TY - JOUR
T1 - Optimal WCDMA network planning by multiobjective evolutionary algorithm with problem-specific genetic operation
AU - Gu, Fangqing
AU - Liu, Hai lin
AU - CHEUNG, Yiu Ming
AU - Xie, Shengli
N1 - Publisher Copyright:
© 2014, Springer-Verlag London.
PY - 2015/12/1
Y1 - 2015/12/1
N2 - The wideband code division multiple access (WCDMA) network planning problem requires to determine the location and the configuration parameters of the base stations (BSs) so as to maximize the capacity and minimize the installation cost. This problem can be formulated as a complex set covering problem. Compared to the classical set covering problems, the coverage area of each BS is unknown in advance. This makes that the selection of each BS location and configuration parameters is determined by the location and configuration parameters of the neighbor BSs. Accordingly, we will conduct a competition and cooperation model based on the re-covered area of the BSs to measure the relationship of the BSs. Then, an efficient genetic operation based on this model is proposed to generate new-quality solutions. Further, four BS configuration parameters, i.e., the antenna height, antenna tilt, sector orientation and pilot signal power, are taken into account as well. Since there are too many combination levels of the configuration parameters, an encoding method based on orthogonal design is presented to reduce the search space. Subsequently, we merge the proposed encoding method and genetic operation into the multiobjective evolutionary algorithm-based decomposition (MOEA/D-M2M) to solve the WCDMA network planning problem. Simulation results show the efficacy of the proposed encoding and genetic operation in comparison with the existing counterpart.
AB - The wideband code division multiple access (WCDMA) network planning problem requires to determine the location and the configuration parameters of the base stations (BSs) so as to maximize the capacity and minimize the installation cost. This problem can be formulated as a complex set covering problem. Compared to the classical set covering problems, the coverage area of each BS is unknown in advance. This makes that the selection of each BS location and configuration parameters is determined by the location and configuration parameters of the neighbor BSs. Accordingly, we will conduct a competition and cooperation model based on the re-covered area of the BSs to measure the relationship of the BSs. Then, an efficient genetic operation based on this model is proposed to generate new-quality solutions. Further, four BS configuration parameters, i.e., the antenna height, antenna tilt, sector orientation and pilot signal power, are taken into account as well. Since there are too many combination levels of the configuration parameters, an encoding method based on orthogonal design is presented to reduce the search space. Subsequently, we merge the proposed encoding method and genetic operation into the multiobjective evolutionary algorithm-based decomposition (MOEA/D-M2M) to solve the WCDMA network planning problem. Simulation results show the efficacy of the proposed encoding and genetic operation in comparison with the existing counterpart.
KW - Multiobjective evolutionary algorithm
KW - Orthogonal design
KW - Set covering problem
KW - Wireless network planning
UR - http://www.scopus.com/inward/record.url?scp=84944514329&partnerID=8YFLogxK
U2 - 10.1007/s10115-014-0799-y
DO - 10.1007/s10115-014-0799-y
M3 - Journal article
AN - SCOPUS:84944514329
SN - 0219-1377
VL - 45
SP - 679
EP - 703
JO - Knowledge and Information Systems
JF - Knowledge and Information Systems
IS - 3
ER -