Enhancing Data Reliability for Fault Tolerant IoT Nodes

Abstract

In the rapidly evolving landscape of the Internet of Things (IoT), where billions of inter- connected devices form a vast network, ensuring data reliability has become paramount. The fault tolerance of IoT nodes is crucial in maintaining the integrity and efficiency of the entire system. IoT nodes must gather and process data from a wide range of existing structures.When the transmission data is faulty it leads to cause for inappropriate actions. Detecting and newlinerecovering the faulty node s data in the network is a challenging task. To maintain high data availability and reliability few approaches are proposed. In the first work, mainly focused on the device layer within IoT, where faults can occur, and introduces the Cold Standby Sparing Redundancy (CSSR) technique to enhance fault tolerance. The technique employs two identi- cal modules as cluster heads, significantly improving system reliability upto 81% as compared with other redundancy approaches. In the second work, a novel data recovery approach has been proposed to recover the faulty node data by using a Redundant Array of Independent newlineDisks (RAID) structure which is used for redundant data storage of the IoT node in the net- work.DRAFT algorithm invokes at IoT node storage unit will achieves 85% data reliability for single node as compared with existing algorithms. In the third work,the proposed approaches leverages spatial-temporal (ST) correlation between IoT nodes, utilizing clustering and a data recovery phase. Missing data is recovered using the ST-hierarchical long short-term mem- newlineory (ST-HLSTM) algorithm, resulting in an impressive 98.5% reliability compared to existing methods newline