Most of Flying Adhoc Networks (FANETs) applications consume GPS based location information for their services, which are also shared in real-time with other UAVs, ground control stations and centralized service operators. GPS spoofing is among the most popular attacks in FANETs that lead the Global Navigation Satellite System (GNSS) receivers to generate false navigation solutions. Several anti-spoofing techniques have been proposed in the literature. However, Conventional detection methods present vulnerabilities against Colluding GPS Spoofing Attack where multiple GPS spoofing signal sources are used. In this paper, we propose a policy-based distributed detection mechanism to face colluding GPS-Spoofing attack in FANETs. Based on Burglary scene and the location of each UAV at the time of the attack, we can distinguish between Active and Passive witnesses that are respectively used to help the target to detect and confirm the presence of GPS spoofing signal using respectively Absolute power and Carrier-to-Noise density ratio. A trust model based on Beta and Weibull Distribution has been adopted as a combination technique of different testimonies to both mitigate the spread of false rumors and classify the different signals. Simulation results depict that our proposal is able to detect and revoke the GPS Spoofing signal with high accuracy reaching the 99% and low communication overhead.