Enhanced Gaussian mixture model of RSSI purification for indoor positioning

Tseng, Chinyang Henry; Yen, Jing-Shyang

doi:10.1016/j.sysarc.2017.10.003

Cited by 20 publications

(14 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The GMM is applied to model the probability distribution of the signal strength for each AP, assuming that the APs are independent at a particular position [ 40 ]. GMM is used to identify the RSS components of multipath decline separated from the line-of-sight (LOS) component in [ 41 ]. Similar work can be found in [ 42 ] where a two-node GMM is used to detect and exclude the outliers, one node for the direct path and the other one for the outliers.…”

Section: Related Workmentioning

confidence: 99%

High-Dimensional Probabilistic Fingerprinting in Wireless Sensor Networks Based on a Multivariate Gaussian Mixture Model

Williams

Moran

et al. 2018

Sensors

View full text Add to dashboard Cite

The extensive deployment of wireless infrastructure provides a low-cost way to track mobile users in indoor environment. This paper demonstrates a prototype model of an accurate and reliable room location awareness system in a real public environment in which three typical problems arise. Firstly, a massive number of access points (APs) can be sensed leading to a high-dimensional classification problem. Secondly, heterogeneous devices record different received signal strength (RSS) levels because of the variations in chip-set and antenna attenuation. Thirdly, APs are not necessarily visible in every scanning cycle leading to missing data issue. This paper presents a probabilistic Wi-Fi fingerprinting method in a hidden Markov model (HMM) framework for mobile user tracking. To account for spatial correlation of the signal strengths from multiple APs, a Multivariate Gaussian Mixture Model (MVGMM) was fitted to model the probability distribution of RSS measurements in each cell. Furthermore, the unseen property of invisible AP was investigated in this research, and demonstrated the efficiency as a beneficial information to differentiate between cells. The proposed system is able to achieve comparable localisation performance. Filed test results achieve a reliable 97% localisation room level accuracy of multiple mobile users in a real university campus Wi-Fi network.

show abstract

Section: Related Workmentioning

confidence: 99%

High-Dimensional Probabilistic Fingerprinting in Wireless Sensor Networks Based on a Multivariate Gaussian Mixture Model

Williams

Moran

et al. 2018

Sensors

View full text Add to dashboard Cite

show abstract

“…In [20], a Wi-Fi based indoor localization method is proposed to adopt assistant nodes with similar RSS sequences as auxiliary nodes to implement the accurate positioning in the complex indoor environment. In [21]- [23], fingerprints can be constructed based on the Voronoi diagram according to the signal propagation model [24], [25] and the signal attenuation parameter [26]- [28]. In a word, these methods could be used to decrease the reliance on the accuracy of the established RPF database.…”

Section: A the Establishment/maintenance Of Rpf Databasementioning

confidence: 99%

A New RSS Fingerprinting-Based Location Discovery Method Under Sparse Reference Point Conditions

Yang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

With the increasing demand for indoor location-based services, the received signal strength fingerprinting-based localization algorithm has become a research focus due to its accuracy and low hardware requirements. However, how to achieve the accurate location discovery relies solely on the received signal strength under the sparse reference points condition, which is the main contribution of this paper. First, the Voronoi diagram is adopted to regionalize the positioning area and form a distributed signal propagation description, which can reduce the influence of environment interference. Second, aiming at the local motion tracking problem, a region-based location search model is constructed to achieve the initial position estimation and provide the motion model for the following optimization of location estimation. Third, in order to reduce the cumulative error caused by the environmental noise and the local optimum problem, the regularized particle filtering algorithm with map-correction is employed to implement the dynamic calibration of the particle updating equation. To verify the proposed algorithm, an indoor wireless experiment system is finally designed in this paper. The experiment results indicate that the proposed algorithm can increase the positioning accuracy by 28.2% compared with the fingerprinting-based localization algorithm when the RPs density is reduced to 0.2/ (0.5m*0.5m). INDEX TERMS Batch gradient descent, indoor positioning, regularized particle filtering, sparse reference points condition, Voronoi diagram.

show abstract

“…One is based on distance, such as received signal strength indicator (RSSI), angle of arrival (AOA), and time difference of arrival (TDOA). The characteristic of these algorithms is that the positioning accuracy is very high, but the peripheral equipment needs to be provided, and the cost is high [8][9][10][11][12]. Another is the range-free location algorithm, such as DV-Hop, centroid location algorithm.…”

Section: Introductionmentioning

confidence: 99%

An improved range-free location algorithm for industrial wireless sensor networks

Wang

Liu

et al. 2020

J Wireless Com Network

View full text Add to dashboard Cite

In the traditional DV-Hop location algorithm, the number of hops within the communication radius is counted as one hop. However, in the case of uneven distribution of nodes, this method will bring a large error to the calculation of hop distance. In order to solve the above problems, an improved DV-Hop location algorithm with multiple communication radius is proposed in this paper. Based on the multi-communication radius location algorithm, the method uses the cosine theorem to correct the hop number and estimated the average hop distance, and optimizes the estimated distance of the unknown node. The algorithm uses multiple communication radius to broadcast the position to obtain the minimum number of hops between the unknown node and the beacon node and adjusts the estimated hop distance by using the cosine theorem after estimating the good hop distance, and finally uses the trilateration method, or the maximum likelihood estimation method estimates the position coordinates. The experimental results show that compared with the traditional DV-Hop algorithm and DDV-Hop algorithm, the improved DV-Hop algorithm improves the positioning accuracy and reduces the average positioning error of the unknown nodes.

show abstract

Enhanced Gaussian mixture model of RSSI purification for indoor positioning

Cited by 20 publications

References 11 publications

High-Dimensional Probabilistic Fingerprinting in Wireless Sensor Networks Based on a Multivariate Gaussian Mixture Model

High-Dimensional Probabilistic Fingerprinting in Wireless Sensor Networks Based on a Multivariate Gaussian Mixture Model

A New RSS Fingerprinting-Based Location Discovery Method Under Sparse Reference Point Conditions

An improved range-free location algorithm for industrial wireless sensor networks

Contact Info

Product

Resources

About