In this paper, the general framework for composite fatigue life prediction is proposed that can incorporate the detected stiffness from structural health monitoring system. First, an in-situ composite fatigue testing with pre-installed piezoelectric sensors is designed and performed to collected sensor signal and overall stiffness measurement. Signal processing techniques are implemented to extract damage identification features. Following this, power law is introduced to express the stiffness degradation mechanism and a general prognosis framework is proposed. Next, the detected overall stiffness degradation is integrated with a Bayesian inference framework for remaining useful life (RUL) prediction. The prognosis performance is validated using prognostics metric. Finally, some conclusions based this work are presented and directions for future work are drawn.