In the modern era of signal processing, digital filters play an important role in real-time applications such as communication, consumer electronics, digital signal processing, audio, etc. In digital filter design, Finite Impulse Response (FIR) filters are highly preferable due to their linear phase and inherent stability. These filters benefit from being time-invariant and simple to implement with minimal computational requirements. Therefore, the hardware security of FIR filters is essential for good performance and reliable results. On the other hand, there is the possibility of hardware threats, such as tampering, reverse engineering, hardware Trojans, etc., as the design of an FIR filter involves many stages. Such hardware attacks on FIR filters can cause several problems, including performance degradation, leakage of confidential information, lack of stability, etc. This study presents the design and implementation of a Trojan-aware FIR filter using Physical Unclonable Functions (PUFs). The key feature of PUFs is that they generate a unique and unpredictable response for each given challenge. In the proposed design, PUFs were used to generate the FIR filter coefficients that are unique and unpredictable by attackers/trojans to improve security. The security of FIR with PUF was tested using ML-based challenges, and the results showed approximately 30% more reliability and consistency compared to the FIR without PUFs.