Optimal Tradeoff Between RT and NRT Services in 3G-CDMA Networks Using Dynamic Fuzzy Q-Learning.
Ridha NasriZwi AltmanHervé DubreilPublished in: PIMRC (2006)
Keyphrases
- dynamic programming
- reinforcement learning
- state space
- fuzzy sets
- dynamic environments
- service oriented
- highly dynamic
- multi agent
- multi user
- membership functions
- service providers
- heterogeneous networks
- fuzzy rules
- access points
- network resources
- service discovery
- mobile networks
- web services
- telecommunication networks
- future internet
- lattice structure
- continuous state spaces
- learning algorithm
- fuzzy numbers
- function approximation
- optimal policy
- end users
- optimal solution
- social networks