Research On Evolutionary Algorithms In Automatic Cell Planning

As the three mobile communication operators, namely China Mobile, China Unicom and China Telecom acquired the operating licenses of TD-LTE, China has already come into the4G era. The multiple advances of4G, such as high downlink speed are attracting more and more users. The increase of users will surely influence network performance. So TD-LTE network optimization is of great importance for now.When the optimization scale in the network is very large, traditional methods which rely on the experience of network engineers cannot meet the increasing requirement of network optimization. Hence the project proposes the Automatic Cell Planning software (ACP). ACP is a piece of network optimization software, which helps users with optimizing certain network quality indexes, such as RSRP and SINR, defined by subscribers, by adjusting antenna parameters, such as azimuth and pilot power, in regions under optimization, using intelligent algorithms.ACP includes six subsystems, namely GUI, GIS, data processing, network, wrapper class and intelligent algorithms. Among them network subsystem is a kernel module which takes parts in network construction and evaluation. Intelligent algorithm subsystem utilizes Immune Genetic Algorithm, Taboo Search Algorithms and etc. to process network optimization.An advantage of ACP is the dimensionality reduction methods, which downsizes the original network optimization problems to increase both the speed and precision of the optimization process.