Among the group of sensors the well-studied biosensors have a particular importance. Biosensors are sensors that contain biologically active ingredients and with those are able to measure selective and reversible concentrations in chemical and biological samples. Unlike sensors developed for diagnostic use, it is vitally important that biosensors are built in an as small as possible, portable device that depicts the adequate sensibility. Furthermore, an important consideration is the lowest possible diagnostic cost, which creates the need for the development of cheap and replaceable sensory parts.During my thesis study I focused on developing and characterizing sensor elements based on the localized surface plasmon resonance (LSPR) principle. I optimized the technology of bonding spherical gold nanoparticles with a diameter 20 nm from colloid solution onto silanized glass surfaces with the goal of maximizing the sensibility for the relative changes of the refractive index of the resulting sensor. During my research, the greatest obstacle was the aggregation of the gold nanoparticles on the silanized glass surface. I carried out different sets of measurements in my studies to understand the underlying causes of the aggregation and the prevention of this pehnomenon. Some of the possible causes are the cleanliness of the glass and the quality of the silan layer. In my work I have compared the behavior and sensitivity of the optimized sensing elements created from colloid with the behavior of surfaces covered with gold and gold-silver nanoparticles created by thermal annealing. While the prevention of the aggregation in case of colloid detached sensing element was successful, the best sample created by this technology was 43,6 nm/RIU, and the sensitivity of the samples created by thermal annealing was 31,51 nm/RIU, however the reproducibility of the latter was significantly better.