FPGA-based SoC architecture for real-time device-free localization using a multistatic US sonar

Device-Free Localization (DFL) defines a novel approach to indoor positioning systems, in which individuals do not need to carry any tags or devices to be located. Whereas most DFL systems rely on radio frequency signals, acoustic-based solutions present an interesting alternative due to their high...

Descripción completa

Detalles Bibliográficos
Autores: García Requejo, Alejandro|||0000-0002-4955-8852, Hernández Alonso, Álvaro|||0000-0001-9308-8133, Pérez Rubio, María del Carmen|||0000-0001-8271-6843
Tipo de recurso: artículo
Fecha de publicación:2025
País:España
Institución:Universidad de Alcalá (UAH)
Repositorio:e_Buah Biblioteca Digital Universidad de Alcalá
Idioma:inglés
OAI Identifier:oai:ebuah.uah.es:10017/67561
Acceso en línea:http://hdl.handle.net/10017/67561
https://dx.doi.org/10.1016/j.measurement.2025.118320
Access Level:acceso abierto
Palabra clave:Device-free localization (DFL)
Real-time localization (RTL)
Ultrasound (US)Field-programmable gate array (FPGA)
System-on-chip (SoC)
Multistatic sonar
Electrónica
Electronics
Descripción
Sumario:Device-Free Localization (DFL) defines a novel approach to indoor positioning systems, in which individuals do not need to carry any tags or devices to be located. Whereas most DFL systems rely on radio frequency signals, acoustic-based solutions present an interesting alternative due to their high accuracy, immunity to electromagnetic interferences, low cost and signal confinement. They also facilitate real-time position estimation, which is essential for localization systems, enabling users to make prompt decisions. For this purpose, this work proposes a real-time DFL system based on acoustic signals at ultrasound frequencies. The system utilizes a ceiling-mounted multistatic sonar configuration to achieve centimetric-level positioning of a person's head within a defined coverage area. This sonar setup consists of one receiver and four emitters, which sequentially emit orthogonal chirp signals. This approach increases the localization update rate and supports the integration of multiple sonar units within a single room. The received signals are acquired through a dedicated analog front-end and processed on a FPGA-based System-on-Chip (SoC) architecture. The combination of programmable logic and a general-purpose processor enables parallel processing of signals from multiple emitters and accelerates the execution of the proposed localization algorithm. Additionally, the platform includes a Wi-Fi module for broadcasting the estimated position to a remote user application. The proposed system has been experimentally validated with the sonar mounted at a height of 2.65 m. Results show localization errors below 13 cm for static points and 10 cm for trajectories in 90% of cases.