Self-reciprocal polynomials and coterm polynomials (Q1650429)

In a previous paper the author introduced the \(n\)-th reversed Dickson polynomial of the \((k+1)\)-th kind \(D_{n,k}(a,x)\) and showed that over a finite field of odd characteristic \[ D_{n,k}(1,x)=\left(\frac{1}{2}\right)^2f_{n,k}(1-4x), \] where \[ f_{n,k}(x)=k\sum_{j\ge 0}\binom{n-1}{2j+1}(x^j-x^{j+1})+2\sum_{j\ge 0} \binom{n}{2j}x^j\in\mathbb Z[x]. \] The fact sometimes \(f_{n, k}(x)\) is self-reciprocal has been useful in the study of these Dickson polynomials. Here the author determines when \(f_{n,k}(x)\) is self-reciprocal. Over \(\mathbb Z\) the results are: For \(n >1\) even, \(f_{n,k}(x)\) is self-reciprocal iff \(k=0\) or \(2\). For \(n>1\) odd, \(f_{n,k}(x)\) is self-reciprocal iff \(k=1\) or \(k=n=3\). There are similar results over \(\mathbb F_p\), \(p\) odd. For a commutative ring $R$, a polynomial \[ f(x)=a_0+a_1x+\cdots +a_{n-1}x^{n-1}\in R[x]/(x^n-1)\] is a \textit{coterm} polynomial if \(a_i=a_{n-i}\) for \(1\le i\le \lfloor \frac{n}{2}\rfloor\). They have been used in the study of DNA codes, see [\textit{E. S. Oztas} et al., Lect. Notes Comput. Sci. 8592, 124--128 (2014; Zbl 1402.94125)]. The author uses self-reciprocal \(f_{n,k}(x)\) to construct coterm polynomials. For example, if \(n\ge 6\) is even, \(k=0\) or \(2\), then \(f_{n,k}(x)-2nx^{n/2-1}\) is coterm.