Improving signal detection accuracy at FC of a CRN using machine learning and fuzzy rules

Abstract

<span>The performance of a cognitive radio network (CRN) mainly depends on the faithful signal detection at fusion center (FC). In this paper, the concept of weighted Fuzzy rule in Iris data classification, as well as, four machine learning techniques named fuzzy inference system (FIS), fuzzy <em>c</em>-means clustering (FCMC), support vector machine (SVM) and convolutional neural network (CNN) are applied in signal detection at FC taking signal-to-interference plus noise ratio of secondary users as parameter. The weighted Fuzzy rule gave the detection accuracy of 86.6%, which resembles the energy detection model of majority rule of FC; however, CNN gave an accuracy of 91.3% at the expense of more decision time. The FIS, FCMC and SVM gave some intermediate results; however, the combined method gave the best result compared to that of any individual technique.</span>

A simplified optimization for resource management in cognitive radio network-based internet-of-things over 5G networks

“…It may also degrade the QoS performance owing to service relocation. IoT is also witnessed to improvise on operationalities while using cognitive radio network (CRN); however, one of the potential challenges in this regard is the degradation of quality of channel in case of more number of secondary users [10]. Apart from this, there is also a need to improve upon the optimization performance of the radio power so that better QoS performance can be delivered.…”