Find Consecutive X-Coordinate algorithm

Let $P = (x,y)$ be a point on the curve $E$ with the $y^2 = x^3 + ax + b$ with Weierstrass equation over the prime field $operatorname{GF}(p)$¹, i.e. it satisfies the curve equation.

The consecutive $x$-coordinate point can be found with the below algorithm.

for i from 1 to number_of_poinst_on_the_curve
    temp = x + i mod p
    if  (temp, y) satisfies the curve equation
       return (temp,y')   
    else
       continue
throw no_consecutive_element

• The $bmod p$ is necessary since we don't want our $x$ has value out of the possible space.

• $y'$ is one of the two possible solitons to the $y^2$, except when $y=mathcal{O}$, the point at the infinity. The educative details of finding the square root in modular arithmetic can be found on Fgrieu's latest answer;

  Is it possible to compute the y-coordinate of a point on SECP256K1, given only the x-coordinate

• The algorithm will terminate since the EC has a finite number of points. The number of points can be

  • limited by Hasse's Theorem: $N - (q+1)| le 2 sqrt{q}$ over finite field with $q$ elements.
  • Or, exactly found by Schoof's algorithm, In the paper Counting Points on Elliptic Curves over Finite Fields, published in 1985,
  • Or, the improved version Schoof–Elkies–Atkin algorithm

---

_{¹ This is used to simplify the algorithm. EC doesn't need to be defined over a prime finite field. In Cryptography, we use the general case of Finite Fields $operatorname{GF}(q=p^m), m in mathbb{Z}, mgeq 1$. Expect the prime field, the other fields have different representation for the elements.}

---

For $y^2 = x^3 + x^2 + x$ over $GF(131)$

E = EllipticCurve(GF(131),[0,1,0,1,0])

P=E.lift_x(0) #lift can throw an error if there is no point with the given input
              #tested before that there is a point with x=0

plotE = E.plot()

for t in srange(1,129):
    try:
        R =  E.lift_x(t)
        plotE += line([P.xy(),R.xy()],color='red')
        P = R
    except ValueError:  
        pass
plotE

update

Though I don't see the direct effect on Cryptography, there are academic works on this;

• Integral Points in Arithmetic Progression on $y^2=x(x^2-n^2)$, by Bremner, Silverman, and Tzanakis, 2000.