A Distributed Denial of Service (DDoS) attack denies authorised users access to a network or service. Entropy-based approaches are commonly employed in Software Defined Networks (SDN) to identify DDOS attacks. This paper presents a dynamic entropy-based technique for detecting DDoS attacks using the programmability feature of SDN and the dynamic aspects of living communication theory. The experimental results demonstrate that dynamic entropy is unaffected by normal traffic but fluctuates significantly in response to an attack. It has been discovered that the proposed technique can efficiently detect DDoS attacks and has a high detection rate.