Blockage detection is a critical functionality for the air interface in modern 5G and future 6G systems operating in millimeter wave (mmWave, 30–300 GHz) and terahertz (0.3–3 THz) frequency bands. In operational systems, blockage has to be detected prior to its occurrence to allow for time to take some actions to avoid the loss of connectivity, e.g., switching over to the back-up link. However, up to date, most of the proposed approaches are reactive detecting blockage only when it already started. In this paper, by utilizing the special signal oscillations occurring just prior to the blockage, we propose a new method for proactive blockage detection. The proposed approach is based on a periodogram of the received signal that can be estimated efficiently using modern signal processing techniques. We then proceed comparing the proposed approach to reactive and proactive methods reported to date using the blockage detection probability as the metric of interest. Our results illustrate that the proposed approach allows to detect blockage with probability one, at least few tens of milliseconds prior to the actual blockage time instant. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.