Table 1.

BLE indoor positioning-related studies.

MAE (m)	Experiment Area (m)	Deployment Requirements	Method and Reference
0.192	$5\times 5$	3 beacons	Kalman/Gaussian filter, trilateration-centroid [10]
0.76	$5.6\times 8.8$	8 fixed beacons	Hybrid, sliding window + KF, trilateration, DR [15]
0.8	$10.5\times 15.6$	22 fixed beacons	BLE fingerprinting [22]
1.0 with $75\%$	$8\times 8$	4 beacons, 5 Wi-Fi AP	Hybrid, SVM, KNN, ML cloud service [13]
1.27	$90\times 37$	48 fixed beacons	Graph optimization, fingerprinting+range based [23]
1.37	$8.4\times 15$	BLE tags and 4 Repeater	Hybrid (Wi − Fi + BLE), fingerprinting, CoO [14]
$\sqrt{2}$ with $94\%$	30 m2	4 fixed beacons	Selection of nearest BLE tag [16]
1.5	$6\times 6$	9 fixed beacons	Trilateration, RSSI filtering [24]
$\mu$:2.29, $\sigma$:1.67	$15\times 16$	1 beacon, 3 receivers	OASLTIP
$2\sqrt{2}$ with $70.2\%$	$5\times 5$	4 fixed beacons, 1 receiver	Two machine learning classifiers [20]
2.89	$12\times 14$	4 fixed beacons	Trilateration, min-max, least square [9]
4.0 with $97.5\%$	$8\times 6$	5 beacons	Particle filter, BLE fingerprinting [18]
4.6 with $90\%$	$16.5\times 17.6$	4 receivers, 1 beacon	Channel diversity, weighted trilateration and KF [21]
1.5 with $91.4\%$	3 (1D)	1 beacon	Low TX powered beacon utilization [17]
0.5	3 (1D)	1 beacon	IPSAPP (A mobile application) [11]