Lithium-ion batteries have become one of the most critical energy storage systems due to their long cycle life and high energy density. Ultrasonic testing technology has been applied to the battery state estimation for the assurance of durability, safety and performance, however, the accuracy and reliability still need to be improved. In this paper, a new signal feature of the ultrasonic signal is proposed for the state of charge (SOC) estimation of batteries. The signal-shear effect is observed during charge/discharge cycles and is explained by the ultrasonic resonance phenomenon, based on which a SOC estimation method is proposed by using damped sine waves as excitations. This signal feature is characterized by the initial rise time (IRT), which is calculated by an algorithm based on the temporal moment and the adaptive moving root mean square. The optimal frequency and damping ratio parameters of the damped sine waves are found by a parameter-sweep study. Based on the experiment results, the proposed IRT feature is highly correlated to the SOC status with satisfying accuracy, monotonicity and linearity. The IRT feature has the additional advantage to be calculated by only a single data acquisition channel, providing a more convenient mean of battery status estimation.