| 1 | """ |
|---|
| 2 | p-adic L-functions of elliptic curves |
|---|
| 3 | |
|---|
| 4 | AUTHORS: |
|---|
| 5 | -- William Stein (2007-01-01): first version |
|---|
| 6 | -- chris wuthrich (22/05/2007): changed minor issued and added supersingular things |
|---|
| 7 | """ |
|---|
| 8 | |
|---|
| 9 | ###################################################################### |
|---|
| 10 | # Copyright (C) 2007 William Stein <wstein@gmail.com> |
|---|
| 11 | # |
|---|
| 12 | # Distributed under the terms of the GNU General Public License (GPL) |
|---|
| 13 | # |
|---|
| 14 | # This code is distributed in the hope that it will be useful, |
|---|
| 15 | # but WITHOUT ANY WARRANTY; without even the implied warranty of |
|---|
| 16 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
|---|
| 17 | # General Public License for more details. |
|---|
| 18 | # |
|---|
| 19 | # The full text of the GPL is available at: |
|---|
| 20 | # |
|---|
| 21 | # http://www.gnu.org/licenses/ |
|---|
| 22 | ###################################################################### |
|---|
| 23 | |
|---|
| 24 | |
|---|
| 25 | from sage.rings.integer_ring import ZZ |
|---|
| 26 | from sage.rings.rational_field import QQ |
|---|
| 27 | from sage.rings.padics.factory import Qp, Zp |
|---|
| 28 | from sage.rings.infinity import infinity |
|---|
| 29 | from sage.rings.all import LaurentSeriesRing, PowerSeriesRing, PolynomialRing, Integers |
|---|
| 30 | |
|---|
| 31 | from sage.rings.integer import Integer |
|---|
| 32 | from sage.rings.arith import valuation, binomial |
|---|
| 33 | |
|---|
| 34 | from sage.structure.sage_object import SageObject |
|---|
| 35 | |
|---|
| 36 | from sage.misc.all import verbose, denominator, get_verbose |
|---|
| 37 | from sage.databases.cremona import parse_cremona_label |
|---|
| 38 | from sage.schemes.elliptic_curves.constructor import EllipticCurve |
|---|
| 39 | import sage.rings.arith as arith |
|---|
| 40 | |
|---|
| 41 | from sage.modules.free_module_element import vector |
|---|
| 42 | import sage.matrix.all as matrix |
|---|
| 43 | import monsky_washnitzer |
|---|
| 44 | from sage.interfaces.all import gp |
|---|
| 45 | from sage.misc.functional import log |
|---|
| 46 | |
|---|
| 47 | class pAdicLseries(SageObject): |
|---|
| 48 | """ |
|---|
| 49 | The p-adic L-series of an elliptic curve. |
|---|
| 50 | |
|---|
| 51 | EXAMPLES: |
|---|
| 52 | An ordinary example: |
|---|
| 53 | sage: e = EllipticCurve('389a') |
|---|
| 54 | sage: L = e.padic_lseries(5) |
|---|
| 55 | sage: L.series(0) |
|---|
| 56 | Traceback (most recent call last): |
|---|
| 57 | ... |
|---|
| 58 | ValueError: n (=0) must be a positive integer |
|---|
| 59 | sage: L.series(1) |
|---|
| 60 | O(T^1) |
|---|
| 61 | sage: L.series(2) |
|---|
| 62 | O(5^4) + O(5^1)*T + (4 + O(5))*T^2 + (2 + O(5))*T^3 + (3 + O(5))*T^4 + O(T^5) |
|---|
| 63 | sage: L.series(3, prec=10) |
|---|
| 64 | O(5^5) + O(5^2)*T + (4 + 4*5 + O(5^2))*T^2 + (2 + 4*5 + O(5^2))*T^3 + (3 + O(5^2))*T^4 + (1 + O(5))*T^5 + (3*5 + O(5^2))*T^6 + (4 + 5 + O(5^2))*T^7 + (2 + 5 + O(5^2))*T^8 + O(5^2)*T^9 + O(T^10) |
|---|
| 65 | |
|---|
| 66 | A prime p such that E[p] is reducible: |
|---|
| 67 | sage: L = EllipticCurve('11a').padic_lseries(5) |
|---|
| 68 | sage: L.series(1) |
|---|
| 69 | 5 + O(5^2) + O(T) |
|---|
| 70 | sage: L.series(2) |
|---|
| 71 | 5 + 4*5^2 + O(5^3) + O(5^0)*T + O(5^0)*T^2 + O(5^0)*T^3 + O(5^0)*T^4 + O(T^5) |
|---|
| 72 | sage: L.series(3) |
|---|
| 73 | 5 + 4*5^2 + 4*5^3 + O(5^4) + O(5^1)*T + O(5^1)*T^2 + O(5^1)*T^3 + O(5^1)*T^4 + O(T^5) |
|---|
| 74 | """ |
|---|
| 75 | def __init__(self, E, p, normalize): |
|---|
| 76 | """ |
|---|
| 77 | INPUT: |
|---|
| 78 | E -- an elliptic curve |
|---|
| 79 | p -- a prime of good reduction |
|---|
| 80 | normalize -- (bool, default: True); whether or not to correctly |
|---|
| 81 | normalize the L-series, up to a power of -1 and 2. |
|---|
| 82 | If False computations may be faster. |
|---|
| 83 | """ |
|---|
| 84 | self._E = E |
|---|
| 85 | self._p = ZZ(p) |
|---|
| 86 | self._normalize = normalize |
|---|
| 87 | if not self._p.is_prime(): |
|---|
| 88 | raise ValueError, "p (=%s) must be a prime"%p |
|---|
| 89 | if E.conductor() % (self._p)**2 == 0: |
|---|
| 90 | raise NotImplementedError, "p (=%s) must be a prime of semi-stable reduction"%p |
|---|
| 91 | |
|---|
| 92 | # this factor adjusts the p-adic L-series so that it is correct for any element of the isogeny class |
|---|
| 93 | crla = parse_cremona_label(E.label()) |
|---|
| 94 | cr0 = Integer(crla[0]).str() + crla[1] + '1' |
|---|
| 95 | E0 = EllipticCurve(cr0) |
|---|
| 96 | q = E0.period_lattice()[0]/E.period_lattice()[0]*E0.real_components()/E.real_components() |
|---|
| 97 | self._quotient_of_periods = QQ(gp.bestappr(q,200)) |
|---|
| 98 | |
|---|
| 99 | self._modular_symbol = E.modular_symbol(sign=1, normalize=normalize) |
|---|
| 100 | |
|---|
| 101 | def elliptic_curve(self): |
|---|
| 102 | """ |
|---|
| 103 | Return the elliptic curve to which this p-adic L-series is associated. |
|---|
| 104 | |
|---|
| 105 | EXAMPLES: |
|---|
| 106 | sage: L = EllipticCurve('11a').padic_lseries(5) |
|---|
| 107 | sage: L.elliptic_curve() |
|---|
| 108 | Elliptic Curve defined by y^2 + y = x^3 - x^2 - 10*x - 20 over Rational Field |
|---|
| 109 | """ |
|---|
| 110 | return self._E |
|---|
| 111 | |
|---|
| 112 | def prime(self): |
|---|
| 113 | """ |
|---|
| 114 | EXAMPLES: |
|---|
| 115 | sage: L = EllipticCurve('11a').padic_lseries(5) |
|---|
| 116 | sage: L.prime() |
|---|
| 117 | 5 |
|---|
| 118 | """ |
|---|
| 119 | return self._p |
|---|
| 120 | |
|---|
| 121 | def _repr_(self): |
|---|
| 122 | """ |
|---|
| 123 | Return print representation. |
|---|
| 124 | |
|---|
| 125 | sage: e = EllipticCurve('37a') |
|---|
| 126 | sage: e.padic_lseries(3) |
|---|
| 127 | 3-adic L-series of Elliptic Curve defined by y^2 + y = x^3 - x over Rational Field |
|---|
| 128 | sage: e.padic_lseries(3,normalize=False) |
|---|
| 129 | 3-adic L-series of Elliptic Curve defined by y^2 + y = x^3 - x over Rational Field (not normalized) |
|---|
| 130 | sage: L = e.padic_lseries(3,normalize=False) |
|---|
| 131 | sage: L.rename('(factor)*L_3(T)') |
|---|
| 132 | sage: L |
|---|
| 133 | (factor)*L_3(T) |
|---|
| 134 | """ |
|---|
| 135 | s = "%s-adic L-series of %s"%(self._p, self._E) |
|---|
| 136 | if not self._normalize: |
|---|
| 137 | s += ' (not normalized)' |
|---|
| 138 | return s |
|---|
| 139 | |
|---|
| 140 | def modular_symbol(self, r): |
|---|
| 141 | """ |
|---|
| 142 | Return the modular symbol used to compute this p-adic |
|---|
| 143 | L-series evaluated at r. |
|---|
| 144 | |
|---|
| 145 | EXAMPLES: |
|---|
| 146 | sage: L = EllipticCurve('11a').padic_lseries(5) |
|---|
| 147 | sage: [L.modular_symbol(r) for r in [0,1/5,oo,1/11]] |
|---|
| 148 | [1/5, 6/5, 0, 0] |
|---|
| 149 | """ |
|---|
| 150 | return self._modular_symbol(r) |
|---|
| 151 | |
|---|
| 152 | def measure(self, a, n, prec): |
|---|
| 153 | r""" |
|---|
| 154 | Return the measure on $\ZZ_p^*$ defined by |
|---|
| 155 | $$ |
|---|
| 156 | \mu_{E,\alpha}^+ ( a + p^n \ZZ_p ) = |
|---|
| 157 | \frac{1}{\alpha^n} \modsym{a}{p^n} - \frac{1}{\alpha^{n+1}} \modsym{a}{p^{n-1}} |
|---|
| 158 | $$ |
|---|
| 159 | that is used to define this $p$-adic $L$-function. |
|---|
| 160 | |
|---|
| 161 | INPUT: |
|---|
| 162 | a -- an integer |
|---|
| 163 | n -- a non-negative integer |
|---|
| 164 | prec -- an integer |
|---|
| 165 | |
|---|
| 166 | EXAMPLES: |
|---|
| 167 | sage: E = EllipticCurve('37a') |
|---|
| 168 | sage: L = E.padic_lseries(5) |
|---|
| 169 | sage: L.measure(1,2, prec=9) |
|---|
| 170 | 1 + 4*5 + 2*5^2 + 4*5^3 + 3*5^4 + 5^5 + 4*5^6 + 4*5^7 + 4*5^8 + O(5^9) |
|---|
| 171 | """ |
|---|
| 172 | try: |
|---|
| 173 | p, alpha, z, w, f = self.__measure_data[(n,prec)] |
|---|
| 174 | except (KeyError, AttributeError): |
|---|
| 175 | if not hasattr(self, '__measure_data'): |
|---|
| 176 | self.__measure_data = {} |
|---|
| 177 | p = self._p |
|---|
| 178 | alpha = self.alpha(prec=prec) |
|---|
| 179 | z = 1/(alpha**n) |
|---|
| 180 | w = p**(n-1) |
|---|
| 181 | f = self._modular_symbol |
|---|
| 182 | self.__measure_data[(n,prec)] = (p,alpha,z,w,f) |
|---|
| 183 | |
|---|
| 184 | if self._E.conductor() % p == 0: |
|---|
| 185 | return z * f(a/(p*w)) |
|---|
| 186 | return z * f(a/(p*w)) - (z/alpha) * f(a/w) |
|---|
| 187 | |
|---|
| 188 | def alpha(self, prec=20): |
|---|
| 189 | r""" |
|---|
| 190 | Return a p-adic root $\alpha$ of the polynomial $x^2 - a_p x |
|---|
| 191 | + p$ with $\ord_p(\alpha) < 1$. In the ordinary case this is |
|---|
| 192 | just the unit root. |
|---|
| 193 | |
|---|
| 194 | INPUT: |
|---|
| 195 | prec -- positive integer, the p-adic precision of the root. |
|---|
| 196 | |
|---|
| 197 | EXAMPLES: |
|---|
| 198 | Consider the elliptic curve 37a: |
|---|
| 199 | sage: E = EllipticCurve('37a') |
|---|
| 200 | |
|---|
| 201 | An ordinary prime: |
|---|
| 202 | sage: L = E.padic_lseries(5) |
|---|
| 203 | sage: alpha = L.alpha(10); alpha |
|---|
| 204 | 3 + 2*5 + 4*5^2 + 2*5^3 + 5^4 + 4*5^5 + 2*5^7 + 5^8 + 5^9 + O(5^10) |
|---|
| 205 | sage: alpha^2 - E.ap(5)*alpha + 5 |
|---|
| 206 | O(5^10) |
|---|
| 207 | |
|---|
| 208 | A supersingular prime. |
|---|
| 209 | sage: L = E.padic_lseries(3) |
|---|
| 210 | sage: alpha = L.alpha(10); alpha |
|---|
| 211 | (1 + O(3^10))*alpha |
|---|
| 212 | sage: alpha^2 - E.ap(3)*alpha + 3 |
|---|
| 213 | (O(3^10))*alpha^2 + (O(3^11))*alpha + (O(3^11)) |
|---|
| 214 | |
|---|
| 215 | A reducible prime: |
|---|
| 216 | sage: L = EllipticCurve('11a').padic_lseries(5) |
|---|
| 217 | sage: L.alpha(5) |
|---|
| 218 | 1 + 4*5 + 3*5^2 + 2*5^3 + 4*5^4 + O(5^5) |
|---|
| 219 | """ |
|---|
| 220 | try: |
|---|
| 221 | return self._alpha[prec] |
|---|
| 222 | except AttributeError: |
|---|
| 223 | self._alpha = {} |
|---|
| 224 | except KeyError: |
|---|
| 225 | pass |
|---|
| 226 | E = self._E |
|---|
| 227 | p = self._p |
|---|
| 228 | a_p = E.ap(p) |
|---|
| 229 | K = Qp(p, prec, print_mode='series') |
|---|
| 230 | |
|---|
| 231 | if E.conductor() % p == 0: |
|---|
| 232 | self._alpha[prec] = K(a_p) |
|---|
| 233 | return K(a_p) |
|---|
| 234 | |
|---|
| 235 | R = ZZ['x'] |
|---|
| 236 | f = R([p, -a_p, 1]) |
|---|
| 237 | if E.is_ordinary(p): |
|---|
| 238 | G = f.factor_padic(p, prec+5) |
|---|
| 239 | for pr, e in G: |
|---|
| 240 | a = -pr[0] |
|---|
| 241 | if a.valuation() < 1: |
|---|
| 242 | self._alpha[prec] = K(a) |
|---|
| 243 | return K(a) |
|---|
| 244 | raise ValueError, "bug in p-adic L-function alpha" |
|---|
| 245 | else: # supersingular case |
|---|
| 246 | f = f.change_ring(Qp(p, prec, print_mode='series')) |
|---|
| 247 | a = f.root_field('alpha', check_irreducible=False).gen() |
|---|
| 248 | self._alpha[prec] = a |
|---|
| 249 | return a |
|---|
| 250 | |
|---|
| 251 | def order_of_vanishing(self, proof=True): |
|---|
| 252 | """ |
|---|
| 253 | Return the order of vanishing of this $p$-adic $L$-series. |
|---|
| 254 | |
|---|
| 255 | The output of this function is provably correct, due to a |
|---|
| 256 | theorem of Kato. This function will terminate if and only if |
|---|
| 257 | the Mazur-Tate-Teitelbaum analogue of the BSD conjecture about |
|---|
| 258 | the rank of the curve is true and the subgroup of elements of |
|---|
| 259 | p-power order in the Shafarevich-Tate group of this curve is |
|---|
| 260 | finite. I.e., if this function terminates (with no errors!), |
|---|
| 261 | then you may conclude that the p-adic BSD rank conjecture is |
|---|
| 262 | true and that the p-part of Sha is finite. |
|---|
| 263 | |
|---|
| 264 | NOTE: currently $p$ must be a prime of good ordinary reduction. |
|---|
| 265 | |
|---|
| 266 | EXAMPLES: |
|---|
| 267 | sage: L = EllipticCurve('11a').padic_lseries(3) |
|---|
| 268 | sage: L.order_of_vanishing() |
|---|
| 269 | 0 |
|---|
| 270 | sage: L = EllipticCurve('11a').padic_lseries(5) |
|---|
| 271 | sage: L.order_of_vanishing() |
|---|
| 272 | 0 |
|---|
| 273 | sage: L = EllipticCurve('37a').padic_lseries(5) |
|---|
| 274 | sage: L.order_of_vanishing() |
|---|
| 275 | 1 |
|---|
| 276 | sage: L = EllipticCurve('43a').padic_lseries(3) |
|---|
| 277 | sage: L.order_of_vanishing() |
|---|
| 278 | 1 |
|---|
| 279 | sage: L = EllipticCurve('37b').padic_lseries(3) |
|---|
| 280 | sage: L.order_of_vanishing() |
|---|
| 281 | 0 |
|---|
| 282 | |
|---|
| 283 | We verify that Sha(E)(p) is finite for p=3,5,7 for the |
|---|
| 284 | first curve of rank 2: |
|---|
| 285 | sage: e = EllipticCurve('389a') |
|---|
| 286 | sage: for p in primes(3,10): |
|---|
| 287 | ... print p, e.padic_lseries(p).order_of_vanishing() |
|---|
| 288 | 3 2 |
|---|
| 289 | 5 2 |
|---|
| 290 | 7 2 |
|---|
| 291 | """ |
|---|
| 292 | try: |
|---|
| 293 | return self.__ord |
|---|
| 294 | except AttributeError: |
|---|
| 295 | pass |
|---|
| 296 | |
|---|
| 297 | if not self.is_ordinary(): |
|---|
| 298 | raise NotImplementedError |
|---|
| 299 | E = self.elliptic_curve() |
|---|
| 300 | if not E.is_good(self.prime()): |
|---|
| 301 | raise ValueError, "prime must be of good reduction" |
|---|
| 302 | r = E.rank() |
|---|
| 303 | n = 1 |
|---|
| 304 | while True: |
|---|
| 305 | f = self.series(n) |
|---|
| 306 | v = f.valuation() |
|---|
| 307 | if v < r: |
|---|
| 308 | raise RuntimeError, "while computing p-adic order of vanishing, got a contradiction: the curve is %s, the curve has rank %s, but the p-adic L-series vanishes to order <= %s"%(E, r, v) |
|---|
| 309 | if v == r: |
|---|
| 310 | self.__ord = v |
|---|
| 311 | return v |
|---|
| 312 | n += 1 |
|---|
| 313 | |
|---|
| 314 | |
|---|
| 315 | def _c_bounds(self, n): |
|---|
| 316 | raise NotImplementedError |
|---|
| 317 | |
|---|
| 318 | def _prec_bounds(self, n,prec): |
|---|
| 319 | raise NotImplementedError |
|---|
| 320 | |
|---|
| 321 | def teichmuller(self, prec): |
|---|
| 322 | r""" |
|---|
| 323 | Return Teichmuller lifts to the given precision. |
|---|
| 324 | |
|---|
| 325 | INPUT: |
|---|
| 326 | prec -- a positive integer. |
|---|
| 327 | |
|---|
| 328 | OUTPUT: |
|---|
| 329 | the cached Teichmuller lifts |
|---|
| 330 | |
|---|
| 331 | EXAMPLES: |
|---|
| 332 | sage: L = EllipticCurve('11a').padic_lseries(7) |
|---|
| 333 | sage: L.teichmuller(1) |
|---|
| 334 | [0, 1, 2, 3, 4, 5, 6] |
|---|
| 335 | sage: L.teichmuller(2) |
|---|
| 336 | [0, 1, 30, 31, 18, 19, 48] |
|---|
| 337 | """ |
|---|
| 338 | p = self._p |
|---|
| 339 | K = Qp(p, prec, print_mode='series') |
|---|
| 340 | return [Integer(0)] + \ |
|---|
| 341 | [a.residue(prec).lift() for a in K.teichmuller_system()] |
|---|
| 342 | |
|---|
| 343 | def _e_bounds(self, n, prec): |
|---|
| 344 | p = self._p |
|---|
| 345 | prec = max(2,prec) |
|---|
| 346 | R = PowerSeriesRing(ZZ,'T',prec+1) |
|---|
| 347 | T = R(R.gen(),prec +1) |
|---|
| 348 | w = (1+T)**(p**n) - 1 |
|---|
| 349 | return [infinity] + [valuation(w[j],p) for j in range(1,min(w.degree()+1,prec))] |
|---|
| 350 | |
|---|
| 351 | def _get_series_from_cache(self, n, prec): |
|---|
| 352 | try: |
|---|
| 353 | return self.__series[(n,prec)] |
|---|
| 354 | except AttributeError: |
|---|
| 355 | self.__series = {} |
|---|
| 356 | except KeyError: |
|---|
| 357 | for _n, _prec in self.__series.keys(): |
|---|
| 358 | if _n == n and _prec <= prec: |
|---|
| 359 | return self.__series[(_n,_prec)].add_bigoh(prec) |
|---|
| 360 | return None |
|---|
| 361 | |
|---|
| 362 | def _set_series_in_cache(self, n, prec, f): |
|---|
| 363 | self.__series[(n,prec)] = f |
|---|
| 364 | |
|---|
| 365 | |
|---|
| 366 | class pAdicLseriesOrdinary(pAdicLseries): |
|---|
| 367 | def series(self, n=2, prec=5): |
|---|
| 368 | r""" |
|---|
| 369 | Return the $n$-th approximation to the $p$-adic $L$-series as |
|---|
| 370 | a power series in $T$ (corresponding to $\gamma-1$ with |
|---|
| 371 | $\gamma=1+p$ as a generator of $1+p\mathbb{Z}_p$). Each |
|---|
| 372 | coefficient is a $p$-adic number whose precision is provably |
|---|
| 373 | correct. |
|---|
| 374 | |
|---|
| 375 | INPUT: |
|---|
| 376 | n -- (default: 2) a positive integer |
|---|
| 377 | prec -- (default: 5) maxima number of terms of the series |
|---|
| 378 | to compute; to compute as many as possible just |
|---|
| 379 | give a very large number for prec; the result will |
|---|
| 380 | still be correct. |
|---|
| 381 | |
|---|
| 382 | ALIAS: |
|---|
| 383 | power_series is identical to series. |
|---|
| 384 | |
|---|
| 385 | EXAMPLES: |
|---|
| 386 | We compute some $p$-adic $L$-functions associated to the elliptic |
|---|
| 387 | curve 11a. |
|---|
| 388 | |
|---|
| 389 | sage: E = EllipticCurve('11a') |
|---|
| 390 | sage: p = 3 |
|---|
| 391 | sage: E.is_ordinary(p) |
|---|
| 392 | True |
|---|
| 393 | sage: L = E.padic_lseries(p) |
|---|
| 394 | sage: L.series(3) |
|---|
| 395 | 2 + 3 + 3^2 + 2*3^3 + O(3^5) + (1 + 3 + O(3^2))*T + (1 + 2*3 + O(3^2))*T^2 + O(3^1)*T^3 + (2*3 + O(3^2))*T^4 + O(T^5) |
|---|
| 396 | |
|---|
| 397 | |
|---|
| 398 | Another example at a prime of bad reduction, where the |
|---|
| 399 | $p$-adic $L$-function has an extra 0 (compared to the non |
|---|
| 400 | $p$-adic $L$-function). |
|---|
| 401 | |
|---|
| 402 | sage: E = EllipticCurve('11a') |
|---|
| 403 | sage: p = 11 |
|---|
| 404 | sage: E.is_ordinary(p) |
|---|
| 405 | True |
|---|
| 406 | sage: L = E.padic_lseries(p) |
|---|
| 407 | sage: L.series(2) |
|---|
| 408 | O(11^4) + (10 + O(11))*T + (6 + O(11))*T^2 + (2 + O(11))*T^3 + (5 + O(11))*T^4 + O(T^5) |
|---|
| 409 | |
|---|
| 410 | We compute a $p$-adic $L$-function that vanishes to order $2$. |
|---|
| 411 | sage: E = EllipticCurve('389a') |
|---|
| 412 | sage: p = 3 |
|---|
| 413 | sage: E.is_ordinary(p) |
|---|
| 414 | True |
|---|
| 415 | sage: L = E.padic_lseries(p) |
|---|
| 416 | sage: L.series(1) |
|---|
| 417 | O(T^1) |
|---|
| 418 | sage: L.series(2) |
|---|
| 419 | O(3^4) + O(3^1)*T + (2 + O(3))*T^2 + O(T^3) |
|---|
| 420 | sage: L.series(3) |
|---|
| 421 | O(3^5) + O(3^2)*T + (2 + 2*3 + O(3^2))*T^2 + (2 + O(3))*T^3 + (1 + 3 + O(3^2))*T^4 + O(T^5) |
|---|
| 422 | """ |
|---|
| 423 | n = ZZ(n) |
|---|
| 424 | if n < 1: |
|---|
| 425 | raise ValueError, "n (=%s) must be a positive integer"%n |
|---|
| 426 | |
|---|
| 427 | |
|---|
| 428 | p = self._p |
|---|
| 429 | #verbose("computing L-series for p=%s, n=%s, and prec=%s"%(p,n,prec)) |
|---|
| 430 | |
|---|
| 431 | bounds = self._prec_bounds(n,prec) |
|---|
| 432 | padic_prec = max(bounds[1:]) + 5 |
|---|
| 433 | verbose("using p-adic precision of %s"%padic_prec) |
|---|
| 434 | |
|---|
| 435 | res_series_prec = min(p**(n-1), prec) |
|---|
| 436 | verbose("using series precision of %s"%res_series_prec) |
|---|
| 437 | |
|---|
| 438 | ans = self._get_series_from_cache(n, res_series_prec) |
|---|
| 439 | if not ans is None: |
|---|
| 440 | verbose("found series in cache") |
|---|
| 441 | return ans |
|---|
| 442 | |
|---|
| 443 | K = QQ |
|---|
| 444 | gamma = K(1 + p) |
|---|
| 445 | R = PowerSeriesRing(K,'T',res_series_prec) |
|---|
| 446 | T = R(R.gen(),res_series_prec ) |
|---|
| 447 | L = R(0) |
|---|
| 448 | one_plus_T_factor = R(1) |
|---|
| 449 | gamma_power = K(1) |
|---|
| 450 | teich = self.teichmuller(padic_prec) |
|---|
| 451 | p_power = p**(n-1) |
|---|
| 452 | |
|---|
| 453 | verbose("Now iterating over %s summands"%((p-1)*p_power)) |
|---|
| 454 | verbose_level = get_verbose() |
|---|
| 455 | count_verb = 0 |
|---|
| 456 | for j in range(p_power): |
|---|
| 457 | s = K(0) |
|---|
| 458 | if verbose_level >= 2 and j/p_power*100 > count_verb + 3: |
|---|
| 459 | verbose("%.2f percent done"%(float(j)/p_power*100)) |
|---|
| 460 | count_verb += 3 |
|---|
| 461 | for a in range(1,p): |
|---|
| 462 | b = teich[a] * gamma_power |
|---|
| 463 | s += self.measure(b, n, padic_prec).lift() |
|---|
| 464 | L += s * one_plus_T_factor |
|---|
| 465 | one_plus_T_factor *= 1+T |
|---|
| 466 | gamma_power *= gamma |
|---|
| 467 | |
|---|
| 468 | # Now create series but with each coefficient truncated |
|---|
| 469 | # so it is proven correct: |
|---|
| 470 | K = Qp(p, padic_prec, print_mode='series') |
|---|
| 471 | R = PowerSeriesRing(K,'T',res_series_prec) |
|---|
| 472 | L = R(L,res_series_prec) |
|---|
| 473 | aj = L.list() |
|---|
| 474 | if len(aj) > 0: |
|---|
| 475 | aj = [aj[0].add_bigoh(padic_prec-2)] + [aj[j].add_bigoh(bounds[j]) for j in range(1,len(aj))] |
|---|
| 476 | L = R(aj,res_series_prec ) * self._quotient_of_periods |
|---|
| 477 | |
|---|
| 478 | self._set_series_in_cache(n, res_series_prec, L) |
|---|
| 479 | |
|---|
| 480 | return L |
|---|
| 481 | |
|---|
| 482 | power_series = series |
|---|
| 483 | |
|---|
| 484 | def is_ordinary(self): |
|---|
| 485 | """ |
|---|
| 486 | Return True if the elliptic that this $L$-function is attached |
|---|
| 487 | to is ordinary. |
|---|
| 488 | |
|---|
| 489 | EXAMPLES: |
|---|
| 490 | sage: L = EllipticCurve('11a').padic_lseries(5) |
|---|
| 491 | sage: L.is_ordinary() |
|---|
| 492 | True |
|---|
| 493 | """ |
|---|
| 494 | return True |
|---|
| 495 | |
|---|
| 496 | def is_supersingular(self): |
|---|
| 497 | """ |
|---|
| 498 | Return True if the elliptic that this L function is attached |
|---|
| 499 | to is supersingular. |
|---|
| 500 | |
|---|
| 501 | EXAMPLES: |
|---|
| 502 | sage: L = EllipticCurve('11a').padic_lseries(5) |
|---|
| 503 | sage: L.is_supersingular() |
|---|
| 504 | False |
|---|
| 505 | """ |
|---|
| 506 | return False |
|---|
| 507 | |
|---|
| 508 | def _c_bound(self): |
|---|
| 509 | try: |
|---|
| 510 | return self.__c_bound |
|---|
| 511 | except AttributeError: |
|---|
| 512 | pass |
|---|
| 513 | E = self._E |
|---|
| 514 | p = self._p |
|---|
| 515 | if E.is_irreducible(p): |
|---|
| 516 | ans = 0 |
|---|
| 517 | else: |
|---|
| 518 | m = E.modular_symbol_space(sign=1) |
|---|
| 519 | b = m.boundary_map().codomain() |
|---|
| 520 | C = b._known_cusps() # all known, since computed the boundary map |
|---|
| 521 | ans = max([valuation(self.modular_symbol(a).denominator(), p) for a in C]) |
|---|
| 522 | self.__c_bound = ans |
|---|
| 523 | return ans |
|---|
| 524 | |
|---|
| 525 | def _prec_bounds(self, n, prec): |
|---|
| 526 | p = self._p |
|---|
| 527 | e = self._e_bounds(n-1, prec) |
|---|
| 528 | c = self._c_bound() |
|---|
| 529 | return [e[j] - c for j in range(len(e))] |
|---|
| 530 | |
|---|
| 531 | |
|---|
| 532 | class pAdicLseriesSupersingular(pAdicLseries): |
|---|
| 533 | def series(self, n=3, prec=5): |
|---|
| 534 | r""" |
|---|
| 535 | Return the $n$-th approximation to the $p$-adic $L$-series as a |
|---|
| 536 | power series in $T$ (corresponding to $\gamma-1$ with |
|---|
| 537 | $\gamma=1+p$ as a generator of $1+p\mathbb{Z}_p$). Each |
|---|
| 538 | coefficient is a $p$-adic number whose precision is probably |
|---|
| 539 | correct. |
|---|
| 540 | |
|---|
| 541 | INPUT: |
|---|
| 542 | n -- (default: 3) a positive integer |
|---|
| 543 | prec -- (default: 5) maxima number of terms of the series |
|---|
| 544 | to compute; to compute as many as possible just |
|---|
| 545 | give a very large number for prec; the result will |
|---|
| 546 | still be correct. |
|---|
| 547 | |
|---|
| 548 | ALIAS: |
|---|
| 549 | power_series is identical to series. |
|---|
| 550 | |
|---|
| 551 | EXAMPLES: |
|---|
| 552 | A superingular example, where we must compute to higher precision to see anything. |
|---|
| 553 | sage: e = EllipticCurve('37a') |
|---|
| 554 | sage: L = e.padic_lseries(3); L |
|---|
| 555 | 3-adic L-series of Elliptic Curve defined by y^2 + y = x^3 - x over Rational Field |
|---|
| 556 | sage: L.series(2) |
|---|
| 557 | O(T^3) |
|---|
| 558 | sage: L.series(4) # takes a long time (several seconds) |
|---|
| 559 | (O(3^1))*alpha + (O(3^2)) + ((O(3^-1))*alpha + (2*3^-1 + O(3^0)))*T + ((O(3^-1))*alpha + (2*3^-1 + O(3^0)))*T^2 + ((O(3^-2))*alpha + (O(3^-1)))*T^3 + ((O(3^-1))*alpha + (3^-1 + O(3^0)))*T^4 + O(T^5) |
|---|
| 560 | sage: L.alpha(2).parent() |
|---|
| 561 | Univariate Quotient Polynomial Ring in alpha over 3-adic Field with capped |
|---|
| 562 | relative precision 2 with modulus (1 + O(3^2))*x^2 + (3 + O(3^3))*x + (3 + O(3^3)) |
|---|
| 563 | """ |
|---|
| 564 | n = ZZ(n) |
|---|
| 565 | if n < 1: |
|---|
| 566 | raise ValueError, "n (=%s) must be a positive integer"%n |
|---|
| 567 | |
|---|
| 568 | p = self._p |
|---|
| 569 | prec = min(p**(n-1), prec) |
|---|
| 570 | bounds = self._prec_bounds(n,prec) |
|---|
| 571 | padic_prec = max(sum(bounds[1:],[])) + 5 |
|---|
| 572 | verbose("using p-adic precision of %s"%padic_prec) |
|---|
| 573 | ans = self._get_series_from_cache(n, prec) |
|---|
| 574 | if not ans is None: |
|---|
| 575 | verbose("found series in cache") |
|---|
| 576 | return ans |
|---|
| 577 | |
|---|
| 578 | alpha = self.alpha(prec=padic_prec) |
|---|
| 579 | K = alpha.parent() |
|---|
| 580 | gamma = 1 + p |
|---|
| 581 | R = PowerSeriesRing(K,'T',prec) |
|---|
| 582 | T = R(R.gen(), prec) |
|---|
| 583 | L = R(0) |
|---|
| 584 | one_plus_T_factor = R(1) |
|---|
| 585 | gamma_power = 1 |
|---|
| 586 | teich = self.teichmuller(padic_prec) |
|---|
| 587 | |
|---|
| 588 | verbose("Now iterating over %s summands"%((p-1)*p**(n-1))) |
|---|
| 589 | verbose_level = get_verbose() |
|---|
| 590 | count_verb = 0 |
|---|
| 591 | for j in range(p**(n-1)): |
|---|
| 592 | s = K(0) |
|---|
| 593 | if verbose_level >= 2 and j/p**(n-1)*100 > count_verb + 3: |
|---|
| 594 | verbose("%.2f percent done"%(float(j)/p**(n-1)*100)) |
|---|
| 595 | count_verb += 3 |
|---|
| 596 | for a in range(1,p): |
|---|
| 597 | b = teich[a] * gamma_power |
|---|
| 598 | s += self.measure(b, n, padic_prec) |
|---|
| 599 | L += s * one_plus_T_factor |
|---|
| 600 | one_plus_T_factor *= 1+T |
|---|
| 601 | gamma_power *= gamma |
|---|
| 602 | |
|---|
| 603 | # Now create series but with each coefficient truncated |
|---|
| 604 | # so it is proven correct: |
|---|
| 605 | L = R(L,prec) |
|---|
| 606 | aj = L.list() |
|---|
| 607 | if len(aj) > 0: |
|---|
| 608 | bj = [aj[0][0].add_bigoh(padic_prec-2) + alpha * aj[0][1].add_bigoh(padic_prec-2)] |
|---|
| 609 | bj += [aj[j][0].add_bigoh(bounds[j][0]) + alpha * aj[j][1].add_bigoh(bounds[j][1]) for j in range(1,len(aj))] |
|---|
| 610 | L = R(bj, prec) |
|---|
| 611 | L = L * self._quotient_of_periods |
|---|
| 612 | self._set_series_in_cache(n, prec, L) |
|---|
| 613 | return L |
|---|
| 614 | |
|---|
| 615 | power_series = series |
|---|
| 616 | |
|---|
| 617 | def is_ordinary(self): |
|---|
| 618 | return False |
|---|
| 619 | |
|---|
| 620 | def is_supersingular(self): |
|---|
| 621 | return True |
|---|
| 622 | |
|---|
| 623 | def _prec_bounds(self, n,prec): |
|---|
| 624 | p = self._p |
|---|
| 625 | e = self._e_bounds(n-1,prec) |
|---|
| 626 | c0 = ZZ(n+2)/2 |
|---|
| 627 | c1 = ZZ(n+3)/2 |
|---|
| 628 | return [[infinity,infinity]] + [[(e[j] - c0).floor(), (e[j] - c1).floor()] for j in range(1,len(e))] |
|---|
| 629 | |
|---|
| 630 | |
|---|
| 631 | def Dp_valued_series(self, n=3, prec=5): |
|---|
| 632 | r""" |
|---|
| 633 | Returns a vector of two components which are p-adic power series. |
|---|
| 634 | The answer v is such that |
|---|
| 635 | $$(1-\varphi)^(-2)* L_p(E,T) = v[1] * \omega + v[2] * \varphi(\omega)$$ |
|---|
| 636 | as an element of the Dieudonne module $D_p(E) = H^1_{dR}(E/\QQ_p)$ where |
|---|
| 637 | $\omega$ is the invariant differential and $\varphi$ is the Frobenius on $D_p(E)$. |
|---|
| 638 | According to the p-adic BSD this function has a zero of order |
|---|
| 639 | rank(E(Q)) and it's leading term is |
|---|
| 640 | \begin{verbatim} |
|---|
| 641 | +- #Sha(E/Q) * Tamagawa product / Torsion^2 * padic height regulator with values in D_p(E). |
|---|
| 642 | \end{verbatim} |
|---|
| 643 | |
|---|
| 644 | |
|---|
| 645 | INPUT: |
|---|
| 646 | n -- (default: 3) a positive integer |
|---|
| 647 | prec -- (default: 5) a positive integer |
|---|
| 648 | |
|---|
| 649 | EXAMPLES: |
|---|
| 650 | sage: E = EllipticCurve('14a') |
|---|
| 651 | sage: L = E.padic_lseries(5) |
|---|
| 652 | sage: L.Dp_valued_series(4) |
|---|
| 653 | (4 + 4*5^2 + O(5^4) + (1 + O(5))*T + (4 + O(5))*T^2 + (1 + O(5))*T^3 + (3 + O(5))*T^4 + O(T^5), O(5^4) + O(5^1)*T + O(5^1)*T^2 + O(5^1)*T^3 + (3 + O(5))*T^4 + O(T^5)) |
|---|
| 654 | """ |
|---|
| 655 | E = self._E |
|---|
| 656 | p = self._p |
|---|
| 657 | lps = self.series(n, prec=prec) |
|---|
| 658 | |
|---|
| 659 | # now split up the series in two lps = G + H * alpha |
|---|
| 660 | R = lps.base_ring().base_ring() # Qp |
|---|
| 661 | QpT , T = PowerSeriesRing(R,'T',prec).objgen() |
|---|
| 662 | G = QpT([lps[n][0] for n in range(0,lps.prec())], prec) |
|---|
| 663 | H = QpT([lps[n][1] for n in range(0,lps.prec())], prec) |
|---|
| 664 | |
|---|
| 665 | # now compute phi |
|---|
| 666 | phi = matrix.matrix([[0,-1/p],[1,E.ap(p)/p]]) |
|---|
| 667 | lpv = vector([G + (E.ap(p))*H , - R(p) * H ]) # this is L_p |
|---|
| 668 | eps = (1-phi)**(-2) |
|---|
| 669 | resu = lpv*eps.transpose() |
|---|
| 670 | return resu |
|---|
| 671 | |
|---|
| 672 | |
|---|
| 673 | def frobenius(self, prec=20, method = "mw"): |
|---|
| 674 | r""" |
|---|
| 675 | This returns a geometric Frobenius $\varphi$ on the Diedonne module $D_p(E)$ |
|---|
| 676 | with respect to the basis $\omega$, the invariant differential and $\eta=x\omega$. |
|---|
| 677 | It satisfies $phi^2 - a_p/p*phi + 1/p = 0$. |
|---|
| 678 | |
|---|
| 679 | INPUT: |
|---|
| 680 | prec -- (default: 20) a positive integer |
|---|
| 681 | method -- either "mw" (default) for Monsky-Washintzer |
|---|
| 682 | or "approx" for the method describedby Bernardi and Perrin-Riou |
|---|
| 683 | (much slower) |
|---|
| 684 | |
|---|
| 685 | |
|---|
| 686 | EXAMPLES: |
|---|
| 687 | sage: E = EllipticCurve('14a') |
|---|
| 688 | sage: L = E.padic_lseries(5) |
|---|
| 689 | sage: phi = L.frobenius(5) |
|---|
| 690 | sage: phi |
|---|
| 691 | [ 2 + 5^2 + 5^4 + O(5^5) 3*5^-1 + 3 + 5 + 4*5^2 + 5^3 + O(5^4)] |
|---|
| 692 | [ 3 + 3*5^2 + 4*5^3 + 3*5^4 + O(5^5) 3 + 4*5 + 3*5^2 + 4*5^3 + 3*5^4 + O(5^5)] |
|---|
| 693 | sage: -phi^2 |
|---|
| 694 | [5^-1 + O(5^4) O(5^4)] |
|---|
| 695 | [ O(5^5) 5^-1 + O(5^4)] |
|---|
| 696 | """ |
|---|
| 697 | E = self._E |
|---|
| 698 | p = self._p |
|---|
| 699 | if method != "mw" and method !="approx": |
|---|
| 700 | raise ValueError, "Unknown method %s."%method |
|---|
| 701 | if method == "approx": |
|---|
| 702 | return self.__phi_bpr(prec=prec) |
|---|
| 703 | if p < 4 and method == "mw": |
|---|
| 704 | print "Warning: If this fails try again using method=\"approx\"" |
|---|
| 705 | Ew = E.integral_weierstrass_model() |
|---|
| 706 | adjusted_prec = monsky_washnitzer.adjusted_prec(p, prec) |
|---|
| 707 | modprecring = Integers(p**adjusted_prec) |
|---|
| 708 | output_ring = Qp(p, prec) |
|---|
| 709 | R, x = PolynomialRing(modprecring, 'x').objgen() |
|---|
| 710 | Q = x**3 + modprecring(Ew.a4()) * x + modprecring(Ew.a6()) |
|---|
| 711 | trace = Ew.ap(p) |
|---|
| 712 | fr = monsky_washnitzer.matrix_of_frobenius(Q, p, adjusted_prec, trace) |
|---|
| 713 | fr = matrix.matrix(output_ring,2,2,fr) |
|---|
| 714 | |
|---|
| 715 | # return a vector for pari's ellchangecurve to pass from e1 to e2 |
|---|
| 716 | def isom(e1,e2): |
|---|
| 717 | if not e1.is_isomorphic(e2): |
|---|
| 718 | raise ValueError, "Curves must be isomorphic." |
|---|
| 719 | usq = (e1.discriminant()/e2.discriminant()).nth_root(6) |
|---|
| 720 | u = usq.sqrt() |
|---|
| 721 | s = (u * e2.a1() - e1.a1() )/ZZ(2) |
|---|
| 722 | r = (usq * e2.a2() - e1.a2() + s**2 + e1.a1()*s)/ZZ(3) |
|---|
| 723 | t = (u**3 * e2.a3() - e1.a3() - e1.a1()*r)/ZZ(2) |
|---|
| 724 | return [u,r,s,t] |
|---|
| 725 | |
|---|
| 726 | v = isom(E,Ew) |
|---|
| 727 | u = v[0] |
|---|
| 728 | r = v[1] |
|---|
| 729 | |
|---|
| 730 | # change basis |
|---|
| 731 | A = matrix.matrix([[u,-r/u],[0,1/u]]) |
|---|
| 732 | frn = A * fr * A**(-1) |
|---|
| 733 | return 1/p*frn |
|---|
| 734 | |
|---|
| 735 | |
|---|
| 736 | # returns the phi using the definition of bernardi-perrin-riou on page 232. |
|---|
| 737 | |
|---|
| 738 | def __phi_bpr(self, prec=0): |
|---|
| 739 | E = self._E |
|---|
| 740 | p = self._p |
|---|
| 741 | if prec > 10: |
|---|
| 742 | print "Warning: Very large value for the precision." |
|---|
| 743 | if prec == 0: |
|---|
| 744 | prec = floor((log(10000)/log(p))) |
|---|
| 745 | verbose("prec set to %s"%prec) |
|---|
| 746 | eh = E.formal() |
|---|
| 747 | om = eh.differential(prec = p**prec+3) |
|---|
| 748 | verbose("differential computed") |
|---|
| 749 | xt = eh.x(prec=p**prec + 3) |
|---|
| 750 | et = xt*om |
|---|
| 751 | # c_(p^k) = cs[k] d... |
|---|
| 752 | cs = [om[p**k-1] for k in range(0,prec+1)] |
|---|
| 753 | ds = [et[p**k-1] for k in range(0,prec+1)] |
|---|
| 754 | delta = 0 |
|---|
| 755 | dpr = 0 |
|---|
| 756 | gamma = 0 |
|---|
| 757 | dga = 0 |
|---|
| 758 | for k in range(1,prec+1): |
|---|
| 759 | # this is the equation eq[0]*x+eq[1]*y+eq[2] == 0 |
|---|
| 760 | # such that delta_ = delta + d^dpr*x ... |
|---|
| 761 | eq = [(p**dpr*cs[k]) % p**k,(-p**dga*ds[k]) % p**k , (delta*cs[k]-gamma*ds[k]-cs[k-1]) % p**k ] |
|---|
| 762 | verbose("valuations : %s"%([x.valuation(p) for x in eq])) |
|---|
| 763 | v = min([x.valuation(p) for x in eq]) |
|---|
| 764 | if v == infinity: |
|---|
| 765 | verbose("no new information at step k=%s"%k) |
|---|
| 766 | else: |
|---|
| 767 | eq = [ZZ(x/p**v) for x in eq] |
|---|
| 768 | verbose("renormalised eq mod p^%s is now %s"%(k-v,eq)) |
|---|
| 769 | if eq[0].valuation(p) == 0: |
|---|
| 770 | l = min(eq[1].valuation(p),k-v) |
|---|
| 771 | if l == 0: |
|---|
| 772 | verbose("not uniquely determined at step k=%s"%k) |
|---|
| 773 | else: |
|---|
| 774 | ainv = eq[0].inverse_mod(p**l) |
|---|
| 775 | delta = delta - eq[2]*ainv*p**dpr |
|---|
| 776 | dpr = dpr + l |
|---|
| 777 | delta = delta % p**dpr |
|---|
| 778 | verbose("delta_prec increased to %s\n delta is now %s"%(dpr,delta)) |
|---|
| 779 | elif eq[1].valuation(p) == 0: |
|---|
| 780 | l = min(eq[0].valuation(p),k-v) |
|---|
| 781 | ainv = eq[1].inverse_mod(p**l) |
|---|
| 782 | gamma = gamma - eq[2]*ainv*p**dga |
|---|
| 783 | dga = dga + l |
|---|
| 784 | gamma = gamma % p**dga |
|---|
| 785 | verbose("gamma_prec increased to %s\n gamma is now %s"%(dga,gamma)) |
|---|
| 786 | else: |
|---|
| 787 | raise RuntimeError, "Bug: no delta or gamma can exist" |
|---|
| 788 | |
|---|
| 789 | # end of approximation of delta and gamma |
|---|
| 790 | delta = Qp(p,dpr)(delta) |
|---|
| 791 | gamma = Qp(p,dga)(gamma) |
|---|
| 792 | verbose("result delta = %s\n gamma = %s\n check : %s"%(delta,gamma, [Qp(3,k)(delta * cs[k] - gamma * ds[k] - cs[k-1]) for k in range(1,prec+1)] )) |
|---|
| 793 | a = delta |
|---|
| 794 | c = -gamma |
|---|
| 795 | d = E.ap(p) - a |
|---|
| 796 | b = (-1/p+a*d)/c |
|---|
| 797 | phi = matrix.matrix([[a,b],[c,d]]) |
|---|
| 798 | return phi |
|---|
| 799 | |
|---|
| 800 | |
|---|
| 801 | def bernardi_sigma_function(self, prec=20): |
|---|
| 802 | r""" |
|---|
| 803 | Return the p-adic sigma function of Bernardi in terms of $z = log(t)$. |
|---|
| 804 | This is the same as padic_sigma with E2 = 0. |
|---|
| 805 | |
|---|
| 806 | EXAMPLES: |
|---|
| 807 | sage: E = EllipticCurve('14a') |
|---|
| 808 | sage: L = E.padic_lseries(5) |
|---|
| 809 | sage: L.bernardi_sigma_function(5) # Todo: some sort of consistency check!? |
|---|
| 810 | z + 1/24*z^3 + 29/384*z^5 - 8399/322560*z^7 - 291743/92897280*z^9 - 4364831/5225472*z^10 + 2172371753/955514880*z^11 - 17875714529/6897623040*z^12 + 2839176621047/1605264998400*z^13 + 32012675789849/10042939146240*z^14 - 367444910151047/89894839910400*z^15 + 973773806885959/241030539509760*z^16 - 33997971208432501/17259809262796800*z^17 - 10331978660756704339/842918229599846400*z^18 + 18601407947897364480389/950670294194847744000*z^19 - 118837570440101901119321/8071784966648129126400*z^20 + O(z^21) |
|---|
| 811 | """ |
|---|
| 812 | E = self._E |
|---|
| 813 | p = self._p |
|---|
| 814 | |
|---|
| 815 | Eh = E.formal() |
|---|
| 816 | lo = Eh.log(prec + 5) |
|---|
| 817 | F = lo.reversion() |
|---|
| 818 | |
|---|
| 819 | S = LaurentSeriesRing(QQ,'z') |
|---|
| 820 | z = S.gen() |
|---|
| 821 | F = F(z) |
|---|
| 822 | xofF = Eh.x(prec + 2)(F) |
|---|
| 823 | #r = ( E.a1()**2 + 4*E.a2() ) / ZZ(12) |
|---|
| 824 | g = (1/z**2 - xofF ).power_series() |
|---|
| 825 | h = g.integral().integral() |
|---|
| 826 | sigma_of_z = z.power_series() * h.exp() |
|---|
| 827 | |
|---|
| 828 | return sigma_of_z |
|---|
| 829 | |
|---|
| 830 | |
|---|
| 831 | def Dp_valued_height(self,prec=20): |
|---|
| 832 | r""" |
|---|
| 833 | Returns the canonical $p$-adic height with values in the Dieudonne module $D_p(E)$. |
|---|
| 834 | It is defined to be |
|---|
| 835 | $$h_{\eta} \cdot \omega - h_{\omega} \cdot \eta$$ |
|---|
| 836 | where $h_{\eta}$ is made out of the sigma function of Bernardi and |
|---|
| 837 | $h_{\omega}$ is $-log^2$. |
|---|
| 838 | The answer $v$ is given as $v[1]*omega + v[2]*eta$. |
|---|
| 839 | The coordinates of $v$ are dependent of the |
|---|
| 840 | Weierstrass equation. |
|---|
| 841 | |
|---|
| 842 | EXAMPLES: |
|---|
| 843 | sage: E = EllipticCurve('53a') |
|---|
| 844 | sage: L = E.padic_lseries(5) |
|---|
| 845 | sage: h = L.Dp_valued_height(7) |
|---|
| 846 | sage: h(E.gens()[0]) |
|---|
| 847 | (2*5 + 3*5^2 + 2*5^3 + 5^4 + 3*5^6 + 3*5^7 + O(5^8), |
|---|
| 848 | 4*5^2 + 4*5^3 + 4*5^5 + 4*5^6 + 2*5^7 + 5^8 + O(5^9)) |
|---|
| 849 | """ |
|---|
| 850 | E = self._E |
|---|
| 851 | p = self._p |
|---|
| 852 | Ehat = E.formal() |
|---|
| 853 | elog = Ehat.log(prec + Integer(3)) |
|---|
| 854 | |
|---|
| 855 | # we will have to do it properly with David Harvey's _DivPolyContext(E, R, Q) |
|---|
| 856 | n = arith.LCM(E.tamagawa_numbers()) |
|---|
| 857 | n = arith.LCM(n, E.Np(p)) # allowed here because E has good reduction at p |
|---|
| 858 | |
|---|
| 859 | if p < 5: |
|---|
| 860 | phi = self.frobenius(min(6,prec),method="approx") |
|---|
| 861 | else: |
|---|
| 862 | phi = self.frobenius(prec+2,method="mw") |
|---|
| 863 | |
|---|
| 864 | def height(P,check=True): |
|---|
| 865 | if P.is_finite_order(): |
|---|
| 866 | return Qp(p,prec)(0) |
|---|
| 867 | if check: |
|---|
| 868 | assert P.curve() == E, "the point P must lie on the curve from which the height function was created" |
|---|
| 869 | Q = n * P |
|---|
| 870 | tt = - Q[0]/Q[1] |
|---|
| 871 | R = Qp(p,prec+5) |
|---|
| 872 | tt = R(tt) |
|---|
| 873 | zz = elog(tt) |
|---|
| 874 | |
|---|
| 875 | homega = -zz**2/n**2 |
|---|
| 876 | |
|---|
| 877 | eQ = denominator(Q[1])/denominator(Q[0]) |
|---|
| 878 | si = self.bernardi_sigma_function(prec=prec+4) |
|---|
| 879 | heta = 2 * log(si(zz)/eQ) / n**2 |
|---|
| 880 | |
|---|
| 881 | R = Qp(p,prec) |
|---|
| 882 | |
|---|
| 883 | return vector([R(heta),-R(homega)]) |
|---|
| 884 | |
|---|
| 885 | return height |
|---|
| 886 | |
|---|
| 887 | |
|---|
| 888 | |
|---|
| 889 | def Dp_valued_regulator(self,prec=20,v1=0,v2=0): |
|---|
| 890 | """ |
|---|
| 891 | Returns the canonical $p$-adic regulator with values in the Dieudonne module $D_p(E)$ |
|---|
| 892 | as defined by Perrin-Riou using the $p$-adic height with values in $D_p(E)$. |
|---|
| 893 | The result is written in the basis $\omega$, $\varphi(\omega)$, and hence the |
|---|
| 894 | coordinates of the result are independent of the chosen Weierstrass equation. |
|---|
| 895 | |
|---|
| 896 | NOTE: |
|---|
| 897 | The definition here is corrected with repect to Perrin-Riou's article |
|---|
| 898 | 'Arithm\'etique des courbes elliptiques \`a r\'eduction supersinguli\`ere en $p$'. |
|---|
| 899 | |
|---|
| 900 | |
|---|
| 901 | EXAMPLES: |
|---|
| 902 | sage: E = EllipticCurve('43a') |
|---|
| 903 | sage: L = E.padic_lseries(7) |
|---|
| 904 | sage: L.Dp_valued_regulator(7) |
|---|
| 905 | (2*7 + 2*7^3 + 2*7^4 + 5*7^5 + 6*7^6 + 2*7^7 + O(7^8), 3*7^2 + 4*7^3 + 3*7^4 + 5*7^5 + 2*7^7 + O(7^8)) |
|---|
| 906 | """ |
|---|
| 907 | |
|---|
| 908 | p = self._p |
|---|
| 909 | E = self._E |
|---|
| 910 | |
|---|
| 911 | h = self.Dp_valued_height(prec=prec) |
|---|
| 912 | |
|---|
| 913 | # this is the height_{v} (P) for a v in D_p |
|---|
| 914 | def hv(vec,P): |
|---|
| 915 | hP = h(P) |
|---|
| 916 | return - vec[0]*hP[1] +vec[1]*hP[0] |
|---|
| 917 | |
|---|
| 918 | # def hvpairing(vec,P,Q): |
|---|
| 919 | # return (hv(vec, P+Q) - hv(vec,P)-hv(vec,Q))/2 |
|---|
| 920 | K = Qp(p, prec) |
|---|
| 921 | |
|---|
| 922 | if v1 ==0 and v2 ==0 : |
|---|
| 923 | v1 = vector([K(0),K(1)]) # that is eta |
|---|
| 924 | v2 = vector([K(-1),K(1)]) # and this is eta-omega. |
|---|
| 925 | # the rest should not depend on this choice |
|---|
| 926 | # as long as it is outside Q_p * omega |
|---|
| 927 | |
|---|
| 928 | rk = E.rank() |
|---|
| 929 | if rk == 0: |
|---|
| 930 | return vector([K(1),K(0)]) |
|---|
| 931 | |
|---|
| 932 | |
|---|
| 933 | basis = E.gens() |
|---|
| 934 | |
|---|
| 935 | def regv(vec): |
|---|
| 936 | M = matrix.matrix(K,rk,rk,0) |
|---|
| 937 | point_height = [hv(vec,P) for P in basis] |
|---|
| 938 | for i in range(rk): |
|---|
| 939 | for j in range(i+1, rk): |
|---|
| 940 | M[i, j] = M[j, i] = (hv(vec,basis[i] + basis[j])- point_height[i] - point_height[j] )/2 |
|---|
| 941 | for i in range(rk): |
|---|
| 942 | M[i,i] = point_height[i] |
|---|
| 943 | |
|---|
| 944 | return M.determinant() |
|---|
| 945 | |
|---|
| 946 | |
|---|
| 947 | def Dp_pairing(vec1,vec2): |
|---|
| 948 | return (vec1[0]*vec2[1]-vec1[1]*vec2[0]) |
|---|
| 949 | |
|---|
| 950 | omega_vec = vector([K(1),K(0)]) |
|---|
| 951 | |
|---|
| 952 | # note the correction here with respect to Perrin-Riou's definition. |
|---|
| 953 | # only this way the result will be indep of the choice of v1 and v2. |
|---|
| 954 | reg1 = regv(v1)/Dp_pairing(omega_vec,v1)**(rk-1) |
|---|
| 955 | |
|---|
| 956 | reg2 = regv(v2)/Dp_pairing(omega_vec,v2)**(rk-1) |
|---|
| 957 | |
|---|
| 958 | |
|---|
| 959 | # the regulator in the basis omega,eta |
|---|
| 960 | reg_oe = (reg1 * v2 - reg2 * v1 ) / Dp_pairing(v2,v1) |
|---|
| 961 | |
|---|
| 962 | if p < 5: |
|---|
| 963 | phi = self.frobenius(min(6,prec),method="approx") |
|---|
| 964 | else: |
|---|
| 965 | phi = self.frobenius(prec+2,method="mw") |
|---|
| 966 | |
|---|
| 967 | c = phi[1,0] # this is the 'period' [omega,phi(omega)] |
|---|
| 968 | a = phi[0,0] |
|---|
| 969 | |
|---|
| 970 | return vector([reg_oe[0] - a/c*reg_oe[1],reg_oe[1]/c]) |
|---|