Software-Defined Networking (SDN) embodies an advanced network architecture, developed to transform traditional networks and adapt them to fulfill contemporary needs. SDN distinguishes itself by decoupling the control and data planes, allowing for more effective monitoring, management, and optimization of network resources. However, the centralized structure of SDN makes it prone to multiple types of cyber-attacks. In particular, Distributed Denial of Service (DDoS) attacks pose a severe threat to the SDN controller. This paper introduces a method for identifying DDoS attacks on SDN controllers using a 1D Convolutional Neural Network (CNN) combined with a unique feature selection strategy termed Integrated Features Selection (IFS). IFS employs four different filter-based feature selection techniques: mutual information, chisquared, ReliefF, and analysis of variance (ANOVA). Our proposed model has demonstrated outstanding performance, achieving a detection accuracy rate of 97.42%, a precision rate of 96.34%, a false positive rate (FPR) of 2.18%, an F- measure of 96.6%, and a detection time of 0.697 seconds. The exceptional effectiveness of our model is attributed to the strength of its innovative feature selection method.