bp.by.bcr(s, lambda ..., analysis.name="nd.dwt", x="mof", block.selection=NULL, criterion.target=NA, max.iter=10000, rel.conv=1e-06, verbose=T, best.basis.fct=best.basis)
wavelet, n.levels
used by the wavelet transform.
dwt
,
nd.dwt
,
wp.table
,
cp.table
,
chirp.table
or
bp.table
as long as the corresponding matrix is the union of orthonormal matrices.
analysis
or a character string: "mof" (method of
frame estimate) or "zero" (zero constant vector) are available. Initial
guess of the wavelet coefficients for the iterative block coordinate relaxation
algorithm.
best.basis
or
best.chirp.rate
.
Bruce, A.G., Sardy, S., and Tseng P. (1998). Block coordinate relaxation
methods for nonparametric signal de-noising. Wavelet Applications, Proceedings
of the SPIE, Orlando, FL, April, 1998.
Sardy, S. (1988). Regularization Techniques for Linear Regression with
a Large Set of Carriers. Ph.D. Thesis. Department of Statistics. University of
Washington, Seattle.
Chen S., Donoho, D.L., and Saunders, M. (1996). Atomic decomposition by basis
pursuit. Technical report, Department of Statistics, Stanford University.
par(mfrow=c(1,2)) xx <- as.vector(make.signal("blocks", n=512, snr=5)) plot(xx, type="l", xlab="noisy blocks", ylab="") bp.hat <- bp.by.bcr(xx, lambda=sqrt(2*log(512)), analysis.name = "wp.table", x = "mof", block.selection = "optimal", criterion.target = NA, max.iter = 10000, rel.conv = 5.e-3, verbose = T, best.basis.fct=best.basis, n.levels=4, wavelet="s8", cost.fun="lp", p=2) plot(bp.hat$s.hat, type="l", xlab="smoothed blocks", ylab="", ylim=range(xx)) par(mfrow=c(1,1)) plot(bp.hat$coef.hat)