Search

In this paper, we propose a new concept of a knowledge management framework to enable a self-optimizing and self-learning for wireless system operation in real time. The framework encapsulates both environment and intelligent agent to reach optimal operation through sensing, perception, reasoning, and learning in a truly autonomous fashion. The agent derives adequate knowledge from previous actions improving the quality of future decisions. Domain experience was provided to guide the agent while exploring and exploiting the set of possible actions in the environment. Thus, it guarantees low-cost learning and achieves a near-optimal network configuration addressing the non-deterministic polynomial-time hardness problem of joint channel and location optimization in a wireless system. Extensive simulations are run to validate its fast convergence, high throughput, and resilience to dynamic interference conditions. We deploy the framework on off-the-shelf wireless devices to propose autonomous self-optimization with knowledge management. [ABSTRACT FROM AUTHOR]

Wireless mesh networks (WMNs) have been extensively studied for nearly two decades as one of the most promising candidates expected to power the high-bandwidth, high-coverage wireless networks of the future. However, consumer demand for such networks has only recently caught up, rendering efforts at optimizing WMNs to support high capacities and offer high QoS, while being secure and fault-tolerant, more important than ever. To this end, a recent trend has been the application of machine learning (ML) to solve various design and management tasks related to WMNs. In this work, key ML techniques are discussed and past efforts applying them in WMNs are analyzed, while noting some existing issues and suggesting potential solutions. Directions are provided on how ML could advance future research. Recent developments in the field are also examined. [ABSTRACT FROM AUTHOR]