Non-localized and localized data storage in large-scale communicating materials: Probabilistic and hop-counter approaches

摘要

The rapid development of Internet of Things has triggered the multiplication of communication nodes based on Radio-Frequency Identification (RFID) and Wireless Sensor Networks (WSNs) in various domains such as building, city, industry, and transport. These communication nodes are attached to a thing or directly included in the material of the thing to form a communicating material. In communicating material, one of the desired objectives is to merge the logical data with its physical material, thus simplifying the monitoring of its life cycle, the maintenance operations, and the recycling process. In this context, the initial form of the communicating material can evolve during its lifecyde. It can be split, aggregated with other materials, or partially damaged. However, the entire information in the material should always be accessible after each change. Thus, the objective of this research is to develop specific algorithms for efficient dissemination of information in the material in order to limit information losses. Two dissemination algorithms hop-counter-based and probabilistic-based are proposed for storing data by using WSNs, and non-localized and localized storage is considered. Non-localized storage ensures that information can be retrieved from each piece of the material by using a uniform data replication process. Localized storage ensures that the information is stored in a limited region of the material. Castalia/OMNeT++ simulator is used to compare the performance of the proposed algorithms with other similar protocols such as DEEP, Supple, and RaWMS.