Wireless jamming is a critical challenge for sprawling and ubiquitous devices, which are connected using radio-frequency (RF) waves. Traditionally direct-sequence spread spectrum (DSSS) and its derivatives have been recognized as an effective jamming resilient technique. However, the effectiveness of DSSS relies on the use of either pre-shared secret code or a large bank of public codes, the generation, distribution, and management of which would be particularly difficult in future large-scale decentralized wireless networks. To tackle this problem, we present a framework which exploits the shared randomness inherent in wireless channels to generate and refresh secret seeds at each communicating node. We highlight why channel randomness cannot be used as it is and develop processing algorithms which ensure that RF based sources are suitable for entropy pooling and random seed generation.