Cell decomposition for two dimensional local fields (Q2462249)

In this paper the author proves several versions of \textsl{cell decomposition} for the field \(K((t))\) of Laurent series over a \(p\)-adic field \(K\). These results extend work of \textit{J. Denef} [J. Reine Angew. Math. 369, 154--166 (1986; Zbl 0584.12015)] and \textit{J. Pas} [J. Reine Angew. Math. 399, 137--172 (1989; Zbl 0666.12014)] concerning cell decomposition for \(p\)-adic fields, which is also the basis for the proofs here. The language (in the sense of mathematical logic) in the present paper is more complicated: it is the so-called `language of 2-valued fields', corresponding to the \(t\)-adic valuation on \(K((t))\) and the \(p\)-adic valuation on \(K\). The basic idea of (one version of) the proven cell decomposition is roughly as follows. Let \(f_1,\dots,f_r\) be polynomials in one variable whose coefficients are definable functions in \(x\in K((t))^m\). Then there exists a \textsl{finite} partition of \(K((t))^m \times K((t))\) in certain `elementary' definable subsets, called \textsl{cells}, such that, for all \((x,T)\) in such a cell, the \(t\)-valuation and angular component of \(f_i(x,T)\) have an `easy and uniform' description.