Companion software for the paper Expansion of the fundamental solution of a second-order elliptic operator with analytic coefficients.
We implement in C++
the construction given in the paper.
g++ -O2 main.cpp -o main
You shall configure the dimension (N
) at the top of the function_representations.hpp file.
Then, in the main()
you shall describe the operator you want to consider.
The example currently implemented is for the operator (in dimension N=3
, with the three variables x_0
, x_1
, x_2
)
Then, executing ./main
will print the first few homogenous terms of the expansion of the fundamental solution for L
.
An operator is a function<AE(AE)>
taking an AsymptoticExpansion=AE
and giving back an AsymptoticExpansion=AE
.
It must be elliptic and its principal symbol must be the same of the Laplacian. See the example in main.cpp
.
The field of coefficient is denoted by F
.
The coefficients defining the operator shall be instances of the class Polynomial
.
A polynomial is a vector of pairs {exponent, coefficient}
.
The exponent is a vector of length N
(where N
is the dimension), and the i-th entry represent the exponent of x_i
.
For example, the polynomial is constructed by {{{0, 0, 0}, F(1)}, {{2, 0, 0}, F(4)}}
.