Dynamic Adjustment of Weighted GCC-PHAT for Position Estimation in an Ultrasonic Local Positioning System

Ultrasonic local positioning systems (ULPS) have been brought to the attention of researchers as one of the possibilities that can be used for indoor localization. Acoustic systems combine a suitable trade-off between precision, ease of development, and cost. This work proposes a method for measurin...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: José Manuel Villadangos, Jesús Ureña, Juan Jesús García-Domínguez, Ana Jiménez-Martín, Álvaro Hernández, Mª Carmen Pérez-Rubio
Formato: article
Lenguaje:EN
Publicado: MDPI AG 2021
Materias:
Acceso en línea:https://doaj.org/article/d943a6f2d46b484cb0d80c8d118307e5
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Descripción
Sumario:Ultrasonic local positioning systems (ULPS) have been brought to the attention of researchers as one of the possibilities that can be used for indoor localization. Acoustic systems combine a suitable trade-off between precision, ease of development, and cost. This work proposes a method for measuring the time of arrival of encoded emissions from a set of ultrasonic beacons, which are used to implement an accurate ULPS. This method uses the generalized cross-correlation technique with PHAT filter and weighting factor <i>β</i> (GCC-PHAT-<i>β</i>). To improve the performance of the GCC-PHAT-<i>β</i> in encoded emission detection, the employment is proposed of mixed-medium multiple-access techniques, based on code division and time division multiplexing of beacon emissions (CDMA and TDMA respectively), and to dynamically adjust the PHAT filter weighting factor. The receiver position is obtained by hyperbolic multilateration from the time differences of arrival (TDoA) between a reference beacon and the rest, thus avoiding the need for receiver synchronization. The results show how the dynamic adaptation of the weighting factor significantly reduces positioning errors from 20 cm to 2 cm in 80% of measurements. The simulated and real experiments prove that the proposed algorithms improve the performance of the ULPS in situations with lower signal-to-noise ratios (SNR) than 0 dB and in environments where the multipath effect makes it difficult to correctly detect the encoded ultrasonic emissions.