Linear independence of finite Gabor systems determined by behavior at infinity (Q2256830)

If \(g\) is a measurable function on \(\mathbb R\) and \(\Lambda=\{(\alpha_k,\beta_k)\}_{k=1}^N\) the finite Gabor system generated by \(g\) and \(\Lambda\) is defined by \(\mathcal G(g, \Lambda)=\{e^{2\pi i \beta_k x}g(x-\alpha_k)\}_{k=1}^N\). It was conjectured by \textit{C. Heil} et al. [Proc. Am. Math. Soc. 124, No. 9, 2787--2795 (1996; Zbl 0859.42023)] that if \(g\in L^2(\mathbb R)\setminus \{0\}\) then \(\mathcal G(g, \Lambda)\) is a linearly independent set. There are a number of partial results in this direction. The authors show that the conjecture holds true for several classes of functions including those square integrable functions whose germs are analytic and belong to a Hardy field closed under translation and for which \(\lim_{x\to \infty} \frac{g(x+\alpha)}{g(x)}\) exits for every \(\alpha\in \mathbb R\) or those decaying faster than any exponential function. Also results when some extra assumption on the set \(\Lambda\) is assumed are given, for instance if at least one of the \(\beta_k\) is different from all the others or for Gabor systems generated by positive functions.