Multi-state prediction: A VOM based user behavior pattern in cognitive internet

Cognitive radio networks enable the unlicensed users to share the spectrum with licensed users, while it makes dynamic environment. The use of the channel by unlicensed users should be sensible of the licensed users in advance whether acceptable interference occurs. Mostly the routing protocol considered the previous state prediction. However, the routing probe and react strategy by unlicensed users sometimes is not only determined by the last one node state. In this paper we introduce a predictive node mobility model which is monitoring a licensed user's mobility, capable of reducing the delay.