Improved Spectrum Usage with Multi-RF Channel Aggregation Technologies for the Next-Generation Terrestrial Broadcasting




  Jordi Joan Giménez Gandia


  David Gomez-Barquero
  Narcis Cardona Marcet


Next-generation terrestrial broadcasting targets at enhancing spectral efficiency to overcome the challenges derived from the spectrum shortage as a result of the progressive allocation of frequencies - the so-called Digital Dividend - to satisfy the growing demands for wireless broadband capacity. Advances in both transmission standards and video coding are paramount to enable the progressive roll-out of high video quality services such as HDTV (High Definition Televison) or Ultra HDTV. The transition to the second generation European terrestrial standard DVB-T2 and the introduction of MPEG-4/AVC video coding already enables the transmission of 4-5 HDTV services per RF (Radio Frequency) channel. However, the impossibility to allocate higher bit-rate within the remaining spectrum could jeopardize the evolution of the DTT platforms in favour of other high-capacity systems such as the satellite or cable distribution platforms. Next steps are focused on the deployment of the recently released High Efficiency Video Coding (HEVC) standard, which provides more than 50% coding gain with respect to AVC, with the next-generation terrestrial standards. This could ensure the competitiveness of the DTT. This dissertation addresses the use of multi-RF channel aggregation technologies to increase the spectral efficiency of future DTT networks. The core of the Thesis are two technologies: Time Frequency Slicing (TFS) and Channel Bonding (CB). TFS and CB consist in the transmission of the data of a TV service across multiple RF channels instead of using a single channel. CB spreads data of a service over multiple classical RF channels (RF-Mux). TFS spreads the data by time-slicing (slot-by-slot) across multiple RF channels which are sequentially recovered at the receiver by frequency hopping. Transmissions using these features can benefit from capacity and coverage gains. The first one comes from a more efficient statistical multiplexing (StatMux) for Variable Bit Rate (VBR) services due to a StatMux pool over a higher number of services. Furthermore, CB allows increasing service data rate with the number of bonded RF channels and also advantages when combined with SVC (Scalable Video Coding). The coverage gain comes from the increased RF performance due to the reception of the data of a service from different RF channels rather that a single one that could be, eventually, degraded. Robustness against interferences is also improved since the received signal does not depend on a unique potentially interfered RF channel. TFS was firstly introduced as an informative annex in DVB-T2 (not normative) and adopted in DVB-NGH (Next Generation Handheld). TFS and CB are proposed for inclusion in ATSC 3.0. However, they have never been implemented. The investigations carried out in this dissertation employ an information-theoretical approach to obtain their upper bounds, physical layer simulations to evaluate the performance in real systems and the analysis of field measurements that approach realistic conditions of the network deployments. The analysis report coverage gains about 4-5 dB with 4 RF channels and high capacity gains already with 2 RF channels. This dissertation also focuses on implementation aspects. Channel bonding receivers require one tuner per bonded RF channel. The implementation of TFS with a single tuner demands the fulfilment of several timing requirements. However, the use of just two tuners would still allow for a good performance with a cost-effective implementation by the reuse of existing chipsets or the sharing of existing architectures with dual tuner operation such as MIMO (Multiple Input Multiple Output).