel conditions. The entropy is used to measure the randomness of the received signals and the FP can be determined by the decision of the sample which is followed by a great entropy decrease. The SVM regression is employed to perform the ranging-error mitigation by the modeling of the regressor between the characteristics of received signals and the ranging error. The presented numerical simulation results show that the proposed approach achieves significant performance improvements in the CM1 to CM4 channels of the IEEE 802.15.4a standard, as compared to conventional approaches.

No MeSH data available.


Related in: MedlinePlus