Summary In this work, privacy‐preserving distributed deep learning (PPDDL) is re‐visited with a specific application to diagnosing long‐term illness like diabetic retinopathy. In order to protect the privacy of participants datasets, a multi‐key PPDDL solution is proposed which is robust against collusion attacks and is also post‐quantum robust. Additionally, the PPDDL solution provides robust network security in terms of integrity of transmitted ciphertexts and keys, forward secrecy, and prevention of man‐in‐the‐middle attacks and is extensively verified using Verifpal. Proposed solution is evaluated on retina image datasets to detect diabetic retinopathy, with deep learning accuracy results of 96.30%, 96.21% and 96.20% for DDL, DDL + SINGLE and DDL + MULTI scenarios respectively. Results from our simulation indicate that accuracy of the PPDDL is maintained while protecting the privacy of the datasets of participants. Our proposed solution is also efficient in terms of the communication and run‐time costs.

Fog computing allows data to be processed on the network edge without reaching the cloud infrastructure to reduce latency and network bandwidth. However, it is not without its security challenges as existing security protocols, implemented in the fog, do not fully cater for the mobility and heterogeneity of the fog, especially on resource-constrained fog nodes. As such, this increases latency and overhead on those nodes which also affects the fog. This paper investigates the possibility of creating a one-time pad (OTP)-based encryption protocol with no packet loss; lesser time and energy overheads as compared to protocols that have been proposed by existing research. The protocol will be tested on wireless sensor nodes, which are resource constrained, and the outcome monitored. The OTPs will be generated using a random number generator within the nodes. Outcomes are positive and can be implemented on resource-constrained fog nodes.