Electrical Network Frequency (ENF) analysis has emerged as a powerful tool for forensic investigations, providing timestamp and geolocation information of audio or video recordings based on the unique frequency pattern of the power grid. However, the forensic usability of ENF signals extracted from recordings can be compromised by noise, leading to inaccurate results. This paper proposes a novel approach to searching with masking for ENF signals. This algorithm acknowledges the potential inaccuracies in ENF extraction without sacrificing signal length. The proposed method presents significant implications for forensic investigation. It can effectively search for ENF patterns in noisy signals, increasing both the accuracy and the reliability of the ENF analysis by providing a valuable tool for law enforcement agencies and forensic experts. To support the validity of the method, we test it on different public datasets and demonstrate a noteworthy improvement in performance.