Telecommunications network node and methods

Abstract

A telecommunications network node and methods are disclosed. The method is for generating, by genetic programming, evolved algorithms for adjusting base station transmission power to control coverage of a cell to assist in providing desired base station operating characteristics. The method comprises the steps of: generating, using predetermined functions and terminals defined in a functions and terminals list, a plurality of evolved algorithms each of which determine whether, for any particular base station operating conditions, to adjust said base station transmission power; determining a fitness level indicative of each evolved algorithm's ability to adjust base station transmission power to control coverage to achieve said desired base station operating characteristics under expected operating conditions; and iteratively performing said steps of generating and determining to generate further evolved algorithms using at least one of said evolved algorithms determined to have achieved a particular fitness level. In this way, it can be seen that algorithms can be assembled and then tested to see how suitable they are at controlling base station power to achieve particular operating characteristics under particular operating conditions. Those algorithms which are determined to be the best at achieving those characteristics under those particular operating conditions may then be used to generate further algorithms which, in turn, are also then assessed. Hence, those individual algorithms which are suited to the operating characteristics and the operating conditions can rapidly be generated thereby avoiding the need to manually design new algorithms which are specialised to particular environments.

Description

technical field [0001] The invention relates to a telecommunications network node and method. Background technique [0002] Telecommunications networks are increasingly capable of self-configuration, self-organization and self-adaptation. As telecommunication networks increase in size and complexity, there is an impetus to realize these so-called "autonomous" properties in a decentralized manner, ie, where each node can act independently using only local information. [0003] Therefore, there is a growing need to develop algorithms from X, ie, algorithms for network node adaptation, which must work without global information about the network and cooperative centralized control of network nodes. [0004] Known solutions are autonomous algorithms designed by skilled engineers based on specific assumptions about the network that may not be realistic, so after they are implemented in the network, the algorithms often need to be evaluated, revised and perfected. This can be a ...

AI Technical Summary

Problems solved by technology

Due to the wide variation in the operating environments of the nodes, performance is degraded as general optimization algorithms perform less well than more specific algorithms for specific problems or operating environments

Also, conversely, specialized algorithms don't perform as well when applied outside of the specific area they're designed for

Images