Rajeev Gopal

Software Defined Satellite Networks

Software Defined Satellite Network (SDSN), similar to SDN [1], decouples data plane functions from control plane functions. SDSN benefits from logically centralized network state knowledge and decision making that enables optimal resource allocations for dynamic packet processing and transmission. An SDSN network node uses forwarding tables configured by a centralized system controller to govern packet routing with embedded Layer 2 switch without requiring elaborate Layer 3 control plane software implementation in every satellite node. Besides upper layer packet queuing and switching, SDSNs also involve radio transmission links and encompass associated modulation, coding, and resource allocation functions with dynamic control. SDSN architectural concepts can be illustrated with the SPACEWAY system which uses a GEO satellite comprising Ka band spot beams and a 10 Gbps packet processing switch. The onboard switching function is orchestrated by a ground-based system controller, with centralized support for addressing, routing, and packet flow management. The SDSN building blocks and performance objectives can be extended to address inter-satellite packet routing using a constellation of satellites with inter-satellite links and enhanced routing and resource management function at the controller. Besides GEO, LEO, and MEO satellites the SDSN architecture and techniques for addressing, routing, QoS, traffic engineering, and resource management can also be utilized for aerial and high altitude networking platforms.

Architectures for next generation high throughput satellite systems

This paper introduces architectures for next-generation high throughput satellite HTS systems comprising various satellite payload options, ground terminal advances, and scalable system-level software control and management techniques. It describes a model to estimate aggregate system capacity as a function of radio band, available spectrum, spot beams, waveforms, and payload capability, including antenna size, power, and digital/analog connectivity across various links and availability objectives. This system model has been used to evaluate aggregate capacity of representative Ka-Band low earth orbit and geosynchronous orbit systems. A system implementation approach is described for next-generation HTS systems based on widely used Industry standards. Modulation and coding techniques are based on Digital Video Broadcasting - S2 extensions DVB-S2X, which comprises spectrally efficient modulation schemes combined with low-rate codes. Several implementation technologies are analyzed related to configurable onboard payload and ground-based, software-defined resource control and management, key enablers of next-generation HTS systems. Basic architectural building blocks are introduced for design of end-to-end systems across low earth orbit, medium earth orbit, and geosynchronous orbit satellite constellations, with and without onboard processing and inter-satellite links, and including several efficient scenarios to achieve lossless handovers. Copyright

Net-centric satellite mesh architecture for emergency communication

This paper introduces a net-centric architecture based on ETSI and TIA Regenerative Satellite Mesh-A (RSM-A) standard that uses satellites with on-board packet switching. An Internet Engineering Task Force (IETF) IP protocol family-based extension of RSM-A for enhanced user interoperability, combined with multiple spot beam satellite implementation facilitates higher capacity, easy deployment and full mesh terminal-to-terminal operation that is required for managing emergency situations. This Net-centric Satellite Mesh (NSM) architecture adequately addresses the current and evolving networking needs for emergency-management information systems that can generate high data rate and real-time multi-media traffic. The NSM architecture supports the evolving IETF Emergency Telecommunications Service standards with packet flow-level admission control for guaranteed packet quality-of-service needed by real-time unicast and multicast traffic over satellite networks. The multi-plane architectural discussion is supported with measurements of selected end-to-end application behavior that validate the technical capabilities of NSM by using a pioneering Ka band satellite system, based on RSM-A, which has recently been deployed in North America. Copyright

Cross-plane information sharing for QoS in satellite–terrestrial integrated packet networks

This paper introduces a standards-based Quality-of-Service (QoS) architecture for satellite and terrestrial IP packet network integration to facilitate the evolution of the next generation integrated networks. Cross-plane information sharing can extend the reach of cross-layer improvements, typically applied for enhancing throughput and robustness in the wireless networks. Besides supporting real-time interactive applications for all users, such a QoS architecture can also provide a prioritization framework for enhanced service assurance. The satellite packet transport operates in a spectral, power, and hardware resource constrained environment. It needs specific functions for packet classification, scheduling, and propagation, all operating within the broader context of dynamic resource allocations at the media access layer for transporting individual packets and their aggregates. Additional information, made available to these networking functions across the data, control, and management planes can facilitate key decision making at the individual packet flow level with the use of formal connection admission control (CAC). Our QoS scheme links the data and management planes with the use of IP differentiated services (Diffserv) information in the packet headers. Originally designed for packet traffic aggregates, Diffserv can be leveraged to provide QoS for the individual packet flows at specific data rates and packet drop sensitivity. This QoS architecture has been validated with a pioneering regenerative packet processing satellite system and it is also suitable for other networks that support flow-level CAC, operating under dynamic network capacity management. Copyright