The objective of this research is to establish phase as a reliable and informative representation for wearable cardiovascular signals. These signals are quasiperiodic, with recurring cycles whose duration and waveform morphology vary with physiology, activity, and measurement conditions. Phase provides a natural representation of progression through these changing cycles, enabling the timing of recurring physiological events and the structure of the waveform between them to be studied within a common framework. Two complementary notions of phase are particularly relevant to cardiac signals. Cardiac phase describes progression between recurring physiological landmarks, such as successive R-peaks in an ECG, and therefore captures the timing of the cardiac cycle. Morphological phase, in contrast, is derived directly from the waveform through its analytic representation and describes progression through the signal's oscillatory morphology. Although both characterize cyclic progression, they encode different aspects of cardiovascular dynamics. Our completed research develops methods that make these two representations more reliable and useful for physiological signal analysis. HIKAF leverages cardiac cycle timing and HR information to track mechanical events in the SCG despite variations in physiology and signal quality. The CGFM method represents physiological waveforms in terms of morphological phase and amplitude to learn transformations between signal modalities while preserving their cyclic structure. The StHT improves morphological phase estimation by identifying and locally correcting instabilities that arise in low-amplitude regions. Collectively, these studies establish tools for robust event tracking, waveform transformation, and phase estimation, providing the methodological foundation for investigating how different notions of phase interact. The proposed research will examine how cardiac and morphological phase relate within and across cardiac cycles. We hypothesize that differences between them capture physiological information that neither represents alone. We will first characterize whether these relationships are consistent across individuals, sessions, sensing modalities, and physiological conditions, and separate physiological differences from variability due to waveform morphology, signal quality, and phase estimation. We will then evaluate whether the relationships reflect physiological state and support downstream monitoring tasks, beginning with controlled recordings and extending to additional datasets and free-living recordings where appropriate.