This website requires JavaScript.

FedBA: Non-IID Federated Learning Framework in UAV Networks

Pei LiZhijun LiuLuyi ChangJialiang PengYi Wu
Oct 2022
摘要
With the development and progress of science and technology, the Internet ofThings(IoT) has gradually entered people's lives, bringing great convenience toour lives and improving people's work efficiency. Specifically, the IoT canreplace humans in jobs that they cannot perform. As a new type of IoT vehicle,the current status and trend of research on Unmanned Aerial Vehicle(UAV) isgratifying, and the development prospect is very promising. However, privacyand communication are still very serious issues in drone applications. This isbecause most drones still use centralized cloud-based data processing, whichmay lead to leakage of data collected by drones. At the same time, the largeamount of data collected by drones may incur greater communication overheadwhen transferred to the cloud. Federated learning as a means of privacyprotection can effectively solve the above two problems. However, federatedlearning when applied to UAV networks also needs to consider the heterogeneityof data, which is caused by regional differences in UAV regulation. Inresponse, this paper proposes a new algorithm FedBA to optimize the globalmodel and solves the data heterogeneity problem. In addition, we apply thealgorithm to some real datasets, and the experimental results show that thealgorithm outperforms other algorithms and improves the accuracy of the localmodel for UAVs.
展开全部
图表提取

暂无人提供速读十问回答

论文十问由沈向洋博士提出,鼓励大家带着这十个问题去阅读论文,用有用的信息构建认知模型。写出自己的十问回答,还有机会在当前页面展示哦。

Q1论文试图解决什么问题?
Q2这是否是一个新的问题?
Q3这篇文章要验证一个什么科学假设?
0
被引用
笔记
问答