Shore protection in confined water bodies is of major importance in Switzerland where many lakes suffer severe shore erosion. It occurs mainly in shallow zones due to several reasons such as: the fluctuations of the water level , the wind-wave impact during major wind events, and the increasing effect of incident waves due to solid reenforcement by non-adapted protection measures. In order mitigate this increased erosion and enhance the shore stability by sustainable coastal structures, pioneer and soft measures have been built during the past decades. Although such solutions proved their efficiency, they raised a new challenge for engineers: how to build soft and porous media that take into consideration the hydraulic and hydrodynamic conditions of the project site, and enhance its efficiency in sand trapping and shoreline accretion. By means of experimental and numerical modelling, the performance of a typical soft measure, called brushwood fence, was investigated in hydrodynamic conditions similar to those found in reality. A downscaled model of 1 : 10 was built on a fixed bed, and then tested in a wave tank of 10 m length and 6 m width under incoming perpendicular and regular waves. The response of the structure regarding the transmission coefficient has been analyzed for five dimensionless variables: 1) the relative freeboard of the structure Rc/Hi, 2) its relative height h/Hi, 3) the relative wave number kd, 4) the wave steepness Hi /gT2, and 5) the porosity of the structure p. The analysis of their effect on the performance of the protection structure allowed the establishment of an empirical relationship for the transmission coefficient. It takes into account the geometrical characteristics of the structure as well as the local hydrodynamic conditions. The response of the structure has to be treated separately in relation to its immersion condition. When the structure is submerged, every dimensionless variable has a differentiated effect on the transmission coefficient in comparison to the condition when the structure is emerging. The relative freeboard, wave steepness and the porosity are the three key variables in the response of the structure regarding wave damping. The efficiency of detached brushwood fences built on a movable bed was tested under the action of waves with fixed characteristics. The bed material was made of a granular, fine sand with a diameter d50 = 0.18 mm. The evolution of the shoreline and the transmission coefficient of the structure were determined using a first series of tests. It was demonstrated that time has negative impact on wave damping, at least during the period where the bed in the vicinity of the structure evolves to its equilibrium state. A second series of tests established the effect of the distance S of the detached structure (length B) from the shoreline on its evolution and on the deposited and eroded sand volumes at the leeside of the structure. It was proven that the wave motion through the porous media hinders the formation of significant salients and prevents the formation of tombolos when the structure was very close to the shoreline. Thus, for high ratio B/S = 1.48 no tombolo was formed in line with non porous breakwaters. The deposited and eroded volumes were also evaluated in a third series of tests with the presence of a single gap in the brushwood fences. The effect of the width of the gap on these volumes was analyzed. The results proved that the main deposition area is located between the gap and the shoreline where sediments are transported by the diffracted waves at the edges of the brushwood fences. The highest deposition rate at this location corresponds to the ratio B/S = 0.25. Using the solver of the elliptic mild slope equation developed under Mike 21, the wave field in an enclosed area surrounded by a porous structure was numerically investigated. The calibration of some major variables in the numerical model such as wave breaking and bottom friction parameters, based on the experimental results, proved the adequacy of the selected numerical scheme. The effect of a single and double gap in a linear infinite porous media was afterwards evaluated and diffraction diagrams were built for the enclosed wave field. A rule was proposed to use a relative gap width G/Wmax less then 0.12. The wave field for two gaps is different and significantly influenced by the spacing between the gaps. For low spacing values (Es/G = 2), waves in the middle of the protected area are high along the structure and very low close to the shore. For high spacing values (Es/G = 5), the wave field is also significantly deformed. However, values of Es/G comprised between 3 and 4 seem to be most appropriate since the corresponding wave field is less deformed. The two gaps configuration does not significantly increase the residual total energy behind the structure. It is relatively constant with spacing variations between two gaps and increased slowly with the increase of a single gap width. The experimental observations and the numerical results were successfully applied in Mörigen bordering Lake Biel. The wave fields were calculated numerically behind a series of segmented brushwood fences. Several wind-wave regimes and varying water levels were analyzed to optimize the performance of the porous protection structures.