In recent years, the global demand for high-rate, high-reliability, and
wide-coverage wireless communications has continued to grow. Traditional
satellite communications are faced with bottlenecks such as spectrum scarcity,
insufficient single-satellite capacity, and poor adaptability to highly dynamic
channels. Relying on the advantages of wide-area coverage and multi-satellite
cooperation, low Earth orbit (LEO) satellite constellations can construct
distributed satellite MIMO systems and form virtual large-scale antenna arrays,
significantly improving system capacity and reliability. However, such systems
are characterized by large spatiotemporal scales, and their channels exhibit
strong heterogeneity in dimensions such as delay, frequency, and polarization,
which are difficult to accurately evaluate using existing methods. To this end,
this paper proposes a large-spatiotemporal-scale channel characteristic analysis
method for distributed LEO satellite MIMO systems. A satellite-terrestrial
full-link channel model is established, and quantitative characterization of
multi-dimensional heterogeneity is realized based on statistical boundary
analysis. Combined with the H3 global grid system, large-scale simulations are
carried out to statistically analyze the distribution laws of delay differences,
frequency offsets, and polarization angle deviations under different user
configurations. The unique value of this analysis method lies in that its output
results can be used to determine the upper limit of buffer capacity required for
the receiver to complete coherent combining of multi-channel signals and the
frequency offset correction range required for signal acquisition, provide a
quantitative basis for optimizing satellite set selection, and offer a solid
theoretical model and data support for performance evaluation and parameter
design of multi-satellite cooperative systems.