Erdős Problem #1188 — Minimal Distinct Covering Systems

At least 73 years old · documented by

A congruence class (n,a)(n,a) represents the integers congruent to a(modn)a\pmod n, where 2≤n2\le n and 0≤a<n0\le a<n. A finite set of such classes is a minimal distinct covering system if every integer belongs to at least one class, no two classes have the same modulus, and no proper subset still covers all integers. Let F(x)F(x) be the number of minimal distinct covering systems whose moduli satisfy 2≤n≤x2\le n\le x. Estimate F(x)F(x). The conjectured estimate is

log⁡ ⁣(log⁡F(x))log⁡x⟶1(x→∞).\frac{\log\!\bigl(\log F(x)\bigr)}{\log x}\longrightarrow 1\qquad (x\to\infty).
References

Progress summary

Refreshed
Claimed progress

The count is known to grow much faster than any polynomial, but no close estimate is known.

The problem asks for the growth of the number F(x)F(x) of minimal distinct covering systems whose moduli are at most xx. Erdős posed a related counting question in 1952 and 1980; the minimal-system formulation is now treated as an alternative interpretation.

Known results

  • Hough, 2015: the minimum-modulus theorem implies F(x)→∞F(x)\to\infty.
  • An elementary construction gives F(x)≫log⁡xF(x)\gg\log x.
  • The trivial upper bound is F(x)≤exp⁡(O(xlog⁡x))F(x)\leq\exp\big(O(x\log x)\big).
  • A separate result counts systems by their number of members, not by the modulus cutoff xx.

2024 superpolynomial lower bound

Balister, Bollobás, Morris, Sahasrabudhe, and Tiba constructed enough systems to prove F(x)≥exp⁡((log⁡x)3−o(1))F(x)\geq\exp\big((\log x)^{3-o(1)}\big). Thus F(x)F(x) grows faster than every polynomial, but no matching upper bound or complete asymptotic estimate has been reported.

Current status (as of June 2025): F(x)F(x) is known to diverge and to satisfy a superpolynomial lower bound, while its sharp growth rate remains open.

Sources

Solutions 0

No solutions have been posted yet.