Â Â Over the past decades, network security issues have been raised due to the rapid growth of internet. Security has been becomes an important challenge to secure the network from cyber-attacks. Various tools that are required for security purposes are firewalls, passwords, IDS for the detection of attacks and prevent it from sending out the harmful traffic to the traffic. This paper present statistical based intrusion detection technique based on chi square to detect DDoS attack by calculating the expected and observed frequencies. In this case traffic is containing in a dataset of certain interval. Traffic is contained from a backscatter dataset. This approach can effectively detect the DDoS attack.