1: \begin{abstract}

2: We propose a group information geometry approach (GIGA) for ultra-massive multiple-input multiple-output (MIMO) signal detection. 

3: The signal detection task is framed as computing the approximate marginals of the {\textsl{a posteriori}} distribution of the transmitted data symbols of all users.

4: With the approximate marginals, we perform the maximization of the {\textsl{a posteriori}} marginals (MPM) detection to recover the symbol of each user.

5: Based on the information geometry theory and the grouping of the components of the received signal, three types of manifolds are constructed and the approximate  {\textsl{a posteriori}} marginals are obtained through  \textsl{m}-projections.

6: The Berry-Esseen theorem is introduced to offer an approximate calculation of the \textsl{m}-projection, while its direct calculation is exponentially complex.

7: In most cases, more groups, less complexity of GIGA.

8: However, when the number of groups exceeds a certain threshold, the complexity of GIGA starts to increase.

9: %Further, damped updating is applied in GIGA to improve its convergence.

10: Simulation results confirm that the proposed GIGA achieves better bit error rate (BER) performance within a small number of iterations, which demonstrates that it can serve as an efficient detection method in ultra-massive MIMO systems.

11: \end{abstract}

12: