Table A3.

Summary of reviewed articles.

Art	Year	Est	AP	rPoint	fMap	fmRoom	mAlg	sAlg	mError	oError	sType
[106]	2019	exp	4	42	80 ${\mathrm{m}}^2$	N	RBF	LM	1.42 $\mathrm{m}$	RMSE = 1.459 m	RSSI
[107]	2019	exp	$N/A$	300	302 ${\mathrm{m}}^2$	Y	SVM		4.6 $\mathrm{m}$		RSSI
[60]	2019	exp	5	10	NULL	N	RF			Accuracy = 97.5% in 2 m	RSSI
[108]	2019	exp	8	107	512 ${\mathrm{m}}^2$	Y	K-ELM			RMSE = 1.7123 m std = 2.418 m	RSSI
[109]	2019	exp	9	96	560 ${\mathrm{m}}^2$	Y	QKMMCC			average = 0.76m	RSSI
[65]	2019	pMap	520	993	UJIIndoorLoc	Y	RNN			Accuracy = 87.41%floor std = 0.83 m	RSSI
		exp	7	$N/A$	4 Rooms	Y	RNN			Accuracy = 95.8% std = 0.60 m	RSSI
[83]	2019	pMap	520	993	UJIIndoorLoc	Y	RNN		4.2 $\mathrm{m}$	std = 3.2 m	RSSI
		exp	6	365	336 ${\mathrm{m}}^2$	Y	RNN		0.75 $\mathrm{m}$	std = 0.64 m	RSSI
[110]	2019	exp	9	261	300 ${\mathrm{m}}^2$	N	BPNN		2.7 $\mathrm{m}$	Accuracy = 90%	RSSI
[111]	2019	exp	8	66	736 ${\mathrm{m}}^2$	Y	SDA		3.7 $\mathrm{m}$	Accuracy = 84%	RSSI
[112]	2019	exp	1	42	50 ${\mathrm{m}}^2$	N	CNN		0.46 $\mathrm{m}$	without obstacles	RSSI
		exp	1	42	50 ${\mathrm{m}}^2$	N	CNN		1.11 $\mathrm{m}$	with some obstacles	RSSI
[113]	2019	exp	1	15	20 ${\mathrm{m}}^2$	Y	MLP		1.42 $\mathrm{m}$		RSSI
		exp	1	15	20 ${\mathrm{m}}^2$	Y	CNN		1.67 $\mathrm{m}$		RSSI
		exp	1	15	14.4 ${\mathrm{m}}^2$	N	MLP		1.43 $\mathrm{m}$		RSSI
		exp	1	15	14.4 ${\mathrm{m}}^2$	N	CNN		1.51 $\mathrm{m}$		RSSI
[114]	2019	exp	258	9	125 ${\mathrm{m}}^2$	Y	CNN		3.91 $\mathrm{m}$	Accuracy = 84%	RSSI
[63]	2019	pMap	$N/A$	$N/A$	NULL	Y	BPNN	ACO		Accuracy = 91.4%	RSSI
[115]	2019	pMap	520	993	UJI Library	Y	CNN, GRP		3.6 $\mathrm{m}$	90% less 2m	RSSI
[42]	2019	exp	1	25	26.4 ${\mathrm{m}}^2$	N	BPNN	PCA-PD	1.42 $\mathrm{m}$	std = 1.1511 m	CSI
[84]	2019	exp	$N/A$	20	1200 ${\mathrm{m}}^2$	Y	MLP	SDAE	3.05 $\mathrm{m}$	1day	RSSI
		exp	$N/A$	57	2400 ${\mathrm{m}}^2$	Y	MLP	SDAE	3.39 $\mathrm{m}$	2 days	RSSI
		pMap	520	993	UJIIndoorLoc	Y	MLP	SDAE	5.64 $\mathrm{m}$	10 days	RSSI
[116]	2019	pMap	520	993	UJIIndoorLoc	Y	VAE			RMSE = 4.65 m	RSSI
[117]	2019	exp	6	49	1600 ${\mathrm{m}}^2$	Y	DNN		0.95 $\mathrm{m}$	Open Doors	RSSI
		exp	6	49	1600 ${\mathrm{m}}^2$	Y	DNN		1.26 $\mathrm{m}$	Closed Doors	RSSI
[118]	2019	exp	4	228	1200 ${\mathrm{m}}^2$	Y	ANN		1.22 $\mathrm{m}$		RSSI
		exp	$N/A$	$N/A$	$N/A$	Y	ANN		1.90 $\mathrm{m}$		RSSI
[119]	2019	exp	7	25	1728 ${\mathrm{m}}^2$	N	RNN	LSTM	1.05 $\mathrm{m}$	std = 0.8856 m	RSSI
[120]	2019	exp	15	71	4000 ${\mathrm{m}}^2$	Y	NN	GA	3.47 $\mathrm{m}$		RSSI
[121]	2019	exp	4	50	1100 ${\mathrm{m}}^2$	Y	BGM		2.9 $\mathrm{m}$		RSSI
[122]	2019	exp	122	48	629 ${\mathrm{m}}^2$	Y	DNN		2.64 $\mathrm{m}$		RSSI
		exp	59	139	65 ${\mathrm{m}}^2$	N	DNN		1.21 $\mathrm{m}$		RSSI
[123]	2018	pMap	520	993	UJIIndoorLoc	Y	CNN			95.76% floor level	RSSI
[124]	2018	pMap	7	1000	Rajen Bhatt	Y	RF			98.3% floor level	RSSI
[125]	2018	exp	20	2100	8250 ${\mathrm{m}}^2$	Y	DNN		3.95 $\mathrm{m}$	std = 2.72 m	RSSI
[126]	2018	exp	16	202	806 ${\mathrm{m}}^2$	Y	SMN	PCA	1.85 $\mathrm{m}$	std = 1.04 m	RSSI
[127]	2018	pMap	520	993	UJIIndoorLoc	Y	DQN			78.79% in 1 m	RSSI
[37]	2018	exp	50	180	75 ${\mathrm{m}}^2$	Y	RF		1.29 $\mathrm{m}$	90% in 3 m	RSSI
[128]	2018	exp	$N/A$	$N/A$	NULL	Y	DNN			83.6% floor with people, 99.6% without	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.