Millimeter wave (mmWave) has been identified to be a promising technology in the beyond 5G wireless communications. Specifically, from the perspective of wireless localization, due to the short wavelength of mmWave signals, the arrays of antennas employed in mmWave communication systems can resolve the angle of arrival (AoA) and angle of departure (AoD) of signals with high precision. This fine angular resolution can aid in pinpointing the location of devices or objects. Furthermore, mmWave systems typically use beamforming to focus the transmission energy in specific directions, which not only increases the signal strength in particular directions but also helps in determining the directions of signal sources with higher accuracy. While mmWave signals face challenges in terms of penetration through obstacles, they are highly reflective. By analysing these reflections, it is possible to extract the information about the environment and enhance localization accuracy through channel estimation. However, localization in mmWave systems also face the challenges, such as, signal blockages and the requirement for more complex and high-power hardware. The line-of-sight (LoS) nature of mmWave can be both an advantage (in terms of resolution) and a disadvantage (in terms of obstructions). In this thesis, we first provide a literature review in terms of the research background for the mmWave systems. Then, to reduce the power consumption at receiver for localization, on the one hand, the compressed sensing aided sampled channel estimation based localization scheme in mmWave systems is studied. On the other hand, the few-bit ADCs and beamspace quantization error mitigation are investigated. Furthermore, to overcome the LoS blockage problem, reconfigurable intelligent surface (RIS) is employed to create virtual LoS (VLoS) path, when the LoS link state is unknown or totally blocked.