Technical Note: \cal Q-Learning
Machine Learning
| Hi-index | 0.00 |
Capacity calculations reveal that required data rates in urban environments are in the order of 1 Gbps/km2, which is beyond any forward looking state-of-the-art. Following the BuNGee [1] design approach, we propose in this paper to follow an aggressive joint access and backhaul design which allows one powerful hub base station to feed several distributed access base stations at the same time as these communicate with associated user − all using the same band. The hence resulting interference environment is extremely dynamic and has eluded a viable communication strategy thus far. We propose however a novel distributed cognitive mechanism, which facilitates the capacity needs to be satisfied at these dynamics whilst being of low complexity and high reliability. Simulation results corroborate our findings.