<p>Let <i>p</i> be a prime; using modular polynomial <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\Phi _p\)</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mi mathvariant="normal">Φ</mi> <mi>p</mi> </msub> </math></EquationSource> </InlineEquation>, Satoh (J Ramanujan Math Soc 15(4):247–270, 2000), Vercauteren (Computing zeta functions of curves over finite fields. PhD Thesis, Katholieke Universiteit Leuven, 2003), Gaudry (Algorithmes de comptage de points d’une courbe définie sur un corps fini, 2004, <a href="http://www.loria.fr/gaudry/publis/pano.pdf">http://www.loria.fr/gaudry/publis/pano.pdf</a>) developed several algorithms to compute the canonical lift of an ordinary elliptic curve <i>E</i> over <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(\mathbb {F}_{p^n}\)</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mi mathvariant="double-struck">F</mi> <msup> <mi>p</mi> <mi>n</mi> </msup> </msub> </math></EquationSource> </InlineEquation> with <i>j</i>-invariant not in <InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(\mathbb {F}_{p^2}\)</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mi mathvariant="double-struck">F</mi> <msup> <mi>p</mi> <mn>2</mn> </msup> </msub> </math></EquationSource> </InlineEquation>. When <i>p</i> is constant, the best variant has complexity <InlineEquation ID="IEq4"> <EquationSource Format="TEX">\(\tilde{O}(n m)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mover accent="true"> <mi>O</mi> <mo stretchy="false">~</mo> </mover> <mrow> <mo stretchy="false">(</mo> <mi>n</mi> <mi>m</mi> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation> bit operations to lift <i>E</i> to <i>p</i>-adic precision&#xa0;<i>m</i>. As an application, lifting <i>E</i> to precision <InlineEquation ID="IEq5"> <EquationSource Format="TEX">\(m=O(n)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>m</mi> <mo>=</mo> <mi>O</mi> <mo stretchy="false">(</mo> <mi>n</mi> <mo stretchy="false">)</mo> </mrow> </math></EquationSource> </InlineEquation> allows to recover its cardinality in time <InlineEquation ID="IEq6"> <EquationSource Format="TEX">\(\tilde{O}(n^2)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mover accent="true"> <mi>O</mi> <mo stretchy="false">~</mo> </mover> <mrow> <mo stretchy="false">(</mo> <msup> <mi>n</mi> <mn>2</mn> </msup> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation>. However, taking <i>p</i> into account the complexity is <InlineEquation ID="IEq7"> <EquationSource Format="TEX">\(\tilde{O}(p^2 n m)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mover accent="true"> <mi>O</mi> <mo stretchy="false">~</mo> </mover> <mrow> <mo stretchy="false">(</mo> <msup> <mi>p</mi> <mn>2</mn> </msup> <mi>n</mi> <mi>m</mi> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation>, so Satoh’s algorithm can only be applied to small&#xa0;<i>p</i>. We propose in this paper two variants of these algorithms, which do not rely on the modular polynomial, for computing the canonical lift of an ordinary curve. Our new method yields a complexity of <InlineEquation ID="IEq8"> <EquationSource Format="TEX">\(\tilde{O}(p n m)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mover accent="true"> <mi>O</mi> <mo stretchy="false">~</mo> </mover> <mrow> <mo stretchy="false">(</mo> <mi>p</mi> <mi>n</mi> <mi>m</mi> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation> to lift at precision&#xa0;<i>m</i>, and even <InlineEquation ID="IEq9"> <EquationSource Format="TEX">\(\tilde{O}(p^{0.5} nm)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mover accent="true"> <mi>O</mi> <mo stretchy="false">~</mo> </mover> <mrow> <mo stretchy="false">(</mo> <msup> <mi>p</mi> <mrow> <mn>0.5</mn> </mrow> </msup> <mi>n</mi> <mi>m</mi> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation> when we are provided a rational point of <i>p</i>-torsion on the curve. This allows us to extend Satoh’s point counting algorithm to larger&#xa0;<i>p</i>.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Towards computing canonical lifts of ordinary elliptic curves in medium characteristic

  • Abdoulaye Maïga,
  • Damien Robert,
  • Djiby Sow

摘要

Let p be a prime; using modular polynomial \(\Phi _p\) Φ p , Satoh (J Ramanujan Math Soc 15(4):247–270, 2000), Vercauteren (Computing zeta functions of curves over finite fields. PhD Thesis, Katholieke Universiteit Leuven, 2003), Gaudry (Algorithmes de comptage de points d’une courbe définie sur un corps fini, 2004, http://www.loria.fr/gaudry/publis/pano.pdf) developed several algorithms to compute the canonical lift of an ordinary elliptic curve E over \(\mathbb {F}_{p^n}\) F p n with j-invariant not in \(\mathbb {F}_{p^2}\) F p 2 . When p is constant, the best variant has complexity \(\tilde{O}(n m)\) O ~ ( n m ) bit operations to lift E to p-adic precision m. As an application, lifting E to precision \(m=O(n)\) m = O ( n ) allows to recover its cardinality in time \(\tilde{O}(n^2)\) O ~ ( n 2 ) . However, taking p into account the complexity is \(\tilde{O}(p^2 n m)\) O ~ ( p 2 n m ) , so Satoh’s algorithm can only be applied to small p. We propose in this paper two variants of these algorithms, which do not rely on the modular polynomial, for computing the canonical lift of an ordinary curve. Our new method yields a complexity of \(\tilde{O}(p n m)\) O ~ ( p n m ) to lift at precision m, and even \(\tilde{O}(p^{0.5} nm)\) O ~ ( p 0.5 n m ) when we are provided a rational point of p-torsion on the curve. This allows us to extend Satoh’s point counting algorithm to larger p.