Spectral determination of semi-regular polygons (Q6158160)

In this paper the characterization of a class of convex polygons by their area, perimeter and a third geometric quantity is considered. To illustrate the main result, we first introduce two definitions. For a convex \(n\)-sided polygon \(P\), define \(S(P)\) as \[ S(P):=\sum_{k=1}^n \frac{\pi^2-\theta_k^2}{\theta_k}, \] where \(\theta_k\in(0,\pi)\) with \(1\leq k\leq n\) are the interior angles of the polygon \(P\). A convex polygon is called semi-regular if it is circumscriptible, and its angles are all equal but possibly one which is larger than the rest. The main result of this paper says that: Let \(P\) be a semi-regular polygon and \(Q\) a convex polygon. If they have the same area, perimeter and quantity \(S\), then they are congruent. It is already known that in the class of polygons with a fixed number of sides, regular polygons are characterized by their area and perimeter. The new result says that when the number of sides is allowed to vary, the area and perimeter are not enough. One needs a third geometric quantity. The proof of this result is achieved by considering a constrained maximization problem whose maximizer happens to be a semi-regular polygon. As an application, the authors obtain the following spectral determination result: If \(\Omega\) is a convex piecewise smooth bounded planar domain, possibly with straight corners, and its Dirichlet or Neumann spectrum coincides with that of a semi-regular polygon \(P\), then \(\Omega\) is congruent to \(P\).