Privacy is becoming increasingly relevant in society. One reason is the growth in digital networking of people, partly resulting from the increased use of end devices. According to Adam Moore, the definition of privacy includes the "right to control access to and uses of places, bodies, and personal information" [Moo08]. This definition assumes the users themselves can govern the flow of their personal data. In many cases, however, user data is still being stored without consent, partly for commercial purposes or misused by other third parties. One specific case is the Facebook-Cambridge Analytica scandal in 2018 [CG18]. In order to protect personal data, the EU introduced a uniform set of rules: The General Data Protection Regulation (GDPR) [EU16]. Implemented in 2018, it aims to ensure privacy at an early stage of software development ("privacy by design"), and data protection as a default setting ("privacy by default"). In software development, architectural patterns are used for designing a software architecture, each representing ä package of design decisions that is found repeatedly in practice, has known properties that permit reuse, and describes a class of architectures" [BCK03]. These patterns do not address privacy, or address it insufficiently. The following two questions arise: Which architectural patterns exist that implement privacy requirements in a software architecture? How can these patterns be selected in a given application context? This thesis proposes privacy-preserving architectural patterns implementing privacy requirements from early on in the software development process. Furthermore, a methodology is presented that assigns appropriate patterns to the respective use case. A use case demonstrates the applicability of the methodology. Finally, the the presented architectural patterns and methodology are discussed.