Research Topic #1: Software-defined optical wireless communications
Visible Light Communications (VLC) is a new technology that enables the simultaneous provision of illumination and data services either indoors and outdoors. The key principle behind this technology consists in modulating the intensity of the light emitted by an LED lamp according to the data message to be transmitted. At the receiver side, these variations on the optical power are detected using a Photodetector. As the intensity of the visible light signal changes faster than the response time of the human eye, the level of illumination in the room remains constant from the end user’s perception.
Figure 1: Indoor VLC system composed of a Centralized Processing Unit (CPU), a wired fronthaul, and multiple LED panels that coordinate their transmission to serve multiple users in the same room.
Though VLC technology has been around for quite a long time, it has only gained momentum recently, due to the replacement of conventional incandescent and fluorescent lights with new low-cost high-power white LED lamps. Most of the research done so far in this area has focused on the data rate that VLC can support from a link-level perspective, using different LED technologies. Unfortunately, the sum data rate that multiple VLC cells can provide in the same service area has not been characterized in detail, particularly when they have the chance to coordinate their transmissions to serve multiple users with random locations.
CTTC is a pioneering institution in the area of optical wireless, and has developed one of the first VLC demonstrators in the world based on IEEE 802.15.7 standard. This development was done using the Software Defined Radio (SDR) concept, where digital signal processing functions were implemented in a general purpose processor, and the signal acquisition and conversion was done using flexible programmable hardware (USRPs).
Figure 2: SDR implementation of a VLC system using flexible programmable hardware (USRPs) and off-the-shelf electro-optical components (i.e., phosphor-converted LEDs and low-cost Photodetectors).
The current version of the IEEE 802.15.7 standard is out-of-date as it fails to consider the latest technological developmentsin the field of optical wireless. New enhancements that improve the Key Performance Indicators (KPIs) of VLC technology have been identified in the literature. However, to accelerate their incorporation into the revised versions of VLC standards, a SDR-based validation of these concepts is very important to check that the advertised performance gains are actually attainable in practice.
Research Topic #2: Integration of Optical and Wireless technologies for future access networks
The communication network and service environment that are foreseen for year 2020 will be much richer and more complex than the one observed so far in 4G. Among these differences, 5G networks will have to support about 1000 times higher mobile data volume per unit area, 10-100 times more data traffic per user, 10-1000 times higher number of devices, 10 times longer battery life, and 5 times less End-to-End (E2E) latency than today. Trying to address these challenges, the current trend consists in separating the digital Baseband Unit (BBU) and analog Radio Unit (RU) processing of a traditional Base Station (BS) into two parts, implementing the so-called Centralized Radio Access Network (C-RAN) architecture. A critical part of this architecture is the fronthaul network, which connects the Centralized Base Station (CBS) with the Remote Antenna Units (RAUs). To implement it, optical fibres are the natural choice as they introduce low latency and support a large bandwidth.
Figure 3: Distributed RAN (left-hand side) and Centralized RAN (right-hand side) architectures. The ‘fronthaul’ is the network segment that appears in a C-RAN to connect the BBU with the RU.
As the number of radio signals to be transported between the CBS and the RAUs grows large, the utilization of contemporary fronthaul technologies (such as CPRI) becomes more difficult. This is because Digital Radio-over-Fibre (RoF) solutions have been originally designed to be transparent to the transported radio signal, introducing a negligible distortion at the cost of expanding the bandwidth notably and increasing the processing delay. On the other hand, the transportation of analog baseband radio signals does not expand the bandwidth, does not introduce processing delay in the fronthaul, and keeps the cost of the RAU at the lowest. However, the main challenge that Analog RoF links have is the distortion that the different optical and electrical components of the fronthaul network introduce.
As the assumption of an ideal fronthaul will not be realistic any more for future C-RANs, novel design paradigms are needed, where the impairments introduced by both fibre fronthaul and wireless air-interface are jointly taken into account.
Figure 4: Joint optimization of the physical-layer processing to maximize the end-to-end data rate of the hybrid optical-wireless link that is configured (left-hand side). Equivalent Amplify-and-Forward relaying system that is configured in the proposed C-RAN architecture (right-hand side). IM: Intensity Modulation. CBS: Centralized Base Station. RAU: Remote Antenna Unit.
Research Topic #3: New transmission methods to enable wireless access and backhauling over THz band frequencies
Terahertz (THz) band communication (0.1−10 THz) has the potential to alleviate the spectrum scarcity and capacity limitations that contemporary wireless systems experience. Though the THz band offers an extremely broad bandwidth, the transmission distance is typically short due to very high path loss attenuation that is experienced in this frequency range. This is because, apart from high spreading loss, there is also strong molecular absorption that depends on the mixture of gases and water vapor that is present in the atmosphere. This effect results in a distance-dependent frequency-selective path loss that grows notably with distance. So far, most of the research done on THz communications considered the molecular absorption as an impairment that needs to be combatted. However, if this phenomenon is addressed from a different perspective, it can be effectively exploited when performing the self-organization of the ultra-dense small cell network environment.
Figure 5: Molecular absorption loss (dB/km) as function of the carrier frequency (GHz). Different transmission windows are configured in practice according to the total absorption loss that is added on top of the conventional spreading loss that is experienced. If this phenomenon is properly exploited, longer transmission distances as well as denser deployment of small cells can be effectively supported.
Most of contemporary research on THz communication has focused on characterizing the supported data rate in point-to-point links, where the usable bandwidth reduces notably for distances beyond few meters. However, such an approach limits the potential of THz technology, as longer transmission distances are feasible if multiple transmission points cooperate by taking into account the distance-frequency-dependent molecular absorption. That is, a much better performance is possible if transmit power and the beamforming weights per carrier are jointly selected by transmission points belonging to the same cluster. Similarly, strong co-channel interference can be controlled in ultra-dense small cell networks if the carrier allocation per cell is smartly selected according to the distance towards adjacent base stations.
The exploration of distance-aware transmission methods, as well as radio resource management schemes that take into account the effect of molecular absorption, will be of key importance to take advantage of the new spectral resources that THz frequency bands offer beyond the horizon that Millimeter Wave Communication has set. With these schemes, THz band will gain importance not only short-range point-to-point communication, but also for longer distance links either in the access and fronthaul segments of future Centralized RANs.