Simulation and analysis of space-time correlated spiky sea clutter with KK distribution

High-fidelity simulation of space-time correlated sea clutter with high-intensity spikes has always been a challenge. Currently popular statistical models, such as the K and Pareto distributions, offer simple analytical forms and are commonly applied in simulating sea clutter using memoryless nonlinear transforms (MNLT) based on inverse complementary cumulative distribution functions (CCDFs). However, these models fall short in representing the highintensity spikes and space-time correlated texture of real-world sea clutter. More sophisticated models, such as the KK and KA distributions, better characterize these features. However, the absence of closed-form expressions for their CCDF inverses hinders simulation. To address this issue, we propose a linear interpolation approach to efficiently approximate the CCDF inverse. This study rigorously evaluates the relative computational error of the proposed approach and provides practical recommendations for selecting the appropriate sampling grid size. Simulations are conducted based on parameters estimated from the Council for Scientific and Industrial Research (CSIR) Fynmeet radar data. It is shown that the simulated sea clutter closely matches the KK distribution and accurately captures both the spike and texture characteristics, validating the efficacy of the approach. Moreover, the proposed approach is extendable to other complex distribution models, such as the KA and 3MD distributions, facilitating the calculation of CCDF inverses for more intricate statistical distributions.