Method and system for packet classificiation with reduced memory space and enhanced access speed

A method and system for packet classification is proposed for applications such as firewalls, intrusion detection, policy-based routing, and network service differentiations, within network systems such as Internet or intranet/extranet systems. The proposed method and system is characterized by the use of protocol-oriented rule rearrangement, aggregated bit vectors (ABV) and folded bit vectors (FBV), an ABV-FBV index table dataset whose data structure is based on a featured split full-tree schema, and a DCBV (Don't-Care Bit Vector) dataset for packet classification. The combination of these features allows the packet classification to be implemented with a reduced amount of memory and access time during operation.