5th Generation (5G) mobile communication systems operating at around 28 GHz have the potential to be applied to simultaneous localization and mapping (SLAM). Most existing 5G SLAM studies estimate environment as many point targets, instead of extended targets. In this paper, we focus on the performance analysis of 5G SLAM for multiple extended targets. To evaluate the mapping performance of multiple extended targets, a new mapping error metric, named extended targets generalized optimal sub-pattern assignment (ET-GOPSA), is proposed in this paper. Compared with the existing metrics, ET-GOPSA not only considers the accuracy error of target estimation, the cost of missing detection, the cost of false detection, but also the cost of matching the estimated point with the extended target. To evaluate the performance of 5G signal in SLAM, we analyze and simulate the mapping error of 5G signal sensing by ET-GOPSA. Simulation results show that, under the condition of SNR = 10 dB, 5G signal sensing can barely meet to meet the requirements of SLAM for multiple extended targets with the carrier frequency of 28 GHz, the bandwidth of 1.23 GHz, and the antenna size of 32.

翻译：第五代（5G）移动通信系统工作在28 GHz频段，具有应用于同时定位与地图构建（SLAM）的潜力。现有大多数5G SLAM研究将环境估计为多个点目标，而非扩展目标。本文聚焦于多扩展目标场景下5G SLAM的性能分析。为评估多扩展目标的建图性能，本文提出一种新的建图误差度量——扩展目标广义最优子模式分配（ET-GOPSA）。与现有度量相比，ET-GOPSA不仅考虑了目标估计的精度误差、漏检代价和虚警代价，还纳入了估计点与扩展目标匹配的代价。为评估5G信号在SLAM中的性能，我们利用ET-GOPSA对5G信号感知的建图误差进行了分析与仿真。仿真结果表明，在信噪比为10 dB的条件下，5G信号感知在载频28 GHz、带宽1.23 GHz、天线尺寸为32时，勉强能满足多扩展目标SLAM的需求。