Rethinking cellular network planning and optimization.

To meet the ever increasing demand of wireless services for high data rates and mobility, mobile communication systems have evolved from the first generation cellular network to the current 4G one. The core purpose of a cellular network is to provide users with guaranteed QoS and seamless coverage throughout the service area. In the early stage, this task is achieved by installing cells at candidate sites and configuring their parameters as optimally as possible. Generally, each cell is assigned a dedicated operating frequency, and its adjacent cells operate on different frequencies to avoid inter-cell interference. As the evolution of the cellular network proceeds, advanced signal processing techniques, such as coordinated multipoint transmission and reception, and inter-cell interference coordination, are rising as promising solutions to improve the system performance, which on the other hand inevitably increase the CAPEX and OPEX of the cellular system. Besides, as the deployment of cells becomes denser and denser, it is more and more difficult and expensive to obtain capacity or QoS gains with these signal processing techniques. In this article, we rethink the cellular network planning issue in the context of heterogeneous networking, which is widely accepted as a cost-efficient paradigm to enhance the performance of the cellular system. We point out that the essential objectives of designing a cellular network, coverage and capacity, can also be achieved in heterogeneous networking scenarios using a cutting-edge territory division technique, which divides a service region into multiple subregions with almost equal traffic loads. We develop a dynamic cellular network planning framework that can significantly reduce the CAPEX and OPEX of the system. The QoS for users can also be enhanced while shifting away from employing complex and costly signal processing techniques.