Multi-Attribute Top-k Query Processing Leveraging Caching Mechanism in IoT Sensing Networks
Lizhi Zhang · 2020
The large-scaled and multifarious smart things generate huge sensory data in the Internet of Things (IoT). The collaboration of cloud computing and edge computing is acquired to support industrial applications to process the continuous real-time monitoring and query, and bring the query results to a centralized entity (e.g., cloud). In the context of edge-cloud collaborative architecture, answering the preference top-k query is a challenging issue. To address this issue, we propose a method to improve the performance of answering continuous preference top-k queries in IoT sensing networks. Specifically, a hierarchical region quadtree is constructed to support efficient query processing by eliminating invalid data at branch nodes. The cloud divides the query into different types according to their preferences, and a responsible edge node caches data records for different types of queries based on popularity. The efficient filter thresholds of top-k queries generate from the cache. To further reduce the transmission of invalid data records, a grid index scheme is developed. Experiments indicate our proposed approach is promising in reducing the energy cost of network transmission.