The rapid development of mobile communication technology has promoted the emergence of mobile edge computing (MEC), which allows mobile users to transfer their computing tasks to nearby edge servers to reduce access latency. In the actual MEC environment, the signal coverage areas of edge servers usually overlap partially, and users in the overlapped areas can choose to connect to one of the edge servers that cover them. How to allocate these users will seriously affect MEC performance. To solve this issue, we focus on the overlapped area user allocation (OAUA) problem in the MEC environment and model it as a multi-objective optimization problem. The objective is to balance the workload among edge servers and minimize the access delay between users and edge servers. Pareto model is universal for solving multi-objective optimization problems. However, the traditional method has high computational complexity to find the Pareto boundary. Therefore, we propose a Pareto boundary search algorithm based on convex hull to reduce the complexity of the algorithm. Since the Pareto boundary is a set of optimal solutions, which contains multiple optimal solutions, we further propose to use the principal component analysis algorithm to find the most suitable solution from the Pareto boundary as the final user allocation strategy. Our experiments use real data sets and compare the performance with several other baseline methods to verify the effectiveness of our proposed solution.

移动通信技术的快速发展推动了移动边缘计算（MEC）的出现，它允许移动用户将他们的计算任务转移到附近的边缘服务器，以减少访问延迟。在实际的MEC环境中，边缘服务器的信号覆盖区域通常会部分重叠，重叠区域的用户可以选择连接到覆盖它们的边缘服务器之一。如何分配这些用户将严重影响 MEC 性能。为了解决这个问题，我们关注 MEC 环境中的重叠区域用户分配 (OAUA) 问题，并将其建模为多目标优化问题。目标是平衡边缘服务器之间的工作负载，并最大限度地减少用户和边缘服务器之间的访问延迟。帕累托模型是解决多目标优化问题的通用模型。然而，传统方法寻找帕累托边界具有很高的计算复杂度。因此，我们提出了一种基于凸包的帕累托边界搜索算法，以降低算法的复杂度。由于帕累托边界是一组最优解，其中包含多个最优解，我们进一步提出使用主成分分析算法从帕累托边界中寻找最合适的解作为最终的用户分配策略。我们的实验使用真实数据集并将性能与其他几种基线方法进行比较，以验证我们提出的解决方案的有效性。由于帕累托边界是一组最优解，其中包含多个最优解，我们进一步提出使用主成分分析算法从帕累托边界中寻找最合适的解作为最终的用户分配策略。我们的实验使用真实数据集并将性能与其他几种基线方法进行比较，以验证我们提出的解决方案的有效性。由于帕累托边界是一组最优解，其中包含多个最优解，我们进一步提出使用主成分分析算法从帕累托边界中寻找最合适的解作为最终的用户分配策略。我们的实验使用真实数据集并将性能与其他几种基线方法进行比较，以验证我们提出的解决方案的有效性。