This paper presents a study undertaken to evaluate the sensor systems that were shortlisted to be used in the development of a portable respiratory-gated transcutaneous auricular vagus nerve stimulation (taVNS) system. To date, all published studies assessing respiratory-gated taVNS have been performed in controlled laboratory environments. This limitation arises from the reliance on non-portable sensing equipment, which poses significant logistical challenges. Therefore, we recognised a need to develop a portable sensor system for future research, enabling participants to perform respiratory-gated stimulation conveniently from their homes. This study aimed to measure the accuracy of an in-ear and a fingertip-based photoplethysmography (PPG) sensor in measuring cardiac vagal tone relevant heart rate variability (HRV) parameters of root mean square of successive R-R interval differences (RMSSDs) and the high-frequency (HF) component of HRV. Thirty healthy participants wore the prototype sensor equipment and the gold standard electrocardiogram (ECG) equipment to record beat-to-beat intervals simultaneously during 10 min of normal breathing and 10 min of deep slow breathing (DSB). Additionally, a stretch sensor was evaluated to measure its accuracy in detecting exhalation when compared to the gold standard sensor. We used Bland-Altman analysis to establish the agreement between the prototypes and the ECG system. Intraclass correlation coefficients (ICCs) were calculated to establish consistency between the prototypes and the ECG system. For the stretch sensor, the true positive rate (TPR), false positive rate (FPR), and false negative rate (FNR) were calculated. Results indicate that while ICC values were generally good to excellent, only the fingertip-based sensor had an acceptable level of agreement in measuring RMSSDs during both breathing phases. Only the fingertip-based sensor had an acceptable level of agreement during normal breathing in measuring HF-HRV. The study highlights that a high correlation between sensors does not necessarily translate into a high level of agreement. In the case of the stretch sensor, it had an acceptable level of accuracy with a mean TPR of 85% during normal breathing and 95% during DSB. The results show that the fingertip-based sensor and the stretch sensor had acceptable levels of accuracy for use in the development of the respiratory-gated taVNS system.