Superhydrophobic surfaces are proven to be capable of reducing the skin friction in laminar and turbulent flows. These surfaces consist of micro/nano-scale hydrophobic roughness features which make the surface render a non-wetting property due to the entrainment of air pockets between the solid surface and the liquid phase. This shear free air-water interface reduces the frictional drag force. This flow control method has two distinct effects in turbulent flow: drag reduction due to effective slip velocity and drag reduction associated with the modification of the turbulent flow structures (Rastegari & Akhavan, 2015). In the current research, the turbulent structure of the inner layer of a turbulent channel flow over a non-wetted superhydrophobic (SHO) surface with random pattern is experimentally studied. The results are compared with the wetted counterpart and also a smooth reference surface. Two planar particle image velocimetry (PIV) measurements are carried out in the streamwise/spanwise and streamwise/wall-normal planes. The vector fields are obtained from both ensemble averaging and individual cross-correlations of double-frame PIV images. The results showed a small increase (~5%) of the mean velocity profile at y+=10 over the non-wetted surface in comparison with the wetted and the smooth surfaces. Up to 15% reduction of normal and shear Reynolds stresses is observed in the inner layer over the non-wetted SHO surface. The wetted SHO counterpart demonstrates no effect on the mean velocity and Reynolds stresses in comparison with the smooth surface implying that the surface is hydrodynamically smooth. A noticeable suppression of the sweep and ejection events, increase of the spanwise spacing of the low and high speed streaks, and attenuation of vortical structures are observed over the non-wetted SHO. These indicate attenuation of the turbulence regeneration cycle due to the slip boundary condition over the non- iii wetted SHO surfaces with random texture. Tomographic PIV (tomo-PIV) and 3D particle tracking velocimetry (3D-PTV) as three-dimensional flow measurement techniques can unravel the relevant physics by revealing the flow modifications across the third dimension. The performance of these measurements is evaluated through comparison with DNS data in the literature. The results show that 3D-PTV is more accurate compared to tomo-PIV especially in near-wall region where noise increases for all PIV measurements.