The rapid growth in network technology and data traffic has made traditional network architectures inadequate in meeting the needs of modern users, leading to the adoption of Software Defined Networking (SDN) as a solution. Despite its benefits, SDN's centralized control, complexity, lack of standardization, inadequate visibility, authentication, and authorization make it vulnerable to security threats such as Distributed Denial of Service (DDoS). In SDN networks, DDoS attacks can overwhelm networks with traffic, causing reduced performance or downtime in the data or controller planes. DDoS attacks within SDN networks can be categorized into three main types volumetric attacks, which inundate networks with substantial traffic, state exclusion attacks which take advantage of vulnerabilities in TCP packet processing and application layer attacks which specifically target services with malevolent activity. The objective of the proposed system is to provide an alternative solution on detecting UDP flood-based Distributed Denial of Service (DDoS) attacks in Software-Defined Networking (SDN) data planes. This will be achieved by incorporating mirror ports as inputs to a machine learning module. Additionally, the machine learning module will have the capability to identify attack patterns within mirrored traffic without causing disruptions to the main network traffic.