Interference suppressions of spread spectrum systems with narrow-band interference

In this project, the worst-case bit-error rate (BER) performance for a non-coherent fast frequency hopping binary frequency-shift keying (FFH/BFSK) system is studied for different diversity-combining receivers. The diversity-combining receivers used are linear-combining (LC) receiver, product-combining (PC) receiver, self-normalizing (SN) receiver and clipper receiver. The FFH/BFSK system is modeled under the conditions of single-tone per band multi-tone jamming (MTJ) and additive white Gaussian noise (AWGN). The BER performance of FFH/BFSK system is studied under Rayleigh, Rician and Nakagami fading channel conditions. The simulation results generated for Rayleigh and Rician fading channel conditions are validated against the analytical and simulation results obtained from existing literature. The above simulation results are then extended to Nakagami fading channel conditions. The number of jamming tones, Q, which results in a worst-case BER performance, is obtained by means of numerical search. Regardless of fading conditions, the BER performance is degraded as the signal-to-noise ratio (SNR) reduces. The BER performance is observed against signal-to-jamming ratio (SJR), with SNR fixed at 13.35 dB and diversity levels at L = 1, 3 and 5 respectively. When effects of fading are not taken into consideration, the LC receiver at L = 1 gives the best BER performance among the three diversity levels. The relative clipping level c of the clipper receiver is chosen to be 1.0. The PC, SN and clipper receiver at L = 3 gives the best BER performance among the three diversity levels. At higher diversity levels, the BER performance becomes poorer when the jamming power is very small, i.e., SJR > 36 dB, due to non-coherent combining loss. Generally, when comparing the four diversity-combining receivers, the clipper receiver gives the best BER performance. However, the clipper receiver requires the information of the desired signal level whereas the other three receivers are completely nonparametric. In case the desired signal level is not known, the SN receiver gives the best BER performance followed by PC and LC receivers. Lastly, the BER performance for diversity-combining FFH/BFSK system is studied over three fading channels. The fading channels are modeled by Rayleigh, Rician and Nakagami fading models with single-tone per band MTJ and AWGN. The simulated results show that the effects of fading on MTJ are insignificant for any diversity-combining receivers. In contrast, the BER performance is degraded as the desired signal amplitude experience more severe fading. The simulation results show that higher diversity levels are more effective than lower diversity levels in combating MTJ and AWGN over fading channels. Generally, when comparing the four receivers, the clipper receiver gives the best BER performance. However, if the desired signal power level is unknown, the LC receiver provides good BER performance when the signal experience Rayleigh fading. When the signal experiences Nakagami or Rician fading, the SN receiver provides the best BER performance under moderate jamming conditions. Under weak jamming conditions, the LC receiver gives the best BER performance. On the other hand, the PC receiver gives the best BER performance for strong jamming.Bachelor of Engineerin