Electrocardiogram (ECG) signal represents a promising and emerging technology for person identification. In this work, several similarity measures with the K-nearest neighbours (KNN) classifier in biometrics will be applied and tested. The tested measures include Shape Exchange Algorithm (SEA), Complexity Invariant Distance, Dynamic Time Warping (DTW) and its variations CIDDTW (Complexity Invariant Distance Dynamic Time Warping) and QPDTW (Quasi-Periodic Time Warping). The biometric techniques primarily rely on using these similarity measures to compare ECG signals without the need to segment them into cycles or extract their features. The proposed biometric algorithms using an international database of ECG signals called MIT/BIH are validated. The obtained results show that the SEA method provides the best results, followed by the hybrid methods QP-DTW and the CID-DTW, which yield better results compared to the DTW and CID methods.