With the development of the Internet of Things (IoT), smart home systems have become an essential component of daily life, greatly enhancing its convenience and comfort. However, issues related to data security and effective identity authentication mechanisms have also emerged. To date, a plethora of research has been conducted in this area. After comprehensive research and comparison, the most refined authentication mechanism currently is the dynamic trusted lightweight authentication mechanism (DTL) based on blockchain. By analyzing the DTL mechanism and using model checking theory and the principles of timed automata, the sequence diagrams of the mechanism were extracted, and variables were constructed. Using UPPAAL, a timed automaton model for the DTL mechanism was established. UPPAAL's verifier was utilized to validate its security properties, such as deadlock-free, confidentiality, integrity, access control, and freshness. It was discovered that there is a deficiency in the consideration of time, specifically in terms of freshness. By analyzing the counterexamples from the sequence diagrams and property verification, corresponding improvements were suggested. The model was subsequently refined and rebuilt to meet the freshness requirements, providing direction for the further enhancement and development of the DTL mechanism.