Non-stationary jammer excision in spread spectrum communications via discrete evolutionary and Hough transforms

• Authors: Luis F. Chaparro;Raungrong Suleesathira
• Venue: Signal Processing
• Year: 2003

Despite the advantages of direct sequence spread spectrum (DSSS) communications, intentional jamming or interference from other users reduces the robustness of the system to interference. Spreading the message by means of a pseudonoise sequence, DSSS increases the bandwidth of the transmitted message making it thus difficult to track the message. Given the ease in tracking jammers in the time-frequency (TF) domain, TF-based excisers applied before despreading enhance the robustness to interferences of DSSS. In this paper, we propose an excision technique for multi-component chirp jammers with constant or time-varying amplitudes. Our procedure initially estimates the instantaneous frequency (IF) of each of the jammer components using a combination of the discrete evolutionary and the Hough transforms. These local linear IF estimates are efficiently obtained, and improved by recursively matching the instantaneous phase of each of the components. Using the estimated IFs, the received message is projected onto a semi-stationary domain where the desired jammer component appears as a low-frequency signal Synthesis of the jammer components is performed by first processing the signal with well-understood procedures such as linear filtering or singular value decomposition, and then chirping the resulting signal. Subtracting the synthesized jammer from the received signal permits the correlator to eliminate the remaining jamming and channel noise interferences. Our procedure applies to multi-component chirp jammers with constant or time-varying amplitudes and instantaneous frequencies not parametrically modeled. A general statistical analysis of the process is developed based on the signal to interference and noise ratio and the probability of bit error. Performance of the proposed excisers is quantified by means of simulations of different cases, and the results used to verify the validity of our statistical analysis and to compare our results with those from existing projection methods.