Performance analysis for parallel grouping‐based labeled multi‐Bernoulli filter

•Firstly, we propose a hypothesis weight related grouping criterion for the label space and measurement set that generalizes MHD-based grouping of [1].•Secondly, we characterize theoretical performance of the PG-LMB filter in terms of L1-error between the IG-IA and the exact GLMB posterior utilizing the property of the proposed grouping criterion.•Thirdly, we reveal the “low-dimensional” compression property of the IG-IA, utilizing which, the PG-LMB filter has the potential to achieve better tracking performance than the LMB filter without grouping, under limited computational resources.