This paper investigates the data aggregation problem in a multi-layer mobile edge computing (MEC)-enabled unmanned aerial vehicle (UAV) systems. When UAV users transmit data to the central operator node for analysis and processing, MEC is a promising paradigm to provide low-delay service for user applications. For UAV users, the quality of experience (QoE) metric can facilitate them to receive more satisfying services according to different task requirements. In this paper, a QoE-based utility model is proposed to guarantee the service quality of each user-layer UAV. Note that in the multi-layer aggregation framework, there are three types of UAVs, e.g., users, helpers, and central operator. Then, an optimization problem is formulated to improve the QoE of users and reduce the cost simultaneously. And the optimization variables are aggregate node selection and resource allocation. Next, we exploit a low complexity matching game method to complete the selection of aggregation nodes and fair resource allocation, e.g., virtual machine (VM) resources. The simulation results demonstrate that the proposed scheme is superior to other traditional schemes.