References | Purpose | Solution | Performance/findings |
---|---|---|---|
Guo et al. [88] | Indoor navigation | Wi-Fi fingerprints | (−) The user should guess his direction at junctions |
Wu et al. [91] | Indoor positioning | Wi-Fi fingerprint spatial gradient. | (+) Reduced the positioning errors caused by fluctuations in RSSI |
Wu et al. [106] | Indoor localization | BLE, Magnetic field, PDR, Trilateration, Particle filter | (+) Reduced the issues associated with the unwanted shaking of smartphones |
Li et al. [103] | Indoor navigation | Wi-Fi, magnetic fingerprint, PDR-based Kalman filter | (+) The system is scalable |
Lee et al. [109] | Indoor localization | BLE (Trilateration algorithm) and PDR | (−) Limitation of sensors |
Campana et al. [114] | Evaluation of machine learning algorithms for positioning | BLE fingerprints, Random Forest and Bayes classifier | (+) Random Forest algorithm increased the accuracy by 30% with respect to the Bayesian classifier |
Abu et al. [121] | Wayfinding in library for people with VI | Localization: Wi-Fi fingerprints, Book finding: Bluetooth + RFID, Dijkstra’s algorithm | (+) Decimeter level accuracy |
Tsirmpas et al. [124] | Indoor navigation system for people with VI | RFID for localization, ultrasonic rangefinder, and IMU for safe navigation | (+) 99% Success rate during real time testing |
Lin and Guo [125] | Indoor positioning | Frequency shift caused in the RFID system | (+) System overhead is reduced |
Loconsole et al. [126] | Indoor navigation system for people with VI | Kalman filter to reduce gravity and sensor baizes, RFID to reduce drift errors, Dijkstra’s algorithm | (+) Drift errors got reduced significantly |
Xu et al. [127] | To solve the fluctuations in RFID-based localization systems | RFID and Bayesian algorithm integrated with K-NN | (+) The average location estimation error was 15 cm |
Ganti et al. [128] | Indoor navigation system | VLC (Trilateration algorithm), sequential important particle filter, and Kalman filter | (+) The particle filter method is better for user tracking |
Jayakody et al. [129] | Indoor navigation system for people with VI | Visible light, geomagnetic sensor, and sonar for orientation and obstacle determination | (+) Implementation cost low |
Martinez-Sala et al. [23] | Indoor navigation system for people with VI | UWB (TDOA, AOA approach) and A* algorithm | (+) Centimeter level accuracy (< 20 cm) |
Fan et al. [131] | To reduce the error in UWB and IMU-based navigation system | UWB (TDOA approach) and IMU, double state adaptive Kalman filter | (+) Significant reduction in positioning errors |
Murata et al. [116] | Indoor localization for blind navigation | RSSI from BLE beacons, IMUs, and probabilistic localization algorithm | (+) Reduced the error in localization up to 1.5 m |
Ahmetovic et al. [117] | Indoor navigation for people with VI | BLE fingerprinting and IMUs | (−) Current version lack the functionality to notify the users traveling in the wrong path |
Kim et al. [118] | Indoor navigation for people with VI | BLE beacons based proximity detection and IMUs | (+) Proposed system was tested in a complex, highly crowded environment and observed results showed its potential for large scale deployment |
Cheraghi et al. [119] | Indoor navigation for people with VI | BLE beacons based proximity detection and IMUs | (+) Low-cost beacons reduced the implementation cost |
Jin-Woo et al. [96] | Indoor localization | CNN based Wi-Fi fingerprinting approach | (+) The proposed system is robust to small RSS fluctuations |
Mittal et al. [97] | Indoor localization | CNN based Wi-Fi fingerprinting approach in smartphone | (+) Average localization error < 2 m |
Ibrahim et al. [98] | Indoor localization | CNN based Wi-Fi fingerprinting using time series of RSS value as input. | (+) Building and floor prediction accuracy: 100% |