SDN paradigm to support Propagation Impairments Mitigation Techniques in EHF satellite communication networks

Recent reports underline that services such as Fast Internet, Mobile Video and Overthe-top content are growing year after year. It is expected that each user will require 40 Mbps up-link and 120 Mbps down-link by 2021 [1]. This trend implies that terrestrial networks, are expected to be used as the primary means to deliver applications with high availability requirements in many industries such as critical infrastructure, manufacturing, emergency communications, automotive, healthcare, etc. Given the relevance that mobile communications have assumed in recent years, networks are expected to meet challenging new requirements such as higher levels of network availability (at least in the five-nines range, i.e., 99.999% availability), as well as ubiquitous broadband connectivity extended to rural and low-density areas, as well as long-range transportation (e.g., aircraft, trains), among others. In this context, the role of satellite communications assume a prominent importance as terrestrial networks alone are not able to meet these demands both in rural and suburban areas where the quality of service is very low. Therefore the possibility to design Satcom systems with a Tb/s of capacity is now considered as an objective to be reached by 2020- 2025 [2]. Indeed the way that capacity is being brought to the market is changing, reducing the price per bit and making it more attractive for other services such as satellite broadband communications The new generation broadband satellite systems, called High-Throughput Satellites (HTS), exploits the following technologies to achieve these goals: • exploitation of large bandwidth availability in Extremely High Frequency (EHF) [3][4] bands, both in the feeder and user-link; • on-board signal processing techniques. • multiple gateways and multiple-beam user coverage to increase frequency reuse. One of the major obstacles towards the use of the EHF bands, which offer wider bandwidths, is the high degradation due to tropospheric propagation impairments. Already in Ka-band, the rain attenuation may exceed 20 dB. Therefore, Fade Mitigation Techniques (FMT) such as power control and Adaptive Coding and Modulation (ACM), have to be used in order to adapt transmission parameters to variable channel conditions without the need for large static link margins. Even if ACM has been proven to be effective and is currently employed in different satellite communication protocols, such as DVB-S2 [5] and DVB-S2X[6], this technique can not compensate more than a few dBs, thereby motivating the use of multiple GWs for transmit diversity to achieve the required availability on the feeder link (i.e. the link between HTS system gateways and satellite). A technique to overcome this issue and maintain high service availability in HTS systems exploiting EHF, is the of Smart Gateway Diversity (SGD)[7]. SGD is based on a pool of synchronized gateways (GWs) connected via a terrestrial fiber network; when the feeder-link of a GW experiences a deep atmospheric fading, its traffic may be routed to another gateway using the terrestrial network. This approach is effective if the involved gateways are spaced apart for more than 100 Km, so that atmospheric propagation impairments can be considered uncorrelated. Furthermore, SGD allows to increase link availabilities dramatically with a reasonable increase of the costs. In particular, two main schemes for SGD have been introduced: the so-called N +P, which is based on a pool of N active gateways (GWs) and P redundant GWs that are in warm standby. When one of the active GWs is in outage, switching occurs and traffic of the active GW is rerouted to one of the redundant GWs; the N+0 scheme, which is based on a pool of N active GWs working in a load sharing mode so that a spare capacity can be used when one or more GWs are in outage. In both cases the service network must be guaranteed even in the presence of degradation atmospheric phenomena on feeder-links. In fact, the handover decision problem need to be managed and executed. Hence, several criteria need to be considered, such as flow observation, network knowledge, quality of changing channels and the new flow tables that will have to be updated, in real time, for the gateways of the whole system. In this context, recent works have started the discussion of the possible benefits of more open architectures based on Software-Defined Networking (SDN) for satellite communication networks, for example, to achieve high rates in satellite applications with wide-scale coverage and high availability. In brief, the SDN gives a new perspective to the network management in which the control plane is centralized and provides a global knowledge to the network managers. Moreover, the adoption of SDN into the satellite networks can help reducing CAPEX and OPEX, enhancing the performance and the QoS delivered to satellite communication end-users, extending the range of applications of satellite communications, and achieving seamless, flexible and agile integration with terrestrial networks. In particularly, terrestrial networks are widely embracing SDN[47] technologies for easier management of the network by enabling the behavior of various devices to be changed, without depending on the standards of the various suppliers. Therefore, SDN enables a more flexible management of the handover procedure between satellite gateways, and a “more granular” exploitation of resources. SDN could be used in an effective way in SGD systems, providing the possibility to optimize traffic control/switching algorithms, as well as GWs network synchronization and GWs handover management and execution. By bringing benefits both on the better management of the system capacity, SDN can help in reducing the number of gateways and maximizing the capacity in the SGD traffic routing. In the following, the objectives and contributions proposed for the development of this thesis are explained in detail.