No abstract

No abstract

Abstract-The Purpose of homomorphic encryption is to ensure privacy of data in communication, storage or in use by processes with mechanisms similar to conventional cryptography, but with added capabilities of computing over encrypted data, searching an encrypted data, etc. Homomorphism is a property by which a problem in one algebraic system can be converted to a problem in another algebraic system, be solved and the solution later can also be translated back effectively. Thus, homomorphism makes secure delegation of computation to a third party possible. Many conventional encryption schemes possess either multiplicative or additive homomorphic property and are currently in use for respective applications. Yet, a Fully Homomorphic Encryption (FHE) scheme which could perform any arbitrary computation over encrypted data appeared in 2009 as Gentry's work. In this paper, we propose a multi-cloud architecture of N distributed servers to repartition the data and to nearly allow achieving an FHE.

No abstract

Summary Ant colony optimization (ACO) and unequal clustering algorithms in wireless sensor networks (WSNs) prove their efficiency in protracting the network lifetime. However, the existing ACO and unequal clustering algorithms, respectively, do not consider jointly energy efficiency and reliability and focus only on some normal parameters to adjust the cluster radius, then neglecting the cluster head (CH) neighborhood information as it is wise to reduce the cluster radius when there are more neighbor CHs in order to balance the load and energy consumption. To resolve these problems, we propose a fault‐tolerant distributed ACO‐based routing (DACOR) protocol for mitigating the hot spot problem in fog‐enabled WSN architecture. To improve the performance of the network, we propose a multiple fog nodes (FNs) and unequal clustering‐based network model. The proposed model is energy efficient as it avoids repetitive clustering and affects CHs to FNs based on distance. Also, unlike the existing works which use either single FN/sink‐based unequal clustering or multiple FNs/sinks to mitigate hot spot problem, we propose to distribute unequal clustering to multiple FNs (partitions). Additionally, we formulate a different rule to calculate the cluster radius based on significant parameters ensuring energy efficiency and balancing. To route data from source to destination, we devise a new probabilistic formula which considers not only energy efficiency but also reliability. The performance of the proposed DACOR protocol has been investigated under different scenarios through simulations. The results show that the proposed DACOR protocol outperforms the existing protocols in terms of various main metrics.