---
title: Periodic Point Sets under Isometry
url: https://www.emergentmind.com/topics/periodic-point-sets-under-isometry
type: topic
---

# Periodic Point Sets under Isometry

Searching arXiv for the cited papers to ground the article in current records.
Periodic point sets are subsets of Euclidean space obtained by repeating a finite motif by all translations of a lattice, and they model solid crystalline materials whose structures are determined in a rigid form. Their natural equivalence is rigid motion or, more generally, isometry preserving inter-point distances, so the central problem is to study periodic sets through descriptors, metrics, and classification procedures that do not depend on arbitrary choices of unit cell or coordinates [2205.02226]. Recent work in Periodic Geometry treats isometry classes of periodic point sets as a continuous space, motivated by the fact that atomic vibrations and measurement noise make discrete crystallographic classifications unstable under perturbations [2103.02749].

## 1. Crystallographic model and isometric equivalence

A periodic point set in \(\mathbb{R}^n\) is described by a basis \(v_1,\dots,v_n\), the associated lattice
\[
\Lambda = \Big\{\sum_{i=1}^n c_i v_i : c_i\in\mathbb{Z}\Big\},
\]
the unit cell
\[
U(v_1,\dots,v_n)=\Big\{\sum_{i=1}^n c_i v_i : c_i\in[0,1)\Big\},
\]
and a finite motif \(M=\{p_1,\dots,p_m\}\subset U\). The periodic point set is the Minkowski sum
\[
S=M+\Lambda=\{u+v\mid u\in M,\ v\in\Lambda\},
\]
that is, a finite motif repeated at all lattice translations [2205.02226]. Equivalent formulations appear across the literature, including the representation of periodic sets as finite unions of lattice cosets and as periodic Delone sets invariant under a full-rank lattice [1802.02072].

The relevant equivalence notion is isometry of Euclidean space. In \(\mathbb{R}^n\),
\[
\mathrm{Isom}(\mathbb{R}^n)=\{x\mapsto Ax+b: A\in O(n),\ b\in\mathbb{R}^n\},
\]
so translations, rotations, and reflections are all allowed [2205.02226]. Two periodic point sets \(S,S'\subset\mathbb{R}^n\) are considered the same up to rigid motion if there exists an isometry \(\phi\) such that \(\phi(S)=S'\) [2205.02226]. This is stronger than equivalence under a particular unit-cell description, because the same periodic set can be represented by infinitely many bases and motifs, and even primitive-cell descriptions can change discontinuously under perturbations [2103.02749].

This setting suggests a general program: an isometry invariant \(I(S)\) should satisfy \(I(S)=I(S')\) whenever \(S'\) is obtained from \(S\) by an isometry, should vary continuously under perturbations, and ideally should support reconstruction or exact comparison of periodic sets. A recurring theme in the cited work is that past invariants based on symmetry groups, reduced cells, or fixed local cut-offs are discontinuous or representation-dependent, whereas the newer invariants are designed to live naturally on isometry classes [2103.02749].

## 2. Continuous invariants in Periodic Geometry

A central family of invariants is given by density functions. For a periodic set \(S=M+\Lambda\subset\mathbb{R}^n\) with unit cell \(U\), for each integer \(k\ge 0\) and radius \(t>0\), one considers all closed balls of radius \(t\) centered at points of \(S\), lets \(U_k(t)\subset U\) be the region covered by exactly \(k\) balls, and defines
\[
\psi_k[S](t)=\frac{\mathrm{Vol}(U_k(t))}{\mathrm{Vol}(U)}.
\]
The infinite sequence \(\Psi[S]=\{\psi_k[S](t)\}_{k=0}^\infty\) is the density fingerprint [2205.02226]. Because isometries preserve distances, balls, and coverage multiplicities, \(\Psi[S]\) is an isometry invariant [2205.02226]. The monograph on Periodic Geometry presents density functions as one of the key new continuous coordinates on the space of isometry classes of periodic point sets [2103.02749].

A second family is based on ordered nearest-neighbor distances. For a periodic set \(S=L+M\subset\mathbb{R}^n\) with motif \(M=\{p_1,\dots,p_m\}\), if \(d_{ij}\) is the distance from \(p_i\) to its \(j\)-th nearest neighbor in the infinite periodic set, then
\[
AMD_j(S)=\frac{1}{m}\sum_{i=1}^m d_{ij}
\]
defines the Average Minimum Distance of order \(j\) [2009.02488]. These invariants are isometry invariant and satisfy the Lipschitz bound
\[
|AMD_k(S)-AMD_k(Q)|\le 2\,d_B(S,Q)
\]
under bottleneck perturbations [2009.02488]. The same paper shows that AMD can distinguish some periodic sets that have identical density functions and that AMD can be computed in near linear time in the input size [2009.02488].

A third family is the Pointwise Distance Distribution. For a periodic set \(S=\Lambda+M\), one forms for each motif point the ordered distances to its \(k\)-nearest neighbors in the full periodic set, lexicographically sorts the resulting rows, and collapses identical rows with weights. The resulting matrix \(PDD(S;k)\) is an isometry invariant, is Lipschitz continuous under bottleneck perturbations, and is complete for all periodic sets in general position according to the 2021 development cited in the database description [2108.04798]. A later machine-learning adaptation states that PDD is a continuous and generically complete isometry invariant for periodic point sets and that it distinguished all more than 660 thousand periodic crystals in the Cambridge Structural Database as purely periodic sets of points without atomic types [2401.15089].

These invariant families are complementary. Density functions emphasize higher-order overlap geometry of balls, AMD emphasizes ordered nearest-neighbor scales, and PDD encodes weighted local distance profiles. This suggests a layered picture of periodic-point-set classification: coarse and fast descriptors for search, richer fingerprints for discrimination, and complete invariants for exact classification.

## 3. One-dimensional periodic sequences and explicit density formulas

In dimension \(n=1\), a periodic point set is a periodic sequence
\[
S=\{p_1,\dots,p_m\}+\mathbb{Z}\subset\mathbb{R},
\]
with \(0\le p_1<\dots<p_m<1\), period \(1\), and gaps
\[
d_i=p_{i+1}-p_i\in(0,1),\quad p_{m+1}:=p_1+1,\quad \sum_{i=1}^m d_i=1
\]
[2205.02226]. In this setting, balls are intervals, every \(\psi_k(t)\) is piecewise linear in \(t\), and the paper “Density functions of periodic sequences” gives a complete explicit description of \(\psi_k(t)\) for all \(k\) for any 1D periodic sequence [2205.02226].

For the uncovered-region density \(\psi_0\), let \(d_{[1]}\le \dots\le d_{[m]}\) be the sorted gaps. Then \(\psi_0\) is piecewise linear with corner points \((0,1)\) and
\[
\left(\frac12 d_{[i]},\,1-\sum_{j=1}^{i-1}d_{[j]}-(m-i+1)d_{[i]}\right),\qquad i=1,\dots,m,
\]
with last corner \(\left(\frac12 d_{[m]},0\right)\) [2205.02226]. A striking consequence is that \(\psi_0(t)\) is entirely determined by the multiset of gap lengths \(d_i\), not by their order along the line [2205.02226]. This immediately produces many non-isometric sequences with the same \(\psi_0\).

For \(k\ge 1\), each \(\psi_k(t)\) decomposes as a sum of trapezoid functions \(\eta_{k,i}(t)\), one for each contiguous block of \(k\) points. If \(s=\sum_{j=i}^{i+k-2}d_j\) and \(d=\min\{d_{i-1},d_{i+k-1}\}\), then
\[
\psi_k(t)=\sum_{i=1}^m \eta_{k,i}(t),
\]
where each \(\eta_{k,i}\) is determined by the triple \((d_{i-1},s,d_{i+k-1})\), with first and last entries symmetric [2205.02226]. This gives a purely combinatorial description of the 1D geometry in terms of the cyclic gap sequence.

The same paper proves refined symmetry relations:
\[
\psi_{m-k}\Big(\tfrac12-t\Big)=\psi_k(t)\quad\text{for }k=0,\dots,\lfloor m/2\rfloor,\ t\in[0,\tfrac12],
\]
and
\[
\psi_{k+m}\Big(t+\tfrac12\Big)=\psi_k(t)\quad\text{for all }k\ge 0,\ t\ge 0
\]
[2205.02226]. It also derives integrated densities
\[
\rho_k[S]=\int_{-\infty}^{+\infty}\psi_k(t)\,dt,
\]
with closed forms
\[
\rho_k[S]=\frac12\sum_{i=1}^m d_{i-1}d_{i+k-1}\quad (k>0),\qquad
\rho_0[S]=\frac14\sum_{i=1}^m d_i^2
\]
[2205.02226].

The 1D theory is not only descriptive but also classificatory. The paper shows that the full density fingerprint is not complete in dimension \(1\): the homometric pair
\[
S_{15}=\{0,1,3,4,5,7,9,10,12\}+15,\qquad
Q_{15}=\{0,1,3,4,6,8,9,12,14\}+15
\]
satisfies \(\psi_k[S_{15}]=\psi_k[Q_{15}]\) for all \(k\ge 0\), hence \(\Psi[S_{15}]=\Psi[Q_{15}]\), despite being non-isometric [2205.02226]. On the positive side, if all gaps \(d_i\) are distinct, then the first density function \(\psi_1[S](t)\) uniquely determines the sequence \(S\) up to isometry, so \(\psi_1\) is a complete isometry invariant for all generic 1D periodic sequences [2205.02226].

A subsequent extension allows different initial radii \(r_i\), motivated by atomic radii and by continuous events occupying disjoint intervals in time series. The generalized density functions remain isometry invariants, admit explicit 1D formulas, and are strictly stronger than the zero-radius case: they distinguish periodic sequences that have identical densities when all radii vanish [2301.05137]. This suggests that “decorated” periodic point sets can carry stronger continuous fingerprints than bare point sets.

## 4. Metrics and complete classification frameworks

A major development is the passage from invariants to metrics on isometry classes. One line of work introduces exactly computable, continuous metrics on isometry classes of finite and 1-periodic sequences [2205.04388]. In 1D, a periodic sequence with successive distances \(d_1,\dots,d_m\) is classified up to isometry by the lexicographically smallest cyclic-or-reversed list \(SDL(S)\), and the elastic metric
\[
Elm(S,Q)=\min_{\sigma\in C(m)}\|D_S-\sigma(D_Q)\|_\infty
\]
is a genuine metric on isometry classes [2205.04388]. The key novelty in the periodic case is continuity under perturbations that change the minimum period, and the paper proves
\[
Elm(S,Q)\le Elm^o(S,Q)\le 2\,d_B(S,Q)
\]
[2205.04388]. For high-dimensional 1-periodic sequences, the analogous invariant is the time-value invariant \(TVI(S)\), and the corresponding elastic metric is computable in \(O(m^3)\) time [2205.04388].

A broader classification framework is built around the isoset. The monograph “Introduction to Periodic Geometry and Topology” describes the isoset as a complete invariant built from local cluster isometry classes at a stable radius, together with weights, and then equips the space of isometry classes with a continuous metric via Earth Mover’s Distance on these weighted cluster classes [2103.02749]. This program is refined in the bridge-length paper, where the bridge length \(\beta(S)\) is the minimum number \(r>0\) such that any two points \(p,q\in S\) can be connected by a finite sequence with consecutive gaps at most \(r\), equivalently the smallest \(r\) such that the graph \(G_r(S)\) is connected [2410.23288]. The paper proposes a practical algorithm to compute \(\beta(S)\) from a lattice basis and motif using labelled quotient graphs, cycle sums, and Smith Normal Form; if the last added edge has length \(d\), then \(\beta(S)=d\) [2410.23288]. The resulting complexity is
\[
O\big(m^2 a(U)^n N\big),
\]
where \(m=|M|\), \(a(U)=r(U)/h(U)\) is the aspect ratio of the cell, and \(N\) is the time complexity of computing Smith Normal Form [2410.23288]. Since the size and complexity of the isoset strongly depend on the bridge length, this algorithm is described as a key ingredient in a continuous isometry classification of periodic point sets [2410.23288].

PDD provides another route to complete or generically complete classification. The 2021 database-scale study defines \(PDD(S;k)\) by collecting sorted nearest-neighbor distance rows for motif points and collapsing identical rows with weights; it proves invariance under isometry, Lipschitz continuity under bottleneck perturbations, and completeness of PDD for all periodic sets in general position [2108.04798]. A later study adapts PDD for property prediction and states that PDD is a continuous and generically complete isometry invariant for periodic point sets, with practical success on large databases [2401.15089]. More recent work introduces higher-order PDD variants \(PDD^{\{h\}}\), based on averages over \((h+1)\)-tuples, specifically to distinguish all known counter-examples to the completeness of past descriptors [2509.15088]. The same paper proves exact completeness in 1D through the pointwise shift distribution \(PSD(S;m)\), which is a complete invariant under rigid motion for all periodic sequences in \(\mathbb{R}\) and computable in time \(O(m^2)\) [2509.15088].

These developments collectively indicate two complementary directions. One direction emphasizes exact classification by complete invariants such as isosets, bridge-length-based constructions, and PDD in general position. The other emphasizes continuous metrics such as elastic metrics, Earth Mover’s Distance on PDD-type fingerprints, and bottleneck-style metrics on periodic-point-set spaces.

## 5. Dynamical, geometric, and coarse-geometric perspectives

Periodic point sets also appear naturally in topological dynamics. In the framework of uniformly discrete and relatively dense sets, periodic sets are Delone sets, and the hull \(X_P\) is the orbit closure of a point set under translations or under the Euclidean group \(E(d)\) in the vague topology [1210.2955]. Almost repetitivity characterizes minimality of the associated dynamical system, and almost linear repetitivity implies unique ergodicity [1210.2955]. Periodic sets sit at the most regular end of this hierarchy: they are linearly repetitive in a very strong sense, hence their hulls are minimal and uniquely ergodic under both translations and Euclidean motions [1210.2955]. This gives a dynamical interpretation of periodicity under isometry: periodic point sets are not only rigid geometric objects but also highly regular points in natural topological dynamical systems.

Another perspective concerns the large-scale geometry of spaces of periodic point sets themselves. The paper “On the metric spaces of lattices and periodic point sets” considers the spaces \(\PPS(\mathbb{R}^d)\) and \(\Latt(\mathbb{R}^d)\) equipped with bottleneck distance \(d_B\) and Euclidean bottleneck distance \(EB\), where
\[
EB(X,Y)=\inf_{\psi\in \Iso(\mathbb{R}^d)} d_B(X,\psi(Y))
\]
[2310.07594]. For periodic point sets \(X,Y\), the following are equivalent: \(EB(X,Y)<\infty\), \(d_B(X,Y)<\infty\), and \(\den(X)=\den(Y)\) [2310.07594]. Thus density is a complete invariant for finite bottleneck distance in the periodic setting. The same paper studies embeddability into Hilbert space and shows that certain spaces of periodic point sets of fixed density with only a packing bound or only a covering bound do not coarsely embed into any uniformly convex Banach space, in particular not into Hilbert space [2310.07594]. By contrast, subclasses with bounded motif size and covering radius, or bounded unit-cell diameter, are bounded in \(EB\) and hence trivially coarsely embeddable [2310.07594]. A plausible implication is that the global geometry of the moduli space of periodic point sets is far more complicated than the local continuity of individual invariants might suggest.

Energy minimization provides a further geometric viewpoint. A periodic point set \(\Lambda\subset\mathbb{R}^n\) is \(m\)-periodic if it is a finite union of lattice cosets, and energies in the Gaussian core model depend only on pairwise distances, hence are invariant under isometries [1802.02072]. The paper “Local Energy Optimality of Periodic Sets” characterizes periodic point sets that are \(f_c\)-critical for all \(c>0\) in terms of balanced shells and weighted spherical \(2\)-designs [1802.02072]. For the 2-periodic family \(\mathsf{D}_n^+\), shell symmetry and spherical \(3\)-design properties allow explicit Hessian analysis, leading to local \(f_c\)-optimality in odd dimensions \(n\ge 9\) for sufficiently large \(c\) [1802.02072]. This suggests that isometry classification and local energy landscape analysis are tightly linked through shell geometry and symmetry.

## 6. Applications, limitations, and open directions

The applications in crystallography and materials science are immediate. Continuous, isometry-invariant descriptors are intended to compare crystal structures across experiments, simulations, and noise, to detect duplicates and near-duplicates, and to organize large databases by geometry rather than by unstable cell descriptions [2205.02226]. PDD-based representations have already been used to compare nearly 1.5 million crystals from the world’s four largest databases within 2 hours on a modest desktop computer [2108.04798]. The same line of work argues that the PDD will not allow anyone to claim a “new” material as a noisy disguise of a known crystal [2108.04798]. Machine-learning work using PDD reports accuracy on par with state-of-the-art methods while being several times faster in both training and prediction time [2401.15089].

The bridge-length algorithm was tested on a large crystal dataset and is explicitly described as required for an efficient continuous classification of all periodic crystals and as a key step toward inverse design of materials from new invariant values [2410.23288]. Higher-order PDD invariants were designed to distinguish all known counter-examples to the completeness of past descriptors and to confirm thousands of near-duplicates in the world’s largest databases of inorganic crystals within hours on a desktop computer [2509.15088]. These claims indicate a shift from purely theoretical classification to operational data curation.

Several limitations remain explicit in the literature. Density fingerprints are complete in general position in \(\mathbb{R}^3\), but incompleteness appears already in dimension \(1\) [2205.02226]. PDD is complete for all periodic sets in general position, yet higher-order constructions were needed to resolve known counterexamples to the completeness of first-order descriptors [2108.04798]. The problem is open for \(k\)-periodic sequences with \(k\ge 2\) in the metric framework of exactly computable elastic metrics [2205.04388]. Extension of explicit 1D density formulas to dimensions \(n>1\) is also open, because ball intersections become much more complex and the piecewise linear structure disappears [2205.02226].

The open directions stated across the cited papers are coherent. One line asks for stronger or simpler complete invariants and for parameterizations of the space of isometry classes that support reconstruction and inverse design [2103.02749]. Another asks for higher-dimensional generalizations of explicit density formulas, for extremal configurations minimizing \(\rho_k[S]\) or the maximum of \(\psi_k[S](t)\), and for understanding when \(\psi_k\) has a unique local maximum [2205.02226]. A third line concerns broader geometric settings: quasi-periodic structures, Delone sets beyond strict periodicity, substitution tilings with dense tile orientations, and point sets with almost periodic modulations [1210.2955].

Periodic point sets under isometry therefore form a domain in which discrete geometry, crystallography, topology, metric geometry, and data science now interact through a common language: lattices and motifs, rigid-motion equivalence, continuous invariants, and metrics on spaces of structures. The cited works show that this language is already strong enough to support explicit formulas in one dimension, exact algorithms for certain complete invariants, practical large-scale database analysis, and a developing global geometry of the moduli space itself [2103.02749].

Source: https://www.emergentmind.com/topics/periodic-point-sets-under-isometry