1: %\input{tcilatex}
2: %\input{tcilatex}
3:
4:
5: \documentclass{amsart}
6: \usepackage{amssymb}
7:
8: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
9: \usepackage{graphicx}
10: \usepackage{amscd}
11:
12: %TCIDATA{OutputFilter=LATEX.DLL}
13: %TCIDATA{Created=Fri Dec 15 08:49:14 2000}
14: %TCIDATA{LastRevised=Thu Sep 25 20:04:59 2003}
15: %TCIDATA{<META NAME="GraphicsSave" CONTENT="32">}
16: %TCIDATA{<META NAME="DocumentShell" CONTENT="Journal Articles\AMS Journal Article">}
17: %TCIDATA{Language=American English}
18: %TCIDATA{CSTFile=amsartci.cst}
19:
20: \newtheorem{theorem}{Theorem}
21: \theoremstyle{plain}
22: \newtheorem{acknowledgement}{Acknowledgement}
23: \newtheorem{algorithm}{Algorithm}
24: \newtheorem{axiom}{Axiom}
25: \newtheorem{case}{Case}
26: \newtheorem{claim}{Claim}
27: \newtheorem{conclusion}{Conclusion}
28: \newtheorem{condition}{Condition}
29: \newtheorem{conjecture}{Conjecture}
30: \newtheorem{corollary}{Corollary}
31: \newtheorem{criterion}{Criterion}
32: \newtheorem{definition}{Definition}
33: \newtheorem{example}{Example}
34: \newtheorem{exercise}{Exercise}
35: \newtheorem{lemma}{Lemma}
36: \newtheorem{notation}{Notation}
37: \newtheorem{problem}{Problem}
38: \newtheorem{proposition}{Proposition}
39: \newtheorem{remark}{Remark}
40: \newtheorem{solution}{Solution}
41: \newtheorem{summary}{Summary}
42: \numberwithin{equation}{section}
43: \numberwithin{theorem}{section}
44: \numberwithin{lemma}{section}
45: \numberwithin{proposition}{section}
46: \numberwithin{corollary}{section}
47: \numberwithin{remark}{section}
48: \numberwithin{example}{section}
49: \numberwithin{definition}{section}
50: \newcommand{\thmref}[1]{Theorem~\ref{#1}}
51: \newcommand{\secref}[1]{\S\ref{#1}}
52: \newcommand{\lemref}[1]{Lemma~\ref{#1}}
53: \input{tcilatex}
54:
55: \begin{document}
56: \title[ Local Zeta Functions and Linear Feedback Shift Registers]{Computing \ Igusa's Local Zeta Functions of Univariate Polynomials, and
57: Linear Feedback Shift Registers}
58: \author{W. A. Zuniga-Galindo}
59:
60: \begin{abstract}
61: We give a polynomial time algorithm for computing the Igusa local zeta
62: function $Z(s,f)$ attached to a polynomial \ $f(x)\in \mathbb{Z}[x]$, in one
63: variable, with splitting field $\mathbb{Q}$, and a prime number $p$. We also
64: propose a new class of Linear Feedback Shift Registers based on the
65: computation of Igusa's local zeta function.
66: \end{abstract}
67:
68: \address{Department of Mathematics and Computer Science, Barry University, 11300 N.E.
69: Second Avenue\\
70: Miami Shores, Florida 33161, USA}
71: \email{wzuniga@mail.barry.edu}
72: \thanks{Supported by COLCIENCIAS-Grant \# 089-2000. }
73: \subjclass{Primary 11S40, 94A60; Secondary 11Y16, 14GG50}
74: \keywords{Igusa's local zeta function, polynomial time algorithms, one-way functions,
75: linear feedback shift registers}
76: \maketitle
77:
78: \section{Introduction}
79:
80: Let $f(x)\in \mathbb{Z}[x]$, $x=(x_{1},\mathbb{\cdots },x_{n})$ be a
81: non-constant polynomial, and $\ p$ a fixed prime number. We put $%
82: N_{m}(f,p)=N_{m}(f)$ for the number of solutions of the congruence $%
83: f(x)\equiv 0$ mod $p^{m}$ in $(\mathbb{Z}/p^{m}\mathbb{Z})^{n}$, $m\geqq 1$,
84: \ and \ $H(t,f)$ for the Poincar\'{e} series
85: \begin{equation*}
86: H(t,f)=\sum\limits_{m=0}^{\infty }N_{m}(f)(p^{-n}t)^{m}\text{,}
87: \end{equation*}
88: with $t\in \mathbb{C}$, $\mid t\mid <1$, and \ $N_{0}(f)=1$. This paper is
89: dedicated to the computation of the sequence $\left\{ N_{m}(f)\right\}
90: _{m\geqq 0}$ when $f$ is an univariate polynomial with splitting field $%
91: \mathbb{Q}$.
92:
93: Igusa showed that the \ Poincar\'{e} series $H(t,f)$ admits a meromorphic
94: continuation to the complex plane as a rational function of $t$ \cite{I1},
95: \cite{I2}. In this paper we make a first step towards the solution of the
96: following problem: given a polynomial \ $f(x)$\ as above, how difficult is
97: to compute the meromorphic continuation of the \ Poincar\'{e} series $H(t,f)$%
98: ?
99:
100: The computation of the \ Poincar\'{e} series $H(t,f)$\ is equivalent to the
101: computation of Igusa's local zeta \ function $Z(s,f)$, attached to $f$ and $%
102: p $, defined as follows. We denote by $\mathbb{Q}_{p}$ the field of $p-$adic
103: numbers, and by $\mathbb{Z}_{p}$ the ring of $p-$adic integers. For $x\in
104: \mathbb{Q}_{p}$, $v_{p}(x)$ denotes the $p-$adic order of $x$, and $%
105: |x|_{p}=p^{-v_{p}(x)}$ its absolute value. The Igusa local zeta function
106: associated to $f$ and $p$ is defined as follows:
107: \begin{equation*}
108: Z(s,f)=\int_{\mathbb{Z}_{p}^{n}}|f(x)|_{p}^{s}\mid dx\mid
109: ,\,\,\,\,\,\,\,\,s\in \mathbb{C},\,\,
110: \end{equation*}
111: where $\func{Re}(s)>0$, and $\mid dx\mid $ denotes the Haar measure on $%
112: \mathbb{Q}_{p}^{n}$ so normalized that $\mathbb{Z}_{p}^{n}$ has measure $1$.
113: The following relation between $Z(s,f)$\ and $\ H(t,f)$ holds (see \cite{I1}%
114: , theorem 8.2.2):
115: \begin{equation*}
116: H(t,f)=\frac{1-tZ(s,f)}{1-t},\text{ }t=p^{-s}.
117: \end{equation*}
118: Thus, the rationality of $Z(s,f)$\ implies the rationality of the
119: Poincar\'{e} series \ $H(t,f)$, and \ the computation of $H(t,f)$ is \
120: equivalent to the computation of$\ Z(s,f)$. Igusa \cite[theorem 8.2.1]{I1}\
121: showed that the local zeta function $Z(s,f)$ \ admits a meromorphic
122: continuation to the complex plane as a rational function \ of $p^{-s}$.
123:
124: The first result of this paper is a polynomial time algorithm for computing
125: the local zeta function $Z(s,f)$ attached to a polynomial \ $f(x)\in \mathbb{%
126: Z}[x]$, in one variable, with splitting field $\mathbb{Q}$, and a prime
127: number $p$. We also give an explicit estimate for its complexity (see
128: algorithm Compute\_$Z(s,f)$ in section 2, and theorem \ref{theoA}).
129:
130: Many authors have found explicit formulas for $Z(s,f)$, or $H(f,t)$,\ for
131: several classes of polynomials, among them \cite{D2}, \cite{D-H}, \cite{G1},
132: \cite{G2}, [\cite{I3} and the references therein], \cite{S-Z-G}, \cite{Z-G1}%
133: , \cite{Z-G2}. In all these works the computation of $Z(s,f)$, or $H(f,t)$,
134: is reduced to the computation of \ other problems, as the computation of the
135: number of solutions of polynomial equations with coefficients in a finite
136: field. Currently, there is no polynomial time algorithm solving this problem
137: \ \cite{Wa}, \cite{V-K-S}. Moreover, none of the above mentioned works
138: include complexity estimates for the computation of Igusa's local zeta
139: functions.
140:
141: Of particular importance is Denef's explicit formula for \ $Z(s,f)$, when $f$
142: satisfies some generic conditions \ \cite{D2}. This formula involves the
143: numerical data associated to a resolution of singularities of the divisor $%
144: f=0$, and the number of rational points of certain non-singular varieties
145: over finite fields. Thus the computation of $Z(s,f),$ for a generic
146: polynomial $f$, is reduced to the computation of \ the numerical data
147: associated to a resolution of singularities of the divisor $f=0$, and the
148: number of solutions of non-singular polynomials over finite fields.
149: Currently, it is unknown if these problems can be solved in polynomial time
150: on a Turing machine. However, during the last few years important
151: achievements have been obtained in the computation of resolution of
152: singularities of polynomials \cite{B-M}, \cite{B-S}, \cite{B-S1}, \cite{Vi}.
153:
154: The computation of the Igusa local zeta function for an arbitrary polynomial
155: seems to be an intractable problem on a Turing machine. For example, for $%
156: p=2 $, the computation of the number of solutions of a polynomial equation
157: with coefficients in $\mathbb{Z}/2\mathbb{Z}$ is an $\mathbf{NP}-$complete
158: problem on a Turing Machine \cite[page 251, problem AN9]{G-J}. Then \ in the
159: case of $2-$adic numbers, the computation of \ the Igusa local zeta function
160: is an $\mathbf{NP}-$complete problem.
161:
162: Recently, Anshel and Goldfeld have shown the existence of a strong
163: connection between the computation of zeta functions and cryptography \cite
164: {A-G}. Indeed, they proposed a new class of candidates for one-way functions
165: based on global zeta functions. \ A one-way function is a function $\ F$
166: such that for each $x$ in the domain of $F$, it is easy to compute $F(x)$;
167: but for essentially all $y$ in the range of $F$, it is an intractable
168: problem to find an $x$ \ such that $y=F(x)$. These functions play a central
169: role, from a practical and theoretical point of view, in modern
170: cryptography. Currently, there is no guarantee that one-way functions exist
171: even if $\mathbf{P}\neq \mathbf{NP}$. Most of the present candidates for
172: one-way functions are constructed on the intractability of problems like
173: integer factorization and discrete logarithms \cite{G-L-N}. Recently, P.
174: Shor has introduced a new approach to attack these problems \cite{S}.
175: Indeed, Shor have shown that on a quantum computer the integer factorization
176: and discrete logarithm \ problems can be computed in polynomial time.
177:
178: We set
179: \begin{equation*}
180: \mathcal{H}=\{H(t,f)\mid f(x)\in \mathbb{Z}[x]\text{, in \ one variable,
181: with splitting field }\mathbb{Q}\},
182: \end{equation*}
183: and $N^{\infty }$ $\left( \mathbb{Z}\right) $ for the set \ of finite
184: sequences of integers. For each positive integer $u$ and a prime number $p$,
185: we define
186: \begin{equation*}
187: \begin{array}{cccc}
188: F_{u,p}: & \mathcal{H} & {\ \rightarrow } & \mathbb{N}^{\infty }\left(
189: \mathbb{Z}\right) \\
190: & H(t,f) & {\ \rightarrow } & \{N_{0}(f,p),N_{1}(f,p),\cdots ,N_{u}(f,p)\}%
191: \text{.}
192: \end{array}
193: \end{equation*}
194:
195: Our second result asserts that \ $F_{u,p}(H(t,f))$ \ can be computed in
196: polynomial time, for every $H(t,f)$\ in $\mathcal{H}$ (see theorem \ref
197: {theoB}). It seems interesting \ to study the complexity on a Turing machine
198: of the following \ problem: given a list of positive integers $%
199: \{a_{0},a_{1},\cdots ,a_{u}\}$, how difficult is it to determine whether or
200: not there exists a Poincar\'{e} series $H(t,f)=\sum_{m=0}^{\infty
201: }N_{m}(f)(p^{-1}t)^{m}$, such that $a_{i}=N_{i}(f)$, $i=1$,$\cdots $, $u$?
202:
203: Currently, the author does not have any result about the complexity of the
204: above problem, however the mappings $F_{u,p}$ can be considered as new \
205: class of stream ciphers (see section 8).
206:
207: \section{The Algorithm \ Compute\_$Z(s,f)$}
208:
209: In this section we present a polynomial time algorithm, Compute\_$Z(s,f)$, \
210: that solves the following problem: given \ a polynomial $f(x)\in \mathbb{Z}%
211: [x]$, in one variable, whose splitting field is $\mathbb{Q}$, \ find an
212: explicit expression for the meromorphic continuation of $Z(s,f)$. The
213: algorithm is as follows.
214:
215: \textbf{Algorithm\ Compute\_}$Z(s,f)$
216:
217: Input\ : A \ polynomial $f(x)\in \mathbb{Z}[x],$\ in one variable, whose
218: splitting field is $\mathbb{Q}$.
219:
220: Output : A rational function of $p^{-s}$\ that is the meromorphic \qquad
221: continuation of $Z(s,f)$.
222:
223: \begin{enumerate}
224: \item Factorize $f(x)$ in $\mathbb{Q}[x]$: $f(x)=\alpha
225: _{0}\prod\limits_{i=1}^{r}(x-\alpha _{i})^{e_{i}}\in \mathbb{Q}[x].\qquad $
226:
227: \item Compute
228: \begin{equation*}
229: l_{f}=\left\{
230: \begin{array}{cc}
231: 1+\max \{v_{p}(\alpha _{i}-\alpha _{j})\mid i\neq j\text{, }1\leq i\text{, }%
232: j\leq r\}\text{,} & \text{ if }r\geqq 2\text{;} \\
233: 1\text{, } & \text{\ if \ \ }r=1\text{.}
234: \end{array}
235: \right.
236: \end{equation*}
237:
238: \item Compute the $p-$adic expansions of the numbers $\alpha _{i}$, $%
239: i=1,2,\cdots ,r$\ modulo $p^{l_{f}+1}$.
240:
241: \item Compute the tree $T(f,l_{f})$\ associated to $f(x)$\ and $p$ (for the
242: definition of $T(f,l_{f})$ see (\ref{tree})).
243:
244: \item Compute the generating function $G(s,T(f,l_{f}),p)$\ attached to $%
245: T(f,l_{f})$ (for the definition of $G(s,T(f,l_{f}),p)$\ see (\ref{Gfunction}%
246: )).
247:
248: \item Return $Z(s,f)=G(s,T(f,l_{f}),p)$.
249:
250: \item End
251: \end{enumerate}
252:
253: In section 6, we shall give a proof of the correctness \ and a \ complexity
254: estimate for the algorithm Compute\_$Z(s,f)$. The first step in our
255: algorithm is accomplished by means \ of the factoring algorithm by A.K.
256: Lenstra, H. Lenstra and L. Lov\'{a}sz \cite{L-L-L}. If $d_{f}$ denotes the
257: degree of $f(x)=\sum\nolimits_{i}a_{i}x^{i}$, and
258: \begin{equation*}
259: \shortparallel f\shortparallel =\sqrt{\sum_{i}a_{i}^{2}},
260: \end{equation*}
261: then\ \ the \ mentioned factoring algorithm needs \ $O\left(
262: d_{f}^{6}+d_{f}^{9}(\log \shortparallel f\shortparallel )\right) $
263: arithmetic operations, and the integers on which these operations are
264: performed each have a binary length \ \ $O\left( d_{f}^{3}+d_{f}^{2}(\log
265: \shortparallel f\shortparallel )\right) $ \cite[theorem 3.6]{L-L-L}.
266:
267: The steps $2$, $3$, $4$, $5$ reduce \ in polynomial time the computation of $%
268: Z(s,f)$ to the computation of a factorization of $f(x)$ over $\mathbb{Q}$.
269: This reduction is accomplished by constructing a weighted tree from the $p-$%
270: adic expansion of the roots of $\ f(x)$ modulo a \ certain power of $p$ (see
271: section 4), and then \ \ associating a generating function to this tree (see
272: section 5). Finally, \ we shall prove that the generating function \
273: constructed in this way coincides with the local zeta function of $f(x)$
274: (see section 5).
275:
276: \section{$p$-adic Stationary Phase Formula}
277:
278: Our main tool in the effective computing of Igusa's local zeta function \ of
279: a polynomial in one variable will be the $p-$adic stationary phase formula,
280: abbreviated SPF \cite{I3}. This formula is a recursive procedure for
281: computing local zeta functions. By using this procedure it is possible to
282: compute the local zeta functions for many classes of polynomials [\cite{I3}
283: and the references therein], \cite{S-Z-G}, \cite{Z-G1}, \cite{Z-G2}, \cite
284: {ZG3}.
285:
286: Given a polynomial $f(x)\in $ $\mathbb{Z}_{p}[x]\setminus $ $p\mathbb{Z}%
287: _{p}[x]$, we denote by $\overline{f(x)}$ its reduction modulo $p\mathbb{Z}%
288: _{p}$, i.e., the polynomial obtained by reducing the coefficients of $f(x)$
289: modulo $p\mathbb{Z}_{p}$. \ We define for each $x_{0}\in $ $\mathbb{Z}_{p}$,
290: \begin{equation*}
291: f_{x_{0}}(x)=p^{-e_{x_{0}}}f(x_{0}+px)\text{,}
292: \end{equation*}
293: where $e_{x_{0}}$\ is the minimum \ order of $p$\ \ \ in the coefficients of
294: $f(x_{0}+px)$. Thus $f_{x_{0}}(x)\in \mathbb{Z}_{p}[x]\setminus $ $p\mathbb{Z%
295: }_{p}[x]$. We shall call the polynomial $f_{x_{0}}(x)$\ the \textit{%
296: dilatation} of $f(x)$\ at $x_{0}$. We also define
297: \begin{equation*}
298: \nu (\overline{f})=\text{Card}\{\overline{z}\in \mathbb{F}_{p}\mid \overline{%
299: f}(\overline{z})\neq 0\},
300: \end{equation*}
301: \begin{equation*}
302: \delta (\overline{f})=\text{Card}\{\overline{z}\in \mathbb{F}_{p}\mid
303: \overline{z}\text{ is a simple root of }\overline{f}(\overline{z})=0\}.
304: \end{equation*}
305:
306: We shall use $\{0,1,\cdots ,p-1\}\subseteq \mathbb{Z}_{p}$ as a set of
307: representatives of the elements of $\mathbb{F}_{p}=\mathbb{Z}/$ $p\mathbb{Z}%
308: =\{\overline{0},\overline{1},\cdots ,\overline{p-1}\}$. Let $S=S(f)$ denote
309: the subset of $\{0,1,\cdots ,p-1\}\subseteq \mathbb{Z}_{p}$ which is mapped
310: bijectively \ by the canonical homomorphism $\mathbb{Z}_{p}\rightarrow
311: \mathbb{Z}_{p}/p\mathbb{Z}_{p}$ to the set of roots of \ $\overline{f}(%
312: \overline{z})=0$ with multiplicity greater than or equal to two.
313:
314: With all the above notation we are able to state the $p-$adic stationary
315: phase formula for polynomials in one variable.
316:
317: \begin{proposition}[{\protect\cite[theorem 10.2.1]{I1}}]
318: Let $f(x)\in $ $\mathbb{Z}_{p}[x]\setminus $ $p\mathbb{Z}_{p}[x]$ be a
319: non-cons\-tant polynomial. Then
320: \begin{equation*}
321: Z(s,f)=p^{-1}\nu (\overline{f})+\delta (\overline{f})\frac{(1-p^{-1})p^{-1-s}%
322: }{(1-p^{-1-s})}+\sum\limits_{\xi \in S}p^{-1-e_{\xi }s}\int\limits_{\mathbb{Z%
323: }_{p}}\mid f_{\xi }(x)\mid _{p}^{s}dx.
324: \end{equation*}
325: \end{proposition}
326:
327: The following example illustrates the use \ of the $p-$adic stationary phase
328: formula, and \ also the \ basic aspects of our algorithm for computing $%
329: Z(s,f)$.
330:
331: \subsection{Example\label{ex2}}
332:
333: Let \ $f(x)=(x-\alpha _{1})(x-\alpha _{2})^{3}(x-\alpha _{3})(x-\alpha
334: _{4})^{2}(x-\alpha _{5})$ be a polynomial such that $\alpha _{1}$, $\alpha
335: _{2}$ , $\alpha _{3}$, $\alpha _{4}$ , $\alpha _{5}$ are integers having the
336: following $\ p-$adic expansions:
337: \begin{equation*}
338: \alpha _{1}=a+dp+kp^{2},
339: \end{equation*}
340: \begin{equation*}
341: \alpha _{2}=a+dp+lp^{2},
342: \end{equation*}
343: \begin{equation*}
344: \alpha _{3}=b+gp+mp^{2},
345: \end{equation*}
346: \begin{equation*}
347: \alpha _{4}=c+hp+np^{2},
348: \end{equation*}
349: \begin{equation*}
350: \alpha _{5}=c+hp+rp^{2},
351: \end{equation*}
352: where \ the $p$-adic digits $a$, $b$, $c$, $d$, $g$, $h$, $l$, $m$, $n$, $r$
353: belong to $\{0,1,\mathbb{\cdots },p-1\}$. We assume the $p$-adic digits to
354: be different by pairs. The local zeta function $Z(s,f)$ will be computed by
355: using SPF iteratively.
356:
357: By applying SPF with $\overline{f(x)}=(x-\overline{a})^{4}(x-\overline{b})(x-%
358: \overline{c})^{3}$, $\nu (\overline{f})=p-3$, $\delta (\overline{f}%
359: )=1,S=\{a,c\}$, $f_{a}(x)=p^{-4}f(a+px)$, and $f_{c}(x)=p^{-3}f(c+px)$, we
360: obtain that
361:
362: \begin{eqnarray}
363: Z(s,f) &=&p^{-1}(p-3)+\frac{(1-p^{-1})p^{-1-s}}{1-p^{-1-s}}+p^{-1-4s}\int_{%
364: \mathbb{Z}_{p}}|f_{a}(x)|_{p}^{s}\mid dx\mid \notag \\
365: &&+p^{-1-3s}\int_{\mathbb{Z}_{p}}|f_{c}(x)|_{p}^{s}\mid dx\mid \text{.}
366: \label{step8}
367: \end{eqnarray}
368: We apply SPF to the integrals involving $f_{a}(x)$ and $f_{c}(x)$\ in (\ref
369: {step8}). First, we consider the integral corresponding to \ $f_{a}(x)$.
370: Since $\overline{f_{a}(x)}=(x-\overline{d})^{4}(\overline{a}-\overline{b})(%
371: \overline{a}-\overline{c})^{3}$, $S=\{d\}$, $f_{a,d}(x)=p^{-4}f_{a}(d+px)$, $%
372: \nu (\overline{f_{a}})=p-1$, and $\delta (\overline{f_{a}})=0$, it follows
373: from (\ref{step8}) using SPF that
374:
375: \begin{eqnarray}
376: Z(s,f) &=&p^{-1}(p-3)+\frac{(1-p^{-1})p^{-1-s}}{1-p^{-1-s}}%
377: +p^{-1}(p-1)p^{-1-4s} \notag \\
378: &&+p^{-2-8s}\int_{\mathbb{Z}_{p}}|f_{a,d}(x)|_{p}^{s}\mid dx\mid
379: +p^{-1-3s}\int_{\mathbb{Z}_{p}}|f_{c}(x)|_{p}^{s}\mid dx\mid \text{.}
380: \label{step9}
381: \end{eqnarray}
382: Now, we apply SPF to the integral involving $f_{c}(x)$ in (\ref{step9}).
383: Since $\overline{f_{c}(x)}$ $=(\overline{c}-\overline{a})^{4}(\overline{c}-%
384: \overline{b})(x-\overline{h})^{3}$, $S=\{h\},f_{c,h}(x)=p^{-3}f_{c}(h+px)$, $%
385: \nu (\overline{f_{c}})=p-1$, and $\delta (\overline{f_{c}})=0$, it follows
386: from (\ref{step9}) using SPF that
387: \begin{eqnarray}
388: Z(s,f) &=&p^{-1}(p-3)+\frac{(1-p^{-1})p^{-1-s}}{1-p^{-1-s}}%
389: +p^{-1}(p-1)p^{-1-4s} \notag \\
390: &&\,+p^{-2-8s}\int_{\mathbb{Z}_{p}}|f_{a,d}(x)|_{p}^{s}\mid dx\mid
391: +p^{-1}(p-1)p^{-1-3s} \notag \\
392: &&\,+p^{-2-6s}\int_{\mathbb{Z}_{p}}|f_{c,h}(x)|_{p}^{s}\mid dx\mid \text{.}
393: \label{step10}
394: \end{eqnarray}
395:
396: By applying SPF to the integral involving $f_{a,d}(x)$ in (\ref{step10}),
397: with $\overline{f_{a,d}(x)}=(x-\overline{k})(x-\overline{l})^{3}(\overline{d}%
398: -\overline{b})(\overline{d}-\overline{c})^{3}$, $S=\{k,l\}$, $%
399: f_{a,d,k}(x)=p^{-1}f_{a,d}(k+px)$, $|f_{a,d,k}(x)|_{p}^{s}=|x|_{p}^{s}$, $%
400: f_{a,d,l}(x)=p^{-3}f_{a,d}(l+px)$, $|f_{a,d,l}(x)|_{p}^{s}=|x|_{p}^{3s}$,$\
401: \nu (\overline{f_{a,d}})=p-2$, and $\delta (\overline{f_{a,d}})=1$, we
402: obtain \ that
403:
404: \begin{eqnarray}
405: Z(s,f) &=&p^{-1}(p-3)+\frac{(1-p^{-1})p^{-1-s}}{1-p^{-1-s}}%
406: +p^{-1}(p-1)p^{-1-4s} \notag \\
407: &&\,+p^{-1}(p-1)p^{-1-3s}+p^{-1}(p-2)p^{-2-8s}+\frac{(1-p^{-1})p^{-3-9s}}{%
408: 1-p^{-1-s}} \notag \\
409: &&\,+\frac{(1-p^{-1})p^{-3-11s}}{1-p^{-1-3s}}+p^{-2-6s}\int_{\mathbb{Z}%
410: _{p}}|f_{c,h}(x)|_{p}^{s}\mid dx\mid \text{.} \label{step11}
411: \end{eqnarray}
412: Finally, by applying SPF to the integral \ involving $f_{c,h}(x)$ in (\ref
413: {step11}), we obtain that
414: \begin{eqnarray}
415: Z(s,f) &=&p^{-1}(p-3)+\frac{(1-p^{-1})p^{-1-s}}{1-p^{-1-s}}%
416: +p^{-1}(p-1)p^{-1-4s} \notag \\
417: &&\,+p^{-1}(p-1)p^{-1-3s}+p^{-1}(p-2)p^{-2-8s}+\frac{(1-p^{-1})p^{-3-9s}}{%
418: 1-p^{-1-s}} \notag \\
419: &&\,+\frac{(1-p^{-1})p^{-3-11s}}{1-p^{-1-3s}}+p^{-1}(p-2)p^{-2-6s}+\frac{%
420: (1-p^{-1})p^{-3-7s}}{1-p^{-1-s}} \notag \\
421: &&+\frac{(1-p^{-1})p^{-3-8s}}{1-p^{-1-2s}}\text{.} \label{step12}
422: \end{eqnarray}
423:
424: \begin{remark}
425: \label{remark}If $\alpha =\frac{a}{b}\in \mathbb{Q}$, and $v_{p}(\alpha )<0$%
426: , then
427: \begin{equation}
428: \mid x-\alpha \mid _{p}=\mid \alpha \mid _{p},\text{ for every }x\in \mathbb{%
429: Z}_{p}. \label{const}
430: \end{equation}
431: On the other hand, a polynomial of the form
432: \begin{equation*}
433: f(x)=\alpha _{0}\prod\limits_{i=1}^{r}(x-\alpha _{i})^{e_{i}}\in \mathbb{Q}%
434: [x],
435: \end{equation*}
436: can be decomposed as $f(x)=\alpha _{0}f_{-}(x)f_{+}(x)$, where
437: \begin{equation}
438: f_{-}(x)=\prod\limits_{\{\alpha _{i}\mid v_{p}(\alpha _{i})<0\}}(x-\alpha
439: _{i})^{e_{i}},\text{ and \ }f_{+}(x)=\prod\limits_{\{\alpha _{i}\mid
440: v_{p}(\alpha _{i})\geqq 0\}}(x-\alpha _{i})^{e_{i}}. \label{pol1}
441: \end{equation}
442: From (\ref{const}) and (\ref{pol1}) follow that \
443: \begin{equation*}
444: Z(s,f)=\mid \alpha _{0}\prod\limits_{\{\alpha _{i}\mid v_{p}(\alpha
445: _{i})<0\}}\alpha _{i}{}^{e_{i}}\mid _{p}^{s}Z(s,f_{+}).
446: \end{equation*}
447: Thus, from a computational point of view, we may assume without loss of
448: generality that all roots of $f(x)$ are $p-$adic integers.
449: \end{remark}
450:
451: \section{Trees and $p$-adic Numbers}
452:
453: The tree \ $U=U(p)$ of residue classes modulo powers of a given prime number
454: $p$ is defined as \ follows. Consider the diagram
455: \begin{equation*}
456: \{0\}=\mathbb{Z}/p^{0}\mathbb{Z}\underleftarrow{\text{ \ }\phi _{1}}\text{ \
457: }\mathbb{Z}/p^{1}\mathbb{Z}\text{ \ }\underleftarrow{\phi _{2}}\text{ \ }%
458: \mathbb{Z}/p^{2}\mathbb{Z}\text{ }\underleftarrow{\phi _{3}}\cdots
459: \end{equation*}
460: where $\phi _{l}$\ the \ are the natural homomorphisms. The vertices of \ $U$
461: are the elements of $\mathbb{Z}/p^{l}\mathbb{Z}$, for \ $l=0,1,2,\mathbb{%
462: \cdots }$, and the directed edges are $u\rightarrow v$ where $u\in \mathbb{Z}%
463: /p^{l}\mathbb{Z}$ and \ $\phi _{l}(u)=v$, for some $l>0$. Thus $U$ is a
464: rooted tree with root $\{0\}$. Exactly one directed edge emanates \ from
465: each vertex of $U$; except from the vertex $\{0\}$, from which no edge
466: emanates. In addition, every vertex is the end point of exactly $p$ directed
467: edges.
468:
469: Given two \ vertices $u$, $v$ the notation $u>v$ will mean that there is a
470: sequence of vertices and edges of the form
471: \begin{equation*}
472: u\rightarrow u^{(1)}\rightarrow \cdots \rightarrow u^{(m)}=v\text{.}
473: \end{equation*}
474: The notation $u\geqq v$ \ will mean that $u=v$ or $u>v$. The\textit{\ level}
475: $l(u)$ of a vertex $u$ is $m$ if $u\in \mathbb{Z}/p^{m}\mathbb{Z}$. The
476: \textit{valence} $Val(u)$ of a vertex $u$ is defined as the number of
477: directed edges whose end point is $u$.
478:
479: A subtree, or simply a tree, is defined as a nonempty subset $T$ of vertices
480: of $U$, such that when $u\in T$ and $u>v$, then \ $v\in T$. Thus \ $T$
481: together with the directed edges $u\rightarrow v$, where $u,v\in T$ , is
482: again a tree with root $\{0\}$.
483:
484: A tree $T$ is named a \textit{weighted tree}, if there exists a weight
485: function $W:T\rightarrow \mathbb{N}$. The value $W(u)$ is called the weight
486: of vertex $u$.
487:
488: If \ $x\in \mathbb{Z}_{p}$, and $x_{l}$ denotes its residue class modulo $%
489: p^{l}$, then every vertex of $U$ is of the type $x_{l}$ with $l\in \mathbb{N}
490: $.
491:
492: A \textit{stalk} is defined as a tree $K$ having at most one vertex at each
493: level. Thus a stalk is either finite, of the type
494: \begin{equation*}
495: \{0\}\longleftarrow u^{(1)}\longleftarrow \cdots \longleftarrow u^{(l)},
496: \end{equation*}
497: or infinite, of the type
498: \begin{equation*}
499: \{0\}\longleftarrow u^{(1)}\longleftarrow \cdots \text{.}
500: \end{equation*}
501: Clearly a finite stalk may be written as \
502: \begin{equation*}
503: \{0\}\longleftarrow x_{1}\longleftarrow \cdots \longleftarrow x_{l},
504: \end{equation*}
505: with $x\in \mathbb{Z}$, and infinite stalks as
506: \begin{equation*}
507: \{0\}\longleftarrow x_{1}\longleftarrow x_{2}\longleftarrow \cdots ,
508: \end{equation*}
509: with $x\in \mathbb{Z}_{p}$. Thus there is a $1-1$ correspondence between \
510: infinite stalks and $p-$adic integers.
511:
512: \subsection{Tree Attached to a Polynomial}
513:
514: Let
515: \begin{equation}
516: f(x)=\alpha _{0}\prod\limits_{i=1}^{r}(x-\alpha _{i})^{e_{i}}\in \mathbb{Q}%
517: [x] \label{pol}
518: \end{equation}
519: be a non-constant polynomial, in one variable,\ of degree $d_{f}$, such that
520: $v_{p}(\alpha _{i})\geqslant 0$, $i=1,2,\mathbb{\cdots },r$. \ We \
521: associate to $f(x)$ and a \ prime number $p$ the integer
522: \begin{equation*}
523: l_{f}=\left\{
524: \begin{array}{cc}
525: 1+\max \{v_{p}(\alpha _{i}-\alpha _{j})\mid i\neq j\text{, }1\leq i\text{, }%
526: j\leq r\}\text{,} & \text{if \ }r\geqq 2\text{;} \\
527: 1\text{, } & \text{if \ \ }r=1\text{.}
528: \end{array}
529: \right.
530: \end{equation*}
531: We set
532: \begin{equation*}
533: \alpha _{i}=a_{0,i}+a_{1,i\text{ }}p+\mathbb{\cdots }+a_{j,i}\text{ }p^{j}+%
534: \mathbb{\cdots }+a_{l_{f},i}\text{ }p^{l_{f}}\text{ mod }p^{l_{f}+1}\text{, }
535: \end{equation*}
536: $a_{j,i}\in \{0,1,\mathbb{\cdots },p-1\}$, $j=0,1,\mathbb{\cdots },l_{f}$, $%
537: i=1,2,\mathbb{\cdots },r$, for the $p-$adic expansion modulo $p^{l_{f}+1}$
538: of \ $\alpha _{i}$. We attach a weighted tree $T(f$, $l_{f})$ to $f$ as
539: follows:
540: \begin{equation}
541: T(f,l_{f},p)=T(f,l_{f})=\bigcup\limits_{i=1}^{r}K(\alpha _{i},l_{f}),
542: \label{tree}
543: \end{equation}
544: where $K(\alpha _{i},l_{f})$\ denotes the stalk corresponding to the $p-$%
545: adic expansion of $\alpha _{i}$\ modulo $p^{l_{f}+1}$. Thus $T(f$, $l_{f})$\
546: is a rooted tree. We introduce a weight function on $T(f$, $l_{f}),$\ by
547: defining the weight of a vertex $u$ of level $m$ as
548: \begin{equation}
549: W(u)=\left\{
550: \begin{array}{cc}
551: \sum\limits_{\{i\mid \alpha _{i}\equiv u\text{ mod }p^{m}\}}e_{i}, & \text{%
552: if }\ m\geqq 1\text{;} \\
553: 0, & \text{if }\ m=0\text{.}
554: \end{array}
555: \right. \label{weight}
556: \end{equation}
557:
558: Given a vertex $u\in T(f,l_{f})$, we define the stalk generated by $u$\ to
559: be
560: \begin{equation*}
561: B_{u}=\{v\in T(f,l_{f})\mid u\geqq v\}.
562: \end{equation*}
563: \ We associate a weight $W^{\ast }(B_{u})$ to \ $B_{u}$ as follows:
564: \begin{equation}
565: W^{\ast }(B_{u})=\sum\limits_{v\in B_{u}}W(v). \label{weightstalk}
566: \end{equation}
567:
568: \subsection{Computation of Trees Attached to \ Polynomials}
569:
570: Our next step is to show that a tree $T(f,l_{f})$ attached to a polynomial $%
571: f(x)$, of type (\ref{pol}), can be computed in polynomial time. There are
572: well known programming techniques \ to construct and manipulate trees and
573: forests (see e.g. \cite[Volume 1]{K}), for this reason, we shall focus on
574: showing \ that such computations can be carry out in polynomial time, and
575: set aside the\ implementation details of a particular algorithm for this
576: task. We shall include in the computation of \ $T(f,l_{f})$, the computation
577: of the weights of the stalks generated by its vertices; because all these
578: data will be used in the computation of the local zeta function of $f$.
579:
580: \begin{proposition}
581: \label{prop4.1}The computation of a tree $T(f,l_{f})$ attached to a
582: polynomial $f(x)$, of type (\ref{pol}), \ from the \ $p-$adic expansions
583: modulo $p^{l_{f}+1}$\ of its roots
584: \begin{equation*}
585: \alpha _{i}=a_{0,i}+a_{1,i\text{ }}p+\mathbb{\cdots }+a_{l_{f},i}p^{l_{f}}%
586: \text{ mod }p^{l_{f}+1}\text{ }
587: \end{equation*}
588: and multiplicities $e_{i}$, $i=1,2,\mathbb{\cdots },r$, involves \ $O(l_{f%
589: \text{ }}^{2}d_{f}^{3})$ arithmetic operations on integers with binary
590: length
591: \begin{equation*}
592: O(\max \{\log p,\log (l_{f}d_{f})\}).
593: \end{equation*}
594: \end{proposition}
595:
596: \begin{proof}
597: We assume that $T(f,l_{f})$ is finite set of the form
598: \begin{equation}
599: T=\{\text{Level}_{0}\text{,}\cdots \text{,Level}_{j}\text{,}\cdots \text{%
600: ,Level}_{l_{f}+1}\}\text{,} \label{tipotree}
601: \end{equation}
602: where \ Level$_{j}$ represents the set of all vertices with level $j$. Each
603: Level$_{j}$ is a set of the form
604: \begin{equation*}
605: Level_{j}=\{u_{j,1},\mathbb{\cdots },u_{j,i},\mathbb{\cdots },u_{j,m_{j}}\},
606: \end{equation*}
607: and each $u_{j,i}$ is a weighted vertex for every $i=1,\mathbb{\cdots },m_{j}
608: $. A \ weighted vertex $u_{j,i}$ is a set of the form
609: \begin{equation*}
610: u_{j,i}=\{\text{ }W(u_{j,i})\text{, }Val(u_{j,i}),W^{\ast }(B_{u_{j,i}})\}%
611: \text{,}
612: \end{equation*}
613: where $W(u_{j,i})$ \ is the weight of $u_{j,i}$, $Val(u_{j,i})$ is its
614: valence, and $W^{\ast }(B_{ui})$ is \ the weight of \ stalk $B_{u_{j,i}}$.
615: The weight of the stalk generated by \ $u_{j,i}$ can be written as
616: \begin{equation*}
617: W^{\ast }(B_{u_{j,i}})=\sum\limits_{v\in B_{u_{j,i}}}W(v).
618: \end{equation*}
619:
620: For the computation of a vertex \ $u_{j,i}$\ of level $j$, we proceed as
621: follows. We put $I=\{1,2,\mathbb{\cdots },r\},$ and
622: \begin{equation*}
623: M_{j}=\{\alpha _{i}\text{ mod }p^{j}\mid i\in I\}.
624: \end{equation*}
625: For each \ $0\leq j\leq l_{f}+1$, we compute a partition of $I$ of type
626: \begin{equation}
627: I=\bigcup\limits_{i=1}^{l_{j}}I_{j,i}, \label{parti}
628: \end{equation}
629: such that
630: \begin{equation*}
631: \alpha _{t}\text{ mod }p^{j}=\alpha _{s}\text{ mod }p^{j},
632: \end{equation*}
633: for every $t,s\in I_{j,i}$. Each subset $I_{j,i}$\ corresponds \ to a vertex
634: $u_{j,i}$\ of level $j$. This \ computation requires\ $O(l_{f}r^{2})$
635: arithmetic operations on integers with binary length $O(\log p)$. Indeed,
636: the cost of computing a ``yes or no'' answer for the question: $\alpha _{t}$
637: mod $p^{j}=\alpha _{s}$ mod $p^{j}?$ is $O(j)$ comparisons of integers with
638: binary length $O(\log p)$. In the worst case, there are $r$ vectors $M_{j}$,
639: \ and the computation of partition (\ref{parti}), for a fixed $j$, involves
640: the comparison of $\alpha _{t}$ with $\alpha _{l}$ for $l=t+1,t+2,\mathbb{%
641: \cdots },r$. This computation requires\ $O(jr^{2})$ arithmetic operations on
642: integers with binary length $O(\log p)$. Since $j\leqq l_{f}+1$, the
643: computation of partition (\ref{parti}) requires\ $O(l_{f}r^{2})$ arithmetic
644: operations on integers with binary length $O(\log p)$.
645:
646: The weight of the vertex \ $u_{j,i}$\ is given by the expression
647: \begin{equation*}
648: W(u_{j,i})=\sum\limits_{k\in I_{j,i}}e_{k}.
649: \end{equation*}
650: Thus the computation of the weight of a vertex requires $O(r)$ additions of
651: integers with binary length $O(\log d_{f}r)$.
652:
653: For the computation of the valence of \ $u_{j,i}$, we proceed as follows.
654: The valence of $u_{j,i}$ \ can be expressed as
655: \begin{equation*}
656: Val(u_{j,i})=\text{Card}\{I_{j+1,l}\mid I_{j+1,l}\subseteq I_{j,i}\},
657: \end{equation*}
658: where $I_{j+1,l}$ runs through all possible sets that correspond to the \
659: vertices $u_{j+1,l}$, with level $j+1$. Thus the computation of $Val(u_{j,m})
660: $\ involves \ the computation of $\ $a ``yes or no'' answer for the question
661: $I_{j+1,l}\subseteq I_{j,i}$? The computation of a\ ``yes or no'' answer \
662: involves $O(r)$\ comparisons of integers with binary length $O(\log r)$.
663: Therefore the computation of $Val(u_{j,i})$\ involves $O(r)$\ comparisons
664: and $O(r)$ additions of integers with binary length $O(\log r)$.
665:
666: For the computation of the weight of \ $B_{u_{j,i}}$, we observe that $%
667: W^{\ast }(B_{u_{j,i}})$ is given by the formula
668: \begin{equation*}
669: W^{\ast }(B_{u_{j,i}})=\sum\limits_{l=0}^{j-1}\sum\limits_{I_{j,i}\subseteq
670: I_{l,k}}W(I_{l,k}),
671: \end{equation*}
672: where $W(I_{l,k})=W(v_{l,k})$, and $v_{l,k}$\ is \ the vertex corresponding
673: to $I_{l,k}$. Thus the computation of $W^{\ast }(B_{u_{j,i}})$ involves $%
674: O(l_{f})$ additions of integers with binary length $O(\log (l_{f}$ $d_{f}))$%
675: , and $O(l_{f}$ $r)$ comparisons of integers with binary length $O(\log r)$.
676:
677: From the above reasoning follows that the computation of a vertex of a tree $%
678: T(f,l_{f})$ \ involves at most $O(l_{f}$ $r^{2})$ arithmetic operations
679: (additions and comparisons) on integers with binary length $O(\max \{\log p,$
680: $\log ($ $l_{f}$ $d_{f})\})$. Finally, since the number of vertices of $%
681: T(f,l_{f})$ is at most $O(l_{f}$ $d_{f})$, it follows that \ the computation
682: of a tree of type $T(f,l_{f})$ involves $O(l_{f}^{2}$ $d_{f}^{3})$
683: arithmetic operations on integers with binary length $O(\max \{\log p,$ $%
684: \log ($ $l_{f}$ $d_{f})\})$.
685: \end{proof}
686:
687: \section{Generating Functions and Trees}
688:
689: In this section we attach to a weighted tree $T(f,l_{f})$ and a \ prime $p$
690: a generating function $G(s,T(f,l_{f}),p)\in \mathbb{Q}(p^{-s})$ defined \ as
691: follows.
692:
693: We set
694: \begin{equation*}
695: \mathcal{M}_{T(f,l_{f})}=\left\{ u\in T(f,l_{f})\left|
696: \begin{array}{c}
697: W(u)=1\text{, and there no exists }v\in T(f,l_{f}) \\
698: \text{with }W(v)=1\text{, such that }u>v.
699: \end{array}
700: \right. \right\} ,
701: \end{equation*}
702:
703: and
704:
705: \begin{equation*}
706: L_{u}(p^{-s})=\left\{
707: \begin{array}{cc}
708: \frac{(1-p^{-1})p^{-l(u)-W^{\ast }(B_{u})s}}{(1-p^{-1-W(u)s})}\text{, \ if}
709: & l(u)=1+l_{f}\text{, and \ }W(u)\geqq 2\text{;} \\
710: & \\
711: p^{-1}(p-Val(u))p^{-l(u)-W^{\ast }(B_{u})s}\text{, \ if} & 0\leqq l(u)\leqq
712: l_{f}\text{, and \ }W(u)\neq 1\text{;} \\
713: & \\
714: \frac{(1-p^{-1})p^{-l(u)-W^{\ast }(B_{u})s}}{1-p^{-1-s}}\text{, \ \ if} &
715: u\in \mathcal{M}_{T(f,l_{f})}\text{;} \\
716: & \\
717: 0\text{,\ \ \ if} & W(u)=1\text{, and }u\notin \mathcal{M}_{T(f,l_{f})}\text{%
718: .}
719: \end{array}
720: \right.
721: \end{equation*}
722:
723: With all the above notation, we define the generating function attached to $%
724: T(f,l_{f})$ and $p$\ \ as
725: \begin{equation}
726: G(s,T(f,l_{f}),p)=\sum\limits_{u\in T(f,l_{f})}L_{u}(p^{-s})\text{.}
727: \label{Gfunction}
728: \end{equation}
729:
730: Our next goal is to show that $G(s,T(f,l_{f}),p)=Z(s,f)$. The proof of this
731: fact requires the following preliminary result.
732:
733: \begin{proposition}
734: The generating function attached to a tree $T(f,l_{f})$ and a prime $p$
735: satisfies
736: \begin{eqnarray}
737: G(s,T(f,l_{f}),p) &=&p^{-1}\nu (\overline{f})+\delta (\overline{f})\frac{%
738: (1-p^{-1})p^{-1-s}}{(1-p^{-1-s})} \notag \\
739: &&+\sum\limits_{\xi \in S}p^{-1-e_{\xi }s}G(s,T(f_{\xi },l_{f}-1),p).
740: \label{ident}
741: \end{eqnarray}
742: \end{proposition}
743:
744: \begin{proof}
745: Let $A_{f}=\{u\in T(f,l_{f})\mid l(u)=1$, $W(u)=1\},$ and $B_{f}=\{u\in
746: T(f,l_{f})\mid l(u)=1$, $W(u)\geqq 2\}$. We have the following \ \ partition
747: for $T(f,l_{f}):$%
748: \begin{equation}
749: T(f,l_{f})=\{0\}\bigcup A_{f}\bigcup \left( \bigcup_{u\in B_{f}}T_{u}\right)
750: \text{,} \label{parti1}
751: \end{equation}
752: with
753: \begin{equation*}
754: T_{u}=\{v\in T(f,l_{f})\mid v\geqq u\}.
755: \end{equation*}
756: Each $T_{u}$\ is a rooted tree with root $\{u\}$.\ \ From \ partition (\ref
757: {parti1}) and the definition of \ $G(s,T(f,l_{f}),p)$, it follows that
758: \begin{eqnarray}
759: G(s,T(f,l_{f}),p) &=&p^{-1}\left( p-Val(\{0\})\right) +\text{Card}\{A_{f}\}%
760: \frac{(1-p^{-1})p^{-1-s}}{(1-p^{-1-s})}+ \notag \\
761: &&\sum\limits_{u\in B_{f}}G(s,T_{u}), \label{ident0}
762: \end{eqnarray}
763: with $G(s,T_{u})=\sum\limits_{v\in T_{u}}L_{v}(p^{-s})$.
764:
765: Since there exists a bijective correspondence between the roots of $%
766: \overline{f}(x)\equiv 0$ mod $p$ and the vertices $\ $of $T(f,l_{f})$\ with
767: level $1$, $\ \ \ \ \ $%
768: \begin{equation}
769: \ \ p-Val(\{0\})=\nu (\overline{f})\text{, and \ Card}\{A_{f}\}=\delta (%
770: \overline{f}). \label{ident1}
771: \end{equation}
772:
773: Now, if the vertex $u$ corresponds to the root \ $\overline{f}(\xi )\equiv 0$
774: mod $p$, $\ $\ then
775: \begin{equation}
776: T_{u}=\left( \bigcup\limits_{\{\alpha _{i}\mid \alpha _{i}\equiv \xi \text{
777: mod }p\}}K(\alpha _{i},l_{f})\right) \setminus \{0\}.
778: \end{equation}
779: On the other hand, $\ $we have that
780: \begin{equation}
781: T(f_{\xi },l_{f}-1)=\bigcup\limits_{\{\alpha _{i}\mid \alpha _{i}\equiv \xi
782: \text{ mod }p\}}K(\frac{\alpha _{i-\xi }}{p},l_{f}-1).
783: \end{equation}
784: Now \ we remark that the map $\alpha _{i}\rightarrow \frac{\alpha _{i}-\xi }{%
785: p}$ induces a isomorphism between the trees $T_{u}$ and \ $T(f_{\xi
786: },l_{f}-1)$, that preserves the weights of the vertices; and \ thus we may
787: suppose that $T_{u}$ $=$\ $T(f_{\xi },l_{f}-1)$.\ The level function $l_{T}$
788: of $T(f_{\xi },l_{f}-1)$ is related to the level function $l_{T_{u}}$ of $%
789: T_{u}$ by means of the equality $l_{T}$ $-l_{T_{u}}=-1$. In addition, $B_{f}=
790: $ $S$, where $S$ is the subset of $\{0,1,\mathbb{\cdots },p-1\}\subseteq
791: \mathbb{Z}_{p}$ whose reduction modulo $pZ_{p}$ is equal to the set of roots
792: of \ $\overline{f}(\xi )=0$ with multiplicity greater or equal than two. \
793: Therefore, it holds that \
794: \begin{equation}
795: G(s,T_{u})=p^{-1-e_{\xi }s}G(s,T(f_{\xi },l_{f}-1),p). \label{ident2}
796: \end{equation}
797: The result follows from (\ref{ident0}) by the identities (\ref{ident1}) and (%
798: \ref{ident2}).
799: \end{proof}
800:
801: \begin{lemma}
802: \label{lemma}Let $p$ be a fixed prime number and $v_{p}$ the corresponding $%
803: p-$adic valuation, and
804: \begin{equation*}
805: f(x)=\alpha _{0}\prod\limits_{i=1}^{r}(x-\alpha _{i})^{e_{i}}\in \mathbb{Q}%
806: [x]\setminus \mathbb{Q},
807: \end{equation*}
808: a polynomial such that $v_{p}(\alpha _{i})\geqq 0$, for $i=1,\cdots ,r$.
809: Then
810: \begin{equation*}
811: Z(s,f)=G(s,T(f,l_{f}),p).
812: \end{equation*}
813: \end{lemma}
814:
815: \begin{proof}
816: We proceed by induction on $l_{f}$.
817:
818: Case\ $l_{f}=1$
819:
820: If $r=1$ the proof follows immediately, thus we may assume that $r\geqq 2$.
821: Since $l_{f}=1$, it holds that $v_{p}(\alpha _{i}-\alpha _{j})=0$, for every
822: $i$, $j$, satisfying $i\neq j$, and thus \ $\overline{\alpha _{i}}\neq
823: \overline{\alpha _{j}}$, if \ $i\neq j$. By applying SPF, we have that
824: \begin{equation}
825: Z(s,f)=p^{-1}\nu (\overline{f})+\delta (\overline{f})\frac{(1-p^{-1})p^{-1-s}%
826: }{(1-p^{-1-s})}+\sum\limits_{\xi \in S}p^{-1-e_{\xi }s}\frac{(1-p^{-1})}{%
827: (1-p^{-1-e_{\xi }s})},
828: \end{equation}
829: where each $e_{\xi }=e_{j}\geqq 2$, for some $j$, and \ $\alpha _{j}=\xi
830: +p\beta _{j}$.
831:
832: \ On the other hand, $T(f,l_{f})$ \ is a rooted tree with $r$ vertices $v_{j}
833: $, satisfying $l(v_{j})=1$, and $W(v_{j})=e_{j}$,\ for $j=1,\cdots ,r$.
834: These observations allow one to deduce that $Z(s,f)=G(s,T(f,l_{f}),p)$.
835:
836: By induction hypothesis, we may assume that \ $Z(s,f)=G(s,T(f,l_{f}),p)$,
837: for every polynomial $f$ satisfying both the \ hypothesis of the lemma, and
838: the condition \ $1\leqq l_{f}\leqq k$, $k\in \mathbb{N}$.
839:
840: Case\ $l_{f}=k+1$, $k\in \mathbb{N}$
841:
842: Let $f(x)$ \ be \ a polynomial satisfying the lemma's \ hypothesis, and \ $%
843: l_{f}=k+1$, $k\geqq 1$. By applying SPF, we obtain that
844: \begin{equation}
845: Z(s,f)=p^{-1}\nu (\overline{f})+\delta (\overline{f})\frac{(1-p^{-1})p^{-1-s}%
846: }{(1-p^{-1-s})}+\sum\limits_{\xi \in S}p^{-1-e_{\xi }s}\int \mid f_{\xi
847: }(x)\mid _{p}^{s}dx. \label{eq1}
848: \end{equation}
849: Now, since $l_{f_{\xi }}=l_{f}-1$, for every \ $\xi \in S$, it follows from
850: the induction hypothesis applied to each \ $f_{\xi }(x)$ in (\ref{eq1}),
851: that
852: \begin{equation}
853: Z(s,f)=p^{-1}\nu (\overline{f})+\delta (\overline{f})\frac{(1-p^{-1})p^{-1-s}%
854: }{(1-p^{-1-s})}+\sum\limits_{\xi \in S}p^{-1-e_{\xi }s}G(s,T(f_{\xi
855: },l_{f}-1),\text{ }p). \label{eq3}
856: \end{equation}
857: Finally, from\ identity (\ref{ident}), and (\ref{eq3}), we conclude that
858: \begin{equation}
859: Z(s,f)=G(s,T(f,l_{f}),p). \label{eq4}
860: \end{equation}
861: \end{proof}
862:
863: The following proposition gives a complexity estimate for the computation of
864: $G(s,T(f,l_{f}),p)$.
865:
866: \begin{proposition}
867: \label{prop4.2}The computation of the generating function \
868: \begin{equation*}
869: G(s,T(f,l_{f}),p)
870: \end{equation*}
871: from $T(f,l_{f})$, involves $O(l_{f}$ $d_{f})$ arithmetic operations on
872: integers with binary length \ $O(\max \{\log p,$ $\log ($ $l_{f}$ $d_{f})\})$%
873: .
874: \end{proposition}
875:
876: \begin{proof}
877: This is a consequence of proposition \ref{prop4.1},\ and the definition of
878: generating function.
879: \end{proof}
880:
881: \section{Computation of $p$-adic Expansions}
882:
883: In this section we estimate the complexity of the steps $2$ and $3$ in \ the
884: algorithm Compute\_$Z(s,f)$.
885:
886: \begin{proposition}
887: \label{prop4.3}Let
888: \begin{equation*}
889: B=\max_{
890: \begin{array}{c}
891: {1\leq i,}\text{ }{j\leq r} \\
892: {i\neq j}
893: \end{array}
894: }{\{}\mid c_{j,i}\mid \text{,}\mid d_{j,i}\mid \text{ \ }{\mid }\text{ }%
895: \alpha _{j}-\alpha _{i}=\frac{c_{j,i}}{d_{j,i}}\text{, }c_{j,i}\text{, }%
896: d_{j,i}\in \mathbb{Z}\setminus \{0\}{\}}.
897: \end{equation*}
898: The computation of the integer $l_{f}$ involves $O(d_{f}^{2}\frac{\log B}{%
899: \log p})$ arithmetic operations on integers with binary length $O(\max
900: \{\log B,\log p\})$.
901: \end{proposition}
902:
903: \begin{proof}
904: First, we observe that for $c\in \mathbb{Z}\setminus \{0\}$, \ the
905: computation of $v_{p}(c)$ \ involves $O(\frac{\log \mid c\mid }{\log p})$
906: divisions of integers of binary length $O(\max \{\log \mid c\mid ,$ $\log
907: p\})$. Thus the computation of $v_{p}(\frac{c}{d})=v_{p}(c)-v_{p}(d),$ \
908: involves $O(\frac{\max \{\log \mid c\mid ,\text{ }\log \mid d\mid \}}{\log p}%
909: )$ divisions and subtractions of integers with binary length
910: \begin{equation*}
911: O(\max \{\log \mid c\mid ,\log \mid d\mid ,\log p\}).
912: \end{equation*}
913: From these observations follow that the computation of $v_{p}(\alpha
914: _{j}-\alpha _{i})$, $i\neq j$, $1\leq i$, $j\leq r$, involves $O(r^{2}\frac{%
915: \log B}{\log p})$ arithmetic operations on integers with binary length $%
916: O(\max \{\log B,\log p\})$. Finally, the computation of the maximum of the $%
917: v_{p}(\alpha _{j}-\alpha _{i})$, $i\neq j$, $1\leq i$, $j\leq r$, involves $%
918: O(\log r)$ comparisons of integers with binary length $O(\max \{\log B,\log
919: p\})$. Therefore the computation of the integer $l_{f}$ involves at most $%
920: O(d_{f}^{2}\frac{\log B}{\log p})$ arithmetic operations on integers with
921: binary length $O(\max \{\log B,\log p\})$.
922: \end{proof}
923:
924: \begin{proposition}
925: Let \ $p$ be a fixed prime and $\gamma =\frac{c}{b}\in \mathbb{Q}$, with $%
926: c,b\in \mathbb{Z}\setminus \{0\}$, and $v_{p}(\gamma )\geqq 0$. The $p-$adic
927: expansion
928: \begin{equation*}
929: \gamma =a_{0}+a_{1\text{ }}p+\mathbb{\cdots }+a_{j}p^{j}+\mathbb{\cdots }%
930: +a_{m}p^{m},
931: \end{equation*}
932: modulo $p^{m+1}$ involves $O(m+\log (\max \{\mid b\mid ,$ $p\}))$\
933: arithmetic operations on integers with binary length $O(\max \{\log \mid
934: c\mid ,\log \mid b\mid ,\log p\})$.
935: \end{proposition}
936:
937: \begin{proof}
938: Let $y\in \{1,\cdots ,p-1\}$ be an integer \ such that $yb\equiv 1$ mod $p$.
939: This integer \ can be computed \ by means of the Euclidean algorithm in $%
940: O(\log (\max \{\mid b\mid ,p\}))$ \ arithmetic operations \ involving
941: integers of binary length $O(\max \{\log \mid b\mid ,\log p\})$ (cf.
942: \cite[Volume 2, section 4.5.2]{K}).
943:
944: We set $\gamma =\gamma _{0}=\frac{c}{b}$, $c_{0}=c$, and define \ $%
945: a_{0}\equiv yc$ mod $p$. With this notation, the $p-$adic digits $a_{i},i=1$,%
946: $\cdots $, $m$, can be computed recursively as follows:
947: \begin{equation*}
948: \gamma _{i}=\frac{\frac{\left( c_{i-1}-a_{i-1}b\right) }{p}}{b}=\frac{c_{i}}{%
949: b},
950: \end{equation*}
951: \begin{equation*}
952: a_{i}=yc_{i}\text{ mod }p.
953: \end{equation*}
954: Thus the computation of the $p-$adic expansion of $\gamma $ \ needs \ $%
955: O(m+\log (\max \{\mid b\mid ,p\}))$\ arithmetic operations on integers with
956: binary length
957: \begin{equation*}
958: O(\max \{\log \mid c\mid ,\text{ }\log \mid b\mid ,\text{ }\log p\}).
959: \end{equation*}
960: \end{proof}
961:
962: \begin{corollary}
963: \label{cor4.4}Let $p$ be a fixed prime number and $v_{p}$ the corresponding $%
964: p-$adic valuation, and
965: \begin{equation*}
966: f(x)=\alpha _{0}\prod\limits_{i=1}^{r}(x-\alpha _{i})^{e_{i}}\in \mathbb{Q}%
967: [x],
968: \end{equation*}
969: a non-constant polynomial such that $v_{p}(\alpha _{i})\geqq 0$, $i=1,\cdots
970: ,r$. The computation of \ the $p-$adic expansions modulo $p^{l_{f}+1}$\ of
971: the roots $\alpha _{i}$, $i=1,2,\cdots ,r$, of $f(x)$\ \ involves $O(d_{f}$ $%
972: l_{f}+d_{f}$ $\log (\max \{B,p\}))$\ arithmetic operations \ on integers
973: with binary length \ $O(\max \{\log B,\log p\})$.
974: \end{corollary}
975:
976: \begin{proof}
977: The corollary \ follows directly from the two previous propositions.
978: \end{proof}
979:
980: \section{Computing local zeta functions of polynomials with splitting $%
981: \mathbb{Q}$}
982:
983: In this section we prove the \ correctness of the algorithm Compute\_$Z(s,f)$
984: and estimate its complexity.
985:
986: \begin{theorem}
987: \label{theoA}The algorithm Compute\_$Z(s,f)$ \ outputs\ the meromorphic
988: continuation of the Igusa local zeta function $Z(s,f)$ of a polynomial \ $%
989: f(x)\in \mathbb{Z}[x]$, in one variable, with splitting field \ $\mathbb{Q}$%
990: . The number of arithmetic operations needed by the algorithm is
991: \begin{equation*}
992: O\left( d_{f}^{6}+d_{f}^{9}\log (\shortparallel f\shortparallel
993: )+l_{f}^{2}d_{f}^{3}+d_{f}^{2}\log \left( \max \{B,p\}\right) \right) ,
994: \end{equation*}
995: and the integers on which these operations \ are performed have a binary
996: length
997: \begin{equation*}
998: O\left( \max \{\log p\text{, }\log l_{f}d_{f}\text{, }\log B\text{, }%
999: d_{f}^{3}+d_{f}^{2}\log (\shortparallel f\shortparallel )\}\right) \text{.}
1000: \end{equation*}
1001: \end{theorem}
1002:
1003: \begin{proof}
1004: By remark (\ref{remark}), we may assume without loss of generality that
1005: \begin{equation*}
1006: f(x)=\alpha _{0}\prod\limits_{i=1}^{r}(x-\alpha _{i})^{e_{i}}\in \mathbb{Q}%
1007: [x]\setminus \mathbb{Q},
1008: \end{equation*}
1009: with $v_{p}(\alpha _{i})\geqq 0$, $i=1,\cdots ,r$. The correctness of the
1010: algorithm follows from lemma \ \ref{lemma}. The complexity estimates are
1011: obtained as follows: the number of arithmetic operations \ needed in the
1012: steps 2 (cf. proposition \ref{prop4.3}), 3 (cf. \ corollary \ref{cor4.4}), 4
1013: (cf. proposition \ref{prop4.1}), 5 (proposition\ \ref{prop4.2}), and 6 is at
1014: most
1015: \begin{equation*}
1016: O\left( l_{f}^{2}d_{f}^{3}+d_{f}^{2}\log \left( \max \{B,p\}\right) \right)
1017: \text{;}
1018: \end{equation*}
1019: and these operations are performed on integers \ whose binary length is at
1020: most
1021: \begin{equation*}
1022: O\left( \max \{\log p\text{, }\log l_{f}d_{f}\text{,}\log B\}\right) .
1023: \end{equation*}
1024: The estimates for the whole algorithm \ follow from the above estimates and
1025: those of \ the factoring algorithm by \ A. K. Lenstra, H. Lenstra and L. Lov%
1026: \'{a}sz (see theorem 3.6 of \cite{L-L-L}).
1027: \end{proof}
1028:
1029: \section{Stream Ciphers and Poincar\'{e} series}
1030:
1031: There is a natural connection between Poincar\'{e} series and stream
1032: ciphers. In order to explain this relation, we recall some basic facts about
1033: stream ciphers \ \cite{R}. \ Let $\mathbb{F}_{p^{n}}$ be a finite field with
1034: $p^{n}$ elements, with $p$ a prime number. \ For any integer $r>0$ and $r$\
1035: fixed elements $q_{i}\in \mathbb{F}_{p^{n}}$, $i=1,\mathbb{\cdots },r$\
1036: (called taps), a Linear Feedback Shift Register, abbreviated LFSR, of length
1037: $r$ consists of $r$ cells with initial contents $\left\{ a_{i}\in \mathbb{F}%
1038: _{p^{n}}\mid i=1,\mathbb{\cdots },r\right\} $. For any $n\geqslant r$, if
1039: the current state is $(a_{n-1},\mathbb{\cdots },a_{n-r})$, then $a_{n}$ is
1040: determined by the linear recurrence relation
1041: \begin{equation*}
1042: a_{n}=-\sum\limits_{i=1}^{r}a_{n-i}q_{i}.
1043: \end{equation*}
1044: The device outputs the rightmost element $a_{n-r}$, shifts all the cells one
1045: unit right, and feeds $a_{n}$ back to the leftmost cell.
1046:
1047: Any configuration of the $r$ cells \ forms a state of the LSFR. If $%
1048: q_{r}\neq 0$, the following polynomial \ $q(x)\in \mathbb{F}_{p^{n}}[x]$ of
1049: degree $r$\ appears in the analysis of LFSRs:
1050: \begin{equation*}
1051: q(x)=q_{0}+q_{1}x+\mathbb{\cdots }.+q_{r}x^{r}\text{ \ with }q_{0}=-1\text{.}
1052: \end{equation*}
1053: This polynomial is called the connection polynomial. An infinite sequence $%
1054: A=\left\{ a_{i}\in \mathbb{F}_{p^{n}}\mid i\in \mathbb{N}\right\} $ has
1055: period $T$ if for any $i\geqslant 0$, \ $a_{i+T}=a_{i}$. Such a sequence is
1056: called periodic. If this is only true for $i$ greater than some index $i_{0}$%
1057: , then the sequence is called eventually periodic. The following facts about
1058: an LFSR of length $r$ are well-known \cite{R}.
1059:
1060: \begin{enumerate}
1061: \item There are only finitely many possible states, and the state with all
1062: the cells zero will produce a $0-$sequence. The output sequence is
1063: eventually periodic and the maximal period is $p^{nr}-1$.
1064:
1065: \item The Poincar\'{e} series $\ g(x)=\sum\limits_{i=0}^{\infty }a_{i}x^{i}$
1066: associated with the output sequence is called the generating function of the
1067: sequence. It is a rational function over $\mathbb{F}_{p^{n}}$\ of the form $%
1068: g(x)=\frac{L(x)}{R(x)}$, with $L(x)$, $R(X)$ $\in \mathbb{F}_{p^{n}}[x]$, $%
1069: deg(R(X))<r$. The output sequence is strictly periodic if and only if $%
1070: deg(L(X))<deg(R(X))$.
1071:
1072: \item There is a one-to-one correspondence between LFSRs of length $r$ with
1073: $q_{r}\neq 0$ and rational functions $\frac{L(x)}{R(x)}$ with $deg(R(X))=r$
1074: and $deg(L(X))<r$.
1075: \end{enumerate}
1076:
1077: We set $\mathbb{F}_{p^{n}}(x)$ for the field of rational functions over $%
1078: \mathbb{F}_{p^{n}}$, and $N^{\infty }(\mathbb{F}_{p^{n}})$ for the set of
1079: sequences \ of the form $\{b_{0},\mathbb{\cdots },b_{u}\}$, $b_{i}\in
1080: \mathbb{F}_{p^{n}}$, $\ 0\leq i\leq u$, $u\in \mathbb{N}$. From the above
1081: considerations, it is possible to identify an LFSR with a function $F_{u}$, $%
1082: u\in \mathbb{N}$, defined as follows:
1083: \begin{equation}
1084: \begin{array}{cccc}
1085: F_{u}: & \mathbb{F}_{p^{n}}(x) & \rightarrow & N^{\infty }(\mathbb{F}%
1086: _{p^{n}}) \\
1087: & \sum\limits_{i=0}^{\infty }a_{i}x^{i} & \rightarrow & \{a_{0},\mathbb{%
1088: \cdots },a_{u}\}.
1089: \end{array}
1090: \label{A}
1091: \end{equation}
1092: We set
1093: \begin{equation*}
1094: \mathcal{H}=\{H(t,f)\mid f(x)\in \mathbb{Z}[x]\text{, in \ one variable,
1095: with splitting field }\mathbb{Q}\},
1096: \end{equation*}
1097: and $N^{\infty }$ $\left( \mathbb{Z}\right) $\ for the set \ of finite
1098: sequences of integers. Also, for each $u\in \mathbb{N}$, and a prime number $%
1099: p$, we define
1100: \begin{equation}
1101: \begin{array}{cccc}
1102: F_{u,p}: & \mathcal{H} & {\ \rightarrow } & \mathbb{N}^{\infty }\left(
1103: \mathbb{Z}\right) \\
1104: & H(t,f) & {\ \rightarrow } & \{N_{0}(f,p),N_{1}(f,p),\mathbb{\cdots }%
1105: ,N_{u}(f,p)\}\text{.}
1106: \end{array}
1107: \label{B}
1108: \end{equation}
1109: Thus the mappings $F_{u,p}$ can be seen as LFSRs, or stream ciphers, over $%
1110: \mathbb{Z}$. If we replace each $N_{u}(f,p)$ by its binary representation,
1111: then the $\ F_{u,p}$ are LFSRs. For practical purposes it is necessary that $%
1112: F_{u,p}$ \ can be computed efficiently, i.e., in polynomial time. With the
1113: above notation our second result is the following.
1114:
1115: \begin{theorem}
1116: \label{theoB} For every $H(t,f)$\ $\in $ $\mathcal{H}$, the computation of \
1117: $F_{u,p}(H(t,f))$ involves $O(u^{2}d_{f}l_{f})$ arithmetic operations, and
1118: the integers on which these operations are performed have binary length
1119: \begin{equation*}
1120: O(\max \{\text{ }(l_{f}+u)\log p,\text{ }\log (d_{f}l_{f})\}).
1121: \end{equation*}
1122: \end{theorem}
1123:
1124: The proof of this theorem will be given at the end of this section. This
1125: proof requires some preliminary results. We set $t=q^{-s}$, and
1126: \begin{equation*}
1127: Z(s,f)=Z(t,f)=\sum\limits_{m=0}^{\infty }c_{m}(f,p)t^{m},
1128: \end{equation*}
1129: with $c_{m}(f,p)=vol(\{x\in \mathbb{Z}_{p}\mid v_{p}(f(x))=m\})$.
1130:
1131: \begin{proposition}
1132: Let $f(x)\in \mathbb{Z}[x]\setminus \mathbb{Z}$ be a polynomial in one
1133: variable and $p$ a prime number. The following formula holds for $N_{n}(f,p)$%
1134: :
1135: \begin{equation}
1136: N_{n}(f,p)=\left\{
1137: \begin{array}{ccc}
1138: 1\text{,} & \text{if} & n=0\text{;} \\
1139: p^{n}\left( 1-\sum\limits_{j=1}^{n}c_{j-1}(f,p)\right) \text{,} & \text{if}
1140: & n\geqslant 1\text{.}
1141: \end{array}
1142: \right. \label{For1}
1143: \end{equation}
1144: \end{proposition}
1145:
1146: \begin{proof}
1147: The result follows by comparing \ the coefficient of $t^{n}$ of the series
1148: \begin{equation*}
1149: \sum\limits_{n=0}^{\infty }\frac{N_{n}(f,p)}{p^{n}}t^{n}\text{ \ \ and \ }%
1150: \sum\limits_{n=0}^{\infty }d_{n}t^{n}\text{,}
1151: \end{equation*}
1152: in the following equality :
1153: \begin{equation*}
1154: H(t,f)=\sum\limits_{n=0}^{\infty }\frac{N_{n}(f,p)}{p^{n}}t^{n}=\frac{%
1155: 1-t\left( \sum\limits_{m=0}^{\infty }c_{m}(f,p)t^{m}\right) }{1-t}%
1156: =\sum\limits_{n=0}^{\infty }d_{n}t^{n}.
1157: \end{equation*}
1158: \end{proof}
1159:
1160: We associate to each $u\in T(f,l_{f})$, and $j\in \mathbb{N}$, a rational
1161: integer $a_{j}(u)$ defined as follows:
1162:
1163: \begin{equation}
1164: a_{j}(u)=\left\{
1165: \begin{array}{cc}
1166: \frac{(p-1)}{p^{l(u)+1+y(u)}}\text{, \ if} &
1167: \begin{array}{c}
1168: l(u)=1+l_{f}\text{, }W(u)\geqq 2\text{, }j=W^{\ast }(B_{u})+y(u)W(u)\text{,}
1169: \\
1170: \text{for some }y(u)\in \mathbb{N}\text{;}
1171: \end{array}
1172: \\
1173: & \\
1174: \frac{(p-Val(u))}{p^{l(u)+1}}\text{, \ if} & 0\leqq l(u)\leqq l_{f}\text{, }%
1175: W(u)\neq 1\text{, }j=W^{\ast }(B_{u})\text{;} \\
1176: & \\
1177: \frac{(p-1)}{p^{l(u)+1+y(u)}}\text{, \ if} &
1178: \begin{array}{c}
1179: u\in \mathcal{M}_{T(f,l_{f})},j=W^{\ast }(B_{u})+y(u)\text{,} \\
1180: \text{for some }y(u)\in \mathbb{N}\text{;}
1181: \end{array}
1182: \\
1183: & \\
1184: 0\text{, \ \ \ \ \ \ \ \ \ \ \ if} & W(u)=1\text{, and }u\notin \mathcal{M}%
1185: _{T(f,l_{f})}\text{;} \\
1186: & \\
1187: 0\text{, \ \ \ \ \ \ \ \ \ \ \ \ } & \text{in other cases.}
1188: \end{array}
1189: \right. \label{For2}
1190: \end{equation}
1191:
1192: \begin{proposition}
1193: Let $f(x)\in \mathbb{Z}[x]\setminus \mathbb{Z}$ be a polynomial in one
1194: variable, with splitting field $\mathbb{Q}$, and $p$ a prime number. The
1195: following formula holds:
1196: \begin{equation}
1197: c_{j}(f,p)=\sum\limits_{u\in T(f,l_{f})}a_{j}(u),\text{ }j\geqq 0.
1198: \label{For4}
1199: \end{equation}
1200: \end{proposition}
1201:
1202: \begin{proof}
1203: As a consequence of lemma (\ref{lemma}), we have the following identity:
1204: \begin{equation}
1205: Z(t,f)=\sum\limits_{u\in T(f,l_{f})}L_{u}(t), \label{For4a}
1206: \end{equation}
1207: with
1208: \begin{equation}
1209: L_{u}(t)=\left\{
1210: \begin{array}{cc}
1211: \frac{(p-1)t^{W^{\ast }(B_{u})}}{p^{l(u)+1}(1-p^{-1}t^{\text{ }W(u)})}\text{%
1212: , \ if} & l(u)=1+l_{f}\text{, }W(u)\geqq 2\text{;} \\
1213: & \\
1214: \frac{(p-Val(u))}{p^{l(u)+1}}t^{W^{\ast }(B_{u})}\text{, \ if} & 0\leqq
1215: l(u)\leqq l_{f}\text{, }W(u)\neq 1\text{;} \\
1216: & \\
1217: \frac{(p-1)t^{W^{\ast }(B_{u})}}{p^{l(u)+1}(1-p^{-1}t)}\text{, \ if} & u\in
1218: \mathcal{M}_{T(f,l_{f})}\text{;} \\
1219: & \\
1220: 0\text{, \ \ \ \ \ if} & W(u)=1\text{, and }u\notin \mathcal{M}_{T(f,l_{f})}.
1221: \end{array}
1222: \right. \label{For5}
1223: \end{equation}
1224:
1225: The result follows by comparing the coefficient of $t^{j}$ in the series $%
1226: Z(t,f)=\sum\limits_{m=0}^{\infty }c_{m}(f,p)t^{m}$, and $Z(t,f)=\sum%
1227: \limits_{u\in T(f,l_{f})}L_{u}(t)$.
1228: \end{proof}
1229:
1230: \begin{proposition}
1231: \label{prop8}Let $f(x)\in \mathbb{Z}[x]\setminus \mathbb{Z}$ be a polynomial
1232: in one variable, with splitting field $\mathbb{Q}$, and $p$ a prime number.
1233:
1234: \begin{enumerate}
1235: \item The computation \ of \ $N_{n}(f,p)$, $n\geqq 1$, \ from \ the $%
1236: c_{j-1}(f,p)$, $j=1,\mathbb{\cdots },n$, involves \ $O(n)$ \ \ arithmetic
1237: operations on integers with binary length \ $O(n\log p)$.
1238:
1239: \item The computation \ of \ $c_{j}(f,p)$, $j\geqq 0$, \ from \ $Z(t,f)$,
1240: involves \ $O(d_{f}l_{f})$ \ \ arithmetic operations on integers \ with
1241: binary length \
1242: \begin{equation*}
1243: \mathit{O}(\max \{(j+l_{f})\log p,\log p,\log (d_{f}l_{f})\}).
1244: \end{equation*}
1245:
1246: \item The computation \ of any $N_{n}(f,p)$, $n\geqq 1$, from $Z(t,f)$,
1247: involves $O(nd_{f}l_{f})$ \ arithmetic operations on integers with binary
1248: length \
1249: \begin{equation*}
1250: \mathit{O}(\max \{(n+l_{f})\log p,\log (d_{f}l_{f})\}).
1251: \end{equation*}
1252: \end{enumerate}
1253: \end{proposition}
1254:
1255: \begin{proof}
1256: (1) By (\ref{For2})\ and (\ref{For4}), $c_{j}(f,p)=\frac{v_{j}}{p^{m_{j}}}$,
1257: $v_{j},m_{j}\in \mathbb{N}$. In addition,
1258: \begin{equation*}
1259: c_{j-1}(f,p)=p^{-j+1}N_{j-1}(f,p)-p^{-j}N_{j}(f,p).
1260: \end{equation*}
1261: Thus $p^{n}c_{j-1}(f,p)\in \mathbb{N}$, for $\ j=1,\mathbb{\cdots },n$, and
1262: \ $m_{j}\leq $\ $n,$ for $\ j=1,\mathbb{\cdots },n$. From (\ref{For1}), it
1263: follows that $\ $%
1264: \begin{equation}
1265: N_{n}(f,p)=p^{n}-\sum\limits_{j=1}^{n}p^{n}c_{j-1}(f,p),\text{ }n\geqslant 1.
1266: \label{For3}
1267: \end{equation}
1268: The above formula implies that the computation of \ $N_{n}(f,p)$, $n\geqq 1$%
1269: , \ from \ the $c_{j-1}(f,p)$, $j=1,\mathbb{\cdots },n$, involves \ $O(n)$ \
1270: \ arithmetic operations on integers with binary length \ $O(n\log p)$.
1271:
1272: (2) The \ computation of $a_{j}(u)$ from $L_{u}(t)$ (i.e. from $Z(t,f)$, cf.
1273: (\ref{For4a}))\ involves $O(1)$ arithmetic operations (cf. (\ref{For2}), (%
1274: \ref{For5})) \ on integers of binary length $O(\max \{\log p$, $\log
1275: (d_{f}l_{f})\})$. Indeed, since the numbers $l(u)$, $W^{\ast }(B_{u})$, $%
1276: W(u) $, $u\in T(f,l_{f})$ are involved in this computation, we know \ by
1277: proposition \ref{prop4.1} that their binary length is bounded by $O(\max
1278: \{\log p$, $\log (d_{f}l_{f})\})$.
1279:
1280: The cost of computing \ $c_{j}(f,p)$ from $L_{u}(t)$, $u\in T(f,l_{f})$
1281: (i.e. from $Z(t,f)$) is bounded by the number of vertices of $T(f,l_{f})$ \
1282: multiplied by an upper bound \ for the cost of computing $a_{j}(u)$ from $%
1283: L_{u}(t)$, for any $j$, and $u$ (cf. (\ref{For4})). Therefore, \ from the
1284: previous discussion the cost of computing $c_{j}(f,p)$ from $Z(t,f)$ is
1285: bounded by $O(d_{f}$ $l_{f})$ \ \ arithmetic operations. These arithmetic
1286: operations are performed on integers of binary length bounded \ by $O(\max
1287: \{(j+l_{f})\log p,\log p,$ $\log (d_{f}l_{f})\}).$ Indeed, the binary
1288: lengths of the numerator and the denominator of \ $a_{j}(u)+a_{j}(u^{^{%
1289: \prime }})$, $u$, $u^{^{\prime }}\in T(f,l_{f})$ are bounded by $%
1290: (l_{f}+1+j)\log p$ (cf. (\ref{For2})). Thus, the mentioned arithmetic
1291: operations \ for calculating $c_{j}(f,p)$ from\ $L_{u}(t)$ are performed on
1292: integers whose binary length is bounded by \ $O(\max \{(j+l_{f})\log p,\log
1293: p,$ $\log (d_{f}l_{f})\})$.
1294:
1295: (3) The third part \ follows the first and second parts by (\ref{For3}).
1296: \end{proof}
1297:
1298: \subsection{Proof of Theorem \ref{theoB}}
1299:
1300: The theorem follows from proposition \ref{prop8} (3).
1301:
1302: \begin{thebibliography}{99}
1303: \bibitem{A-G} Anshel, M., and Goldfeld, D., Zeta functions, one-way
1304: functions and pseudorandom number generators, Duke Math. J., 88, 2 (1997),
1305: 371-390.
1306:
1307: \bibitem{B-M} Bierstone., \ E., Milman, P., Canonical desingularization in
1308: characteristic zero by blowing up the maximum strata of a local invariant,
1309: Invent. Math. 128 (1997), 207-302.
1310:
1311: \bibitem{B-S} Bodn\'{a}r, G\'{a}bor; Schicho, Josef. A computer program for
1312: the resolution of singularities. Resolution of singularities (Obergurgl,
1313: 1997), 231--238, Progr. Math., 181, Birkh\"{a}user, Basel, 2000.
1314:
1315: \bibitem{B-S1} Bodn\'{a}r, G., Schicho, J., \ Automated resolution of
1316: singularities for hypersurfaces, Journal of Symbolic Computation 30 (2000),
1317: 401-428.
1318:
1319: \bibitem{D1} Denef J., \ Report on Igusa's local zeta function,
1320: S\'{e}minaire Bourbaki 1990/1991 (730-744) in Asterisque 201-203 (1991),
1321: 359-386.
1322:
1323: \bibitem{D2} Denef J., On the degree of Igusa's \ local zeta functions,
1324: Amer. Math. J. 109 (1987), \ 991-1008.
1325:
1326: \bibitem{D-H} Denef J., Hoornaert Kathleen, \ Newton polyhedra and Igusa
1327: local zeta function, Journal of Number Theory 89 (2001), 31-64.
1328:
1329: \bibitem{K} Knuth, D., The art of computer programming, 3 volumes,
1330: Addison-Wesley, 1999.
1331:
1332: \bibitem{G-J} Garey, M. R., Johnson, D. S., Computers and intractability: A
1333: guide to the theory of $\mathbf{NP}$-Completeness, 1979, \ W. H. Freedman
1334: and Company, New York.
1335:
1336: \bibitem{G1} Goldman, Jay R., \ Numbers of solution of congruences:
1337: Poincar\'{e} series for strongly nondegenerate forms, Proc. Amer. Math.
1338: Soc., 87 (1983), 586-590.
1339:
1340: \bibitem{G2} Goldman, Jay R., \ Numbers of solution of congruences:
1341: Poincar\'{e} series for algebraic curves, Adv. in Math. 62 (1986), 68-83.
1342:
1343: \bibitem{G-L-N} Goldreich, O., Levin, L. A., and Nisan, N., On constructing
1344: 1-1 one-way functions, preprint available at
1345: http://www.wisdom.weizmann.ac.il/~\symbol{126}oded/cryptography.html.
1346:
1347: \bibitem{G-K-L} Goldreich, O., \ Krawczyk, H., \ Luby, M., \ On the
1348: existence of pseudorandom number generators, SIAM J. on Computing, Vol. 22
1349: (1993), 1163-1175.
1350:
1351: \bibitem{I1} Igusa, Jun-Ichi, An introduction to the theory of local zeta
1352: functions, AMS /IP studies in \ advanced mathematics, v. 14, 2000.
1353:
1354: \bibitem{I2} Igusa, J., \ Complex powers and asymptotic expansions, I
1355: Crelles J. Math., 268/269 (1974), 110-130; II, ibid., 278/279 (1975),
1356: 357-368.
1357:
1358: \bibitem{I3} Igusa, J., A stationary phase formula for $p-$adic integrals \
1359: and its applications, in Algebraic Geometry and its Applications,
1360: Springer-Verlag (1994), 175-194.
1361:
1362: \bibitem{L-L-L} Lenstra, A.K., Lenstra, H.W., Lov\'{a}sz, L., Factoring
1363: polynomials with rational coefficients, Math. Ann. 261 (1982), 515-534.
1364:
1365: \bibitem{R} Ruepel R., Analysis and design of stream ciphers,
1366: Springer-Verlag, New York, 1986.
1367:
1368: \bibitem{S-Z-G} Saia, M.J., Zuniga-Galindo, W.A., Local zeta functions,
1369: Newton polygons and non degeneracy conditions, to appear in Trans. Amer.
1370: Math. Soc.
1371:
1372: \bibitem{S} P. Shor, Algorithms for quantum computation, discrete
1373: logarithms, and factoring, in Proceedings of 35th Annual Symposium on
1374: Foundations of Computer Science, IEEE Computer Society Press, Los alamitos,
1375: California, 1994, 124-134.
1376:
1377: \bibitem{Vi} Villamayor, O., Constructiveness of Hironaka's resolution,
1378: Ann. Scient. Ecole Norm. Sup. 4 (1989), 1-32.
1379:
1380: \bibitem{V-K-S} von zur Gathen, J., Karpinski, M., Shparlinski, I.,
1381: Counting curves and their projections, in Proceedings ACM\ STOC\ 93, 805-812.
1382:
1383: \bibitem{Wa} Daqing Wan, Algorithmic theory of zeta functions over finite
1384: fields,to appear in MSRI Computational Number Theory Proceedings.
1385:
1386: \bibitem{Z-G1} Zuniga-Galindo W. A., Igusa's local zeta functions of
1387: semiquasihomogeneous polynomials, Trans. Amer. Math. Soc. 353, (2001),
1388: 3193-3207.
1389:
1390: \bibitem{Z-G2} Zuniga-Galindo W. A., Local zeta functions and Newton
1391: polyhedra, \ to appear in Nagoya Math. J.
1392:
1393: \bibitem{ZG3} Zuniga-Galindo W.A., Local zeta function for non-degenerate
1394: homogeneous mappings, preprint 2003.
1395: \end{thebibliography}
1396:
1397: \end{document}
1398: