TOPICS
Search

Lattice Covering Radius


The covering radius of a full-rank point lattice L subset R^n is

 mu(L)=max_(x in R^n)min_(v in L)||x-v||_2.

Equivalently, it is the smallest r>=0 such that the closed Euclidean balls of radius r centered at the points of L cover R^n.

The lattice covering radius problem takes a rational invertible matrix B and a positive rational number r and asks whether mu(BZ^n)<=r. Vallentin (2026) proved that this decision problem is Pi_2^P-complete, making it complete for the second level of the polynomial hierarchy.

Vallentin (2026) states that generative AI developed the hardness reduction and drafted and revised the paper under his direction. He then digested and simplified the proof and takes responsibility for its correctness.


See also

Closest Vector Problem, Point Lattice, Voronoi Cell

Explore with Wolfram|Alpha

References

Vallentin, F. "The Complexity of Computing the Covering Radius of a Euclidean Lattice." 28 Sep 2026. https://arxiv.org/abs/2609.35027.

Cite this as:

Weisstein, Eric W. "Lattice Covering Radius." From MathWorld--A Wolfram Resource. https://mathworld.wolfram.com/LatticeCoveringRadius.html

Subject classifications