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.

Given an elliptic curve with a Weierstrass equation, is there any easy way to see whether it has got an isogeny of low degree?

share|improve this question
    
Is this elliptic curve defined over $\mathbb{Q}$? Are you looking for $\mathbb{Q}$-isogenies? –  Álvaro Lozano-Robledo Mar 7 '12 at 15:16
    
Yes, for a start. –  Alan Lee Mar 7 '12 at 15:47

1 Answer 1

up vote 2 down vote accepted

You can use the so-called classical modular polynomials $\Phi_n(x,y)$. Suppose we have an elliptic curve $E$ with $j$-invariant $j(E)$, and suppose there is an isogeny $\phi:E\to E'$, to a second elliptic curve $E'$ with $j$-invariant $j(E')$, and the isogeny has degree $n$. Then, $\Phi_n(j(E),j(E'))=0.$

So, if you have an elliptic curve $E$ and you know its $j$-invariant $j(E)$, then you can study the polynomials $$p_{E,n}(y)= \Phi_n(j(E),y),$$ for each $n\geq 1$, since the roots of these polynomials are $j$-invariants of curves isogenous to $E$.

For instance, let $E/\mathbb{Q}: y^2=x^3+x^2+x$, with $j(E)=2048/3$. We know that $E$ is $2$-isogenous to $E':y^2=x^3-2x^2-3x$, via $\phi(x,y)=(y^2/x^2,y(1-x^2)/x^2).$ Let us see that we could retrieve $E'$ from the classical modular polynomials $\Phi_2(x,y)$. This polynomial is given by $$\Phi_2(x,y)=x^3 - x^2y^2 + 1488x^2y - 162000x^2 + 1488xy^2 + 40773375xy + 8748000000x + y^3 - 162000y^2 + 8748000000y - 157464000000000.$$ When we evaluate $x=2048/3$ we obtain: $$p_{E,2}(y)=y^3 + 3489968y^2/9 + 111828246784y/3 - 4092314705809408/27,$$ and $p_{E,2}(y)$ factors as: $$p_{E,2}(y)=(y-35152/9)(y^2 + 391680y + 116417691904/3).$$ Thus, $E$ is isogenous to a curve $E'$ with $j$-invariant $j(E')=35152/9$. Now you can check that the curve $E'$ given above has this $j$-invariant. There are $2$ other $j$-invariants of curves that are $2$-isogenous to $E$, but those are defined over quadratic extensions of $\mathbb{Q}$.

share|improve this answer
    
That answers my question perfectly. But just how are classical modular polynomials defined? –  Alan Lee Mar 7 '12 at 18:27
    
That's a different question! It is a polynomial relation between the functions $j(\tau)$ and $j(n\tau)$. –  Álvaro Lozano-Robledo Mar 7 '12 at 20:28
    
Is there an explicit expression for these polynomials (at least for low-index ones)? –  Alan Lee Mar 8 '12 at 21:30
    
Yes, see Magma's documentation: magma.maths.usyd.edu.au/magma/handbook/text/1421#15581 –  Álvaro Lozano-Robledo Mar 8 '12 at 23:01
    

Your Answer

 
discard

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.