We address the secure pairing of mobile devices based on accelerometer data under various transportation environments, e.g., train, tram, car, bike, walking, etc. As users commonly commute by several transportation modes, extracting session keys from various scenarios to secure the private network of user’s devices or even the public network formed by devices belonging to distinct users that share the same location is crucial. The main goal of our work is to establish the amount of entropy that can be collected from these environments in order to determine concrete security bounds for each environment. We test several signal processing techniques on the extracted data, e.g., low-pass and high-pass filters, then apply sigma-delta modulation in order to expand the size of the feature vectors and increase both the pairing success rate and security level. Further, we bootstrap secure session keys by the use of existing cryptographic building blocks EKE (Encrypted Key Exchange) and SPEKE (Simple Password Exponential Key Exchange). We implement our proof-of-concept application on Android smart-phones and take benefit from numerical processing environments for the off-line analysis of the collected datasets.