A statistical approach could be used to allow a communication system to carry out self-optimisation, according to research published in the International Journal of Autonomous and Adaptive Communications Systems.
Jose Aguilar, Kristell Aguilar, and José Torres of the Universidad de Los Andes, in Mérida, Venezuela, have proposed an autonomic communication system based on a Bayesian network and an ontology, which allows the system itself to make reconfiguration decisions. The ontology provides the necessary knowledge about performance factors and the relationships within the network. The statistical system then adjusts factors within the system to optimise performance based on that knowledge, acting as a stochastic reasoning mechanism.
The team has demonstrated in simulations how their approach improves performance significantly particularly for scenarios where there are high reconfiguration requirements. The adaptive self-optimising approach improves flow rate, reduces loss rate, and minimises delay within the system.
Aguilar, J., Aguilar, K. and Torres, J. (2019) ‘Design of an autonomic communication system’, Int. J. Autonomous and Adaptive Communications Systems, Vol. 12, No. 4, pp.299–330.