Multi-Feature Sensor Similarity Search for the Internet of Things

• DOI: 10.1587/transcom.2017ebp3221
• Publication Date: 2017-12-07T22:10:28Z
• Authors: Suyan LIU; Yuanan LIU; Fan WU; Puning ZHANG
• Keywords: Similarity (geometry); Identification; Feature (linguistics)

The tens of billions devices expected to be connected the Internet will include so many sensors that demand for sensor-based services is rising. task effectively utilizing enormous numbers deployed daunting. need automatic sensor identification has expanded research on similarity searches. Things (IoT) features massive non-textual dynamic data, which raising critical challenge efficiently and searching selecting most related a need. Unfortunately, single-attribute searches are highly inaccurate when among similar attribute values. In this paper, we propose group-fitting correlation calculation algorithm (GFC) can identify clusters sensors. GFC method considers multiple attributes (e.g., humidity, temperature) calculate similarity; thus, it performs more accurate than do existing solutions.