Efficient Path Reconstruction in Prehistoric Human Migration: An Adaptive Dijkstra's Algorithm Based on Wavelet Compression for Topographic Data

The reconstruction of prehistoric migration routes requires the accurate evaluation of effective distances across complex topographies.

Traditional Least-Cost Path Analyses (LCPA) using high-resolution elevation models often lead to dense computational grids, making algorithms like Dijkstra's prohibitively slow for large-scale archaeological modelling.

In this paper, we present a novel approach using adaptive wavelet methods to dynamically compress topographic data.

By retaining high resolution only in areas of high topographic complexity (e.g., mountain passes) and smoothing homogeneous regions, we significantly accelerate the Dijkstra algorithm without losing essential routing topology.

We demonstrate the efficacy of this method using migration scenarios.