ElasticWISP: Energy-Proportional WISP Backhaul Networks

<p>The provision of rural broadband infrastructure is a challenge for network operators across the globe, irrespective of their size. Wireless Internet Service Providers (WISPs) have shown that the small-scale deployment of wireless broadband infrastructure is a viable alternative to relying on cellular network providers for remote coverage. However, WISPs must often resort to using off-grid renewable energy sources such as solar energy for powering network sites, often resulting in undesirable, low-performance backhaul radios being used between sites out of concern for excessive energy consumption. The challenges of managing performant wireless backhaul networks in respect to energy constraints at remote, off-grid sites informs the need for energy-proportional design. Backhaul radios typically used by WISPs are not energy-proportional, meaning they use a consistent amount of energy, irrespective of wireless link utilisation. Using data from a real WISP network, diurnal traffic patterns show that WISP networks could benefit from energy-proportional design, without having to sacrifice performance. To encourage the development of high-performance, energy-proportional WISP backhaul networks, ElasticWISP, an optimisation architecture that reduces network-wide backhaul energy consumption while satisfying the user-demand for traffic, is introduced. ElasticWISP dynamically controls the configuration of backhaul radios based on bandwidth demands and the network-wide energy consumption of these radios. Through simulations driven by real WISP topology and data traffic, results show that ElasticWISP can offer energy savings of approximately 65% when WISP operators follow the proposed backhaul design methodology. Finally, a lightweight Multiprotocol Label Switching (MPLS)-based traffic engineering scheme, based on Segment Routing, is proposed. The implementation, named Segment Routing over MPLS (SR-MPLS), keeps traffic engineering path-state within each packet, meaning per-flow state is only held at SR-MPLS ingress routers. The lightweight approach of SR-MPLS also eliminates the otherwise necessary network-wide label flooding of traditional Segment Routing, making it ideal for bandwidth-sensitive wireless backhaul networks. Evaluation of SR-MPLS shows that it can perform as well as – and sometimes better than – competitor schemes.</p>

ElasticWISP: Energy-Proportional WISP Backhaul Networks

<p>The provision of rural broadband infrastructure is a challenge for network operators across the globe, irrespective of their size. Wireless Internet Service Providers (WISPs) have shown that the small-scale deployment of wireless broadband infrastructure is a viable alternative to relying on cellular network providers for remote coverage. However, WISPs must often resort to using off-grid renewable energy sources such as solar energy for powering network sites, often resulting in undesirable, low-performance backhaul radios being used between sites out of concern for excessive energy consumption. The challenges of managing performant wireless backhaul networks in respect to energy constraints at remote, off-grid sites informs the need for energy-proportional design. Backhaul radios typically used by WISPs are not energy-proportional, meaning they use a consistent amount of energy, irrespective of wireless link utilisation. Using data from a real WISP network, diurnal traffic patterns show that WISP networks could benefit from energy-proportional design, without having to sacrifice performance. To encourage the development of high-performance, energy-proportional WISP backhaul networks, ElasticWISP, an optimisation architecture that reduces network-wide backhaul energy consumption while satisfying the user-demand for traffic, is introduced. ElasticWISP dynamically controls the configuration of backhaul radios based on bandwidth demands and the network-wide energy consumption of these radios. Through simulations driven by real WISP topology and data traffic, results show that ElasticWISP can offer energy savings of approximately 65% when WISP operators follow the proposed backhaul design methodology. Finally, a lightweight Multiprotocol Label Switching (MPLS)-based traffic engineering scheme, based on Segment Routing, is proposed. The implementation, named Segment Routing over MPLS (SR-MPLS), keeps traffic engineering path-state within each packet, meaning per-flow state is only held at SR-MPLS ingress routers. The lightweight approach of SR-MPLS also eliminates the otherwise necessary network-wide label flooding of traditional Segment Routing, making it ideal for bandwidth-sensitive wireless backhaul networks. Evaluation of SR-MPLS shows that it can perform as well as – and sometimes better than – competitor schemes.</p>

Goodput Modelling and Optimisation of Channel Assignment For Planning IEEE 802.11Wireless Backhaul Networks

<p>IEEE 802.11Wireless backhaul networks (WBNs) provide scalable and cost-effective solutions for interconnecting small-cell networks and backbone networks or Internet. With newer and farther reaching applications being developed in IEEE 802.11 WBNs, such as smart grids and intelligent transportation systems, users expect high goodput and better fairness. However, some performance issues in IEEE 802.11 protocols such as border effect, exposed nodes and hidden nodes are exacerbated as network densification occurs, leading to goodput degradation and severe unfairness such as flow starvation (extreme low goodput). These issues may cause an IEEE 802.11 WBN to form a bottleneck and impact the overall network performance. Therefore, in-depth study is required in order to improve the IEEE 802.11 WBN planning to achieve better goodput and fairness. This research aims to improve IEEE 802.11 WBN planning through goodput modelling and optimising channel assignment. A novel simple goodput distribution model is proposed to predict goodput and fairness in IEEE 802.11 WBNs. Simulation results show that the proposed goodput model accurately predicts goodput with consideration of carrier sensing effect and traffic demands. Based on this goodput model, a new interference model is proposed to more realistically reflect both local and global interference in IEEE 802.11 WBNs. With the proposed interference model, two anti-starvation channel assignments have been developed to prevent flow starvation. Simulation validations show that the new anti-starvation channel assignments effectively prevent flow starvation and improve network fairness in IEEE 802.11 WBNs. This research also optimises channel assignment to achieve desired fairness and goodput. A multi-objective optimisation problem is formulated and a new fitness function is designed to evaluate a channel allocation with accurate prediction of goodput and fairness. Utilising the new fitness function, two multi-objective channel assignments have been developed to achieve both fairness and goodput. Compared with existing channel assignments through simulation, the proposed multi-objective channel assignments provide a set of feasible solutions that meet desired fairness and goodput. This research helps network planners or service providers to improve the IEEE 802.11 WBN planning from predicting network performance to optimising goodput and fairness. The proposed goodput model, interference model, and fitness function are also useful for node placement, and optimising routing and scheduling in IEEE 802.11 WBNs.</p>

Goodput Modelling and Optimisation of Channel Assignment For Planning IEEE 802.11Wireless Backhaul Networks

<p>IEEE 802.11Wireless backhaul networks (WBNs) provide scalable and cost-effective solutions for interconnecting small-cell networks and backbone networks or Internet. With newer and farther reaching applications being developed in IEEE 802.11 WBNs, such as smart grids and intelligent transportation systems, users expect high goodput and better fairness. However, some performance issues in IEEE 802.11 protocols such as border effect, exposed nodes and hidden nodes are exacerbated as network densification occurs, leading to goodput degradation and severe unfairness such as flow starvation (extreme low goodput). These issues may cause an IEEE 802.11 WBN to form a bottleneck and impact the overall network performance. Therefore, in-depth study is required in order to improve the IEEE 802.11 WBN planning to achieve better goodput and fairness. This research aims to improve IEEE 802.11 WBN planning through goodput modelling and optimising channel assignment. A novel simple goodput distribution model is proposed to predict goodput and fairness in IEEE 802.11 WBNs. Simulation results show that the proposed goodput model accurately predicts goodput with consideration of carrier sensing effect and traffic demands. Based on this goodput model, a new interference model is proposed to more realistically reflect both local and global interference in IEEE 802.11 WBNs. With the proposed interference model, two anti-starvation channel assignments have been developed to prevent flow starvation. Simulation validations show that the new anti-starvation channel assignments effectively prevent flow starvation and improve network fairness in IEEE 802.11 WBNs. This research also optimises channel assignment to achieve desired fairness and goodput. A multi-objective optimisation problem is formulated and a new fitness function is designed to evaluate a channel allocation with accurate prediction of goodput and fairness. Utilising the new fitness function, two multi-objective channel assignments have been developed to achieve both fairness and goodput. Compared with existing channel assignments through simulation, the proposed multi-objective channel assignments provide a set of feasible solutions that meet desired fairness and goodput. This research helps network planners or service providers to improve the IEEE 802.11 WBN planning from predicting network performance to optimising goodput and fairness. The proposed goodput model, interference model, and fitness function are also useful for node placement, and optimising routing and scheduling in IEEE 802.11 WBNs.</p>

Topology Control for Resource Management in Microwave Backhaul Networks

<p>Microwave backhaul networks are the dominant technology used to connect together access and core networks for their flexibility and cost-effectiveness in deployment. Unfortunately, microwave backhaul networks are susceptible to interference and are statically managed leading to poor Quality of Service (QoS) in the form of high delays and loss as well as being inefficient on energy. The use of Software Defined Networking (SDN) is proposed to address these problems by dynamically managing resources to work around the interference and remove static allocations. Two new algorithms, CUT and OptiCUT were designed to compute an optimal topology, to minimise loss and delay while at the same time reducing power consumption.</p>

Topology Control for Resource Management in Microwave Backhaul Networks

<p>Microwave backhaul networks are the dominant technology used to connect together access and core networks for their flexibility and cost-effectiveness in deployment. Unfortunately, microwave backhaul networks are susceptible to interference and are statically managed leading to poor Quality of Service (QoS) in the form of high delays and loss as well as being inefficient on energy. The use of Software Defined Networking (SDN) is proposed to address these problems by dynamically managing resources to work around the interference and remove static allocations. Two new algorithms, CUT and OptiCUT were designed to compute an optimal topology, to minimise loss and delay while at the same time reducing power consumption.</p>

A Survey on Integrated Access and Backhaul Networks

Benefiting from the usage of the high-frequency band, utilizing part of the large available bandwidth for wireless backhauling is feasible without considerable performance sacrifice. In this context, integrated access and backhaul (IAB) has been proposed by the Third Generation Partnership Project (3GPP) to reduce the expenses related to the deployment of fiber optics for 5G and beyond networks. In this paper, first, a brief introduction of IAB based on the 3GPP release is presented. Then, the existing research on IAB networks based on 3GPP specifications and possible non-3GPP research extensions are surveyed. The research on non-3GPP extensions includes the integration of IAB networks with other advanced techniques beyond the currently defined protocol stacks, such as the integration of IAB to cache-enabled, optical communication transport, and non-terrestrial networks. Finally, the challenges and opportunities related to the development and commercialization of the IAB networks are discussed.