Department of Mathematics and Statistics University of Alaska Fairbanks, USA
The Korteweg-de Vries equation and Hankel operators
As well-known, many problems in the theory of completely integrable systems can be formulated in terms of Riemann-Hilbert boundary problems. This has been used (explicitly or implicitly) since the late 1980s. On the other hand, it is also well-known that the Riemann-Hilbert problem is closely related to the theory of Hankel and Toeplitz operators. Moreover, since the 1960s (and implicitly even earlier) the former has stimulated the latter. But, surprisingly enough, while having experienced a boom at the same time, soliton theory and the theory of Hankel and Toeplitz operators have not shown much of direct interaction.
In the KdV context, we construct a Hankel operator which symbol is conveniently represented in terms of the scattering data for the Schrodinger operator associated with the initial data. Thus the spectral properties of this Schrodinger operator can be directly translated into the spectral properties of the Hankel operator. The latter then yield properties of the solutions to the KdV equation through explicit formulas. This allows us to recover and improve on many already known results as well as a variety of new ones. The main feature of this approach is that it applies to large classes of initial data far beyond the classical realm. For instance, we can handle low regularity initial data, lift any decay assumption at minus infinity, and significantly relax the decay at plus infinity. In this talk we discuss some representative results in this context focusing on well-posedness issues and basic properties of underlying solutions.
Our approach is not restricted to the KdV. Moreover, we believe that the interplay between soliton theory and Hankel operators may be even more interesting and fruitful for some other integrable systems with richer than KdV structures.