Employing Preprocessor Filters to Suppress the Effect of Impulsive Noise on Communications Systems at Millimeter-Wave Frequencies

With the explosive growth of mobile traffic demand, the contradiction between capacity requirements and spectrum shortage becomes increasingly prominent. The bottleneck of wireless bandwidth becomes a key problem of the Fifth Generation (5G) wireless networks that could offer high level of reliability, low levels of latency and constant connectivity anytime, anywhere. As one of its most promising potential technologies, Millimeter wave (mmWave) communication endorses 5G goals through the utilization of 30 gigahertz (GHz)-300 GHz frequency band. The high frequencies of mmWave technology allocate more bandwidth to provide multi-gigabit communication services and reinforce their applicability in cellular communications and radar alongside employing high gain steerable antennas. However, using such frequencies and multi antennas may offer signal attenuation, blockage through obstacles, besides a noise behavior. Noise over mmWave system can be any sort of undesired fluctuation, nevertheless, the catastrophic presence of Impulsive Noise (IN) at high frequencies band whether it is a man-made noise, electromagnetic interference, non-Gaussian clutter, or nonstationary chaos, will shed the light on how much it affects the mmWave system. In this thesis, we examine IN impact on mmWave system and suppressing its effect through implementing clipping, blanking and hybrid preprocessors. Hybrid gains the lowest bit error rate over clipping and blanking in variant impulsive environments and at different Signal to Noise Ratio (SNR) values. At several impulsive cases, the optimal threshold values were obtained for all filters by proposing a new thresholding mechanism to eliminate IN components and maximize SNR performance. Blanking has the potential to become the preferred filter in achieving better spectral efficiency than clipping for mmWave system at different SNR and probability of IN occurrence values. The thesis shows that the use of clipping, blanking and hybrid filters with the optimal threshold values reduces the unfavorable effect of IN and improves system performance significantly. Keywords: fifth generation, millimeter wave, impulsive noise, clipping, blanking, hybrid.