This paper describes a new signal processing technique for understanding the dynamics of time-varying signals in vehicles: Hyperstate analysis. Vehicle noise and vibration are examples of randomly-varying transient or non-stationary signals that are not effectively analyzed with classical spectral analysis techniques. By the use of nested Hidden Markov Models, Hyperstate analysis explicitly identifies transient and nonstationary behavior on many time scales for better signal discrimination. It uses a probability-based framework that allows for automated, objective classification of noisy signals. This paper describes the general Hyperstate modeling and filtering techniques. Results illustrating the application of Hyperstate analysis for multi-scale characterization and classification of data that represent the engine starting sequence from different types of vehicles are presented. This preliminary work demonstrates that Hyperstate analysis discerns similarities and differences in randomly-varying signals of this type, and can perform effective automatic, objective classification and signal decomposition for NVH (noise, vibration, and harshness) studies.