DENSITY IN DB 
DENSITY IN DB 
THEORETICAL SPECTRUM OF AUTO- ESTIMATED SPECTRUM OF SYNTHESIZED ESTIMATED SPECTRUM OF SYNTHESIZED 
REGRESSIVE MODEL WITH ORDER 2 DATA WITH AR-2 DATA WITH AR-2 
2.0 2.0 N = 1000 20 N = 1000 
AR-10 LAG = 50 
HANNING WINDOW 
1.0 1.0 
0.0 0 
-1.0 -1.0 
0 0.10 0.20 0.30 0.40 0.50 0 0.10 0.20 0.30 0.40 
—= —~ WwW — WwW 
ESTIMATED SPECTRUM OF SYNTHESIZED ESTIMATED SPECTRUM OF SYNTHESIZED ESTIMATED SPECTRUM OF SYNTHESIZED 
DATA WITH AR-2 DATA WITH AR-2 DATA WITH AR-2 
20 N = 1000 20 N = 1000 20 N = 1000 
AR-2 (MINIMUM AIC) AR-20 LAG = 100 
HANNING WINDOW 
0 0.10 0.20 0.30 0.40 0.50 0 0.10 0.20 0.30 0.40 0.50 
(a) AR method (optimum). (b) AR method (not optimum). (c) B—T method. 
Fig. 7.5 Comparison of AR-model fitting method and B—T method and the effect of 
changing orders of AR-model fitted to the simulated AR-2 process. 
(From Yamanouchi, et al.52) 
231 
