RSSI Prediction in WiFi Considering Realistic Heterogeneous Restrictions

Abstract

Received Signal Strength Indicator is an indication of power level being received by a wireless device. This parameter for Wireless Fidelity has never been standardized, so any manufacturer implements it measuring as he knows better doing. This is a problem because different scenarios can be found based in several elements like: the particular Wireless Fidelity Network Interface implementation, the driver, the operating system and software monitoring implementation. This makes the utilization of the Received Signal Strength Indicator to implement higher level functions in wireless networks (like roaming management or localization of terminals) be a headache in realistic scenarios. This heterogeneity of devices, drivers… makes that the application of the Received Signal Strength Indicator be not general. In this paper we present the comparison of our new model of Received Signal Strength Indicator prediction in several realistic and heterogeneous scenarios. The experimental results indicate that we obtain in general, better results than the other models.

A Machine-Learning-Based Access Point Selection Strategy for Automated Guided Vehicles in Smart Factories

“…In wireless LAN, information on the RSSI is included in the beacon packets sent by each AP at a 100 ms interval, providing a mechanism to let the terminals know the signal strength before they connect. W. Rehan et al compiled multiple insights [13], but there are several limitations to using RSSI for estimating link quality in environments with interference [14][15][16]. Subsequent studies reported that it had a specific effect as an indicator for detecting the decrease in communication link reliability at an early stage, especially when targeting AGVs [17,18].…”

Posicionamiento en espacios interiores con Android, Bluetooth y RSSI

“…En estos casos los costos suelen ser menores dado que la infraestructura es mucho más ligera. En esta línea es frecuente el uso de radiofrecuencia y se encuentran investigaciones que exploran WiFi (Suárez et al, 2014), ZigBee (Larranaga, Muguira, López-Garde, & Vázquez, 2010), RFID (Krukowski & Arsenijevic, 2010) y Bluetooth (Torstensson, 2016) entre otras. Todas las técnicas presentan ventajas y desventajas (Mainetti et al, 2014) que, según el entorno y las necesidades específicas del proyecto, pueden resultar de utilidad.…”

“…Una sola medición, luego, es insuficiente para obtener una distancia precisa. En la experimentación 2 se modificó la aplicación para realizar más de una iteración de toma de datos y promediarlos (Suárez et al, 2014). La misma idea fue aplicada en la investigación realizada por Torstensson (2016) en donde se promediaron 3 tomas de medidas.…”

Feature Learning for Fingerprint-Based Positioning in Indoor Environment

“…Chen et al [41] proposed a fuzzy logic-based system to solve this problem. Suárez et al [42] proposed a linear regression model to predict Wi-Fi RSSI with considering different situations (e.g., heterogeneous devices, drivers, and operating systems). Obviously, this approach is impractical for rapidly increasing new kinds of devices.…”