Table A2.

Summary of reviewed articles.

Art	Year	Est	AP	rPoint	fMap	fmRoom	mAlg	sAlg	mError	oError	sType
[23]	2020	pMap	520	993	UJIIndoorLoc	N	KNN, LR, SVM, RF			RMSE = 1.87 m	RSSI
[55]	2020	exp	6	112	460 ${\mathrm{m}}^2$	Y	capsnet		0.68 $\mathrm{m}$		RSSI
[31]	2020	exp	8	133	512 ${\mathrm{m}}^2$	N	Deep Fuzzy Forest		1.36 $\mathrm{m}$	RMSE = 1.79 m	RSSI
[52]	2020	exp	1	32	50 ${\mathrm{m}}^2$	N	CNN		1.77 $\mathrm{m}$		CSI
		exp	1	24	40 ${\mathrm{m}}^2$	N	CNN		1.16 $\mathrm{m}$		CSI
		exp	1	66	49 ${\mathrm{m}}^2$	N	CNN		2.54 $\mathrm{m}$		CSI
[97]	2020	exp	6	50	60 ${\mathrm{m}}^2$	N	RF	Bernoulli distribution		RMSE = 2.50 m	RSSI
[98]	2020	exp	25	240	315 ${\mathrm{m}}^2$	N	RF	Co-forest	2.44 $\mathrm{m}$		RSSI
		exp	5	$N/A$	NULL	N	RF		4.44 $\mathrm{m}$		RSSI
[24]	2020	pMap	7	1000	Rajen Bhatt	Y	MLP			Accuracy = 94.4%	RSSI
[25]	2020	pMap	520	993	UJIIndoorLoc	Y	CNN			Accuracy = 88%	RSSI
[99]	2020	exp	195	300	800 ${\mathrm{m}}^2$	N	DNN	HMM	1.22 $\mathrm{m}$	RMSE = 1.43 m	RSSI
[32]	2020	exp	3	56	87.75 ${\mathrm{m}}^2$	N	DNN	LC	0.78 $\mathrm{m}$	std = 1.96 m	CSI
[28]	2020	exp	4	236	1148 ${\mathrm{m}}^2$	Y	BPNN	GA-PSO	0.22 $\mathrm{m}$		RSSI
[26]	2020	exp	10	102	568.4 ${\mathrm{m}}^2$	Y	LSTM	LF-D	1.48 $\mathrm{m}$		RSSI
		exp	30	353	2750 ${\mathrm{m}}^2$	Y	LSTM		1.75 $\mathrm{m}$		RSSI
[27]	2020	pMap	$N/A$	$N/A$	Cramariuc	Y	SEQ2SEQ	LSTM	5.5 $\mathrm{m}$		RSSI
		pMap.	$N/A$	$N/A$	Cramariuc	Y	SEQ2SEQ		3.08 $\mathrm{m}$		RSSI
[100]	2020	pMap	$N/A$	$N/A$	IPIN2016	Y	CNN, LSTM		4.93 $\mathrm{m}$		RSSI
		pMap	$N/A$	$N/A$	IPIN2016	Y	CNN, LSTM		5.4 $\mathrm{m}$		RSSI
		pMap	520	993	UJI Library	Y	CNN, LSTM		3.2 $\mathrm{m}$		RSSI
		pMap	520	993	UJI Library	Y	CNN, LSTM		4.98 $\mathrm{m}$		RSSI
[56]	2020	exp	5	22	293 ${\mathrm{m}}^2$	Y	DNN			Accuracy = 95.45% in 3.65 × 3.65 m	RSSI
[29]	2020	exp	$N/A$	157	5500 ${\mathrm{m}}^2$	Y	RNN	DL	3.05 $\mathrm{m}$	std = 2.818 m	RSSI
		pMap	520	993	UJIIndoorLoc	Y	RNN		4.92 $\mathrm{m}$	std = 3.719 m	RSSI
		sim	4	00	1681 ${\mathrm{m}}^2$	Y	RNN	DL	2.42 $\mathrm{m}$–2.92 $\mathrm{m}$		RSSI
[101]	2020	sim	54	54	10,000 m ${}^2$	N	MLP		3.35 $\mathrm{m}$		RSSI
[9]	2020	exp	3	7	25 ${\mathrm{m}}^2$	Y	DNN	RESNET	0.11 $\mathrm{m}$	RMSE = 0.08 m	SNR
[102]	2020	pMap	$N/A$	40	UJI Library	N	CNN	SVR	2.15 $\mathrm{m}$		RSSI
[66]	2019	exp	3	30	540 ${\mathrm{m}}^2$	N	DBN	cross entropy and the mean squared		NULL	RSSI
[34]	2019	exp	2	59	125 ${\mathrm{m}}^2$	Y	SVM		0.7 $\mathrm{m}$		RSSI
[57]	2019	exp	$N/A$	206	NULL	Y	DNN	Stacked AutoEncoder		Accuracy = 85%	RSSI
[35]	2019	exp	1	100	100 ${\mathrm{m}}^2$	N	SVM		1.9 $\mathrm{m}$	std = 0.07 m	CSI
[103]	2019	exp	$N/A$	$N/A$	NULL	NULL	CNN			RMSE = 0.31 m	RSSI
[104]	2019	exp	1	$N/A$	63 ${\mathrm{m}}^2$	Y	SVM			96.4%	RSSI
		exp	1	$N/A$	63 ${\mathrm{m}}^2$	Y	MLP			96.5%	RSSI
[53]	2019	exp	$N/A$	$N/A$	NULL	NULL	SVM			RMSE = 0.42 m	CSI
[58]	2019	exp	16	83	305 ${\mathrm{m}}^2$	Y	DNN		2 $\mathrm{m}$		RSSI
[59]	2019	pMap	520	993	UJIIndoorLoc	Y	CNN			Accuracy = 95.92%	RSSI
		pMap	309	3951	Tampere	Y	CNN			Accuracy = 94.13%	RSSI
[105]	2019	exp	6	300	300 ${\mathrm{m}}^2$	N	MEA-BP		0.72 $\mathrm{m}$		RSSI
[67]	2019	exp	50	$N/A$	NULL	NULL	ELM			NULL	RSSI
[61]	2019	exp	256	74	1664 ${\mathrm{m}}^2$	Y	CNN			Accuracy = 95.4% in 4 m	RSSI
[62]	2019	exp	54	180	1209 ${\mathrm{m}}^2$	Y	RDF			Accuracy = 89% at room level	RSSI
[64]	2019	exp	256	74	300 ${\mathrm{m}}^2$	Y	CNN		1.46 $\mathrm{m}$	Accuracy = 94% std = 2.24 m	RSSI

art: Article; mAlg: Main algorithm used; est: Experimental or pMapulated study; sAlg: Other algorithms used in the study; AP: APs used; mError: Mean Error; rPoint: Reference Points used in offline phase; oError: Other metrics reported in the study; fMap: Size of experimental room or radio-map used; sType: Signal type used; fmRoom: Rooms used in exp/pMap.