With the widespread application of wireless sensor networks in complex environments, the traditional DV-Hop localization algorithm exhibits low accuracy under conditions such as uneven node density and obstacle interference. This study proposes an improved DV-Hop algorithm (IOWDV-Hop) to address this issue. The IOWDV-Hop integrates an optimal weighting function with an anchor node selection strategy. This method improves the accuracy of node localization by introducing error-minimizing weights and a constraint-based mechanism for selecting anchor nodes. Experiments were conducted using different irregular network structures for comparative testing. The results showed that the proposed algorithm performed best in a “Tˮ-shaped network. It achieved an average positioning error of only 19%, which was nearly a 60% reduction compared to the traditional DV-Hop algorithm. The study demonstrates that this method possesses stronger robustness and practicality. It provides both theoretical support and engineering value for high-precision positioning in wireless sensor networks.