Cluster Angular Spread and Mean AoA Estimation for 3GPP CDL Channels: A CRLB Analysis

Multiple-input multiple-output (MIMO) mode selection is strongly affected by spatial correlation, which is dictated by angular statistics of the propagation environment. In the 3GPP clustered delay line (CDL) channel model, the cluster angular spread is a fundamental yet underexplored parameter that directly affects spatial correlation and achievable MIMO rank. This work primarily addresses the estimation of the cluster angular spread σAoA, and the mean angle-of-arrival µAoA. A CDL-consistent estimation framework is developed by combining a proposed delay-domain cluster identification method with conventional Bartlett power angular spectrum processing. Cramér-Rao lower bounds (CRLBs) for the joint estimation of σAoA and µAoA are derived to benchmark performance. Simulations under 3D-UMa NLOS show dominant clusters having higher signal-tonoise ratio (SNR) approaching the CRLB across wide SNR and array-size ranges.