Uniqueness theorems for Fourier transform (Q1918683)

Let \(f\in L_1(\mathbb{R})\) and \(\widehat f\in L_2(\mathbb{R})\) be its Fourier transform. The author obtains some uniqueness theorems for the Fourier transform. Theorem 1: Assume the set \(E\subset\mathbb{R}\) has the property that together with any point \(x\), \(E\) contains all the points \(x+ n\sqrt{2\pi}\), \(n\in\mathbb{Z}\). Then the assertion \(f(x)= \widehat f(x)= 0\) almost everywhere (a.e.) on \(\mathbb{R}\setminus E\) implies \(f(x)= 0\) a.e. on \(\mathbb{R}\). Theorem 2: Let \(E_\alpha= \bigcup_{n\in\mathbb{Z}} \left({\pi\over\alpha} n-\alpha,{\pi\over\alpha} n+\alpha\right)\), where \(\alpha\in \left(0,\sqrt{{\pi\over 2}}\right)\). If \(f(x)= 0\) a.e. on \(\mathbb{R}\setminus E_\alpha\) for some \(\alpha= \sqrt{{\pi\over 2n}}\), \(n= 2,3,\dots\), then \[ \int_{E_\alpha} |\widehat f(x)|^2 dx= {1\over n} \int_{E_\alpha}| f(x)|^2dx. \]