Abstract: The Distance Vector Hop (DV-Hop) localization algorithm, as a simple and efficient localization algorithm, may have significant localization errors in wireless sensor networks with uneven node distribution. In order to improve the localization accuracy of the algorithm, an improved sparrow search optimized DV-Hop localization algorithm is proposed. Firstly, a chaotic sequence is generated through Cat chaotic mapping to initialize the sparrow population, improving the algorithm's early search ability. Secondly, based on the hop count and hop distance data obtained by the traditional DV-Hop algorithm, combined with the convex programming algorithm, a search box area for unknown nodes can be established, effectively reducing the early search range of the sparrow search algorithm. Finally, the Cauchy-Gaussian mutation strategy is introduced to optimize the optimal sparrow individual and eliminate the poorer individuals, enhancing the algorithm's ability to jump out of local optima. The simulation experiment takes into account the radio irregularity and introduces the radio irregularity model instead of the ideal model. The average positioning error of unknown nodes is selected as the experimental evaluation index. The simulation results show that under the same experimental environment, compared with the conventional DV-Hop and the others, the mean location error decreased by approximately 18.60%, 10.21%, 8.32% and 4.77%, respectively.