Enhanced security of random seed DSSS algorithms against seed jamming attacks

Researchers have recently studied random spread spectrum techniques to protect the wireless broadcast communications from reactive jamming attacks in traditional Direct Sequence Spread Spectrum (DSSS) networks. They proposed mechanisms to eliminate the pre-shared key vulnerability by generating different code sequences for each message using random seeds and disclosing the seeds at the end of the message. In this paper, we present a new type of jamming attack called a seed jamming attack for the fixed message size and these seed disclosure schemes are vulnerable to it. The seed jamming attack focuses on jamming any part of the random seeds of the messages. Their sizes are relatively small and their position in messages is known to the public a priori. Thus, the receivers cannot despread any part of the messages due to the failure of regenerating proper code sequences with the corrupted seed. Jamming the seed precludes the use of any possible FEC since the receiver cannot decode any bits in the message. To overcome the seed attack, we propose an advanced random seed DSSS (ARS-DSSS) scheme which strengthens the previous algorithm called DSD-DSSS [4] by using an additional location seed. The new seed avoids the seed jamming attack by using variable message sizes instead of using known fixed message sizes while incurring almost no additional performance overhead. Our security analysis and implementation results demonstrate how to defeat the seed jamming attacks and how to reduce the computation overhead of DSD-DSSS from the cardinality of seed code set C_e to 1.