R/DataAvailability.R
In overeem11/RAINLINK: Retrieval algorithm for rainfall mapping from microwave links in a cellular communication network.
Defines functions
DataAvailability
Documented in
DataAvailability
## The RAINLINK package. Retrieval algorithm for rainfall mapping from microwave links
## in a cellular communication network.
##
## Version 1.3
## Copyright (C) 2022 Aart Overeem
##
## This program is free software: you can redistribute it and/or modify
## it under the terms of the GNU General Public License as published by
## the Free Software Foundation, either version 3 of the License, or
## (at your option) any later version.
##
## This program is distributed in the hope that it will be useful,
## but WITHOUT ANY WARRANTY; without even the implied warranty of
## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
## GNU General Public License for more details.
##
## You should have received a copy of the GNU General Public License
## along with this program. If not, see <http://www.gnu.org/licenses/>.
#' Function for plotting data availability of commercial microwave link network.
#' @description Function for plotting data availability of commercial microwave link data. This function provides the following figures:
#' \enumerate{
#' \item Bar plot with percentage of sub-links for bins of data availability (%).
#' \item Bar plot with percentage of link paths for bins of data availability (%).
#' \item Plot with average number of sub-links and link paths for each time interval of the sampling strategy.
#' }
#' In addition, the Average number of sub-links and link paths is printed to the screen.
#'
#' Does not depend on sampling strategy.
#'
#' The input microwave link data do not have to be sorted chronologically. Full-duplex links will give two data entries, these will both be used.
#'
#' Data availability is computed over the number of time intervals in the dataset as obtained from DateTime.
#' In case a time interval is missing, i.e. not present in DateTime, it is not included in the computations.
#' Hence, the computed availability does only refer to the time intervals for which a DateTime entry is present in the dataset.
#' In the computation of availabilities only links or paths are counted as available in case the link-derived rainfall intensities
#' are equal to or larger than 0 (mm h\eqn{^{-1}}). I.e., NA values in Rmean are taken into account in the computation of data availability.
#' So data availability is always computed over all time intervals in DateTime. Note that NA values do not exist in the DateTime object if function
#' ''PreprocessingMinMaxRSL'' has been run and note that Data must be preprocessed, because Rmean is used.
#'
#' @param Data Data frame with microwave link data (use data(Linkdata) to load example data).
#' @param cex.axis The magnification to be used for axis annotation relative to the current setting of ''cex'' in file FigNameTimeseriesAvailability.
#' @param cex.lab The magnification to be used for x and y labels relative to the current setting of ''cex'' in file FigNameTimeseriesAvailability.
#' @param FigNameBarplotAvailabilityLinks Name of file with bar plot with percentage of links for bins of link orientation. The extension must be ''.pdf''.
#' @param FigNameBarplotAvailabilityLinkPaths Name of file with bar plot with percentage of links for bins of link path length. The extension must be ''.pdf''.
#' @param FigNameTimeseriesAvailability Name of file with bar plot with percentage of links for bins of microwave frequency. The extension must be ''.pdf''.
#' @param ps integer; the point size of text (but not symbols) in file FigNameTimeseriesAvailability.
#' @param Rmean Vector of link-derived rainfall intensities (mm h\eqn{^{-1}}) with length equal to Data.
#' @param TimeZone Time zone of data (e.g. "UTC").
#' @return Figures with data availability of commercial microwave link network.
#' @export DataAvailability
#' @examples
#' data(Linkdata)
#' DataAvailability(Data=Linkdata,cex.axis=0.9,cex.lab=1.15,
#' FigNameBarplotAvailabilityLinks="Barplot_Availability_Links.pdf",
#' FigNameBarplotAvailabilityLinkPaths="Barplot_Availability_LinkPaths.pdf",
#' FigNameTimeseriesAvailability="TimeseriesAvailability.pdf",ps=18,Rmean=Rmean,TimeZone="UTC")
#' @author Aart Overeem
#' @references ''ManualRAINLINK.pdf''
#'
#' Overeem, A., Leijnse, H., and Uijlenhoet, R., 2016: Retrieval algorithm for rainfall mapping from microwave links in a
#' cellular communication network, Atmospheric Measurement Techniques, 9, 2425-2444, https://doi.org/10.5194/amt-9-2425-2016.
DataAvailability <- function(Data,cex.axis,cex.lab,FigNameBarplotAvailabilityLinks,FigNameBarplotAvailabilityLinkPaths,FigNameTimeseriesAvailability,ps,Rmean,TimeZone="UTC",verbose=TRUE)
{
# Bar plot data availability per sub-link.
# Set link IDs and time intervals
Data$ID <- as.character(Data$ID)
IDLink <- unique(Data$ID)
N_links <- length(IDLink)
# Compute number of time intervals in dataset.
t <- sort(unique(Data$DateTime))
N_t <- length(t)
# Compute availability per sub-link.
PercAvail <- c(NA)
for (i in 1:N_links)
{
cond <- which(Data$ID==IDLink[i] & Rmean>=0)
PercAvail[i] <- length(cond) / N_t * 100
}
StepPercAvail <- 10
pdf(FigNameBarplotAvailabilityLinks,family="Times")
par(pty="s")
par(ps=24)
par(mar=c(5,5,1,1)+0.1)
PercAvaildata <- c(NA)
NrAvailClass <- 0
for (availclass in seq(StepPercAvail,100,StepPercAvail))
{
NrAvailClass <- NrAvailClass + 1
if (availclass==StepPercAvail)
{
# This is done to make sure that 0 availability is also shown in the graph.
PercAvaildata[NrAvailClass] <- length(PercAvail[PercAvail>=(availclass-StepPercAvail)&PercAvail<=availclass])
}
if (availclass>StepPercAvail)
{
PercAvaildata[NrAvailClass] <- length(PercAvail[PercAvail>(availclass-StepPercAvail)&PercAvail<=availclass])
}
}
total=sum(PercAvaildata)
perc=100*PercAvaildata/total
ylimbarplot <- c(0,max(perc))
barplot(perc,xlab = "Data availability sub-links (%)", ylab = "Percentage",
xaxt="n",tcl=.5,ylim=ylimbarplot,cex.lab=1.3)
at <- 0.1
axis(side = 1, at = at, labels = 0, cex.axis = 0.6)
for (availclass in seq(StepPercAvail,100,StepPercAvail))
{
at <- at + 1.2
axis(side = 1, at = at, labels = availclass, cex.axis = 0.6)
}
legend("topleft",legend=substitute("Mean" == value*"%",list(value=formatC(mean(PercAvail), format="f", digits=2))),bty="n",cex=1)
dev.off()
# Bar plot data availability per link path.
# Set link IDs and time intervals
# Determine coordinates of middle of link:
X_middleLinks <- (Data$XStart + Data$XEnd) / 2
Y_middleLinks <- (Data$YStart + Data$YEnd) / 2
# Determine unique middle of links:
Coord_dataf <- data.frame(cbind(X_middleLinks, Y_middleLinks))
Coord_uniq <- unique(Coord_dataf)
N_linkpaths <- length(Coord_uniq[,1])
# Compute number of time intervals in dataset.
t <- sort(unique(Data$DateTime))
N_t <- length(t)
# Compute availability per link path:
PercAvail <- c(NA)
for (i in 1:N_linkpaths)
{
cond <- which(X_middleLinks==Coord_uniq[i,1] & Y_middleLinks==Coord_uniq[i,2] & Rmean>=0)
# Determine number of intervals with data, i.e. where the link path has at least 1 observation of Rmean>=0:
NumberIntervals <- length(unique(Data$DateTime[cond]))
PercAvail[i] <- NumberIntervals / N_t * 100
}
StepPercAvail <- 10
pdf(FigNameBarplotAvailabilityLinkPaths,family="Times")
par(pty="s")
par(ps=24)
par(mar=c(5,5,1,1)+0.1)
PercAvaildata <- c(NA)
NrAvailClass <- 0
for (availclass in seq(StepPercAvail,100,StepPercAvail))
{
NrAvailClass <- NrAvailClass + 1
if (availclass==StepPercAvail)
{
# This is done to make sure that 0 availability is also shown in the graph.
PercAvaildata[NrAvailClass] <- length(PercAvail[PercAvail>=(availclass-StepPercAvail)&PercAvail<=availclass])
}
if (availclass>StepPercAvail)
{
PercAvaildata[NrAvailClass] <- length(PercAvail[PercAvail>(availclass-StepPercAvail)&PercAvail<=availclass])
}
}
total=sum(PercAvaildata)
perc=100*PercAvaildata/total
ylimbarplot <- c(0,max(perc))
barplot(perc,xlab = "Data availability link paths (%)", ylab = "Percentage",
xaxt="n",tcl=.5,ylim=ylimbarplot,cex.lab=1.3)
at <- 0.1
axis(side = 1, at = at, labels = 0, cex.axis = 0.6)
for (availclass in seq(StepPercAvail,100,StepPercAvail))
{
at <- at + 1.2
axis(side = 1, at = at, labels = availclass, cex.axis = 0.6)
}
legend("topleft",legend=substitute("Mean" == value*"%",list(value=formatC(mean(PercAvail), format="f", digits=2))),bty="n",cex=1)
dev.off()
# Plot with average number of sub-links and link paths for each time interval of the sampling strategy.
UniqueTimestamp <- sort(unique(Data$DateTime))
# Determine for each unique timestamp the number of sub-links with path-average rainfall intensities.
NumberSubLinks <- NumberPaths <- c(NA)
s <- 0
for (Timestamp in UniqueTimestamp)
{
s <- s + 1
cond <- which(Data$DateTime==Timestamp & Rmean>=0)
NumberSubLinks[s] <- length(cond)
# Determine coordinates of middle of link:
X_middleLinks <- (Data$XStart[cond] + Data$XEnd[cond]) / 2
Y_middleLinks <- (Data$YStart[cond] + Data$YEnd[cond]) / 2
# Determine unique middle of links:
Coord_dataf <- data.frame(cbind(X_middleLinks, Y_middleLinks))
Coord_uniq <- unique(Coord_dataf)
# Compute average number of unique link paths:
NumberPaths[s] <- (length(Coord_uniq[, 1]))
}
Timestamp_data = strptime(UniqueTimestamp,"%Y%m%d%H%M",tz=TimeZone)
pdf(FigNameTimeseriesAvailability,width=13.2,height=4.4,family="Times")
par(mar=c(3,3.8,1.8,0.1))
par(ps=ps)
plot(Timestamp_data,NumberPaths,xlab="",ylab="",type="p",ylim=c(0,max(NumberSubLinks)),mgp=c(1.8,0.5,0),xaxt="n",yaxt="n",lwd=1,lty=1,col="gray",pch=16,cex=0.4)
par(new=TRUE)
plot(Timestamp_data,NumberSubLinks,xlab="Date",ylab="Mean number per interval",type="p",ylim=c(0,max(NumberSubLinks)),cex.lab=cex.lab,cex.axis=cex.axis,col="black",mgp=c(1.8,0.5,0),lwd=1,lty=1,pch=16,cex=0.4)
legend("bottomright",legend=c("Sub-links","Link paths"),col=c("black","gray"),pch=c(16,16),bty="n",cex=1.15)
dev.off()
print(paste("Average number of sub-links:",mean(NumberSubLinks)))
print(paste("Average number of link paths:",mean(NumberPaths)))
# A missing interval means that no timestamp and associated number of link (paths) is available, i.e. also no "NA" value is provided.
# In such a case, the other intervals are correctly plotted, because the horizontal axis is an object of class "POSIXlt" (a date time object).
}
overeem11/RAINLINK documentation built on July 8, 2023, 5:53 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions DataAvailability
Documented in DataAvailability
## The RAINLINK package. Retrieval algorithm for rainfall mapping from microwave links
## in a cellular communication network.
##
## Version 1.3
## Copyright (C) 2022 Aart Overeem
##
## This program is free software: you can redistribute it and/or modify
## it under the terms of the GNU General Public License as published by
## the Free Software Foundation, either version 3 of the License, or
## (at your option) any later version.
##
## This program is distributed in the hope that it will be useful,
## but WITHOUT ANY WARRANTY; without even the implied warranty of
## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
## GNU General Public License for more details.
##
## You should have received a copy of the GNU General Public License
## along with this program. If not, see <http://www.gnu.org/licenses/>.
#' Function for plotting data availability of commercial microwave link network.
#' @description Function for plotting data availability of commercial microwave link data. This function provides the following figures:
#' \enumerate{
#' \item Bar plot with percentage of sub-links for bins of data availability (%).
#' \item Bar plot with percentage of link paths for bins of data availability (%).
#' \item Plot with average number of sub-links and link paths for each time interval of the sampling strategy.
#' }
#' In addition, the Average number of sub-links and link paths is printed to the screen.
#'
#' Does not depend on sampling strategy.
#'
#' The input microwave link data do not have to be sorted chronologically. Full-duplex links will give two data entries, these will both be used.
#'
#' Data availability is computed over the number of time intervals in the dataset as obtained from DateTime.
#' In case a time interval is missing, i.e. not present in DateTime, it is not included in the computations.
#' Hence, the computed availability does only refer to the time intervals for which a DateTime entry is present in the dataset.
#' In the computation of availabilities only links or paths are counted as available in case the link-derived rainfall intensities
#' are equal to or larger than 0 (mm h\eqn{^{-1}}). I.e., NA values in Rmean are taken into account in the computation of data availability.
#' So data availability is always computed over all time intervals in DateTime. Note that NA values do not exist in the DateTime object if function
#' ''PreprocessingMinMaxRSL'' has been run and note that Data must be preprocessed, because Rmean is used.
#'
#' @param Data Data frame with microwave link data (use data(Linkdata) to load example data).
#' @param cex.axis The magnification to be used for axis annotation relative to the current setting of ''cex'' in file FigNameTimeseriesAvailability.
#' @param cex.lab The magnification to be used for x and y labels relative to the current setting of ''cex'' in file FigNameTimeseriesAvailability.
#' @param FigNameBarplotAvailabilityLinks Name of file with bar plot with percentage of links for bins of link orientation. The extension must be ''.pdf''.
#' @param FigNameBarplotAvailabilityLinkPaths Name of file with bar plot with percentage of links for bins of link path length. The extension must be ''.pdf''.
#' @param FigNameTimeseriesAvailability Name of file with bar plot with percentage of links for bins of microwave frequency. The extension must be ''.pdf''.
#' @param ps integer; the point size of text (but not symbols) in file FigNameTimeseriesAvailability.
#' @param Rmean Vector of link-derived rainfall intensities (mm h\eqn{^{-1}}) with length equal to Data.
#' @param TimeZone Time zone of data (e.g. "UTC").
#' @return Figures with data availability of commercial microwave link network.
#' @export DataAvailability
#' @examples
#' data(Linkdata)
#' DataAvailability(Data=Linkdata,cex.axis=0.9,cex.lab=1.15,
#' FigNameBarplotAvailabilityLinks="Barplot_Availability_Links.pdf",
#' FigNameBarplotAvailabilityLinkPaths="Barplot_Availability_LinkPaths.pdf",
#' FigNameTimeseriesAvailability="TimeseriesAvailability.pdf",ps=18,Rmean=Rmean,TimeZone="UTC")
#' @author Aart Overeem
#' @references ''ManualRAINLINK.pdf''
#'
#' Overeem, A., Leijnse, H., and Uijlenhoet, R., 2016: Retrieval algorithm for rainfall mapping from microwave links in a
#' cellular communication network, Atmospheric Measurement Techniques, 9, 2425-2444, https://doi.org/10.5194/amt-9-2425-2016.
DataAvailability <- function(Data,cex.axis,cex.lab,FigNameBarplotAvailabilityLinks,FigNameBarplotAvailabilityLinkPaths,FigNameTimeseriesAvailability,ps,Rmean,TimeZone="UTC",verbose=TRUE)
{
# Bar plot data availability per sub-link.
# Set link IDs and time intervals
Data$ID <- as.character(Data$ID)
IDLink <- unique(Data$ID)
N_links <- length(IDLink)
# Compute number of time intervals in dataset.
t <- sort(unique(Data$DateTime))
N_t <- length(t)
# Compute availability per sub-link.
PercAvail <- c(NA)
for (i in 1:N_links)
{
cond <- which(Data$ID==IDLink[i] & Rmean>=0)
PercAvail[i] <- length(cond) / N_t * 100
}
StepPercAvail <- 10
pdf(FigNameBarplotAvailabilityLinks,family="Times")
par(pty="s")
par(ps=24)
par(mar=c(5,5,1,1)+0.1)
PercAvaildata <- c(NA)
NrAvailClass <- 0
for (availclass in seq(StepPercAvail,100,StepPercAvail))
{
NrAvailClass <- NrAvailClass + 1
if (availclass==StepPercAvail)
{
# This is done to make sure that 0 availability is also shown in the graph.
PercAvaildata[NrAvailClass] <- length(PercAvail[PercAvail>=(availclass-StepPercAvail)&PercAvail<=availclass])
}
if (availclass>StepPercAvail)
{
PercAvaildata[NrAvailClass] <- length(PercAvail[PercAvail>(availclass-StepPercAvail)&PercAvail<=availclass])
}
}
total=sum(PercAvaildata)
perc=100*PercAvaildata/total
ylimbarplot <- c(0,max(perc))
barplot(perc,xlab = "Data availability sub-links (%)", ylab = "Percentage",
xaxt="n",tcl=.5,ylim=ylimbarplot,cex.lab=1.3)
at <- 0.1
axis(side = 1, at = at, labels = 0, cex.axis = 0.6)
for (availclass in seq(StepPercAvail,100,StepPercAvail))
{
at <- at + 1.2
axis(side = 1, at = at, labels = availclass, cex.axis = 0.6)
}
legend("topleft",legend=substitute("Mean" == value*"%",list(value=formatC(mean(PercAvail), format="f", digits=2))),bty="n",cex=1)
dev.off()
# Bar plot data availability per link path.
# Set link IDs and time intervals
# Determine coordinates of middle of link:
X_middleLinks <- (Data$XStart + Data$XEnd) / 2
Y_middleLinks <- (Data$YStart + Data$YEnd) / 2
# Determine unique middle of links:
Coord_dataf <- data.frame(cbind(X_middleLinks, Y_middleLinks))
Coord_uniq <- unique(Coord_dataf)
N_linkpaths <- length(Coord_uniq[,1])
# Compute number of time intervals in dataset.
t <- sort(unique(Data$DateTime))
N_t <- length(t)
# Compute availability per link path:
PercAvail <- c(NA)
for (i in 1:N_linkpaths)
{
cond <- which(X_middleLinks==Coord_uniq[i,1] & Y_middleLinks==Coord_uniq[i,2] & Rmean>=0)
# Determine number of intervals with data, i.e. where the link path has at least 1 observation of Rmean>=0:
NumberIntervals <- length(unique(Data$DateTime[cond]))
PercAvail[i] <- NumberIntervals / N_t * 100
}
StepPercAvail <- 10
pdf(FigNameBarplotAvailabilityLinkPaths,family="Times")
par(pty="s")
par(ps=24)
par(mar=c(5,5,1,1)+0.1)
PercAvaildata <- c(NA)
NrAvailClass <- 0
for (availclass in seq(StepPercAvail,100,StepPercAvail))
{
NrAvailClass <- NrAvailClass + 1
if (availclass==StepPercAvail)
{
# This is done to make sure that 0 availability is also shown in the graph.
PercAvaildata[NrAvailClass] <- length(PercAvail[PercAvail>=(availclass-StepPercAvail)&PercAvail<=availclass])
}
if (availclass>StepPercAvail)
{
PercAvaildata[NrAvailClass] <- length(PercAvail[PercAvail>(availclass-StepPercAvail)&PercAvail<=availclass])
}
}
total=sum(PercAvaildata)
perc=100*PercAvaildata/total
ylimbarplot <- c(0,max(perc))
barplot(perc,xlab = "Data availability link paths (%)", ylab = "Percentage",
xaxt="n",tcl=.5,ylim=ylimbarplot,cex.lab=1.3)
at <- 0.1
axis(side = 1, at = at, labels = 0, cex.axis = 0.6)
for (availclass in seq(StepPercAvail,100,StepPercAvail))
{
at <- at + 1.2
axis(side = 1, at = at, labels = availclass, cex.axis = 0.6)
}
legend("topleft",legend=substitute("Mean" == value*"%",list(value=formatC(mean(PercAvail), format="f", digits=2))),bty="n",cex=1)
dev.off()
# Plot with average number of sub-links and link paths for each time interval of the sampling strategy.
UniqueTimestamp <- sort(unique(Data$DateTime))
# Determine for each unique timestamp the number of sub-links with path-average rainfall intensities.
NumberSubLinks <- NumberPaths <- c(NA)
s <- 0
for (Timestamp in UniqueTimestamp)
{
s <- s + 1
cond <- which(Data$DateTime==Timestamp & Rmean>=0)
NumberSubLinks[s] <- length(cond)
# Determine coordinates of middle of link:
X_middleLinks <- (Data$XStart[cond] + Data$XEnd[cond]) / 2
Y_middleLinks <- (Data$YStart[cond] + Data$YEnd[cond]) / 2
# Determine unique middle of links:
Coord_dataf <- data.frame(cbind(X_middleLinks, Y_middleLinks))
Coord_uniq <- unique(Coord_dataf)
# Compute average number of unique link paths:
NumberPaths[s] <- (length(Coord_uniq[, 1]))
}
Timestamp_data = strptime(UniqueTimestamp,"%Y%m%d%H%M",tz=TimeZone)
pdf(FigNameTimeseriesAvailability,width=13.2,height=4.4,family="Times")
par(mar=c(3,3.8,1.8,0.1))
par(ps=ps)
plot(Timestamp_data,NumberPaths,xlab="",ylab="",type="p",ylim=c(0,max(NumberSubLinks)),mgp=c(1.8,0.5,0),xaxt="n",yaxt="n",lwd=1,lty=1,col="gray",pch=16,cex=0.4)
par(new=TRUE)
plot(Timestamp_data,NumberSubLinks,xlab="Date",ylab="Mean number per interval",type="p",ylim=c(0,max(NumberSubLinks)),cex.lab=cex.lab,cex.axis=cex.axis,col="black",mgp=c(1.8,0.5,0),lwd=1,lty=1,pch=16,cex=0.4)
legend("bottomright",legend=c("Sub-links","Link paths"),col=c("black","gray"),pch=c(16,16),bty="n",cex=1.15)
dev.off()
print(paste("Average number of sub-links:",mean(NumberSubLinks)))
print(paste("Average number of link paths:",mean(NumberPaths)))
# A missing interval means that no timestamp and associated number of link (paths) is available, i.e. also no "NA" value is provided.
# In such a case, the other intervals are correctly plotted, because the horizontal axis is an object of class "POSIXlt" (a date time object).
}
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.