<p>Constructing an effective and scalable anti-counterfeiting system against cyber-physical attacks has become an increasingly crucial problem. While mass production and digitalization across many domains profoundly transformed our society with an abundance of goods and hyper-connectivity, they have also made it harder to track products and associated information, thereby raising authentication risks. In response to such risks, leveraging random physical traces that are generated as byproducts in manufacturing processes has gained great attention as a robust and efficient replacement for existing approaches. Yet, as current methods of generating such inherently unique patterns rely on complex process setups and specific materials, they face significant limitations to be applied and scaled up as a product authentication system in various manufacturing industries. To address this issue, we propose a Physically Unclonable Identifier (PUID) that exploits the inherent randomness and process characteristics of Pulsed Cold Spray (PCS) to fabricate a unique physical identifier. In addition, we develop a complementary framework that leverages the resulting spectral features for reliable product identification. Specifically, our framework utilizes Implicit Neural Representations (INRs) and Fast Fourier Transform (FFT)-based cross-correlation as key strategies to create, register, manage, and authenticate PUIDs. Through experiments, we validate the robustness, applicability, and scalability of our approach by rapidly generating 54 PUIDs on an arbitrary substrate and successfully authenticating each one. These results highlight the significant potential of PUIDs for deployment across various industries.</p>