Abstract: In missile defense systems, rapid warhead identification and threat assessment are critical. Existing methods often consider the missile bus as clutter and rely on the warhead′s micro-motion signatures for recognition and early warning. However, factors such as range measurement accuracy, warhead size, and motion patterns limit the precision of micro-motion curve extraction. Furthermore, feature parameter estimation typically depends on prior information, constraining identification and assessment performance. To address these issues, an approach is proposed that transforms the bus from a nuisance into a resource by leveraging its feature parameters to assist warhead identification and support threat capability assessment. Specifically, a multi-dimensional feature parameter estimation method for tumbling cylindrical targets based on initial-value-aided phase ranging is developed. The main contributions are as follows: 1.A initial-value-aided phase ranging technique is proposed to mitigate off-grid issues and improve the precision and robustness of phase measurements by finely extracting first-value micro-ranges and refining overlapping regions through gradient-based reassociation.2.A two-step sum-difference estimation algorithm is introduced to eliminate interference terms and enable accurate multidimensional feature parameter estimation without relying on prior information, by exploiting the sum and difference properties of micro-motion curves and their derivatives. Extensive experiments validate that the proposed method achieves accurate and robust multidimensional feature parameter estimation for tumbling cylindrical targets, offering strong technical support for true warhead recognition and threat assessment.