No label defined (Q4025972)

[For the entire collection see Zbl 0745.00060.] Let \(p\) be a prime integer and \(\mathbb{F}_ q\) the finite field with \(q=p^ n\) elements. Let \(A=\mathbb{F}_ q[T]\) and let \(L\) be any extension field of \(k=\mathbb{F}_ q(T)\). Let \(\tau\) be the \(q\)-th power mapping of \(\mathbb{G}_ a\) and let \(L\{\tau\}\) be the ring of algebraic endomorphisms of \(\mathbb{G}_ a\) generated by \(\tau\); so \(L\{\tau\}\) is the ring of endomorphisms \(\left\{\sum^ m_{j=0}\alpha_ j\tau^ j\mid\alpha_ i\in L\;\forall i\right\},\) subject to the commutation relation \(\tau\cdot\alpha=\alpha^ q\cdot\tau\), \(\alpha\in L\). A Drinfeld \(A\)- module is then an \(\mathbb{F}_ q\)-linear injection \(\varphi\) of \(A\) into \(L\{\tau\}\), \(a\mapsto\varphi_ a\), such that \(\varphi_ T=T\tau^ 0+\cdots+a_ d\tau^ d\), \(d>0\), \(a_ d\neq 0\). We say that \(d\) is the rank of \(\varphi\). Now let \(L\) be a finite extension of \(K=\mathbb{F}_ q((1/T))\). If we equip \(\mathbb{F}_ q((1/t))\) with the topology arising from making \(1/T\) small, then one is also able to topologize the algebraic closure \(\overline K\) of \(K\) in a canonical fashion. An \(L\)-lattice is defined to be a finite rank \(A\)-submodule of the separable closure \(L^{sep}\subset\overline K\) which is stable under \(\text{Gal}(L^{sep}/L)\). Drinfeld established that such lattices of rank \(d\) correspond to Drinfeld modules of rank \(d\) via an exponential function (which is similar to the classical exponential function). Suppose now that we replace \(\mathbb{F}_ q\) by any commutative field \(F\), \(K\) by \(F((1/T))\), and \(\tau\) by a continuous endomorphism \(\sigma\) of \(F((1/T))\). Much of the theory can be carried over in this much more general context. So one has the exponential, logarithm, etc., in this more general context. Under certain conditions one can also get (\(\sigma\)-) transcendency results in the manner of Wade and Dubovitskaya. For more along these lines the reader may consult the author's ``Galois calculus and Carlitz exponentials'' which has appeared in [The Arithmetic of Function Fields, Proc. Workshop Ohio State Univ. 1991, 33-50 (1992)]. Finally, as is well-known, the theory of Drinfeld modules is valid in much greater generality than just the polynomial ring \(\mathbb{F}_ q[T]\). Indeed, let \({\mathcal C}\) be any smooth, geometrically connected curve over \(\mathbb{F}_ q\) and let \(\infty\in{\mathcal C}\) be a closed point. Then one can develop the theory for the affine ring of \({\mathcal C}-\infty\). It thus makes sense to ask whether the present theory can also be developed in this great generality.