The physical properties of voice propagation play a crucial role in enhancing the security of voice processing systems, particularly in driverless cars. By incorporating these properties, systems can accurately identify legitimate voice commands. For instance, the direction of voice propagation can help distinguish the driver's voice from passengers', even if passengers attempt to mimic the driver's voice. This is achieved through the use of multiple microphones to detect the time difference of arrival (TDoA) of the voice signal, which varies based on the speaker's position relative to the microphones. Additionally, the system can analyze the signal's frequency and time-domain features to detect any anomalies indicative of malicious voice commands. By combining these physical and signal characteristics, driverless cars can effectively authenticate voice commands, ensuring the system responds only to authorized users, thereby enhancing overall security.