Under conditions of a priori noise uncertainty, it is effective to use adaptive noise-abatement techniques to improve noise protection of continuous automatic cab signaling (CACS) receivers. The purpose of this article is to study the possibilities of an adaptive signal-processing technique, adaptive noise filtration (AF), providing better noise resistance of the CACS signals. To achieve this goal, the following problems are solved: a mathematical model of the CACS receivers' input signals is developed, a functional circuit and adaptation algorithm of a basic and a modified multichannel CACS receivers' adaptive filters is substantiated, and a computer simulation of adaptive filters is performed, confirming the accuracy of the obtained results and effectiveness of such filters for noise reduction. A distinctive feature of the proposed filters, which significantly expands their functionality, is a priori information on the presence of breaks in the CACS signals used in the adaptive algorithms.