Optimising CDMA Cell Planning with Soft Handover

• Published in: Wireless personal communications 2013, Vol.68 (2), p.321-347
• Main Authors: Ghosh, Sasthi C., Whitaker, Roger M., Allen, Stuart M., Hurley, Steve
• Format: Article
• Language: English
• Subjects: Algorithmics. Computability. Computer arithmetics; Applied sciences; Code division multiple access; Communications Engineering; Computer Communication Networks; Computer science; control theory; systems; Computer systems and distributed systems. User interface; Density; Engineering; Exact sciences and technology; Inclusions; Lower bounds; Mathematical models; Networks; Optimization; Radiocommunications; Signal,Image and Speech Processing; Software; Telecommunications; Telecommunications and information theory; Theoretical computing; Universal Mobile Telecommunications System
• ISSN: 0929-6212; 1572-834X
• DOI: 10.1007/s11277-011-0454-9

Soft handover (SHO) is one of the fundamental features of code division multiple access (CDMA) systems such as universal mobile telecommunication system (UMTS), and it is affected by the placement and density of cells. Inclusion of soft handover in optimization models for UMTS and CDMA cell site selection and configuration has previously been very limited but it is important for coverage because it can provide gain to the user. Some authors have excluded SHO on the basis of tractability while others have found that omitting SHO in planning gives adequate solutions. As such the incorporation of SHO remains an important component for definitive investigation in optimisation models for cell planning. In this paper we focus on the problem and effect of including SHO in cell planning optimisation. We introduce a new cell planning optimisation model that explicitly incorporates SHO and reduces computational complexity. Exact results can be obtained when the orthogonality factor is zero, while a conservative approximation of interference is used to generate lower bounds on coverage in the general case. We demonstrate the tractability of this model and show that it leads to improved lower bounds for coverage maximisation in network planning.