eptransf {EPT} | R Documentation |
Ensemble Patch Transform of a Signal
Description
This function performs ensemble patch transform of a signal for a size parameter.
Usage
eptransf(tindex = NULL, signal, type = "rectangle", tau,
process = c("average", "average"), pquantile = c(0, 1), equantile = c(0, 1),
gamma = 1, boundary = "symmetric")
Arguments
tindex |
time index at which a signal is observed. When it is |
signal |
a set of data or a signal observed at time |
type |
patch type of |
tau |
size parameter for ensemble patch transform. |
process |
specifies transform types for patch and ensemble processes:
|
pquantile |
quantiles for lower and upper envelopes of patch transform. When it is |
equantile |
quantiles for lower and upper envelopes of ensemble patch transform. |
gamma |
controls the amount of envelope magnitude. |
boundary |
specifies boundary condition from |
Details
This function performs ensemble patch transform of a signal for a size parameter tau
, and
produces statistics and envelopes for ensemble patch transform.
When process[1]
is "average"
or "median"
, outputs related to envelopes are defined as NULL
.
When process[2]
is "envelope"
, outputs, pstat
and Epstat
, are defined as NULL
.
Value
tindex |
time index at which a signal is observed. |
signal |
a set of data or a signal observed at time |
pstat |
centrality of patch transform for size parameter |
Epstat |
centrality of ensemble patch transform for size parameter |
psd |
standard deviation of patch transform for size parameter |
Epsd |
standard deviation of ensemble patch transform for size parameter |
pL |
lower envelope of patch transform for size parameter |
pU |
upper envelope of patch transform for size parameter |
pM |
mean envelope, |
pR |
distance between lower and upper envelopes, |
EpL |
lower envelope of ensemble patch transform for size parameter |
EpU |
upper envelope of ensemble patch transform for size parameter |
EpM |
mean envelope, |
EpR |
distance between lower and upper envelopes, |
parameters |
a list of input parameters of |
nlevel |
the number of size parameter |
See Also
Examples
# a doppler signal
n <- 256
tindex <- seq(0, 1, length=n)
j <- 5
f <- 10 * sqrt(tindex*(1-tindex)) * sin((2*pi*(1+2^((9-4*j)/5))) / (tindex+2^((9-4*j)/5)))
fnoise <- f + 0.4 * rnorm(n)
#### Ensemble statistics
op <- par(mfrow=c(5,3), mar=c(2,2,2,1))
layout(matrix(c(1, 1, 1, 2:13), 5, 3, byrow = TRUE))
plot(f, main="a doppler signal", xlab="", ylab="", type='l', ylim=range(fnoise))
points(fnoise); abline(h=0, lty=3)
#### Ensemble Patch Transform
taus <- c(5, 10, 20)
out <- list()
for (i in 1:length(taus))
out[[i]] <- eptransf(signal=fnoise, tau=taus[i], process=c("average", "average"))
for (i in 1:length(taus)) {
plot(out[[i]]$Epstat, type="l", xlab="", ylab="",
main=paste0("ensemble average of patch mean, tau=", taus[i]))
abline(h=0, lty=3)
}
for (i in 1:length(taus))
plot(out[[i]]$Epsd, type='l', xlab="", ylab="",
main=paste0("ensemble average of standard deviation, tau=", taus[i]))
out2 <- list()
for (i in 1:length(taus))
out2[[i]] <- eptransf(signal=fnoise, tau=taus[i], process=c("envelope", "average"))
for (i in 1:length(taus)) {
plot(out2[[i]]$EpM, type="l", col="red", xlab="", ylab="",
ylim=range(c(out2[[i]]$EpU,out2[[i]]$EpL)),
main=paste0("ensemble average of mean envelope, tau=", taus[i]))
points(fnoise, cex=0.1)
abline(h=0, lty=3); lines(out2[[i]]$EpU); lines(out2[[i]]$EpL)
}
for (i in 1:length(taus))
plot(out2[[i]]$EpR, type='l', xlab="", ylab="",
main=paste0("ensemble average of envelope distance, tau=", taus[i]))
par(op)