To deal with the wireless big data's significant challenges of
network management and control architecture, modeling techniques
and algorithm design, we presented a multiple cognitive
agent-based divide-and-conquer management and control architecture.
Furthermore, a Markovian game-theoretic framework was
proposed to model the state big data-based decision-making
problem. Then, we investigated various state information dependent
learning methodologies, in particular, we concentrated on
construction of state space, state transition computation, and
convergence verification distributed Q-learning technique. This
work will facilitate the design of suitable architecture and algorithm
with effective mathematical modeling techniques and effi-
cient learning methodologies.