In UWOC (Underwater Wireless Optical Communication) systems, scattering and absorption occur due to water molecules and suspended particles, resulting in weak signals at the receiver end. In this thesis, a LDPC (Low-Density Parity-Check) code, which is a kind of error-correcting code, was employed in order to compensate for performance loss, and its performance was improved only when the input values of the decoder were soft decision types. However, no algorithm has yet been reported that applies a soft decision technique for the M-ary PPM (Pulse Position Modulation) and QAM (Quadrature Amplitude Modulation) schemes in the case of UWOC. Therefore, we developed a SVG (Soft Value Generator) algorithm in order to use a turbo equalizer, which improves the performance over all of the iterations in the case of the M-ary PPM and QAM schemes. Through simulations, it was confirmed that the proposed method performs better than the conventional hard decision algorithm. Also it evaluated that the performance of the proposed method through four water tank experiments, in which M-ary PPM and QAM data were employed to perform experiments by varying the turbidity and transmission rates in a water tank. In addition, the experiment was conducted to compare performance with changes in water temperature, and to compare green laser with red laser. And an experiment was done to compare when the water was mixed with when it was unmixed. This again showed that the performance of the proposed algorithm is superior to that of the conventional algorithm.