Abstract: Neutron kinetic parameters are crucial for evaluating the dynamic characteristics of nuclear reactors, ensuring safe operation, and optimizing physical startup processes. Accurate measurement of these parameters remains a critical task in reactor physics experiments. As a key direction in advanced nuclear energy systems, lead-bismuth fast reactor (LFR) requires comprehensive studies on their kinetic properties. The effective fraction of delayed neutron (β_eff) and neutron generation time (Λ) have been measured by Feynman-α method for the first time at the lead-bismuth zero power of uranium rod gate critical assembly at the China Institute of Atomic Energy, in order to provide basic data for the engineering design of lead-bismuth fast reactors. The experimental methodology integrated neutron activation, full-core transport simulation, and noise analysis techniques. Initially, the detection efficiency of BF₃ proportional counters was calibrated via a gold foil activation method. Absolute neutron flux density was determined by irradiating gold foils at high-power critical conditions and measuring their activity using high-purity germanium (HPGe) detector. MCNP code was employed to establish the spatial distribution of neutron flux and fission rates, enabling the derivation of total core fission rate (F) from localized neutron flux measurements. A reactor noise measurement system, comprising ³He detectors, national instruments (NI) data acquisition cards, and pulse signal processing modules, was developed to collect neutron count time-series under subcritical conditions. Four subcritical configurations with effective multiplication factor (k_eff) in the vicinity of 0.98 to 0.99 were established by adjusting fuel loading. For each configuration, Feynman-α method was applied to the variance-to-mean ratio of neutron counts to extract the prompt neutron decay constant (α), β_eff, and Λ. The critical prompt neutron decay constant (α_c=(141.22±19.05) s⁻¹) was deduced from the relationship between the measured α value and the detector count rate. Combining experimental α and F with theoretical models, β_eff and Λ were calculated. The results show that β_eff=0.006 203±0.000 276 and Λ=(43.926±2.325) μs, demonstrating relative deviations of −14.14% and −13.68%, respectively, from MCNP-predicted values (0.007 225 for β_eff and 50.89 μs for Λ). These discrepancies highlight uncertainties of theoretical calculations but confirm the feasibility of the Feynman-α method for LFR kinetic parameter measurement. Key challenges include the indirect determination of total fission rates in subcritical states and statistical fluctuations in neutron pulse timing. Increasing the detector count rate and refined detector calibration are proposed to reduce uncertainties. The study successfully establishes a technical framework for neutron noise analysis in LFR, providing critical experimental data to support the design and safety evaluation of generation Ⅳ nuclear systems. Future work will focus on enhancing measurement precision and extending the method to other advanced reactor configurations.