A classification-based sensor data processing method for the internet of things assimilated wearable sensor technology

Internet of Things (IoT) based wearable healthcare data monitoring is an emerging application of smart medicines. IoT-based communication and information exchange methods make it adaptable for the recent sixth-generation computing systems. The terahertz and large-scale processing of the 6th generation communication and computation technology is assimilated in the wearable sensor data exchange process. In this article, a fault-tolerant data processing method (FTDPM) is proposed to handle uneven sensor data. The non-uniform and unsynchronised observation interval-based sensor data is analyzed for its impact on healthcare recommendations. The reliability in the recommendation ensures the precise need for sensor data processing, mitigating the faults. In this reliability estimation process, a support vector machine classifier is used. This learning classifier differentiates the uniform and non-uniform sensor data traffic for providing reliable recommendations. The data from the uniform process is replicated in the non-uniform sequence for recommendation filling. This is carried out based on marginal classification and near-to-reliable data as classified using SVM. The IoT wearable alliance is exploited using the 6G communication paradigms in handling monitored data. The proposed method's performance is verified using the metrics processing time, recommendation failure, processing complexity, and recommendation response time.

Automated summary

This article proposes a fault-tolerant data processing method for handling uneven sensor data, and uses a support vector machine classifier to provide reliable recommendations. The method can be applied in IoT wearable technology, and its performance is verified using multiple metrics.