daudaudaudau said:
Obviously such a problem has infinitely many solutions.
Could you state one for general [tex]f[/tex] and [tex]g[/tex]?What I would do first is assume we can separate [tex]K(x,t)=F(x)G(t)[/tex]
And also assume that [tex]f[/tex] and [tex]F[/tex] are differentiable.
Then you find that [tex]\frac{f(x)}{F(x)} = \int^R_x G(t)g(t)dt[/tex]
let [tex]\hat{f} = \frac{f(x)}{F(x)}[/tex]
and taking the derivative with the FTC you get that:
[tex]\hat{f}' = -G(x)g(x)[/tex]
After some calculation you can devise the non-linear ODE:
[tex]F'(x) - \frac{F^2(x)G(x)g(x)}{f(x)} - \frac{f'(x)F(x)}{f(x)} = 0[/tex]
So with specific choices for [tex]G[/tex] i.e. [tex]G=\frac{1}{F}[/tex] or[tex]G=\frac{1}{F^2}[/tex] you can get certain solutions.
for example the choice [tex]G=\frac{1}{F}[/tex] leads to the solution:
[tex]K(x,t) = e^{h(x)-h(t)}[/tex]**
where [tex]h(y)=\int^y_a \frac{g(s)+f'(s)}{f(s)}ds[/tex] for any [tex]a[/tex] in the shared domain of [tex]f,g,and\ f'[/tex]**this is actually: [tex]K(x,t)=e^{\int^x_t\frac{g(s)+f'(s)}{f(s)}ds[/tex]
which is just one solution to this problem.
edit: the solution I gave for K can be multiplied by any constant, presumably determined by [tex]R[/tex]