Quantum Algorithm for Polynomial Root Finding Problem
Guodong Sun, Shenghui Su
College of Computer Science
Beijing University of Technology
Beijing, China
sgd-150@163.com
Maozhi Xu
School of Mathematics Sciences
Peking University
Beijing, China
Mzxu@pku.com
Abstract—Quantum computation is a new computing model
based on fundamental quantum mechanical principle.
Grover’s algorithm finds the solution for a searching problem
in the square root time of exhaustive search. Brassard, Hoyer,
Tapp’s algorithm counts the number of solutions for a
searching problem. Through exploiting the two quantum
algorithms, we propose a quantum algorithm for solving a new
cryptography problem----polynomial root finding problem,
which could be used to design a cryptosystem. The algorithm
will take O(
/Mt) steps for finding one of the t solutions to the
problem, where M is the modular of the equation. The success
rate of the algorithm is a constant and the cost of the algorithm
depends on the calculations of modular exponentiation and the
number of iterations.
Keywords-Polynomial root finding problem; Quantum
searching; Quantum counting; Signature algorithm
I. INTRODUCTION
In the past few years, quantum computation [1] based
on fundamental quantum mechanical principle has attracted
numerous attractions of people all over the world. The
concept of quantum computation was firstly put forward by
R.Feynman [2]. In early 1980s, he pointed out that
computation in general could be done more efficiently if it
made use of these quantum physical phenomena. After that,
the first great quantum algorithm was proposed by Shor. In
1994, Shor [3, 4] invented a polynomial time quantum
mechanical algorithm for solving the factorization problem
and discrete logarithm problem which were and still are
widely believed to have no polynomial solution on a
classical computer. The results can be used to break the
widely used RSA cryptosystem [5] and Diffie-Hellman key-
exchange protocol. However, Shor’s algorithm is not a
universal algorithm and it can only solve a few computing
problem. After the pioneering work of Shor, in 1996,
Grover [6, 7] designed another quantum mechanical
algorithm for searching a marker element in an unsorted
database, providing a square root speedup for exhaustive
search over classical algorithms. In the same year, Brassard,
Hoyer and Tapp [8, 9] presented the idea of quantum
counting which combined Shor’s factoring algorithm and
Grover’s searching algorithm. These two quantum
algorithms have been widely used to many cryptography
applications, such as the cryptanalysis of block ciphers [10]
and the exhaustive attack of DES [11].
Nowadays, with the development of the cryptography
technology such as encryption, key distribution, digital
signatures, and one-way functions, there are many
applications such as secure encryption, signing contracts,
and electronic voting in the information society. As is
known to all, the security of these activities is guaranteed by
public-key cryptography [12]. A great deal of public-key
cryptosystems are believed to be secure against the classical
computer, such as RSA and ECC [13]. However, in the case
of a quantum computer is used, these cryptosystems are
thoroughly unsecure due to the invention of Shor’s
algorithm. Besides, the strength of ciphers is challenged in
virtue of Grover’s searching algorithm. In order to confront
the challenge of the quantum computer in the future, we
must find as more quantum-resistant public-key
cryptosystems as possible to ensure the information security
in the quantum computer world.
In this paper, a new cryptographic problem named
polynomial root finding problem, of which the hardness is
the basis of the signature algorithm for the public-key
cryptosystem REESSE1+ [14], is considered under the
attack of a quantum computer. Specifically, we try to solve
the polynomial root finding problem with the help of
Grover’s quantum searching algorithm and Brassard, Hoyer
and Tapp’s quantum counting algorithm.
Throughout the paper, unless otherwise specified, all the
variables and constants are positive integers. n 80 is the
length of cipher, the sign % represents ‘modulo’, M is a large
prime and
M denotes M-1, lg x means the logarithm of x to
the base 2, |
x| denotes the absolute value of an element x.
II. D
ESCRIPTION OF THE PROBLEM
In this section, we give some definitions of problems
which are relevant to each other. These definitions may not
have be used to a special schema, but they are potential
candidates to design a public-key encryption scheme or a
signature schema.
Definition 1: Given
a ≠ 0, 1, |b| + |c| ≠ 0, and d ≠ 0, solving
ax
n
+ bx
n-1
+cx+d 0(% M ) for x, x
∈
[1, M ], is called the
polynomial root finding problem, shortly PRFP.
From definition 1, we know that the variables
a and d
cannot be zero, b and c cannot be zero simultaneously.
Thus, the PRFP is a special form of a
n-degree polynomial
that the relevant coefficients satisfy some conditions. In
fact, the hardness of PRFP is the basis of the security of the
famous public-key cryptosystem REESSE1+. In
2014 10th International Conference on Computational Intelligence and Security
978-1-4799-7434-4/14 $31.00 © 2014 IEEE
DOI 10.1109/.39
469
2014 10th International Conference on Computational Intelligence and Security
978-1-4799-7434-4/14 $31.00 © 2014 IEEE
DOI 10.1109/CIS.2014.40
469
2014 Tenth International Conference on Computational Intelligence and Security
978-1-4799-7434-4/14 $31.00 © 2014 IEEE
DOI 10.1109/CIS.2014.40
469