Euclidean lattice

A Euclidean lattice is a discrete subgroup $\Lambda \subseteq {\mathbb{R}}^n$ spanning the ambient vector space. Its covolume $m(\Lambda )$ is the volume of a fundamental parallelepiped.

For an integer basis of a Euclidean lattice, the algorithm uses integer column operations and swaps to obtain a size-reduced basis with Gram-Schmidt coefficients $|{\mu }_{ij}|\le 1/2$ and the condition $\delta \Vert b_i^{\ast }{\Vert }^2\le \Vert b_{i+1}^{\ast }{\Vert }^2+{\mu }_{i+1,i}^2\Vert b_i^{\ast }{\Vert }^2$. It preserves the lattice and exposes short vectors and near-relations. It is useful for improving proven Diophantine bounds by encoding scaled logarithms in integer coordinates. A certified lower bound for every nonzero lattice vector is ${\min }_i\Vert b_i^{\ast }\Vert$: in a nonzero integer linear combination, project onto the last nonzero Gram-Schmidt direction. Better-reduced bases improve this lower bound. Rounding and logarithm errors must still be bounded before turning it into a certified lower bound for a logarithmic form.

A complex lattice is a discrete rank-two subgroup of the complex plane, generated by two real-linearly independent periods. An oriented basis can be chosen with $\mathrm{Im}({\omega }_1/{\omega }_2)>0$. Scaling by ${\omega }_2^{-1}$ then gives $\mathbb{Z}\tau +\mathbb{Z}$ for a point in the complex upper half-plane.

Two complex lattices are homothetic when multiplication by one nonzero complex number takes one onto the other. Changes between oriented integral bases have determinant one, so normalized period ratios classify these classes by $S{L}_2(\mathbb{Z})\backslash \mathbb{H}$. The lattice Eisenstein sums scale as $G_k(a\Lambda )=a^{-k}{G}_k(\Lambda )$.

For the skew matrix $S=\left(\begin{array}{cccc}0& 1& 1& 1\\ -1& 0& -1& 1\\ -1& 1& 0& -1\\ -1& -1& 1& 0\end{array}\right)$, both Euclidean lattices reduce to the tetracode modulo $3$. A reflection depending on the reduction coset matches their vectors coordinatewise in absolute value. Their weighted length spectra agree for all positive diagonal axis weights. Rationally independent weights force any possible linear isometry to be a coordinate sign change; the code then forces an overall sign, which does not identify the two lattices.

Choose positive axis weights linearly independent over the rationals, and suppose both integer Euclidean lattices contain $q{\mathbb{Z}}^d$ for some positive integer $q$. A length-preserving linear map sends each $q{e}_i$ to an integer vector with exactly that absolute-coordinate pattern, by the displayed implication. Thus the map must be a diagonal sign change. For the four-coordinate tetracode, preserving the code forces all four signs to agree, because each of its four nonzero projective words has support on a different triple of coordinates. A global sign cannot exchange the two odd-vector parity conditions, proving nonisometry for these weights.

Add respectively the conditions ${\sum }_i{x}_i\equiv 0(\mathrm{mod}4)$ and ${\sum }_i{x}_i-2x_1\equiv 0(\mathrm{mod}4)$ to the displayed common conditions. These define the two integer Euclidean lattices underlying the tetracode length-isospectral lattice construction. They coincide on their all-even vectors and differ on all-odd vectors. Reflecting a coordinate divisible by three preserves the code condition; on odd vectors it changes the coordinate sum by two modulo four, while on even vectors both conditions remain satisfied. This yields the coordinate reflection bijection for tetracode lattices.

For the tetracode length-isospectral lattice construction $L_{\pm }=(\pm 3I+S){\mathbb{Z}}^4$, choose the first coordinate $i$ divisible by three and reverse its sign. Such a coordinate always exists because each nonzero tetracode word has one zero coordinate. The reflection preserves the reduction modulo three and switches the two lattice membership conditions modulo four. It is an involution between $L_{+}$ and $L_{-}$ preserving every absolute coordinate, so arbitrary positive diagonal weights preserve equality of their vector-length multisets.

The covolume of a full Euclidean lattice is the Lebesgue measure of a fundamental parallelepiped in its ambient inner product space. In orthonormal coordinates it is $|\det B|$ for a lattice basis matrix $B$. Scaling the lattice by $t>0$ multiplies the covolume by $t^n$; the dual lattice has reciprocal covolume.

In a rank-two Euclidean lattice, a shortest nonzero vector $v$ and a shortest vector $w\notin \mathbb{Z}v$ form a basis. The shortest vector is primitive. If the vertical coordinate of $w$ has index $m\ge 2$ in the projection lattice, a vector with smaller positive vertical coordinate can be reduced horizontally to length squared at most $(|v{|}^2+|w{|}^2)/4<|w{|}^2$, a contradiction. This lemma need not hold in higher rank.

For a full-rank Euclidean lattice of covolume $A$ in ${\mathbb{R}}^d$, bounded fundamental cells compare lattice points in a radius-$R$ ball with the volumes of balls of radii $R\pm C$. Hence $N(R)=\mathrm{vol}(B_1)R^d/A+O(R^{d-1})$. A lattice norm multiset therefore determines the covolume, which is useful for the spectrum of a flat torus.

The dual lattice of a full-rank Euclidean lattice $\Lambda \subset {\mathbb{R}}^n$ isHere $⟨x,y⟩$ is the Euclidean inner product. If the columns of an invertible matrix $B$ form a basis of $\Lambda =B{\mathbb{Z}}^n$, thenIndeed, $y$ belongs to the dual lattice exactly when $B^{\mathsf{T}}y\in {\mathbb{Z}}^n$; $B^{-\mathsf{T}}$ denotes the transpose of the matrix inverse. The reciprocal covolume follows by taking the determinant. With the Fourier transform kernel $e^{-2\pi i⟨x,y⟩}$, these are precisely the frequencies of the characters of a real torus: translating $x$ by any $\lambda \in \Lambda$ leaves $e^{2\pi i⟨x,y⟩}$ unchanged. In one dimension, $(h\mathbb{Z}{)}^{\vee }=h^{-1}\mathbb{Z}$ for $h>0$: increasing the spacing in physical space decreases the spacing in frequency space.

Every continuous group homomorphism ${\mathbb{R}}^n/\Lambda \to {\mathbb{C}}^{\times }$ is $x+\Lambda \mapsto e^{2\pi i⟨u,x⟩}$ for a unique $u$ in the dual lattice. Compactness forces unit modulus, and lifting the character to the simply connected cover makes its phase a real linear functional. These characters are the frequencies of torus Fourier series.