Bruner-Greenlees conjecture on real connective \(K\)-theory of generalized quaternion groups (Q388825)

For a finite group \(G\), the calculation of the real connective \(K\)-homology \(ko_*(BG)\) is an important ingredient in the investigation of the Gromon-Lawson-Resenberg Conjecture. For \(G = Q_{2n}\), the quartenion group, there are partial calculations for this homology. \textit{R. R. Bruner} and \textit{J. P. C. Greenlees} [Connective real \(K\)-theory of finite groups. Mathematical Surveys and Monographs 169. Providence, RI: American Mathematical Society (AMS) (2010; Zbl 1251.19001), and Mem. Am. Math. Soc. 785, 127 p. (2003; Zbl 1032.19003)] offered an algebraic systematic way to compute \(ko_{*}(BG)\), called the Bruner-Breenlees method. The method use the calculation of \(ku_*(BG)\), \(ku^*(BG)\) and \(ko^*(BG)\) together with certain spectral sequences (the Adams Spectral Sequence, the \(\eta\)-Bockstein Spectral Sequence and the Greenlees Spectal Sequence), together with information on the cohomology of \(BG\) to calculate \(ko_*(BG)\). Bruner and Greenlees gave explicit calculations of \(ko_*(BG)\) for certain classes of finite groups. For the generalized quartenion group \(Q_{2}^{n+2}\), they calculated the complex \(K\)-homology and cohomology of \(BQ_{{2}^{n+2}}\). As a result of a series of calculations by Bruner and Greenlees, they formulated a conjecture for the ko-homology of \(BQ_{{2}^{n+2}}\). In the paper under review the author proves the conjecture. The author uses the suggestions of Bruner-Greenlees for proving the conjecture, using the Bruner-Greenlees method. The main tools used in the understanding of the \(\eta\)-Bockstein Spectral Sequence together with the use of information on the representation theory of \(Q\). The results are consistent with the results of \textit{B. Botvinnik} et al. [J. Differ. Geom. 46, No. 3, 374--405 (1997; Zbl 0929.58011)] who calculated the order of \(ko_{4n-1}(BQ_{2^n})\).