Efficient Confidentiality and Erasure-Coded Storage for Scalable Multi-Cloud Sensor Networks

Abstract

As wireless sensor networks increasingly power smart cities, environmental monitoring, and industrial automation, the need for reliable, scalable, and cost-effective long-term data storage has become critical. This paper presents a novel intercloud storage framework based on two enhanced Byzantine fault-tolerant algorithms—Streaming DepSky-A and Streaming DepSky-CA—which extend the DepSky protocol to support real-time data streaming, confidentiality, and fault tolerance. These models eliminate the need for active intermediary servers while ensuring data availability and security across multiple untrusted cloud providers. Experimental evaluations across Amazon S3, Azure Blob, and Google Cloud Storage demonstrate that the proposed models achieve up to 20% cost savings over traditional replication techniques, 500,000+ measurements per second throughput per VM, and near 100% availability under fault-tolerant quorum conditions. Additionally, response latency across read and write operations was significantly reduced, confirming the model’s suitability for real-time and large-scale deployments. This work contributes to a scalable, vendor-neutral solution to secure and optimize sensor data retention in cloud-based infrastructures.