Limits...
Target localization in wireless sensor networks using online semi-supervised support vector regression.

Yoo J, Kim HJ - Sensors (Basel) (2015)

Bottom Line: Machine learning has been successfully used for target localization in wireless sensor networks (WSNs) due to its accurate and robust estimation against highly nonlinear and noisy sensor measurement.Second, with an extension to online learning, the proposed OSS-SVR automatically tracks changes of the system to be learned, such as varied noise characteristics.The algorithms are evaluated for estimating the unknown location of a mobile robot in a WSN.

View Article: PubMed Central - PubMed

Affiliation: Department of Mechanical and Aerospace Engineering, Seoul National University, 599 Gwanangno, Gwanak-gu, Seoul KS013, Korea. yjh5455@snu.ac.kr.

ABSTRACT
Machine learning has been successfully used for target localization in wireless sensor networks (WSNs) due to its accurate and robust estimation against highly nonlinear and noisy sensor measurement. For efficient and adaptive learning, this paper introduces online semi-supervised support vector regression (OSS-SVR). The first advantage of the proposed algorithm is that, based on semi-supervised learning framework, it can reduce the requirement on the amount of the labeled training data, maintaining accurate estimation. Second, with an extension to online learning, the proposed OSS-SVR automatically tracks changes of the system to be learned, such as varied noise characteristics. We compare the proposed algorithm with semi-supervised manifold learning, an online Gaussian process and online semi-supervised colocalization. The algorithms are evaluated for estimating the unknown location of a mobile robot in a WSN. The experimental results show that the proposed algorithm is more accurate under the smaller amount of labeled training data and is robust to varying noise. Moreover, the suggested algorithm performs fast computation, maintaining the best localization performance in comparison with the other methods.

No MeSH data available.


Related in: MedlinePlus

Comparison of Kalman filter-based localization of SSML, GP and our algorithm.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4507659&req=5

f6-sensors-15-12539: Comparison of Kalman filter-based localization of SSML, GP and our algorithm.

Mentions: We compare the results of the Kalman filter-based localization of SSML, GP, SSC and our algorithm. The experimental trajectory is same as Figure 5a. In this scenario, we use 100 initial labeled training data points and 50 initial unlabeled data points. As shown in Figure 6, all of the Kalman filter-based localization results provide better accuracy than the basic fingerprinting methods in Figure 5b. Furthermore, in Figure 6, the suggested algorithm gives the greatest accuracy when the Kalman filter is combined.


Target localization in wireless sensor networks using online semi-supervised support vector regression.

Yoo J, Kim HJ - Sensors (Basel) (2015)

Comparison of Kalman filter-based localization of SSML, GP and our algorithm.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4507659&req=5

f6-sensors-15-12539: Comparison of Kalman filter-based localization of SSML, GP and our algorithm.
Mentions: We compare the results of the Kalman filter-based localization of SSML, GP, SSC and our algorithm. The experimental trajectory is same as Figure 5a. In this scenario, we use 100 initial labeled training data points and 50 initial unlabeled data points. As shown in Figure 6, all of the Kalman filter-based localization results provide better accuracy than the basic fingerprinting methods in Figure 5b. Furthermore, in Figure 6, the suggested algorithm gives the greatest accuracy when the Kalman filter is combined.

Bottom Line: Machine learning has been successfully used for target localization in wireless sensor networks (WSNs) due to its accurate and robust estimation against highly nonlinear and noisy sensor measurement.Second, with an extension to online learning, the proposed OSS-SVR automatically tracks changes of the system to be learned, such as varied noise characteristics.The algorithms are evaluated for estimating the unknown location of a mobile robot in a WSN.

View Article: PubMed Central - PubMed

Affiliation: Department of Mechanical and Aerospace Engineering, Seoul National University, 599 Gwanangno, Gwanak-gu, Seoul KS013, Korea. yjh5455@snu.ac.kr.

ABSTRACT
Machine learning has been successfully used for target localization in wireless sensor networks (WSNs) due to its accurate and robust estimation against highly nonlinear and noisy sensor measurement. For efficient and adaptive learning, this paper introduces online semi-supervised support vector regression (OSS-SVR). The first advantage of the proposed algorithm is that, based on semi-supervised learning framework, it can reduce the requirement on the amount of the labeled training data, maintaining accurate estimation. Second, with an extension to online learning, the proposed OSS-SVR automatically tracks changes of the system to be learned, such as varied noise characteristics. We compare the proposed algorithm with semi-supervised manifold learning, an online Gaussian process and online semi-supervised colocalization. The algorithms are evaluated for estimating the unknown location of a mobile robot in a WSN. The experimental results show that the proposed algorithm is more accurate under the smaller amount of labeled training data and is robust to varying noise. Moreover, the suggested algorithm performs fast computation, maintaining the best localization performance in comparison with the other methods.

No MeSH data available.


Related in: MedlinePlus