Reinforcement Learning for Routing in Cognitive Radio Ad Hoc Networks
Table 1
CRQ-routing model embedded at SU node .
State
, each state representing a SU destination node . represents the number of SUs in the entire network.
Action
, each action representing the selection of a SU next-hop node along with its operating channel. represents the number of SU ’s neighboring SU nodes.
Cost
represents the link-layer delay incurred to successfully deliver a packet from SU node to SU neighbor node , including retransmission delays as a result of PU-SU packet collision and packet loss.