Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Why must there be a unique solution to a linear constant-coefficient fractional differential equation of order $(n,q)$ with $\lceil\frac{n}{q}\rceil$ initial conditions? (All notation is as in Miller & Ross, An Introduction to the Fractional Calculus and Fractional DEs.)

More specifically, if $n$ and $q$ are positive integers and $v=\frac{1}{q}$ and $N=\lceil nv\rceil$ and $P(x)$ is a monic polynomial of degree $n$, consider the fractional DE $$P(_0D_t^v)y(t)=x(t)$$ $$y(0)=y'(0)=\dots=y^{(N-1)}(0)=0.$$ I can prove that $y(t)=\int_0^tK(t-\xi)x(\xi)d\xi$ is a solution, where $K$ is the fractional Green's function (inverse Laplace transform of $\frac{1}{P(s^v)}$), but Miller & Ross say it's the unique solution, and I can't see why!

I've been thinking of a proof along the lines of:

Assume there are 2 solutions, say $y_1(t)$ and $y_2(t)$. Then $y_1-y_2$ must satisfy the corresponding homogeneous DE, with the same initial conditions. But why does this mean $y_1-y_2\equiv0$? Is there some theorem from ODE theory about the dimension of the solution space which extends to cover fractional DEs - and if so, why does it extend?

Many thanks for any help with this!

share|improve this question
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.