Representations of affine Hecke algebras (Q1337749)

It should be indicated that the concept of an affine Hecke algebra in the present book slightly differs from the usual one. Let \(G\) be a connected reductive group over the complex field \(\mathbb{C}\) and \(T\) a maximal torus of \(G\). Let \(W_ 0\) be the Weyl group of \(G\) with respect to \(T\) and \(X\) the character group of \(T\). The semi-direct product \(W = W_ 0 \ltimes X\) is called an extended affine Weyl group. Then in the present book, the Hecke algebra \(H_ q\) (\(q\) is a non-zero complex number) of \(W\) is called an affine Hecke algebra. This definition of an affine Hecke algebra agrees with the usual one precisely when \(G\) is adjoint. In the representation theory of an affine Hecke algebra \(H_ q\), one of the main themes is to classify all the \(H_ q\)-simple modules. Kazhdan and Lusztig classified all these modules in the cases when the parameter \(q\) is either 1 or is not a root of unity. Ginzburg got some interesting results in this respect. The author considers this theme in the case when \(q \neq 1\) is a root of 1. He shows that when the order of \(q\) is not too small, the representation theory of \(H_ q\) is entirely the same as that in the case when \(q\) is either 1 or is not a root of unity. On the other hand, he shows that it becomes quite different when \(q\) is a root of unity of certain orders. In order to understand the simple \(H_ q\)-modules well, the author studies some based rings related to the asymptotic Hecke algebras of affine Weyl groups. A conjecture of Lusztig relates these based rings to the \(K_ 0\)-rings of certain vector bundles on finite sets. The author verifies this conjecture in some special cases, in particular, in all cases when the concerned affine Weyl group has rank two. The author renames Lusztig's terminology ``\(\phi\)-cell'' by ``generalized cell'' in this book. But I think that it is better to retain the name ``\(\phi\)-cell''. The reason for this is as follows. (a) Such a ``cell'' is heavily dependent on the map \(\phi\). (b) There is no natural relation between such a ``cell'' and an ordinary cell. For example, there are given conditions for the following assertion to hold: the lowest two- sided cell \(c_ 0\) of an affine Weyl group is a generalized two-sided cell (see Chapter 3 of this book). Rui He-bing has found a counter- example in the case when \(W_ a = \langle s,t,r \mid (st)^ 4 = (tr)^ 4 = (sr)^ 2 = s^ 2 = t^ 2 = r^ 2 = 1\rangle\) and \(\phi(r)^{1/2} > \phi(s)^{1/2} \phi(t)\).