Abstract : This paper introduces the Hilbert transform, a commonly used signal processing technique, to construct the Hilbert modulus. Through spectral analysis, it aims to monitor and diagnose rotor faults in squirrel-cage induction motors. This method requires only single-phase sampling, significantly reducing hardware investment. Numerical simulations using Matlab effectively verify the feasibility of this theory in rotor fault diagnosis. Finally, analysis of experimental data from dynamic model experiments shows that this method can sensitively capture fault characteristic components, proving its practicality. Keywords : Rotor bar breakage; Hilbert modulus spectral analysis; Fault diagnosis 1. Introduction The main rotor fault in squirrel-cage induction motors is rotor bar breakage. Among various monitoring methods, the method based on line current spectral analysis has two major advantages: it can be made non-intrusive and is easy to integrate with protection and speed control systems, thus gaining the most widespread research and application. However, this method struggles to highlight the characteristics of rotor bar breakage faults, necessitating the exploration of new methods for various fault diagnoses. Since most rotor fault diagnosis methods require simultaneous sampling of three-phase currents, we aim to achieve the same effect by sampling only a single phase (or line) current. Therefore, the Hilbert transform, a commonly used signal processing technique, is introduced to construct the Hilbert modulus. Spectral analysis of this modulus is then performed to monitor and diagnose rotor faults in squirrel-cage induction motors. This paper employs Hilbert spectral analysis to diagnose rotor bar breakage faults. This method requires only sampling and processing of a single line (or phase) current signal, significantly reducing hardware investment and making software implementation easier. It is suitable for induction motors prone to frequent rotor faults. [Full text of the research on rotor fault diagnosis of induction motors based on spectral analysis available for download.]