On the existence of a primitive quadratic of trace 1 over \(GF(p^ m)\) (Q1823994): Difference between revisions

In J. Comb. Theory, Ser. A 37, 13-21 (1984; Zbl 0547.05020), \textit{S. W. Golomb} conjectured that there exists a primitive quadratic of \(trace\quad 1\) over every \(GF(p^ m)\), i.e. that there is a primitive element in \(GF(p^{2m})\) of \(trace\quad 1\) over \(GF(p^ m)\). For \(p=2\) primitive elements of \(trace\quad 1\) are useful in algebraic coding theory. The author shows that Golomb's conjecture is true. The method of proof relies on estimates for character sums to show that the result is true for sufficiently large \(p^ m\) and then a computer is employed to check the remaining cases. It should be pointed out that in a forthcoming paper \textit{S. D. Cohen} [``Primitive elements and polynomials with arbitrary trace'', Discrete Math. (to appear)] proves, without any computer calculations, that the following more general result holds. Let \(m\geq 2\) and \(t\in GF(q)\) with \(t\neq 0\) if \(m=2\) or if \(m=3\) and \(q=4\). Then there exists a primitive element in \(GF(q^ m)\) with trace t.