Problem 8 - A Computational Introduction to Number Theory and Algebra

Question

Show how to deterministically compute square roots modulo primes \(p \equiv 5(\bmod 8)\) in time \(O\left(\operatorname{len}(p)^{3}\right)\)

Step-by-Step Solution

Answer

Question: Explain the deterministic method to compute square roots modulo primes for prime numbers congruent to 5 modulo 8 within a time complexity of O(len(p)^3). Answer: For a prime number p congruent to 5 modulo 8, follow these steps to deterministically compute square roots modulo primes: 1. Calculate the unique integer x satisfying \(x^2 \equiv -1 \pmod{p}\). 2. Construct the polynomial \(f(z) = z^2 + xz + 1\). 3. Using Lenstra's polynomial division algorithm, find the unique divisor of \(f(z)\) modulo \(p\). 4. Factorize \(f(z) \pmod{p}\) into factors \(h_1(z)\) and \(h_2(z)\) using the GCD with the divisor obtained in step 3. 5. Evaluate \(h_1(1) \pmod{p}\) and \(h_2(1) \pmod{p}\) to obtain results \(r_1\) and \(r_2\). 6. Verify \(r_1^2 \equiv r_2^2 \equiv a \pmod{p}\). If successful, the square root modulo p of a is deterministically computed as \(r_1\) and \(r_2\).

Step 1: Calculate the unique integer x

Let \(p\) be a prime number such that \(p \equiv 5 \pmod{8}\). Calculate the unique integer \(x\) satisfying \(x^2 \equiv -1 \pmod{p}\). This can be done using repeated squaring and calculating \(x = 2^{(p-1)/4} \pmod{p}\).

Step 2: Construct the polynomial

Construct the polynomial \(f(z) = z^2 + xz + 1\) where \(z\) is the variable and \(x\) is the integer obtained in Step 1.

Step 3: Find the unique divisor of f(z) modulo p

By using the Lenstra's polynomial division algorithm, efficiently find the unique divisor (if it exists) of \(f(z)\) modulo \(p\). Let the divisor polynomial be \(g(z)\).

Step 4: Factorize f(z) modulo p

Compute the greatest common divisor (GCD) of \(f(z)\) and \(g(z) \pmod{p}\). If the GCD is a nontrivial factor of \(f(z)\), we can factorize \(f(z) \pmod{p}\). Let the two factors be \(h_1(z)\) and \(h_2(z)\)

Step 5: Calculate the square root modulo p

Substitute \(z = 1\) in the factors \(h_1(z)\) and \(h_2(z)\). The evaluations can be computed as \(h_1(1) \pmod{p}\) and \(h_2(1) \pmod{p}\). Let the results be \(r_1\) and \(r_2\).

Step 6: Verify the results

Verify that \(r_1^2 \equiv r_2^2 \equiv a \pmod{p}\). If the verification is successful, then the square root modulo \(p\) of \(a\) is deterministically computed as \(r_1\) and \(r_2\).

Key Concepts

Lenstra's Polynomial Division Algorithm

When working within the realm of modular arithmetic, particularly when dealing with polynomials modulo a prime number, efficient algorithms for operations such as division and factoring are crucial. Lenstra's polynomial division algorithm is one such method that stands out for its efficiency, especially in the context of factoring polynomials over finite fields.

Henri Lenstra introduced this algorithm as part of an ingenious way to factor large numbers using elliptic curves, but the principles involved are also applied to polynomials. This algorithm helps to quickly find divisors of polynomials modulo a given prime number. In the given exercise, Lenstra's algorithm is used to identify a unique divisor of the polynomial f(z) modulo the prime p.

Understanding the mechanics of Lenstra's algorithm usually requires familiarity with concepts such as Euclidean division of polynomials, modular inverses, and elliptic curves, which are not covered here but are integral to the algorithm's implementation. Generally, it aims to perform polynomial division with fewer operations than the traditional long division method, making it better-suited for computations where primes are large.

For students looking to grasp the fundamental use of this algorithm in computing square roots modulo primes, it's essential to focus on the underlying algebraic relationships and transformations rather than the complex number theory or the algorithmic intricacies.

Repeated Squaring

In the realm of modular arithmetic, finding powers of numbers can become computationally expensive, particularly for large exponents. This is where the technique of repeated squaring comes into play, greatly optimizing the process of exponentiation within a modular context.

As the name suggests, repeated squaring relies on iteratively squaring numbers and reducing them modulo the given prime number. Instead of multiplying a number by itself many times, which can be slow and inefficient, repeated squaring breaks down the exponentiation into a binary operation where each step involves squaring and taking the modulus, if necessary.

For example, to calculate aⁿ mod p, with n being a large exponent, one would express n in binary and square a (mod p), doubling the exponent in each step. As each bit of the exponent is considered, the interim result is squared, and if the bit is a one, the current power of a is multiplied into the running total, with reductions by p occurring when needed. This method is not only efficient but also less prone to overflows and easy to implement.

Repeated squaring is very relevant for the exercise as it is the technique used for the initial calculation of x that solves x² ≡ -1 (mod p) when p ≡ 5 (mod 8). The efficient computation allows us to manage significant figures without performance penalties.

Greatest Common Divisor (GCD)

The greatest common divisor (GCD), also known as the greatest common factor, is the highest number that divides two or more integers without leaving a remainder. In arithmetic, the GCD of two numbers is often found using the Euclidean algorithm, which involves repeated division.

In the modular arithmetic and polynomial context, determining the GCD has critical importance for many algorithms, including factoring polynomials modulo a prime number. In the context of the exercise, after using Lenstra's polynomial division algorithm to find a unique divisor g(z) of the polynomial f(z) modulo p, the next step involves computing the GCD of f(z) and g(z) mod p.

The GCD enables us to determine if there are any nontrivial factors of f(z). If the GCD is a polynomial of degree higher than 0 but lower than the degree of f(z), we have successfully found a factor of f(z). Factoring polynomials over finite fields, like finding the GCD, is a necessary step in many algorithms in number theory and cryptography, and the process reflects the close relationship between arithmetic operations on numbers and their polynomial counterparts.

For students working through the concept, understanding the role of GCD in polynomial arithmetic can provide a foundation for more complex topics in discrete mathematics and algorithm design.