The Extreme Mass-Ratio Inspiral (EMRI) refers to binary system with a mass ratio between 10⁴ and 10⁶, where the smaller object loses energy as it inspirals closer to a massive black hole, emitting gravitational waves. It is estimated that there are 10⁵ cycles during the last year before plunge, providing rich information on the evolution of gravitational wave phases. The motion of the smaller object in the strong gravitational field of the massive black hole can reflect the surrounding spacetime structure. The massive black hole is typically located at the center of a galaxy, in which the galaxy environment leaves traces in the EMRI waveform, thus EMRI can be used to constrain the gravitational theories, no-hair theorem and so on. Multiple sources can provide constraints on the mass and spin distribution of massive black holes, contributing to the understanding of cosmic and galactic evolution. Due to this significance, EMRI has become an important target for space gravitational wave missions such as LISA, Taiji, and TianQin. Consequently, the EMRI data analysis has become a crucial task. However, due to the high dimensionality and complexity of the waveform, relevant methods are still under discussion. This article initially reviews the framework and discussions regarding EMRI data analysis conducted during the Mock LISA Data Challenge (MLDC). It subsequently provides a comprehensive overview of the challenges encountered and the corresponding improvements proposed based on the authors’ research. Finally, the article offers some clues and suggestions regarding potential advancements in EMRI data analysis methods. 

极端质量比旋近EMRI指质量比在10⁴~10⁷之间的双星系统, 其中小质量天体被捕获后绕大质量黑洞旋近, 期间损失的能量以引力波的形式向外辐射。旋近结束前的最后一年估计能产生10⁵个旋近周期, 因此可提供丰富的引力波相位演化信息。小质量天体处在大质量黑洞的强引力环境中, 运动轨迹能反映大质量黑洞周围的时空结构, 其波形可用来限制引力理论和无毛定理等。大质量黑洞一般处在星系中心, 星系所处的天文学环境会在波形中留下痕迹。多个波源能给出大黑洞的质量和自旋分布, 用于限制宇宙演化和星系演化等。基于上述科学意义, EMRI成为空间引力波计划LISA、太极和天琴的重要观测目标, 因此对EMRI的数据分析成为一个重要任务。EMRI波形的高维度和复杂性对数据分析方法要求较高, 目前还没有公认的满意答案。首先回顾LISA模拟数据挑战赛中提出的方法和最近的讨论, 然后总结EMRI数据分析的难点, 并提出改进方法, 最后讨论未来研究发展中可能用到的几个线索。