Code covered by the BSD License
-
[ctimes, cval]=linjpcut(jmpti...
LINJPCUT Truncate piecewise linear functions at a given time.
-
[ctimes, cval]=staircut(jmpti...
STAIRCUT Truncate piecewise constant functions at a given time.
-
[ctimes, le_ij, cle_ij, exi, ...
STTIMESCUT Truncate nondecreasing sequences at a given point.
-
countpath(cdir)
COUNTPATH add the counting processes and random number
-
distrmu(distr, dpar)
DISTRMU a table-lookup function. For a given handle to an external
-
distrstat(distr, dpar)
-
lracc(sc_type, alpha, refvar,...
LRACC Compute the "acceleration factor": the ratio between
-
lrscales(sc_type, alpha, refv...
LRSCALES Compute the time and space scales according to the
-
onoff(nproc, maxtime, on_dist...
ONOFF generate N independent stationary on-off processes. An
-
rencount(nproc, maxtime, dist...
% RENCOUNT Simulate independent renewal counting processes
-
renewpp(nproc, maxtime, distr...
RENEWPP Generate a matrix of N independent renewal point
-
renrew(nproc, maxtime, ren1_d...
% RENREW Generate N independent renewal reward processes
-
renuni(nproc, maxtime)
RENUNI Generate a matrix of N independent renewal counting
-
scalemginfty(sc_type, lambda,...
SCALEMGINFTY Simulate the rescaled and centered cumulative
-
scaleonoff(sc_type, maxtime, ...
SCALEONOFF Simulate a rescaled and centered sum of integrated
-
scaleren(sc_type, alpha, ren_...
SCALEREN Simulate a rescaled and centered sum of independent
-
servpath(sdir)
SERVPATH add the counting processes and random number
-
simbinom(npoints, n, p)
SIMBINOM random numbers from binomial distribution
-
simdiscr(npoints, pdf, val)
SIMDISCR random numbers from a discrete random
-
simexp(M, N, lambda)
SIMEXP random numbers from exponential distribution
-
simgeom(npoints, p)
SIMGEOM random numbers from geometric distribution
-
simlinear(M, N)
-
simmd1(tmax, lambda)
SIMMD1 simulate a M/D/1 queueing system. Poisson arrivals
-
simmg1(tmax, lambda)
SIMMG1 simulate a M/G/1 queueing system. Poisson arrivals
-
simmginfty(tmax, lambda)
SIMMGINFTY simulate a M/G/infinity queueing system. Arrivals are
-
simmm1(n, lambda, mu)
SIMMM1 simulate a M/M/1 queueing system. Poisson arrivals of
-
simpareto(M, N, alpha)
SIMPARETO random numbers from Pareto distribution:
-
simparetonrm(M, N, alpha, gam...
SIMPARETONRM Generate a matrix of random numbers from the
-
simstmginfty(maxtime, lambda,...
SIMSTMGINFTY generate the system size process in a stationary
-
stairintegr(jptimes, fval, st...
% STAIRINTEGR Integrate piecewise constant (stair) functions. The results
-
stairsum(jmptimes, fval)
% STAIRSUM Add piecewise constant (stair) functions.
-
stintresc(jmptimes, fval, ...
STINTRESC Integrate and rescale a piecewice constant function:
-
strescint(jmptimes, fval, ...
STRESCINT Rescale and integrate a piecewice constant function:
-
stsumplot(jptimes, fval, stim...
STSUMPLOT Plot piecewise constant functions and their sum. The
-
telepath(tdir)
TELEPATH add the teletraffic models directories to the path.
-
simgeod1.m
-
View all files
from
Aggregated Teletraffic Models with Long-range Dependence
by Ingemar Kaj Raimundas Gaigalas
Simulates aggregated teletraffic: infinite source Poisson model, on-off model, sum of renewal proc
|
| stairintegr(jptimes, fval, startv)
|
function [ival] = stairintegr(jptimes, fval, startv)
%
% STAIRINTEGR Integrate piecewise constant (stair) functions. The results
% are piecewise linear functions.
% Both the piecewise constant and piecewise linear functions are stored
% in two matrices containing the jump or break points and function
% values at these points respectively. If they have different number of
% jumps, each vector is padded with the last value to have the same
% length. All functions are assumed to be right-continuous.
%
% [ival] = stairintegr(jptimes, fval [, startv])
%
% Inputs:
% jptimes - a matrix of the jump times stored columnwise
% fval - a matrix of function values at the jump points stored
% columnwise
% startv - optional; start value to add to the integral;
% a row vector with length equal to size(jptimes, 2). The
% default value is zeros(1, size(jptimes, 2)).
%
% Outputs:
% ival - a matrix of the values of the integral at the jump
% points. The matrix of break points is equal to jptimes.
%
% See also STAIRCUT, STAIRSUM.
% Authors: R.Gaigalas, I.Kaj
% v2.0 17-Oct-05
% start value is zero if omitted
if (nargin==2)
startv = zeros(1, size(jptimes, 2));
end
% for the linear function F=Int(f) we have delta_F=f(x)*delta_x
ival = cumsum([startv; fval(1:end-1, :).*diff(jptimes)]);
|
|
Contact us at files@mathworks.com