Odor profiles oftwo kind of dry-cured hams(Jinhua and Xuanwei hams) of different grades and years were analyzed byE-nose detection and quantitative descriptive analysis (QDA).The results of sensory evaluation and e-nose analysis were compared. The principal component analysis(PCA) results of e-nose showed that the odor profiles of all kinds of hams can be distinguished significantly, and the distribution of samples on principle component analysis chart were correlated to the results of sensory evaluation. Soft independent modeling of class analogy (SIMCA) wasapplied to establish the recognizing models of hams either on grade or on origin. Results showed that the models had better performance, and most of the rejection rate reached 100%. While the discriminative modelof specialand senior grade Jinhua ham showed unsatisfactoryresult, because of the similar odor between them.