EURASIP Journal on Wireless Communications and Networking

Table 1 Parameter estimation

From: An Unsupervised LLR Estimation with unknown Noise Distribution

			μ_a	σ_a	μ_b	σ_b
α=1.8	γ=0.53	Unsupervised	3.43	0.06	5.73	0.15
		Sup_LS=20000	3.25	0.05	7.59	0.28
		Sup_LS=1200	3.27	0.24	8.50	14.48
		Sup_LS=900	3.27	0.27	11.72	46.15
	γ=0.55	Unsupervised	3.23	0.05	5.61	0.14
		Sup_LS=20000	3.05	0.05	7.62	0.28
		Sup_LS=1200	3.07	0.22	7.97	1.54
		Sup_LS=900	3.07	0.26	10.73	30.41

Comparison of the mean and standard deviation evolution for the parameters (a,b) as a function of the dispersion γ of a S αS noise with α=1.8 for the supervised with different learning sequence sizes and unsupervised optimization

Back to article page