Analysis of Spectrum and Power Performance Using Distributed Base Station Architectures
This thesis has investigated a number of methods for optimizing energy and spectrum performance for 4G commercial radio access applications. The research interest is narrowed down to distributed base station architectures and in particular the remote radio module. Designing energy efficient radio access radio networks becomes a necessity not only due to the high operation and maintenance cost but also because of the major trend of providing ecofriendly solutions across the industry. The benefits of incorporating remote radio modules in next-generation mobile networks were investigated and a comparison with conventional base station architectures was realized. This analysis demonstrated that efficient hardware, intelligent software and self-organized subsystems can result in decreasing substantially power wastes. The advantages of optical fiber as transport medium for relaying baseband modulated signals to remote antenna sites were enlisted and the concept of Fiber To The Antenna (FTTA) was introduced. In terms of efficient hardware, the system requirements of the remote radio module have been analyzed thoroughly and a proposed architecture has been described in detail. In addition, digital signal processing techniques were developed for improving energy and spectrum performance. In particular a novel, lightweight crest factor reduction algorithm has been simulated, implemented in hardware and tested using the radio test platform provided by Radiocomp ApS. Finally, an adaptive polynomial digital predistortion block is proposed based on cartesian to polar conversion. It has been designed and implemented on a low-cost FPGA, overcoming the challenges imposed by the logical size and timing constraints.