In this paper, we present a platform based on a modular approach coupled with powerful algorithms for accurate detection and identification of chemical compounds. The system relies on multi-SAW (Surface Acoustic Wave) sensors that are functionalized differently, resulting in multi-responses that collectively constitute a fingerprint of the chemical compound. A prototype has been developed and the overall system, including the design of SAW module, the acquisition system, learning algorithms and online recognition of various compounds, has been tested and validated. The results showed a reliable and accurate system with a perfect score of 100\% recognition of DMMP.