Single frequency network planning

Both Digital Audio Broadcasting (DAB) and Digital Video Broadcasting (DVB) systems use orthogonal frequency division multiplexing (OFDM) to facilitate single frequency networks (SFNs). In an SFN all transmitters transmit the same information in the same frequency block simultaneously. This transmitter diversity structure makes SFNs very efficient in fading channels with an excellent spectrum saving and very good power utilization. The SFN concept requires a new design methodology to take full advantage of the provided diversity gain. In this thesis we address key problems that emerge in the planning of SFN based broadcasting networks. A consequence of the SFN concept is that the received signal is a superposition of several useful and interfering components. Shadow fading and transmitters contributing both to the useful and to the interfering power introduce considerable correlation among the signal components. We show that conventional methods to calculate the signal-to-interference ratio that do not take such correlations into account introduce significant error in the estimation of the local coverage probability. We propose new methods to reduce the estimation error. The (total) coverage of an SFN is evaluated using testpoints of which selection can be regarded as a two-dimensional sampling problem. Among the investigated sampling techniques systematic sampling proves to be the most efficient. Moreover, the proper testpoint density is shown to be dependent on the terrain characteristics, the propagation model and the total coverage level. We address the cost efficient design of individual SFNs to cover a predefined service area with requirements on the received signal quality and on the interference level. DAB and DVB allow the introduction of personal services, i.e., a part of the digital broadcasting bandwidth is used to serve individual users or a dedicated population. The new capacity constraints enforce the use of several SFNs to cover a given area, introducing frequency assignment problems to SFN design. We formulate the capacity planning as an optimization problem and propose efficient heuristic algorithms that use linear programming steps. Based on a flexible network model, the algorithms can cope with inhomogeneous capacity demand distributions over irregular terrains, which allows their direct application in computer aided SFN planning.