Secure Federated Learning by Power Control for Internet of Drones

Document Type

Article

Publication Date

12-1-2021

Abstract

Fog-aided Internet of Drones (IoD), where massive training data are collected by drones and analyzed in the fog node, can leverage machine learning to provision various services. Aggregating all data in the fog node may incur huge network traffic and drone data privacy leakage. Federated learning (FL) is hence proposed to preserve drone data privacy by performing local training in drones and sharing training model parameters in the fog node without uploading drone raw data. However, drone privacy can still be divulged to ground eavesdroppers by wiretapping and analyzing uploaded parameters during the FL training process. In this paper, we investigate the power control of all drones to maximize the FL system security rate constrained by drone battery capacities and the quality of service (QoS) requirement (i.e., FL training time). We formulate this problem as a non-linear programming problem and design an algorithm to obtain the optimum solutions with a low computational complexity. Extensive simulations are conducted to demonstrate the performance of our proposed algorithm.

Identifier

85105040640 (Scopus)

Publication Title

IEEE Transactions on Cognitive Communications and Networking

External Full Text Location

https://doi.org/10.1109/TCCN.2021.3076167

e-ISSN

23327731

First Page

1021

Last Page

1031

Issue

4

Volume

7

Grant

CNS-1647170

Fund Ref

National Science Foundation

This document is currently not available here.

Share

COinS