Harmonic Analysis on Directed Networks via a Biorthogonal Laplacian Calculus for Non-Normal Digraphs

TL;DR

This paper develops a harmonic analysis framework for directed graphs with non-normal Laplacians, introducing a biorthogonal Fourier transform and stability bounds that account for non-normality effects.

Contribution

It introduces a biorthogonal graph Fourier transform and stability analysis for directed graphs with non-normal Laplacians, advancing spectral methods for directed network analysis.

Findings

Vertex energy equals a Gram-metric quadratic form in BGFT coordinates

Sampling and reconstruction guarantees depend on eigenvector geometry

Filtering robustness correlates with non-normality measures

Abstract

Spectral graph signal processing is traditionally built on self-adjoint Laplacians, where orthogonal eigenbases yield an energy-preserving Fourier transform and a variational frequency ordering via a real Dirichlet form. Directed networks break self-adjointness: the combinatorial directed Laplacian $L=D_{\mathrm{out}}-A$ is generally non-normal, so eigenvectors are non-orthogonal and classical Parseval identities and Rayleigh-quotient orderings do not apply. This paper develops a Laplacian-centric harmonic analysis for directed graphs that remains exact at the algebraic level while explicitly quantifying the geometric distortion induced by non-normality. We (i) define a Biorthogonal Graph Fourier Transform (BGFT) for $L$ using dual left/right eigenbases and show that vertex energy equals a Gram-metric quadratic form in BGFT coordinates, (ii) introduce a directed variational semi-norm $TV_{\mathcal{G}}(x)=\|Lx\|_2^2$ and prove sharp two-sided BGFT-domain bounds controlled by singular values of the eigenvector matrix, and (iii) derive sampling and reconstruction guarantees with explicit stability constants that separate sampling-set informativeness from eigenvector geometry. Finally, we provide reproducible simulations comparing a normal directed cycle to perturbed non-normal digraphs and show that filtering and reconstruction robustness track $\kappa(V)$ and the Henrici departure-from-normality $\Delta(L)$, validating the theoretical predictions.