METHOD AND DEVICE FOR CRYPTOGRAPHIC KEY GENERATION

Abstract

A method and a device for generation of a cryptographic key pair for use in a (generalized) Goldwasser-Micali cryptosystem. The device generates a first prime p≡1 (mod 2k), where k≧1 is an integer, and a second prime q≡3 (mod 4) or q≡1 (mod 4); computes a modulus N=pq; picks an integer y∈N\N, where N is a set of integers whose Jacobi symbol is 1 and N is a set of quadratic residues; and outputs a public key pk={N,y,k} and a private key sk={p,k}.

Description

TECHNICAL FIELD

The present disclosure relates generally to cryptography, and in particular to a cryptosystem based on the Goldwasser-Micali cryptosystem.

BACKGROUND

This section is intended to introduce the reader to various aspects of art, which may be related to various aspects of the present disclosure that are described and/or claimed below. This discussion is believed to be helpful in providing the reader with background information to facilitate a better understanding of the various aspects of the present disclosure. Accordingly, it should be understood that these statements are to be read in this light, and not as admissions of prior art.

The Goldwasser-Micali (GM) cryptosystem is a well-known public key cryptosystem that encrypts one bit of the plaintext at a time; put another way, the message space is {0,1}^k, with k=1.

At EUROCRYPT 2013, Joye and Libert presented a generalized GM cryptosystem in which k bits are encrypted at a time, i.e. the message space is {0,1}k, with k≧1 [see Marc Joye and Benoit Libert. Efficient cryptosystems from 2^k-th power residue symbols. In T. Johansson and P. Nguyen, editors, Advances in Cryptology—EUROCRYPT 2013, volume 7881 of Lecture Notes in Computer Science, pages 76-92. Springer, 2013.]. As in GM, the generalized scheme uses a composite N=pq where p and q are prime, and p,q≡1 (mod 2^k). In addition, y∈_N\_N and the public key is pk={N,y,k} while the private key is sk={p,k}.

Given a plaintext message m=Σ_i=0^k−12ⁱ with m_i∈{0,1}, the corresponding ciphertext c is formed as c=y^mx^2k mod N for some random element x∈_N*. Plaintext message m is then recovered from the ciphertext c as the unique integer in [0,2^k [satisfying the relation

$\begin{array}{cc}{\left[{\left(\frac{y}{p}\right)}_{2^k}\right]}^m\equiv {\left(\frac{c}{p}\right)}_{2^k}\left(\mathrm{mods}p\right).& \left(1\right)\end{array}$

where

${\left(\frac{y}{p}\right)}_{2^k}$

denotes the 2^k-th power residue symbol of y modulo p, defined as

${\left(\frac{y}{p}\right)}_{2^k}=y^{\frac{p-1}{2^k}}$

mods p. (It is noted that a mods p represents the absolute smallest residue of a modulo p, namely, the complete set of absolute smallest residues are: −(p−1)/2, . . . , −1, 0, 1, . . . , (p−1)/2). Doing so 2^nd power residue symbol (i.e., when k=1) boils down to the classical Legendre symbol.) Solving Eq. (1) can be carried out with a variation of the Poligh-Hellman algorithm; see section 3.2 of the previously mentioned article.

The generalized GM cryptosystem as described in the previously mentioned article meets the standard security notion of semantic security under the quadratic residuosity assumption and the squared Jacobi symbol assumption [see the article with the same name by the same authors published in the Cryptology ePrint Archive as Report 2013/435].

In order to facilitate comprehension, the following notation is introduced. As already mentioned, let N=pq be the product of two (odd) primes p and q. The set of integers whose Jacobi symbol is 1 is denoted by _N,

$N_{}=\left\{a\in N_{*}^{}|\left(\frac{a}{N}\right)=1\right\};$

the set of quadratic residues is denoted by _N,

$N_{}=\left\{a\in N_{*}^{}|\left(\frac{a}{p}\right)=\left(\frac{a}{q}\right)=1\right\};$

and the set of integers whose Jacobi symbol is −1 is denoted by _N,

$N_{}=\left\{a\in N_{*}^{}|\left(\frac{a}{N}\right)=-1\right\}.$

It is to be noted that _N is a subset of _N.

Definition 1 (Quadratic Residuosity Assumption).

Let RSAGen be a probabilistic algorithm that, given a security parameter κ, outputs primes p and q such that p, g≡1 (mod 2^k), and their product N=pq. The Quadratic Residuosity (k−QR) assumption asserts that the function Adv_D^k−QR(1^k), defined as the distance

$\mathrm{Pr}\left[\left(x,N\right)=1x\stackrel{R}{\leftarrow }N_{}]-\mathrm{Pr}[\left(x,N\right)=1x\stackrel{R}{\leftarrow }N_{}\backslash N_{}\right]$

is negligible for any probabilistic polynomial-time distinguisher ; the probabilities are taken over the experiment of running (N,p,q)←RSAGen(1^κ) and choosing at random x∈_N and x∈_N\_N.

Definition 2 (Squared Jacobi Symbol Assumption).

Let RSAGen be a probabilistic algorithm that, given a security parameter κ, outputs primes p and g such that p, q≡1 (mod 2^k), and their product N=pq. The Squared Jacobi Symbol (k−SJS) assumption asserts that the function Adv_D^k−SJS(1^κ), defined as the distance

$\mathrm{Pr}\left[\left(y^2\mathrm{mod}N,N\right)=1y\stackrel{R}{\leftarrow }N_{}]-\mathrm{Pr}[\left(y^2\mathrm{mod}N,N\right)=1y\stackrel{R}{\leftarrow }N_{}\right]$

is negligible for any probabilistic polynomial-time distinguisher ; the probabilities are taken over the experiment of running (N,p,q)←RSAGen (1^κ) and choosing at random y∈_N and _N.

The case k=1 corresponds to the GM cryptosystem which has indistinguishable encryptions (semantic security) solely under the standard Quadratic Residuosity assumption.

It is also noted that the generalized GM requires a special prime generation algorithms for the generation of p and q during key generation.

It will therefore be appreciated that it is desirable to provide a generalized GM cryptosystem whose semantic security solely relies on a quadratic residuosity assumption and in which the key generation is simplified.

The present disclosure provides such a generalized GM cryptosystem.

SUMMARY

In a first aspect, the disclosure is directed to a method for generation of a cryptographic key. A device generates a first prime p≡1 (mod 2^k), where k≧1 is an integer; generates a second prime q≡3 (mod 4) or q≡1 (mod 4), q≢1 (mod 2^k); computes a modulus N that is a multiple of the product between the first prime p and the second prime q; picks an integer y∈_N\_N, where _N is the set of integers whose Jacobi symbol is 1 and _N is the set of quadratic residues; and outputs a public key pk={N,y,k}.

In a first embodiment, the device generates a private key sk={p,k}. It is advantageous that the device stores the private key sk.

In a second embodiment, the public key pk is for use in a cryptosystem in which a message m∈, where ={0,1}^k, is encrypted by picking a random x∈_N* and calculating a ciphertext c=y^mx^2k mod N.

In a second aspect, the invention is directed to a cryptographic device comprising a processing unit configured to: generate a first prime p≡1 (mod 2^k), where k≧1 is an integer; generate a second prime q≡3 (mod 4) or q≡1 (mod 4), q≢1 (mod 2^k); compute a modulus N that is a multiple of the product between the first prime p and the second prime q; pick an integer y∈_N\_N, where _N is the set of integers whose Jacobi symbol is 1 and _N is the set of quadratic residues; and output a public key pk={N,y,k}.

In a first embodiment, the processing unit is further configured to generate a private key sk={p,k}. It is advantageous that the processing unit is configured to store the private key sk, preferably in a memory or in the processing unit.

In a second embodiment, the processing unit is configured to output the public key via a communication interface. It is advantageous that the processing unit is configured to use a communication protocol to output the public key via the communication interface.

In a third embodiment, the cryptographic device belongs to one of the group of: a mobile device, a communication device, a game device, a set top box, a TV set, a tablet, a laptop and a cryptographic chip.

In a fourth embodiment, the public key pk is for use in a cryptosystem in which a message m∈, where ={0,1}^k, is encrypted by picking a random x∈_N* and calculating a ciphertext c=y^mx^2k mod N.

In a third aspect, the disclosure is directed to a computer program product storing instructions that, when executed by a processor, perform the method of the first aspect.

BRIEF DESCRIPTION OF DRAWINGS

Preferred features of the present disclosure will now be described, by way of non-limiting example, with reference to the accompanying drawings, in which:

FIG. 1 illustrates a first preferred embodiment; and

FIG. 2 illustrates a second preferred embodiment.

DESCRIPTION OF EMBODIMENTS

It will be shown that, quite surprisingly, a small modification to the key generation procedure of the generalized GM cryptosystem enables its semantic security to rely solely on a quadratic residuosity assumption.

As already mentioned, the generalized GM cryptosystem defines primes p and q such that p,q≡1 (mod 2^k). It is however observed that the decryption process only involves prime p.

This observation makes it possible to develop a new proof technique that allows to prove the semantic security when q≡3 (mod 4). Remarkably, the new security proof (presented hereafter for completeness) solely assumes the quadratic residuosity assumption for RSA moduli N=pq where p≡1 (mod 2^k) and q≡3 (mod 4).

Furthermore, a close inspection of the security proof offered in the article published in the Cryptology ePrint Archive shows that the very same proof carries on when q≡1 (mod 4). The proof crucially requires that the square roots of a square in _N* all have the same Jacobi symbol. This is readily satisfied when −1∈_N, or equivalently when p,q≡1 (mod 4).

Security Analysis (q≡3 (Mod 4))

Let N=pq be the product of two primes p and q with p≡1 (mod 2^k) for some k≧1. Let

$N_{}=\left\{x|\left(\frac{x}{N}\right)=1\right\}$

and _N={x²|x∈_N*}. For 0≦i≦k, consider the subsets D_i of _N given by D_i={y²ⁱ mod N|y∈_N\_N} and define the subgroup of 2^k-th residues R_k={y^2k mod N|y∈_N*}.

Lemma 1.

Let N=pq be the product of two large primes p and q where p≡1 (mod 2^k) for some k≧1 and q≡3 (mod 4). Then, for any w∈_N, letting W:=w²ⁱ⁻¹ for a given 1≦i≦k, gives W∈R_k∪U_j=1^k−1D_j.

Further, if w is uniform over _N, then W is uniform over D_j with probability

$\frac{1}{2^{j-i+1}}$

for i≦j≦k−1 and W is uniform over R_k with probability

$\frac{1}{2^{k-i}}.$

Proof.

It is assumed that w is uniform over _N.The case i=k (which includes the case k=1) yields W=w^2k−1 with w∈_N. It is then readily verified that W is uniform over R_k with probability 1.It is henceforth supposed that i≦k−1 and k≧2. In particular, this implies p≡1 (mod 4) and thus

$\left(\frac{-1}{p}\right)=1.$

Denoting by (ŵ_p, ŵ_q) the CRT representation of a square root ŵ of w (i.e., ŵ_p=ŵ mod p and ŵ_q=w mod q), the four square roots of w modulo N are given by (±ŵ_p,±ŵ_q). Since

$\left(\begin{array}{c}-1\\ q\end{array}\right)=-1,$

it can be assumed without loss of generality that

$\left(\frac{{\hat{w}}_q}{q}\right)=\left(\frac{{\hat{w}}_p}{p}\right),$

or equivalently that ŵ∈_N. If ŵ∈_N the process can be re-iterated, and so on. More generally, t is defined as the largest integer in {1, . . . , k−i} such that w=ŵ^2t for some ŵ∈_N. It is then possible to write W=ŵ^2t+i−1 for some ŵ∈_N. It is worth noting that since t is the largest integer in the set {1, . . . , k−i}, ŵ∈_N only when t=k−i. Defining j=t+i−1 (observe that i≦j≦k−1), gives W=ŵ^2j∈D_j if ŵ∉_N (i.e., ŵ∈_N\_N) and W=ŵ^2k−1∈R_k if ŵ∈_N. The probability that W∈D_j (for i≦j≦k−1) is Pr [w=ŵ^2t and

$\hat{w}\notin N_{}]=\frac{1}{2^t}=\frac{1}{2^{j-i+1}}$

and the probability that W∈R_k is

$\mathrm{Pr}\left[W\notin {\bigcup }_{j=i}^{k-1^{}}D_j\right]=1-{\sum }_{j=i}^{k-1}\frac{1}{2^{j-i+1}}=\frac{1}{2^{k-i}}.$

Theorem 1.

For RSA moduli N=pq such that p≡1 (mod 2^k) and q≡3 (mod 4), the Gap 2^k-Residuosity assumption (defined in the previously mentioned paper published at EUROCRYPT) holds if the k QR assumption (see Definition 1) holds. More precisely, for any Probabilistic Polynolial Time (PPT) distinguisher against the latter, there exists a k−QR distinguisher with comparable running time and for which

${\mathrm{Adv}}_{}^{\mathrm{Gap}-2^k-\mathrm{Res}}\left(1^K\right)\leqq \frac{k+1}{2}·{\mathrm{Adv}}_{ℬ}^{k-\mathrm{QR}}\left(1^K\right).$

[Here the k−QR assumption is defined for RSA moduli N=pq such that p≡1 (mod 2^k) and q≡3 (mod 4).]

Proof.

Let be an adversary against Gap−2^k−Res running in time t. Write:

${\epsilon }_{i}=\mathrm{Pr}\left[\left(x,N\right)=1x\stackrel{R}{\leftarrow }D_i\right],$

for i∈{0, . . . , k−1} and

${\epsilon }_k=\mathrm{Pr}\left[\left(x,N\right)=1x\stackrel{R}{\leftarrow }R_k\right].$

The advantage of against Gap−2^k−Res is:

(1^κ)=|ε₀−ε_k|.

First k distinguishers ₁, . . . , _k against k−QR are constructed as follows. _i takes as input an RSA modulus N=pq, with p≡1 (mod 2^k) and q≡3 (mod 4), and an element w∈_N. Its task is to decide whether w is uniform over _N\_N or uniform over _N. To this ends, _i chooses a random element z_N* . It then defines x=z²ⁱw²ⁱ⁻¹ mod N and feeds with (x,N).

• • If w is uniform over _N\_N, x is clearly uniform over D_i−1. Therefore, in that case, outputs 1 with probability ε_i−1.
  • If w is uniform over _N, outputs 1 with probability

${\sum }_{j=i}^k\frac{1}{2^{j-i+1}}{\epsilon }_j+\frac{1}{2^{k-i}}{\epsilon }_k$

• •  according to Lemma 1.Therefore, the (signed) advantage of _i in solving k−QR is

$a_i={\epsilon }_{i-1}-\left(\sum _{j=i}^k\frac{1}{2^{j-i+1}}{\epsilon }_j+\frac{1}{2^{k-i}}{\epsilon }_k\right)={\epsilon }_{i-1}-\sum _{j=8}^k2^{i-1}{\beta }_j{\epsilon }_j$

with

${\beta }_j=\frac{1}{2^j}$

for j∈{1, . . . , k−1} and

${\beta }_k=\frac{1}{2^{k-1}}.$

The following probability distribution over {1, . . . ,k} is considered:

$\underset{X\stackrel{R}{\leftarrow }}{\mathrm{Pr}}\left(X=i\right)=p_i=\{\begin{array}{cc}\frac{2}{k+1}& \mathrm{if}i=1\\ \frac{1}{k+1}& \mathrm{if}i\ge 2\end{array}.$

An adversary B against k QR is defined as follows: B chooses a random i and feeds _i with its k−QR challenge. The advantage of is:

$\begin{array}{c}{\mathrm{Adv}}_{ℬ}^{k-\mathrm{QR}}\left(1^K\right)=\sum _{i=1}^kp_ia_i\\ =\sum _{i=1}^kp_i{\epsilon }_{i-1}-\sum _{i=1}^kp_i\sum _{j=i}^k2^{i-1}{\beta }_j{\epsilon }_j\\ =\sum _{j=0}^{k-1}p_{j+1}{\epsilon }_j-\sum _{j=1}^k\sum _{i=1}^j2^{i-1}p_i{\beta }_j{\epsilon }_j\\ =p_1{\epsilon }_0+\sum _{j=1}^{k-1}\left(p_{j+1}-\sum _{i=1}^j2^{i-1}p_i{\beta }_j\right){\epsilon }_j-\sum _{i=1}^k2^{i-1}p_i{\beta }_k{\epsilon }_k.\end{array}$

For j∈{1, . . . , k−1},

${\beta }_j=\frac{1}{2^j}$

and:

$\begin{array}{c}\sum _{i=1}^j2^{i-1}p_i{\beta }_j=\sum _{i=1}^j2^{i-j-1}p_i=2^{-j}p_1+\sum _{i=2}^j2^{i-j-1}p_j\\ =2^{1-j}\frac{1}{k+1}\left(1+\sum _{i=2}^j2^{i-2}\right)=2^{1-j}\frac{1}{k+1}2^{j-1}\\ =\frac{1}{k+1}=p_{j+1}.\end{array}$

In addition:

$\begin{array}{c}\sum _{i=1}^k2^{i-1}p_i{\beta }_k=\sum _{i=1}^k2^{i-k}p_i=\frac{1}{k+1}\left(2^{2-k}+\sum _{i=2}^k2^{i-k}\right)\\ =2^{2-k}\frac{1}{k+1}\left(1+\sum _{i=2}^k2^{i-2}\right)\\ =\frac{2}{k+1}.\end{array}$

Therefore, the advantage of is

${\mathrm{Adv}}_{ℬ}^{k-\mathrm{QR}}\left(1^K\right)=\frac{2}{k+1}{\epsilon }_0+0-\frac{2}{k+1}{\epsilon }_k=\frac{2}{k+1}{\mathrm{Adv}}_{}^{\mathrm{Gap}-2^k-\mathrm{Res}}\left(1^K\right),$

which concludes the proof.

FIG. 1 illustrates a first embodiment in which q≡3 (mod 4).

FIG. 1 shows a cryptographic device 110 comprising an interface 111 configured for communication with other devices (not shown), at least one hardware processing unit (“CPU”) 112 and memory 113. The cryptographic device also comprises other necessary hardware and software components such as internal connections, but these are not shown to simplify the illustration. Also shown is a non-transitory computer program storage medium 114 that stores instruction that, when executed by a processing unit, perform the key generation method KeyGen of the first embodiment.

In more detail, the proposed encryption scheme of the preferred embodiment is the tuple (KeyGen, Encrypt, Decrypt) defined as follows:

• KeyGen(1^κ) Given a security parameter κ, KeyGen defines an integer k≧1, randomly generates primes p≡1 (mod 2^k), step S10, and q≡3 (mod 4), step S11, and sets, step S12, N=pq. It also picks, step S13, y∈_N\_N. The public key pk={N,y,k} and the private key sk={p,k} are output, step S14.
• Encrypt(pk,m) Let ={0,1}^k. To encrypt a message m∈ (seen as an integer in {0, . . . , 2^k−1}), Encrypt picks a random x∈_N* and returns the ciphertext c=y^mx^2k mod N.
• Decrypt(sk,c) Given c∈_N* and the private key sk={p,k}, the algorithm first computes

$z={\left(\frac{c}{p}\right)}_{2^k}$

•  and then finds m∈{0, . . . , 2^k−1} such that the relation

${\left[{\left(\frac{y}{p}\right)}_{2^k}\right]}^m=z\left(\mathrm{mods}p\right)$

•  holds.

FIG. 2 illustrates a second embodiment in which q≡1 (mod 4). It is noted that the case where k=2 corresponds to the prior art Generalized GM.

FIG. 2 shows a cryptographic device 120 comprising an interface 121 configured for communication with other devices (not shown), at least one hardware processing unit (“CPU”) 122 and memory 123. The cryptographic device also comprises other necessary hardware and software components such as internal connections, but these are not shown to simplify the illustration. Also shown is a non-transitory computer program storage medium 124 that stores instruction that, when executed by a processing unit, perform the key generation method KeyGen of the first embodiment.

In more detail, the proposed encryption scheme of the second preferred embodiment is the tuple (KeyGen, Encrypt, Decrypt) defined as follows:

• KeyGen(1^κ) Given a security parameter κ, KeyGen defines an integer k≧1, randomly generates primes p≡1 (mod 2^k), step S20, and q≡1 (mod 4), step S21, and sets N=pq; step S22. It also picks y∈_N\_N, step S23. The public key pk={N,y,k} and the private key sk={p,k} are output, step S24.
• Encrypt(pk,m) Let ={0,1}^k. To encrypt a message m∈ (seen as an integer in {0, . . . , 2^k−1}), Encrypt picks a random x∈*_N and returns the ciphertext c=y^mx^2k mod N.
• Decrypt(sk,c) Given c∈*_N and the private key sk={p,k}, the algorithm first computes

$z={\left(\frac{c}{p}\right)}_{2^k}$

•  and then finds m∈{0, . . . , 2^k−1} such that the relation

${\left[{\left(\frac{y}{p}\right)}_{2^k}\right]}^m=z\left(\mathrm{mods}p\right)$

•  holds.

It is noted that q≡1 (mod 2^k) also means that q≡1 (mod 4), but the present method can make it easier to generate q since there are more possibilities. A variant excludes q≡1 (mod 2^k).

In a variant, the modulus N is equal to a integer multiple of the product of the primes p, q.

According to specific embodiments, the interface 111, 121 is can be a wireline interface (for example a bus interface such as USB (Universal Serial Bus)) or a wireless interface (such as a IEEE 802.11 interface, WiFi® or a Bluetooth® interface); the interface can be a wide area network interface, a local area network interface or a HDMI (High Definition Multimedia Interface) interface.

According to different embodiments, the cryptographic device 110, 120 uses a communication protocol such as HTTP, IP, or FLUTE to transmit the public key.

According to different embodiments, the cryptographic device 110, 120 stores the private key in a memory, such as a random access memory (RAM) or in the processing unit 112, 122.

According to different embodiments, the cryptographic device 110, 120 belongs to a set comprising:

• • a mobile device;
  • a communication device;
  • a game device;
  • a set top box;
  • a TV set;
  • a tablet (or tablet computer);
  • a laptop; and
  • a cryptographic chip.

A advantage of the first preferred embodiment, i.e., when q≡3 (mod 4)) is security. The notion of semantic security is already met under a quadratic residuosity assumption. In particular, the squared Jacobi symbol assumption is not necessary.

A second advantage of the first preferred embodiment (and also the second, i.e., when q≡1 (mod 4)) is a simplified key generation process. Only prime p requires a specialized prime generation algorithm as the one described by Joye and Paillier [see Marc Joye and Pascal Paillier. Fast generation of prime numbers on portable devices: An update. In L. Goubin and M. Matsui, editors, Cryptographic Hardware and Embedded Systems—CHES 2006, volume 4249 of Lecture Notes in Computer Science, pages 160-173. Springer, 2006]. It is noted that a random (form-free) prime p is congruent to 3 (mod 4) with a probability of ½ (it is congruent to 1 (mod 4) otherwise). Form-free primes are much easier to generate than other primes.

Each feature disclosed in the description and (where appropriate) the claims and drawings may be provided independently or in any appropriate combination. Features described as being implemented in hardware may also be implemented in software, and vice versa. Reference numerals appearing in the claims are by way of illustration only and shall have no limiting effect on the scope of the claims.

Claims

1. A method for generation of a cryptographic key comprising, in a device: generating a first prime p≡1 (mod 2k), where k≧1 is an integer;generating a second prime q≡3 (mod 4) or q≡1 (mod 4), q≡1 (mod 2k);computing a modulus N that is a multiple of the product between the first prime p and the second prime q;picking an integer y∈N\N, where N is the set of integers whose Jacobi symbol is 1 and N is the set of quadratic residues; andoutputting a public key pk={N,y,k}.

2. The method of claim 1, further comprising generating a private key sk={p,k}.

3. The method of claim 2, further comprising storing the private key sk.

4. The method of claim 1, wherein the public key pk is for use in a cryptosystem in which a message m∈, where ={0,1}k, is encrypted by picking a random x∈N* and calculating a ciphertext c=ymx2k mod N.

5. A cryptographic device comprising at least one processing unit configured to: generate a first prime p≡1 (mod 2k), where k≧1 is an integer;generate a second prime q≡3 (mod 4) or q≡1 (mod 4), q≡1 (mod 2k);compute a modulus N that is a multiple of the product between the first prime p and the second prime q;pick an integer y∈N\N, where N is the set of integers whose Jacobi symbol is 1 and N is the set of quadratic residues; andoutput a public key pk={N,y,k}.

6. The cryptographic device of claim 5, wherein the at least one processing unit is further configured to generate a private key sk={p,k}.

7. The cryptographic device of claim 6, wherein the at least one processing unit is configured to store the private key sk.

8. The cryptographic device of claim 7, wherein the at least one processing unit is configured to store the private key sk in a memory or in the at least one processing unit.

9. The cryptographic device of claim 5, wherein the at least one processing unit is configured to output the public key via a communication interface.

10. The cryptographic device of claim 9, wherein the at least one processing unit is configured to use a communication protocol to output the public key via the communication interface.

11. The cryptographic device of claim 5, wherein the cryptographic device belongs to one of the group of: a mobile device, a communication device, a game device, a set top box, a TV set, a tablet, a laptop and a cryptographic chip.

12. The cryptographic device of claim 5, wherein the public key pk is for use in a cryptosystem in which a message m∈, where ={0,1}k, is encrypted by picking a random x∈N* and calculating a ciphertext c=ymx2k mod N.

13. A non-transitory computer program product storing instructions that, when executed by a at least one processor, perform the method of claim 1.