Impulse-Radio Ultra-Wideband (IR-UWB) technology and time-based ranging methods have gained wide acceptance in indoor localization applications. This technology and these approaches offer the capability to determine the distance and orientation between wireless devices with high accuracy using radio exchanges. These signals propagate between the devices through the air and through the UWB device design specific analog circuit. These RF paths have a direct effect on the Time of Flight (ToF) measurements by introducing delays, which lead to erroneous estimates. To ensure good localization accuracy, these delays must be known and their effect mitigated. In this paper, we present an improvement of the actual antenna delay calibration method initially proposed by DecaWave (now Qorvo). In the proposed algorithm, we apply some well-known principles of Genetic Algorithm (GA) to improve performance. An experimental validation was conducted in an anechoic and office environments to demonstrate the performance of our algorithm. The results and a comparison with the original method are presented.