The Fourth Space Internet Workshop

June 8-10, 2004

Ramada Inn, BWI Airport
7253 Parkway Drive
Hanover, Maryland 21076
(410) 712-4300

NASA Home page

SIW-4 Abstracts


Session Title Presenter
Session A

Space Protocols
MST/MD: a Transport Layer Protocol that improves Large Data Set Transmission over Geo-Stationary Earth Orbit Satellites Paul D. Wiedemeier
and Harry W. Tyrer
Maximizing Data Volume for Direct to Ground Satellite Systems David A. Carek, P.E.
An Intelligent Internet Protocol Architecture for Space-Based Networks Ivan Thomas,
Brian Robinson
Including HDLC Framing in CCSDS Recommendations Keith Hogie
Embedded Transport Agents for Near-Earth Communications Timothy J. Salo
Advancements in SCPS-TP Technology and Deployments Charlie Younghusband,
Eric Edwards,
Joshua Lamorie
Session B

Space Networks
Air Force Satellite Control Network Interoperability Project John Pietras

NASA Space Network IP Services (SNIS)

David J. Israel
Integrated Satellite Control Network
Top-Down and Bottom-Up Implementation
Richard A. Russel
Standard Interface For Satellite IP Networks Enrique Laborde
Session C

Onboard Applications
Building a Reliable Onboard Network Jane K. Marquart
The GPM Core Observatory Jonathan Wilmot
The Trade Between CCSDS and HDLC Framing on Global Precipitation Measurement David F. Everett
An Implementation of the CCSDS File Delivery Protocol
for the GPM Mission
Arturo Ferrer
Design of a Fault-Tolerant Link Establishment Protocol Stephen Horan and Giriprassad Deivasigamani
Session D

Spacecraft Nodes
Current Activities and Status of the ESA-ESTEC Sponsored
Wireless Onboard Spacecraft Working Group
Rodger Magness, et. al.
Update of the SSTL DMC Constellation Chris Jackson
Software-Defined Radio Ground Station for Internet-Protocol Communications to a Low Earth Orbit Nanosatellite Brendan Surrusco
Automated File Transfer and Storage Management
Concepts for Space
Ed Criscuolo
Space-based Telemetry And Range Safety (STARS) Using IP Ed Criscuolo
Session E

Applications Beyond Earth Orbit
Handbook for using IP protocols for Space Missions Keith Hogie

CisLunar BoF Report

Keith Scott
Alaska Beacon - Adapting the Iridium Network
to Enable Search and Rescue
John Bystroff, Ignacio Gómez, Robert Kikta
Session F

Routing & Infrastructure
Does IPv6 Provide Benefits to Space Missions? Allen P. Holtz
Routing in Space: Getting Data from Here to There Allen P. Holtz
Space Communications Emulations Facility (SCEF):  Improving NASA Missions through Emulation James Stegeman
An Efficient Layer 2-Mesh Communications
Protocol for Space Sensor Networks
Esther H. Jennings
Reusable, Open Source, and Internet Enabled Software Development Tools for Efficient Small Satellite Design,
Mission Planning and Analysis
Eric Tapio
Session G

Network Architecture & Emulation
Dynamic Access for a Space Communications Network
with IP Functionality
Hui Zeng and
Michael Hadjitheodosiou*
End-to-End Network Models Encompassing Terrestrial, Wireless, and Satellite Components Ramon P. Williams
Analysis of a Mars Networking Scenario within the Space Based Network Emulator Shaun Endres
An Integrated Model for Using Dynamic Link Space Communication Analysis Tool for High Latency Space Network Emulation of Martian Assets Debra L. Emmons
Web Services for Satellite Emulation Development Kathy J. Liszka
Demonstration of the TASC Rapid Architecture
Development and Evaluation System (TRADES)
Eliud Bonilla

MST/MD: a Transport Layer Protocol that improves Large Data Set Transmission over Geo-Stationary Earth Orbit Satellites


Paul D. Wiedemeier[1] and Harry W. Tyrer[2]


Rural Earth locales are under-served by high-bandwidth terrestrial data transmission infrastructure, and even when available the price is often prohibitive.  Geo-stationary earth orbit (GEO) satellites can provide high-bandwidth data transmission to large geographic areas and are not affected by natural and terrorist events that can render terrestrial data transmission infrastructure unusable.  In specific rural areas, they may be the only option available to mitigate terrestrial data transmission infrastructure destruction and limitations.

The Transmission Control Protocol (TCP) exhibits poor performance when transmitting data sets over GEO satellites.  Our research shows that TCP Reno requires approximately 6110 seconds (101 minutes) to transmit a 20 MB data set (the typical size of a medical image) when the bit error rate is 1.0e-04.  We have designed a Multiple Segment Transmission with Majority Decoding (MST/MD) transport layer protocol to improve the transmission of large data sets over GEO satellites. 

The Multiple Segment Transmission function of our MST/MD transport layer protocol transmits each segment n times (n >= 1 and odd) and uses an 8 MB maximum window size and an 8 MB segment size. We chose an 8 MB size because it (1) is an exponential value of base 2 and (2) represents the largest amount of data that can be in transit over a GEO satellite with a transmission rate of 155.520 Mbps and a round trip transmission delay of 560 ms.

The Majority Decoding function of our MST/MD transport layer protocol uses well-established binomial random variable theory.  A single bit will be corrupted with probability p when it is transmitted over a GEO satellite.  Here, p represents the satelliteŐs bit error rate (0 <= p <= 1).  To increase the probability that the single bit is transmitted correctly, it is transmitted n times (n >= 1 and odd).  A majority decoder at the destination reviews each bit corresponding to the same index from all multiply received segments and determines the correctly transmitted bit as that of the majority.

We have developed three variants of the MST/MD transport layer protocol.  TCP Columbia version 1 transmits each 8 MB segment seven times and waits for either an acknowledgement (ACK) to arrive or a retransmission timeout (RTO) to occur before transmitting the segment again.  TCP Columbia version 2 transmits a single 8 MB segment and then waits for an ACK or RTO.  It retransmits the segment if no ACK arrives or when a RTO occurs.  Otherwise, the transmission was successful.  Using TCP Columbia version 2, segments are transmitted at least once, but at most seven times.  UDP Columbia version 1 transmits each 8 MB segment seven times but does not wait for ACKs or RTOs.

We used the computer network protocol simulator ns-2 version 2.1b9 and a personal computer running Red Hat Linux to test the designs of our three MST/MD transport layer protocol variants.  At zero bit error rate, our simulations show that data sets larger than or equal to 2 MB are transmitted with substantially less delay over a GEO satellite using all three MST/MD transport layer protocol variants versus TCP Reno.  We also determined that TCP Columbia version 2 transmits data sets larger than or equal to 1 KB with less delay than TCP Reno.  For all non-zero bit error rates (1.0e-08 through 1.0e-04), all three MST/MD transport layer protocol variants transmit a 20 MB data set with less delay versus TCP Reno.  Specifically, UDP Columbia version 1 transmits a 20 MB data set in approximately 15 seconds when the bit error rate is 1.0e-04; this represents a 99.7% decrease in transmission time compared to TCP Reno.

The transmission of large data sets (e.g. medical images) over a GEO satellite using TCP takes considerable time, which is due to the high delay and lossy nature of GEO satellites as well as the congestion control and avoidance algorithms used by TCP.  Our research shows that the time required to transmit large data sets over GEO satellites is significantly reduced when the MST/MD transport layer protocol is used, compared to TCP Reno, even when the bit error rate is high.  This will benefit particularly regional hospitals and medical clinics that use GEO satellites to provide their high-bandwidth data transmission infrastructure.


[1] Computer Science Department, University of Missouri @ Columbia, 201 Engineering Building West, Columbia, Missouri, 65211, USA, 573-875-7644,

[2] Electrical and Computer Engineering Department, University of Missouri Đ Columbia, 305 Engineering Building West, Columbia, Missouri, 65211, USA, 573-882-6489,

To Abstract Index





E. Laborde

Hughes Network Systems, Inc.




This presentation is about satellite access networks based on the IP over satellite (IPoS) air interface standard.  The IPoS standard has been developed within the Satellite Communications Division of the Telecommunications Industry Association (TIA), Technical Subcommittee TR 34, and was published as TIA-1008 standard in November 2003.


The presentation will start with a summary of the different access technologies (DSL, cable, wireless, satellite) competing to provide commercial broadband IP services to residential and Small Office, Home Office (SOHO) customers, focusing on satellite access networks.


Included in the presentation will be the challenges involved in the deployment of commercially successful satellite access networks capable of delivering competitive IP services and how the concepts fulfilling these challenges are incorporated in the IPoS satellite interface standard.


The star topology used by IPoS is the current practical solution able to strike price/performance compromises that compete favorably with the other access technologies in implementation of low-cost Internet services to residential and SOHO users.  In IPoS-based satellite access networks, remote terminals located at the user premises are connected by the IPoS interface to a broadband access node, or gateway, using the IPoS air interface over non-regenerative transponders on geostationary satellites.  The gateway aggregates the traffic from the remote terminals and interconnects to the Internet with high-speed terrestrial links.


The description of the network elements and network interfaces, particularly of the protocol used by the IPoS interface between remote terminals and gateway, will be the main topic of the presentation.  The IPoS protocol stack separates satellite-dependent functions from satellite-independent functions or those above the IP layer that run end to end between host computers.


The physical, data link, and network protocol layers included in the satellite-dependent part of the protocol stack will be covered in the final part of the presentation.


The IPoS interface is already deployed in the worldŐs largest satellite access network with 200,000 residential/SOHO users in North America as well as many other satellite networks around the world.  IPoS-based satellite access networks coverage is estimated to be already in excess of 90% of the globe.


Extending the provision of broadband satellite IP-based services to consumers through one industry-accepted standard interface will potentially have a lasting impact in supporting the increasing demands for Internet access services to the home.  These services might change from services offered by network access operators today (such as Web surfing and e-mail) to higher transmission rate services (such as videoconferencing and voice and video streaming), using the broadband multicast/broadcast connectivity provided more effectively by satellite access networks.


To Abstract Index

Design of a Fault-Tolerant Link Establishment Protocol

Stephen Horan and Giriprassad Deivasigamani

Telemetering Center

Klipsch School of Electrical and Computer Engineering

New Mexico State University

Las Cruces, NM  88003-8001

Contact Info: Ph (505) 646-4856; Fax (505) 646-6417



One of the networking technologies that will be required for future satellite cluster missions is the development of a protocol for establishing and managing the communications links between the cluster members over the life of the mission.  This protocol will need to manage the traffic flow, adjust routing tables as nodes move into and out of the cluster, allow for sub-networks in the cluster, and similar activities.  One important constraint in the protocol is that it needs to account for the real probability that link corruption will cause packet losses.  Therefore, no management decisions should be made based on a single occurrence of a fault.  An example of this type of decision is declaring that a link between two nodes is no longer valid.

In this presentation, we will describe the design of an overall protocol for a cluster link establishment and management that accounts for link corruption, node failures, and node reestablishment.   This protocol development is in its initial stages and we will describe the protocol requirements to be met, the underlying approaches, and the testing done to date to validate the approach.


To Abstract Index

Software-Defined Radio Ground Station
for Internet-Protocol Communications
to a Low Earth Orbit Nanosatellite

Nathan F. O'Connor, Brendan S. Surrusco, Sven G. Bilén, and Charles L. Croskey

The Pennsylvania State University, Electrical Engineering Department, University Park, PA 16802

The Local Ionospheric Measurements Satellite (LionSat) mission is a participant in theNanosat-3 (NS-3) program, which is a joint program between the American Institute ofAeronautics and Astronautics (AIAA), the National Aeronautics and Space AdministrationGoddard Space Flight Center (NASA GSFC), the Air Force Office of Scientific Research(AFOSR), and the Air Force Research Labs Space Vehicles Directorate (AFRL/VS). Theobjectives of the NS-3 program are to educate and train the future workforce through a national student satellite design and fabrication competition and to enable small satellite R&D, payload development, integration, and flight test. Also important to the program is the ability to fly new technologies to qualify them for use in space.

Existing space telemetry standards in use by NASA, ESA, NASDA, and other space agencies rely heavily on legacy systems, coding, and methods, placing large demands onmanpower while remaining inflexible to the communications needs of future space missions. Autonomous navigation and control, "target of opportunity" observations, constellation flight, and inter-satellite crosslink communications are fundamental capabilities in NASA's roadmap, to which traditional telemetry methods are not suited. The growth of commercial 3G, 802.11, and other wireless networking technologies provides a basis for a new paradigm in spacecraft communications and operations emerging from secure commercial IP networking.

This presentation will describe the development of an IP end-to-end communications and operations plan for the LionSat mission. We will also describe the software-defined radio (SDR) solution for decoding the RF downlink (i.e., physical layer) from the satellite. SDR is emerging as a powerful and eventually pervasive technology in the communications field. One of the reasons for the current significant interest in SDR is the ability to "reconfigure" communications systems without requiring major hardware changes. The SDR solution first will be developed in the Simulink environment. The Simulink blocks will then be replaced by Xilinx and Texas Instrument blocks which will be used to generate code for the onboard DSP and FPGA of the Lyrtech SignalMaster system.

This presentation will describe the development of an IP end-to-end communications and operations plan for the LionSat mission. We will also describe the software-defined radio (SDR) solution for decoding the RF downlink (i.e., physical layer) from the satellite. SDR is emerging as a powerful and eventually pervasive technology in the communications field. One of the reasons for the current significant interest in SDR is the ability to ŇreconfigureÓ communications systems without requiring major hardware changes. The SDR solution first will be developed in the Simulink environment. The Simulink blocks will then be replaced by Xilinx and Texas Instrument blocks which will be used to generate code for the onboard DSP and FPGA of the Lyrtech SignalMaster system.


The higher layers of the communications protocol for the downlink from LionSat will be a variation of the pervasive Internet Protocol. Standard IP packets from the spacecraft flight computer will be formatted into HDLC frames and sent to the ground. Using the Xilinx FPGA within the SDR the HDLC frames can be processed into any packet format needed to pass the data to a processing station. In the case of the LionSat project the HDLC frames will be reconstructed into Ethernet packets and sent to a desktop computer. The resulting user interface for satellite operators will be as simple as an FTP program or web browser.



To Abstract Index

Alaska Beacon - Adapting the Iridium Network to Enable Search and Rescue


Authors:      John Bystroff, The Boeing Company (Boeing)

Ignacio Gómez, KinetX Inc.

Robert Kikta, NexGen Communications (NexGen)


The Alaska Beacon program demonstrates how the Iridium Satellite Network (ISN) architecture can be leveraged to provide a new service free of cost for a humanitarian purpose. This pilot program allows users of the Iridium system in Alaska to activate a capability in the phone to periodically connect to the network and update location information. In the event the user is reported lost to the Alaska Rescue Coordination Center (ARCC), a relationship has been established with BoeingŐs Iridium Satellite and Network Operations Center to retrieve the location data based on these periodic network connections.


This project involved the development of new software for the Iridium phone by NexGen. It also required establishing new processes and procedures by Boeing to innovatively use network elements already in place to support IridiumŐs Short Burst Data (SBD) service. Additionally, Boeing completed extensive analysis based on knowledge of the satellites to qualify the location information to focus rescuersŐ search efforts. The result of these efforts is an ability to send the ARCC an email response that will provide latitude and longitude boundaries for the subscriberŐs last known locations with a greater than 90% confidence.


This presentation will describe the approach taken to quickly generate the new Alaska Beacon software; the network elements and tools that enabled the team to create the processes; and the organizational relationships leveraged to technically enable the success of this pilot program.



To Abstract Index

Update of the SSTL DMC Constellation

Chris Jackson -

I intend to submit an abstract for SIW4 which will be an update of the SSTL DMC constellation. Spacecraft for this constellation were launched in Nov 2002, and Sept 2003 and are employing IP.


To Abstract Index

Use of Internet Protocols in Transformational Communications MILSATCOM

Authors: Carl Sunshine, Joel Sercel, Jeremy Mineweaser


The Transformational Communications MILSATCOM (TCM) system is being designed to provide a combination of circuit and packet communications services for mobile and deployed military users in the next decade. TCM will extend Internet technology into space, with attention to the extra security, endurance, and priority requirements necessary in a military environment. The briefing will summarize the baseline Government Reference Architecture for the system, with emphasis on meeting the challenges of extending the Internet architecture into a military space environment.


To Abstract Index

Dynamic Access for a Space Communications Network with IP Functionality


Michael Hadjitheodosiou*,
Hui Zeng
Brenda L. Ellis
CSHCN, Institute for Systems Research
University of Maryland,
College Park, MD 20742
E-mail: {michalis, zengh}
Mission Network Applications Branch
NASA Glenn Research Center
21000 Brookpark Road, MS 142-1
Cleveland, OH 44135


The vision for the future space network involves a scenario where all scientific spacecraft form a distributed network to provide real-time information transfer to users on the ground.  This scenario will require sensors and instruments on spacecraft to become addressable nodes in a communication network. To enable this vision, there is critical need for advanced communications and dynamic network connectivity to provide broad coverage and intelligent-based real time data delivery to scientists. The eventual extension of IP functionality to space missions would change the way scientists can access data from space.  To take full advantage of this new functionality the changes cannot be limited to the spacecraft but must also take place in the ground and the infrastructure.  A transition to dynamic missions operation would be desirable, and for this to happen it will be necessary to develop more adaptive ways to share resources and limit costs.

In order to do that we are studying the existing mission traffic and develop simple traffic models of the spacecraft-generated science traffic. Based on them, we proposed a hybrid-mode reservation-based TDMA protocol. Our objective is to provide an optimal or near-optimal utilization and fair allocation of bandwidth of the downlink channel while guaranteeing specific QoS requirements for different service classes. We formulate and study an assignment problem for optimal timeslot scheduling for this protocol. By using simulation, the protocol performance is analyzed and compared with that of the existing static fixed-assignment scheme. The simulation results demonstrate that our frame-based scheme has advantages over a static protocol under bursty traffic load. It utilizes the reservation-mode data slots dynamically by global optimization and guarantees the minimal bandwidth for each spacecraft via the fixed-mode data slots. By addressing these issues we try to contribute in the development of the next generation space network infrastructure that will serve as an enabler for better space exploration.

*Corresponding Author


Dr. Michael Hadjitheodosiou

Center for Satellite & Hybrid Communication Networks ISR,

A.V. Williams Building, University of Maryland, College Park, MD 20742

Tel: +301-405-7904  Fax:+301-314-8586



Additional Author (presentor):
Hui Zeng, Graduate Research Assistant,
Ph.D. Graduate Student
Electical and Computer Engineering
University of Maryland, College Park


To Abstract Index

Building a Reliable Onboard Network

Jane K. Marquart

Systems engineers routinely include the MIL-STD 1553B in the bus design of spacecraft because of the high reliability it provides and the wide use of compatible components available. Attitude control components, such as star trackers, demand strict timing constraints in order to meet pointing requirements of the spacecraft. With rising data rates, however, the 1MByte 1553 bus is hard-pressed to meet these requirements. The choices for a reliable, high-speed bus are limited, so many engineers opt to use the 1553B bus for critical timing data only, adding a second high-speed bus for all other bus traffic. This method works, but two busses add more complexity, testing, cost, etc., to the overall spacecraft. A second option, is to build a reliable, high-speed bus that can handle the escalating data rates AND meet the critical timing requirements demanded by the ACS system. This paper proposes the use of standard Ethernet as the high-speed bus of choice, and describes how the GSFC added reliability features to satisfy onboard timing requirements.

To Abstract Index

Air Force Satellite Control Network Interoperability Project


John Pietras

Global Science and Technology, Inc.


The National Security Space Architect has recommended evolution toward a shared national satellite control network that would provide cost-efficient interoperability between satellite control assets used by the Department of Defense (DoD), the National Aeronautics and Space Administration (NASA), the National Oceanographic and Atmospheric Administration (NOAA), other government agencies, and potentially commercial space users as well.

Achieving such a system would allow any government satellite to be controlled from its associated Satellite Operations Center (SOC) through any government space communication antenna system or Remote Tracking Station (RTS) that is designated to support multi-user operations. The fundamental premise of such an approach is that increased productivity and cost effectiveness of the nation's separate satellite control networks can be achieved via the ability for "cross support" among the various agencies' satellite control resources. A corollary premise is that this interoperability is most possible if there are standard protocols and interfaces between the telemetry, tracking, and commanding (TT&C) resources such that they can be interconnected easily.

In the long term, this interoperability may evolve to be based on end-to-end IP networking. However, the DoD, NASA, and other space-operating organizations have non-IP-enabled spacecraft that will be in operation for the next ten years or longer, and a viable interoperability solution must also address these pre-IP spacecraft. 

The Consultative Committee for Space Data Systems (CCSDS) has developed standards for a suite of Space Link Extension (SLE) services to provide interoperability between SOCs and TT&C systems using Internet-based Wide Area Network (WAN) communications services. The SLE standards are candidates for interoperability between DoD and non-DoD assets. However, national security space activities have demanding requirements for accuracy and completeness of data transfer, controlled delay, time-data correlation, and security, which may not be fully met with the existing CCSDS standards.

The Satellite and Launch Control Systems Program Office of the Air Force Space and Missile Systems Center (SMC) has undertaken an Interoperability Project to evaluate the suitability of SLE and other candidate protocols for supporting national security TT&C functions. Phases 1 and 2 of the Interoperability Project involved the development of adaptations of standard SLE services to support the bitstream-oriented interfaces of current DoD satellite operations, and the testing of those adaptations in an AFSCN environment.

Phase 3 of the Interoperability Project employed commercial, NASA, and NOAA ground stations to support DoD Test and Check-Out (TACO) satellites under the control of the AF Center for Research Support (CERES) SOC. Modified CCSDS SLE data transfer services were used to move command and telemetry streams between the SOC and the ground stations. Also, a prototype scheduling system, based on a draft CCSDS standard for TT&C and SLE service management, was used to schedule support periods from the ground stations. The Phase 3 experiments were generally successful, and they have improved our understanding of how the SLE services can be better used to support DoD satellite missions.

Phase 4 (currently under way) leverages the lessons learned in Phase 3 to refine the adaptation of SLE transfer services to the DoD environment and to impel the transition from experimental prototypes to commercial product-based implementations.  

This presentation summarizes the capabilities exercised in Phase 3 of the Interoperability Project, the results of that period of experimentation, the influence of those results on Phase 4, and the services being prototyped and demonstrated in Phase 4.


To Abstract Index

Space Communications Emulations Facility (SCEF):
Improving NASA Missions through Emulation.


Authors: Rich Slywczak, Allen P. Holtz,
Brenda L. Ellis, Thong Luu,
Fran Lawas-Grodek,
and Cindy Tran
Affiliation: NASA, Glenn Research Center
Presenter: Allen Holtz, NASA, Glenn Research Center
Point of Contact: Rich Slywczak
21000 Brookpark Road, M.S. 54-5
Cleveland, Ohio 44135
Phone: 216.433.3493
Fax: 216.433.8705

NASA scientists are designing more complex and interesting space missions that require new and diverse ways of managing space-based resources.  In the near future, these new missions will challenge the traditional thinking about typical single satellite missions that simply record measurements and then transfer the data to predetermined ground stations.  Communications between satellites will become essential to not only route data between them but also to dynamically share information so that satellites can take more effective measurements.  Mission planners need an effective emulation environment that can model these missions and show the interactions between each of the satellites. 


NASA/Glenn Research Center (GRC) has been developing an emulation environment that will allow mission planners from government, universities or private industries to emulate missions.  Missions can be based on multiple parameters, such as the number of satellites within the mission, communications patterns within the satellites, and the number of relay satellites and ground stations.  Once the missions are emulated, the mission planners will be able to see how effective the satellites communicate within themselves and amongst other satellites based on position and time.  Eventually, researchers will be able to exchange satellite components to hopefully improve timing algorithms, Command and Data Handling (C&DH) software, or scheduling algorithms.  NASA/GRC is currently working with NASA/Ames Research Center (ARC) on modeling intelligent scheduling algorithms.


This presentation will describe the NASA/GRC satellite emulation environment.  It will explore the software and hardware architecture and describe the components of the system.  NASA mission planners will be able to see the advantages of how an emulation system will help them to continue launching successful missions by first resolving the technical risks.  They will be able to use the emulation environment and setup their own scenarios to get the most effectiveness from the system.  Finally, the presentations will describe future enhancements that are being planned for the emulation environment.  The satellite community must become familiar with the tools needed to help them proactively solve their risks before developing missions.  Emulation environments will become one of the more important tools.


To Abstract Index

Does IPv6 Provide Benefits to Space Missions?

Authors: Brenda L. Ellis, Rich Slywczak,
Fran Lawas-Grodek, Cindy Tran,
Larry McFarland, Allen P. Holtz,
and Thong Luu,
Affiliation: NASA Glenn Research Center
Presenter: Brenda L. Ellis, NASA Glenn Research Center
Point of Contact: Brenda L. Ellis
21000 Brookpark Road, M.S. 142-1
Cleveland, Ohio 44135
Phone: 216.433.5214
Fax: 216.433.8000

Future NASA missions are expected to be more complex than those of previous generations. These missions require the use of new functionality which would allow researchers to easily access data aboard a spacecraft. Communication will include a system by which information and resources are dynamically shared.
In order to provide the proper functionality, NASA Glenn Research Center (GRC) has been investigating a satellite network, with IP functionality as a basis of communication. The network will be dynamic and adaptive to ensure simpler access. As the new missions’ architecture designs progress to include IP, it is evident that more addresses will be needed. As with the increased use of the Internet, there became an increase need for security. Such a concern is prevalent in the mission environment as well. The question now arises as to which version of the IP would be more advantageous. Studies have concluded that IPv6 provides features, which includes larger address space, better mobility and security in terrestrial networks. Since IPv6 has proven to be beneficial in terrestrial test beds and simulations, an investigation is currently being conducted to determine if IPv6 is beneficial to the space missions.

This presentation will give an overview of IPv6 and briefly describe its apparent difference from its predecessor, IPv4. It will provide an example of one of the mission architectures that NASA is currently investigating. Finally, the presentation will share views of how beneficial IPv6 can be to such a mission.


To Abstract Index

Routing in Space: Getting Data from Here to There


Authors: Allen Holtz, Richard Slywczak,
Brenda Ellis, Diepchi Tran, Thong Luu, Fran Lawas-Grodek, and Kul Bhasin
Affiliation: NASA, Glenn Research Center
Presenter: Allen Holtz, NASA, Glenn Research Center
Point of Contact: Allen Holtz
21000 Brookpark Road
Cleveland, Ohio 44135
Phone: 216.433.6005
Fax: 216.433.8000

NASA has the mission to "extend human presence across the solar system." First, this will be accomplished by returning to the moon for the first time in many years. From the moon, human presence will continue to Mars and beyond. This presence must be sustained. Consequently, a sustained human presence beyond the orbit of earth will require continuous communications. 

Continuous communication with earth from beyond the earth's orbit can be problematic, especially when the earth based station is no longer in the direct line of sight with t the object in which communication is desired.  This problem can be overcome by dynamically routing communication data among various space based communications objects.  This will enable a crew on Mars to communicate with mission control on earth at any time during the day, eliminating the need to wait until the main communication devices are within the line of sight of each other.


Routing data over multiple satellites is  not straightforward.  Terrestrial based routers are fairly well aware of the routers to which they are connected.  But in space, two satellites must be in the line of sight of each other and their antennas must be pointed toward each other in order to communicate.   In addition, the satellites that are in the line of sight of a satellite at one moment may be different minutes later. 


If we are to extend human presence in space, efficient methods of communication must be developed.  This includes dynamically routing data over satellites and other space based communications objects.  To do this, new routing algorithms that consider the dynamic movement of satellites/routers must be developed and tested.  The NASA Glenn Research Center is developing a test bed that will enable scientists and engineers to develop and test new routing algorithms in a space-like environment.  This presentation will focus on our development of the ability to insert different routing algorithms into the Space Communications Emulation Facility.


To Abstract Index

Web Services for Satellite Emulation Development


Kathy J. Liszka                                                                             Allen P. Holtz

The University of Akron                                                     NASA Glenn Research Center

CAS 227                                                                                             21000 Brookpark Road

Akron, Ohio 44325-4002                                                  Cleveland, Ohio 44135

(216) 433-5110                                                                            (216) 433-6005                                                          




It is critical to capture the requirements of analyzing complex multi-satellite emulation scenarios in a complete and reasonable manner.  Remotely located researchers need a user-friendly interface that collects and automatically encodes their data into a well-formed, complete XML document. We are investigating and developing a web-based architecture for remote access and control of a multi-user, distributed system to emulate space based Internet architectures, backbone networks, formation clusters and constellations. The objective is to define the framework for an open distributed system to make the satellite emulation test-bed accessible through the Internet.


Web services based on standard protocols will be used to control the emulation remotely. We provide an intuitive interface to encode the data into a well-formed and complete XML (Extensible Markup Language) document. XML provides support for portable encoding of data for transfer between dissimilar systems. Scenario specifications include control parameters, network routes, interface bandwidths, delay and bit error rate. Specifications for all satellite, instruments, and ground stations in a given scenario are also included in the XML document.


We use XForms, a recent W3C Recommendation web-based forms language for data collection. Contrary to older forms technology, the interactive user interface makes the science prevalent, not the data representation. Required versus optional input fields, default values, automatic calculations, data validation, and reuse will help researchers quickly and accurately define missions. XML schemas are defined for each test mission to validate data before forwarding it to the emulation facility. New instrument definitions, facilities and mission types can be added to existing schema.


Scenarios are submitted using a Common Gateway Interface (CGI) program. A web server running in the satellite emulation facility executes the CGI program to validate user access rights and schedule emulation resources. Interoperability and dynamic component testing are accomplished with lower level web services.  For example, incorporating standard protocols like the Simple Object Access Protocol (SOAP) means users do not need to change their development environment in order to use the emulation. Scripts isolate one or more components (data collection satellites, high speed communication satellites, ground stations, communication protocols and onboard satellite software) and replace default systems in the emulation system. This presentation will describe these components in the overall design.


To Abstract Index

Handbook for using IP protocols for Space Missions

Operating Missions as Nodes on the Internet (OMNI) Project

Ed Criscuolo - (
Keith Hogie - (
Ron Parise - (


This presentation will provide a summary of a handbook developed at GSFC last year that contains concepts and guidelines for using Internet protocols for space missions.  It will include topics on:


The presentation will also pose questions on what sort of information would be useful in future versions of the document.


To Abstract Index

Including HDLC Framing in CCSDS Recommendations

Operating Missions as Nodes on the Internet (OMNI) Project

Ed Criscuolo - (
Keith Hogie - (
Ron Parise - (


Current Space IP missions use High-Level Data Link Control (HDLC) framing to provide standard serial link interfaces over a space link.  HDLC is the standard framing technique used by all routers over clock and data serial lines and is also the basic framing used in all Frame Relay services which are widely deployed in national and international communication networks.  In late 2003 a presentation was made to CCSDS committees to initiate discussion on including HDLC in the CCSDS recommendations for space systems. 


This presentation will summarize the differences between variable length HDLC frames and fixed length CCSDS frames.  It will also discuss where and how HDLC framing would fit into the overall CCSDS structures.




To Abstract Index

Automated File Transfer and Storage Management Concepts for Space

Operating Missions as Nodes on the Internet (OMNI) Project

Ed Criscuolo - (
Keith Hogie - (
Ron Parise - (


This presentation will summarize work that has been done to prototype and analyze approaches for automated file transfer and storage management for space missions.  The concepts were prototyped in an environment with data files being generated at the target mission rates and stored in onboard files.  The space-to-ground link was implemented using a channel simulator to introduce representative mission delays and errors.  The system was operated for days with data files building up on the spacecraft and periodically being transferred to ground storage during a limited contact time. 


Overall performance was measured to identify limits under which the entire data volume could be transferred automatically while still fitting into the missionŐs limited contact time.  The overall concepts, measurements, and results will be presented.


To Abstract Index

Space-based Telemetry And Range Safety (STARS) Using IP

Operating Missions as Nodes on the Internet (OMNI) Project

Ed Criscuolo - (
Keith Hogie - (
Ron Parise - (


This presentation will describe the design, development, and testing of a system to collect telemetry, format it into UDP/IP packets, and deliver it to a ground test range using standard IP technologies over a TDRSS link.  This presentation will discuss the goal of the STARS IP Formatter along with the overall design.  It will also present performance results of the current version of the IP formatter.  Finally, it will discuss key issues for supporting constant rate telemetry data delivery when using standard components such as PC/104 processors, the Linux operating system, Internet Protocols, and synchronous serial interfaces to.


To Abstract Index

An Efficient Low To Medium Data Rate Network Topology
For Microsats And Nanosats

Jeff Janicik -
Gino Innocenti -

Much work is currently being done in the area of ad-hoc machine-to-machine (M2M) communication and wireless networks for high-speed self-organizing networks. Much of this work can be applied to space networks. This paper will show how SpaceDev is leveraging from advancements in industry networking efforts to arrive at a streamlined approach for a unique situation.

Space wireless networks present some unique problems especially when implemented on small satellites. Among them are power consumption, link quality of service and range. We will address several issues relating to an in-space wireless network for 2 to 6 microsats separated by up to 1000 km. One of the challenges is that the satellites will slew about their elevation and azimuth angles and will require near real time cognizance of each other’s data. Different methods of implementation will be discussed that lead to why our network approach is best suited for a cluster of microsats.


To Abstract Index

Integrated Satellite Control Network
Top-Down and Bottom-Up Implementation

Dr. Richard A. Russel
Chief Architect ­ Satellite Control Network Contract
Northrop Grumman Mission Systems

The Integrated Satellite Control Network (ISCN) will provide interoperability between AFSCN, NASA, NOAA, and Navy antennas. The Satellite Control Network Contract (SCNC) has completed the AFSCN Department of Defense Architecture Framework (DoD AF) essential products that characterize the current AFSCN architecture. By utilizing these products as a starting point, the ISCN architecture can be developed using a Top-Down approach that fully characterizes the Operational Views. SCNC has also developed a detailed Roadmap with supporting Product Descriptions that describe the work needed to evolve the AFSCN. The Roadmap will be used to evolve the AFSCN to ISCN through incremental installations of core capabilities that map to the DoD AF Operational Views. This Bottoms-Up approach will also consider planned upgrades to the AFSCN as well


To Abstract Index


Reusable, Open Source, and Internet Enabled Software Development
Tools for Efficient Small Satellite Design, Mission Planning and Analysis


Author/Presenter: Eric Tapio

Affiliation: Stanford University

Contact Info:  55 Chumasero Drive 9G

San Francisco, California 94132

(415) 335-0737 (phone & fax)


Session Topic: Infrastructure - current & Future Developments & Plans and Implementation



Last year Stanford University's third small satellite, QuakeSat,weighing all of 9.5 lbs. was completed and launched, and is now beingoperated on a daily basis. With a mission to detect earthquakes fromouter space, the complete QuakeSat project from the initial design tothe tested flight unit was completed in just one and a half years.Benefiting from Internet technologies, such utilizing a TCP/IP enabledground station and communication infrastructure, QuakeSat satellite isanother example that small satellites are a viable and cost effectiveplatform for conducting scientific research and space experimentation.Through the use of COTS parts a quick development to flight time isrealizable, resulting a reduction in one of the key drivers in spacemission design - cost.  Nevertheless, further refinement in theefficiency of the initial design processes, and subsystem developmentincluding conducting quantitative satellite design comparisons,variations and utility analyses, as well as subsystem testing andinterfacing still remains a viable challenge in making small satellitedesign to flight time within a year a consistent and repeatable process.


Leveraging off of my design experience with QuakeSat, my research istargeted in further developing and integrating design and analysis toolsthat incorporate Internet TCP/IP technologies to streamline the processof small satellite development, but also aid in mission planning andanalysis.  TCP/IP subsystem design and test tools that were developedfor the QuakeSat project greatly facilitated the effectiveness of thefive-student team to verify the reliability of satellite design, whichwas critical to complete the QuakeSat project in a short period of time.Today, as a part of my research, these tools are tools are being refinedand generalized to support the design efforts of two additional smallsatellite projects at Stanford University, with the intent ofimplementing better processes that succeed in achieving a one yeardesign to launch target timeline.

In my presentation, I discuss these reusable, open source tools beingused and developed, and the Internet enabled infrastructure that Iconstructing to achieve new breakthroughs in the efficient design,development and analysis for small satellites missions to come.


To Abstract Index

NASA Space Network IP Services (SNIS)


David J. Israel
NASA/GSFC Code 567.3
Greenbelt, MD 20771(301) 286-5294


The NASA Space Network (SN) supports a variety of missions using the Tracking and Data Relay Satellite System (TDRSS), which includes ground stations in White Sands, New Mexico and Guam. Space Network IP Services (SNIS) are being developed to support future users with requirements for end-to-end Internet Protocol (IP) communications. These services will support all IP protocols, including Mobile IP, over TDRSS Single Access, Multiple Access, and Demand Access RF links. SNIS will be an operational service for data rates up to 7 Mbps. This presentation will describe the SNIS architecture, present how some SN IP user operational scenarios will be supported, and provide an update on the SNIS

To Abstract Index

The Trade Between CCSDS and HDLC Framing on Global Precipitation Measurement


David F. Everett

Global Precipitation Measurement

Mission System Engineer


mail code 420.2

Greenbelt, MD  20771

(301) 286-1596

fax (301) 286-0232



Global Precipitation Measurement (GPM) will improve climate, weather, and hydro-meteorological forecasts through more frequent and more accurate precipitation measurement from space.  The GPM Core Spacecraft will be built at Goddard Space Flight Center, for a launch around 2010.  GPM compared Consultative Committee for Space Data Systems (CCSDS) framing to High-level Data Link Control (HDLC) framing on the space-to-ground link.  This presentation will describe, from GPM's perspective, the advantages of each approach and the rationale for our selection of HDLC framing.



To Abstract Index

End-to-End Network Models Encompassing Terrestrial, Wireless, and Satellite Components


Ramon P. Williams

Gordon M. Shankman

Christopher M. Ryan

Jonathan M. Pendzick

Chandler L. Boyarko

Charles B. Lambert


Northrop Grumman Information Technology


4801 Stonecroft Boulevard

Chantilly, VA 20151, U.S.A

Development of network models that reflect true end-to-endarchitectures such as the Transformational Communications Architecture needto encompass terrestrial, wireless and satellite component to trulyrepresent all of the complexities in a world wide communications network.Use of best-in-class tools including OPNET, Satellite Tool Kit (STK), PopkinSystem Architect and their well known XML-friendly definitions, such asOPNET Modeler's Data Type Description (DTD), or socket-based data transfermodules, such as STK/Connect, enable the sharing of data betweenapplications for more rapid development of end-to-end system architecturesand a more complete system design.  By sharing the results of andintegrating best-in-class tools we are able to (1) promote sharing of data,(2) enhance the fidelity of our results and (3) allow network andapplication performance to be viewed in the context of the entire enterpriseand its processes.


To Abstract Index

Demonstration of the TASC Rapid Architecture Development and
Evaluation System (TRADES)


Ramon P. Williams

Gordon M. Shankman

Christopher M. Ryan

Jonathan M. Pendzick

Chandler L. Boyarko

Charles B. Lambert


Northrop Grumman Information Technology


4801 Stonecroft Boulevard

Chantilly, VA 20151, U.S.A


We wish to persent several models of End-to-End modeling capabilities that have been developed for the DoD, intelligence, and scientific community.Demonstrate that modeling a space system as a complete end-to-end TCP/IPnetwork is available today and can serve future missions and architectures.


To Abstract Index

An Efficient Layer 2-Mesh Communications Protocol for Space Sensor Networks

L. Clare, E. Jennings, C. Okino and J. Gao
Jet Propulsion Laboratory, 4800 Oak Grove Drive, Pasadena, CA 91109
(818) {354-1650, 354-1390, 393-6668, 354-9528}
email: {Loren.P.Clare, Esther.H.Jennings, Clayton.M.Okino, Jay.L.Gao}

Future space exploration missions will involve multiple spacecraft that cooperatively perform science exploration via multipoint sensing. Use of cross-link communications to relay information to/from Earth users will greatly enhance the efficient and adaptive use of resources, enable time-critical cueing of complementary (e.g., ground-based) assets, and simplify overall systems operation. A layer 2-mesh communications protocol has been designed for space sensor networks that generates the link activation schedule and routes used for efficient traffic relay through the network. The protocol supports anycast transfer to multiple ground stations.

Our model and assumptions are as follows. Due to the large inter-spacecraft distances, directional antennas are used. To keep the cost low, each spacecraft is assumed to be equipped with a single half-duplex transceiver. (However, the protocol can be extended to full-duplex transceivers.) The spacecraft are homogeneous and move along known orbits. Satellites may communicate if they are within a given range and not occluded by Earth, and similarly for satellite-ground station communications; inter-satellite and Earth-satellite maximum range constraints may differ. At any point of time, there is at least one communication path from any spacecraft to Earth. The antenna beamwidths, spacecraft spacing and traffic flows are such that multi-access interference may be ignored.

In a space sensor network, the most common traffic flows are (1) sensor information collection from all spacecraft to the ground station(s), and (2) command/schedule distribution from the ground station(s) to each or all spacecraft. The traffic load distribution is general. These traffic flow patterns can be represented by weighted star graphs. Our main contribution is a heuristic method that prunes the communications topological graph to “map” the traffic graph onto it, deriving which links should be activated and the routes taken to minimize the schedule length. The algorithm finds an embedded tree for each ground station, and then applies the Florens and McEliece scheduling algorithm for tree networks. The underlying tree structure is especially effective in accommodating the large time-bandwidth product links that comprise space-based networks.

The protocol is applicable to any topology. An evaluation has been conducted using a simulation study of 26,000 randomly generated graphs with uniformly distributed random traffic. Latency and throughput performance are derived as a function of the communication ranges and number of base stations. As the communication ranges or the number of base stations increases, the schedule length decreases, although in a non-linear fashion since geometry also plays an important role in traffic load balancing.

The layer 2 mesh protocol may be integrated within a larger space network architecture that also accommodates precision formation flying class distributed spacecraft missions as well as general relay networks supporting heterogeneous space assets and missions.


To Abstract Index

"Advancements in SCPS-TP Technology and Deployments"




Charlie Younghusband (, Eric Edwards (, Joshua Lamorie (

Xiphos Technologies

#800 Đ 3981 St. Laurent Blvd.

Montreal, Quebec

Canada H2W 1Y5




Xiphos will be discussing the current use, new technology development and planned space deployments of its commercial implementation of SCPS-TP: XipLink.  In particular, Xiphos will discuss applications in IP based spacecraft, aircraft and UAVs.  Xiphos will also discuss new technology development; primarily its onboard network controller card, which now supports embedded SCPS-TP/IP networking.  Plans for expected flight operation using this card and SCPS-TP, occurring in late June, will also be described.




To Abstract Index

A cooperative flow control scheme for InterPlaNetary Internet


by Jian Fang and
Dr. Ian F. Akyildiz
Georgia Institute of Technology





With the developments in the space technologies, recently the researchinterests in deep space are raising rapidly. InterPlaNetary Internet isproposed as a network architecture for communications among Mars, Jupiter,Venus, and other planets. InterPlaNetary Internet is mainly characterizedby intermittent connectivity, large and variable delays, asymmetric datarates, and high bit error rates. Even though the communication resourcesare highly scheduled in the InterPlaNetary Internet, congestions can stilloccur due to unexpected link failures and data bursts from opportunisticcontacts, hence the flow control at the bundle layer is an importanceissue for the efficiency of the InterPlaNetary Internet. The mainchallenge of the flow control at the bundle layer is that bundles receivedat a node consume permanent storage and generally cannot be discardedsafely because of the natural of custodial transfer. If a bundle isdropped, the message will be lost and cannot be recovered. In this paper,a cooperative flow control scheme is proposed to solve the flow controlproblem in InterPlaNetary Internet. The network model is based on a newsystem of time intervals. The local information at a node is put into thebundle header and passed to its neighboring nodes. Based on such partialglobal information, each node constructs its own partial global graph,then a rolling horizon algorithm is adopted to schedule the flow in thepartial global graph and to address the dynamic changes in the network.The scheduling is performed by solving the linear programming using aninterior point algorithm. To determine the actual flow over an edge,several neighboring nodes work cooperatively to minimize the deviationsfrom local optimal solutions.  As a result, the cooperative flow controlscheme can avoid local decisions caused by the custodial transfer nature.

To Abstract Index

Title:  Maximizing Data Volume for Direct to Ground Satellite Systems


David A. Carek, P.E.

  NASA Glenn Research Center


David Carek


This presentation examines a conceptual data downlink system developed for the International Space Station for delivering latency tolerant payload data.  The goal of this system is to maximize data volume transferred to the ground.  Various factors influence the quantity of data that can be transmitted over a satellite link, including contact time, transmission rate, protocol efficiency, and the quality of the link.  This presentation looks at the design space associated with these factors and identifies areas that can be optimized to increase data volume.  An overview will be presented on the effects of varying bit error ratios and bit error patterns and how they influence data efficiency relative to higher-level protocols.  In addition, onboard data system bus speeds may limit the selection and applicability of standardized protocols for some high rate applications.


To Abstract Index

Analysis of a Mars Networking Scenario within the Space Based Network Emulator

Shaun Endres and Behnam Malakooti
Case Western Reserve University,
Cleveland, Ohio, 44106
Kul Bhasin and Allen Holtz
NASA Glenn Research Center,
Cleveland, Ohio, 44135

Future deep space missions will benefit from a common, flexible communication network. This network will provide reliable, high speed data delivery to satellites and ground stations positioned around the solar system. A robust architecture, similar to the architecture used in the terrestrial Internet, should be deployed, so that future space missions can cost effectively utilize and extend the capabilities of today’s missions.

In this paper, we present a variety of scenarios built within a network emulator and analyze a future deep space communication architecture based upon the Internet Protocol (IP). This emulator is flexible in the way that it handles the underlying network and the protocols used over the network. Our network emulator is flexible enough to allow modifications to be made to the network, transport, and application layer headers. Here we show the emulator’s flexibility in emulating the underlying network. We present and analyze a couple of scenarios based around communication with Mars. The scenarios analyzed are as follows: a ground station on Earth communicating with a rover on Mars directly, with a rover on Mars through a relay satellite above the rover, and with a rover on Mars through a relay satellite in a Lagrange point.


An Intelligent Internet Protocol Architecture for Space-Based Networks

Behnam Malakooti, Brian Robinson and Ivan Thomas
Case Western Reserve University,
Department of Electrical Engineering and Computer Science
Cleveland, Ohio 44106
Phone: 216-368-4462
E-mail: {Behnam.Malakooti, Brian.Robinson, Ivan.Thomas}


Kul Bhasin and Allen Holtz
NASA Glenn Research Center,
Space Communications
Cleveland, Ohio, 44135
{Kul.B.Bhasin, Allen.P.Holtz}

We present an architecture that extends the capabilities of the Internet Protocol (IP) to facilitate the existence of intelligent decision-making within the network layer. This Intelligent Internet Protocol (IIP) architecture provides functionality that allows for the injecting of intelligent mobile programs within the network layer. These programs are known as intelligent mobile agents (IMAs). An intelligent mobile agent moves from node to node, gathers information, and routes itself based on the gathered information to achieve its purpose. The agent uses the knowledge it acquires to determine future actions. By exploiting IMAs, the IIP architecture will effectively handle volatile, high latency networks; and can seamlessly integrate with IP to enable space-based internet.


Space-BGP: Routing in the Interplanetary Network

Chao Chen and Ian F. Akyildiz

Broadband & Wireless Networking Laboratory

School of Electrical & Computer Engineering

Georgia Institute of Technology, Atlanta, GA 30332

Tel: (404)894-5141, (404)894-6616

Email: _ cchen, ian



The Interplanetary (IPN) network is envisioned by NASA enterprises to provide communication services for scientific data delivery and navigation services for the future deep space missions. The main challenges that affect network layer design in the IPN network are very long and variant propagation delay, intermittent connectivity, and power constraints. Most of these characteristics are unique to the space communication paradigm and lead to different research approaches from those in the terrestrial Internet. The IPN network is composed of multiple autonomous regions (ARs). An AR contains communication entities that are located close (i.e., much shorter than the interplanetary distance) to each other and can communicate among themselves using a single common protocol family. As the routing among different autonomous systems (ASes) in the terrestrial Internet is achieved by the use of the border gateway protocol (BGP), a new routing framework, called Space-BGP, is proposed for routing through different ARs in the IPN network. Inside each AR, specific routing protocols can be developed for intra-AR communications to address its specific challenges. Space-BGP has two integral parts: Space external BGP (SeBGP) and Space interior BGP (SiBGP). The main traffic through the IPN network contains: _ Remote control: The control messages from the Earth to remote nodes at the exploration sites.

_ Data delivery: The scientific data delivery from the exploration site back to the Earth.

According to the traffic types in the IPN network, the following important assumption can be made: The Earth control center is aware of WHEN and WHERE to send the control messages and retrieve the scientific data.

SeBGP addresses the delivery of remote control message and scientific data through ARs in the IPN network. Directional broadcast is proposed for reliable delivery of remote control messages. Specially, paths to the destination are calculated en route by SeBGP based on the predictable AR locations. These paths are used to direct and limit the forwarding area of the control message broadcast. For data delivery from remote exploration site back to the Earth, the route discovery is initiated on-demand by the receiver and routing tables are maintained in soft state at the nodes along the forwarding area. Data delivery follows closer to the optimal path as it approaches the receiver. SiBGP exchanges inter-AR routing information among backbone nodes within an AR and schedule the inter-AR packet transmission. We propose the longest connected queue (LCQ) policy for the contact allocation for border routers and the earliest allocated router (EAR) policy for the scheduling of interregion bundle traffic through an AR. We prove that the combination of LCQ and EAR policies achieves maximum data throughput of an AR and minimum buffering delay of bundles at an AR. Space-BGP is proposed based on the hierarchical architecture and specially address the characteristics of the IPN network. It can be coupled with Space Communication Protocol Standards - Network Protocol (SCPS-NP) by CCSDS or the bundle layer store-and-forward message switching in the Delay-Tolerant Network (DTN) architecture to achieve autonomous routing in the IPN network.

To Abstract Index

ITP: The Integrated Transmission Protocol for InterPlaNetary Internet



Ian F. Akyildiz
Özgür B. Akan

Broadband & Wireless Networking Laboratory
School of Electrical & Computer Engineering
Georgia Institute of Technology, Atlanta, GA 30332
Tel: (404) 894-5141 Fax: (404) 894-7883



The InterPlaNetary (IPN) Internet is currently being envisioned to provide advanced communication and navigation services for the next generation deep space missions. This objective requires the challenges posed by the IPN Internet to be addressed. Current TCP protocols have very poor performance in the IPN Internet characterized by extremely high propagation delays, link errors, intermittent connectivity and asymmetrical bandwidth. Furthermore, these characteristics render the end-to-end congestion control and reliability mechanisms inefficient. Thus, hop-by-hop bundle transport approach is currently being developed for the IPN Internet. However, there is no reliable data transport mechanism to provide hop-by-hop congestion control and reliability functionalities.


In this paper, a new Integrated Transmission Protocol (ITP) is presented for reliable data transport in the IPN Internet. ITP is a unified transmission protocol solution for hop-by-hop congestion control and reliability mechanisms specifically tailored for IPN Internet paths with intermittent connectivity. The objective of ITP is to address the challenges and to achieve high performance reliable data transport on deep space links of the IPN Internet. Exploiting the hop-by-hop nature of the connections, ITP unifies the common functionalities of the conventional transport and link layers. ITP deploys a new rate-based hop-by-hop local flow control (LFC) mechanism; which exploits the local resource availability and traffic information at the receiver in order to provide explicit available bandwidth feedback to the sender. LFC mechanism decouples congestion decision from reliability to avoid the erroneous congestion decisions due to high link errors and avoids the delayed-feedback problem. ITP incorporates the selective-acknowledgment (SACK) based Automatic Repeat reQuest (ARQ) to assure hop-by-hop local packet-level transport reliability. To reduce the effects of blackout conditions on the performance, ITP includes the Blackout Mitigation (BM) procedure. The optimum packet size is analytically obtained to further improve the transmission efficiency over deep space links. Bandwidth asymmetry problem is addressed by the adoption of delayed SACK. Simulation experiments show that the ITP significantly improves the throughput performance and addresses the challenges posed by the IPN Internet.


To Abstract Index

An Integrated Model for Using Dynamic Link
Space Communication Analysis Tool for High Latency
Space Network Emulation of Martian Assets

Yogi Y. Krikorian, Don Lanzinger, and Debra L. Emmons

A model is described in this paper to demonstrate how dynamic communication link information can be used as input into a high latency space network emulation. The dynamic link analysis tool determines the command and telemetry link closure of space and ground assets. These assets could be an Earth orbiting satellite, a Mars orbiting satellite, a lander or rover on Mars, or a ground station on Earth such as the Deep Space Network (DSN). Once Link closure is determined between assets, their relative slant range and propagation delay at each time instant is passed to the high latency space network emulation.

The Aerospace Corporation developed a dynamic link analysis tool to determine link margin for space and ground assets. This tool has currently been successfully applied in support of a number of NASA customers including JPL and NASA Langley Research Center. The model takes into account several dynamic effects in determining link closure such as transmit and receive antenna patterns, space loss, polarization diversity, rain attenuation, and plume attenuation (for space launch vehicles). The trajectory of the launch vehicle or a satellite in addition to its orientation is also taken into account. As a result, a clock and a cone angle are calculated in order to compute the instantaneous antenna gain. Also, a slant range is computed at each instant in time to calculate the corresponding space loss. Finally, a link margin is calculated. Link closure information between assets and their relative slant range is passed to the high latency space network emulation.

One widespread problem in space networks that can be effectively analyzed through software emulation is high latency due to long propagation delays or intermittent connectivity. Intermittent connectivity caused by jamming, weather outages, warfare attrition, or failures result in high latency when messages wait for link availability. This paper presents the applicability of network emulation tools, in particular The Aerospace Corporation Beowulf Network Emulation Tool (BeoNET), to model software emulations. The emulation described in this paper are software programs with strict equivalence to the modeled system. These programs are particularly important to space networking development because space networks are too expensive to deploy incrementally. Many military and NASA communication systems could benefit from standardized solutions to their implicit high latency problems. The protocols of the Delay Tolerant Networking (DTN) Research Group are a good basis for standardized solutions to the high latency problems of these systems. The DTN protocols are currently deficient in the areas of routing, congestion control, and security. The Aerospace Corporation BeoNET tool is being tailored to specifically address these deficiencies.


An Implementation of the CCSDS File Delivery Protocol for the GPM Mission


Art Ferrer

NASA/GSFC, Code 582




Prior GSFC work presented at the "Space Internet Workshop 3" in Cleveland, Ohio,  discussed the use of IP protocols and the Multicast Dissemination  Protocol (MDP) in a "realistic" lab environment.  However, a thorough consideration of both lifecycle support and GPM mission requirements determined that the CCSDS File Delivery Protocol was a better fit for actual flight use.  This presentation discusses the issues involved with the selection of CFDP over MDP, and the modifications that are currently being made to  GSFC CFDP application software and flight software components to support GPM mission specific requirements. 


To Abstract Index

Embedded Transport Agents for Near-Earth Communications

Timothy J. Salo
Architecture Technology Corporation
9971 Valley View Road
Eden Prairie, MN 55344
(952) 829-5864 x133

Embedded Transport Agents will implement new intelligence within the transport layer that will continuously adapt the behavior of transport protocols to the unique requirements and highly dynamic environment of near-Earth space communications. While the initial focus of this project is on enhancing the behavior of the Internet-standard Transmission Control Protocol (TCP), these technologies are equally applicable to other transport protocols, such as the Space Communications Protocol Specification (SCPS) Transport Protocol (SCPS-TP).

Embedded Agents can use exogenous information, such as satellite ephemeris, ground station locations, or the current characteristics of the communications path, to tune the behavior of TCP to match current or predicted conditions. A low-bandwidth secondary data channel associated with, but otherwise independent of, an existing TCP session will enable a pair of Embedded Agents to coordinate their activities. The initial use of Embedded Agents will be to enable TCP to gracefully maintain sessions, transparently to applications, across temporary losses of communications, such as when a satellite passes between ground stations. An Embedded Agent may, for example, inform the local TCP that the session should be suspended because communications is about to be lost due to the motion of the satellite. It may also inform the remote Embedded Agent of the period of time for which communications is expected to be lost, perhaps enabling the remote device to enter a low-power sleep mode for that period. By extending the transport protocol in a manner that is transparent to higher-level protocols, this approach will improve near-Earth communications for all protocols that use TCP, such as FTP or HTTP, and will eliminate the need to modify these applications for space communications.

Embedded Agents provide an ideal platform to support a wide range of TCP "autotune" capabilities. They can easily be extended to mitigate other effects of near-Earth communications, such as variable latencies or bit error rates. Agents could use information maintained by the TCP protocol, such as retransmission timeouts, recent round-trip times, or changes in the round trip times. Likewise, they could use information from other protocol layers, such as transmission errors, the loss of carrier or the results of periodic end-to-end tests initiated by the agents. For example, Embedded Agents could detect a single-hop path between a satellite and a gateway collocated with a ground station, and employ optimizations that might be inappropriate for use across the wider Internet.

This work, funded by NASA Glenn Research Center under SBIR Phase I contract number NNC04CA52C, will develop an architecture and protocols for Embedded Transport Agents and demonstrate a proof-of-concept Embedded Agent. The Contracting Officer Technical Representative (COTR) is Rich Slywczak,, (216) 433-3493.


Jonathan Wilmot

NASA/GSFC Code 580.0

March 26, 2004




                    The GPM Core Observatory is base-lining the use of IP in Space. Additional, the work being done at GSFC's Ethernet Development Laboratory will be used as the basis for GPM's on-board network.

                    The end-to-end technical status of this implementation will be discussed as well as its operational concepts. Issues of peer-to-peer vs centrally controlled networks will be investigated.


To Abstract Index

Current Activities and Status of the ESA-ESTEC Sponsored Wireless Onboard Spacecraft Working Group


SIW Main Page

Responsible NASA Official: Gary Meyers
Technical Contact Person: James L. Rash @ Goddard Space Flight Center
Curator: Owen Kardatzke

Privacy, Security, Notices