Overview

This documentation was generated with the assistance of AI. Please report any inaccuracies.

irurueta-navigation-indoor is a Java library for indoor positioning based on GNSS/INS-style navigation techniques applied to radio signals (Wi-Fi access points, Bluetooth beacons, and similar sources). Where GPS is unavailable, this library estimates a device’s position and/or the location of the radio sources themselves from the signals a device observes indoors.

Why it exists

Indoor environments block satellite navigation, so indoor positioning instead relies on radio signals already present in a building: Wi-Fi access points and Bluetooth beacons. This library turns raw signal readings — signal strength (RSSI) or distance (ranging) measurements — into position estimates, using well-established estimation and robust-statistics techniques (least squares, non-linear optimization, LMedS, MSAC, PROSAC) so that noisy or partially wrong measurements don’t derail the result.

Core concepts

The library is organized around three complementary problems, implemented respectively in the com.irurueta.navigation.indoor.fingerprint, com.irurueta.navigation.indoor.position, and com.irurueta.navigation.indoor.radiosource packages:

Fingerprint-based positioning: Signal readings collected at known locations are stored as Fingerprints. A new, unlocated fingerprint is then matched against that database — e.g. via WeightedKNearestNeighboursPositionSolver or a FingerprintPositionEstimator — to estimate the device’s position without needing to know where the radio sources themselves are.
Direct (lateration) positioning: When the radio sources' locations are known, RssiPositionEstimator and RangingPositionEstimator solve directly for a device’s position from RssiReading/RangingReading measurements, using linear or non-linear multilateration.
Radio source estimation: RadioSourceEstimator and its variants solve the inverse problem: given readings taken from multiple known locations, estimate an unknown radio source’s own position and transmission parameters (transmit power, path-loss exponent).

A RadioSource (implemented by Beacon and WifiAccessPoint) represents any transmitting device involved in these estimations.

flowchart LR subgraph Fingerprinting A[Signal readings at known locations] --> B[Fingerprint database] C[New unlocated fingerprint] --> D[k-NN / fingerprint position estimator] B --> D D --> E[Estimated device position] end subgraph Direct lateration F[RSSI / ranging readings] --> G[Known radio source locations] G --> H[Rssi / Ranging position estimator] F --> H H --> E end subgraph Radio source estimation I[Readings from known locations] --> J[Radio source estimator] J --> K[Estimated source position + tx power + path-loss exponent] end

Path-loss distance model

Several estimators convert a received signal strength into an estimated distance using a log-distance path-loss model. Given a measured RSSI, the transmit power, and a path-loss exponent , the estimated distance is:

where:

depends on the signal’s frequency and the speed of light . Robust variants of the estimators (LMedS, MSAC, PROSAC) apply this model while tolerating outlier readings.

API reference

Main classes used on this page, linked to their source code and Javadoc:

Class Source Javadoc

Class	Source	Javadoc
`Beacon`	Source	Javadoc
`Fingerprint`	Source	Javadoc
`FingerprintPositionEstimator`	Source	Javadoc
`RadioSource`	Source	Javadoc
`RadioSourceEstimator`	Source	Javadoc
`RangingPositionEstimator`	Source	Javadoc
`RangingReading`	Source	Javadoc
`RssiPositionEstimator`	Source	Javadoc
`RssiReading`	Source	Javadoc
`WeightedKNearestNeighboursPositionSolver`	Source	Javadoc
`WifiAccessPoint`	Source	Javadoc

Beacon

Fingerprint

FingerprintPositionEstimator

RadioSource