Protocols for Inter-Satellite Communication in a Formation Flying System

The simulated performance comparisons between IEEE 802.11a and wireless 1394, which has not yet been standardized, for use in a formation flying satellite system are presented. The system collects data in the form of files and relays the files to a terrestrial station. We consider two configurations for the formation flying system. The difference between the configurations is the traffic load simulated within the formation. The performance of 802.11a and W1394 are compared for the two configurations. From the simulation results, it may be concluded that 802.11a should be used within the formation when data throughput is the primary concern. However, when the client needs short response times W1394 may be the appropriate choice. INTRODUCTION In order to reduce the cost and design time for formation flying satellite missions, commercial offthe-shelf (COTS) communication technologies and their hardware and protocols could be used. For these missions, it is advantageous if the upper layer protocols are compatible with the terrestrial Internet. Thus, TCP/IP for the transport and networking layers, respectively, are obvious choices. Data link and physical layer protocols that may be used include IEEE 802.11 and the yet to be standardized version of wireless IEEE 1394, which is referred to as W1394 in this paper. These formation flying missions, also known as constellation systems, have several advantages over single mission systems. First, having more than one satellite provides redundancy in the case of a node failure. Second, using autonomous nodes that exchange and compile data can increase the throughput of data to the ground. Third, spatial coverage for applications such as distributed aperture remote sensing may not be accomplishable using a large, heavy, single-mission satellite. There are several requirements that need to be fulfilled by any communications system to be used in formation flying missions. These requirements are based on the intensive data collection and near realtime control communications. The first requirement is high data rates in the range of multiple Mbps. The satellites in the system will have to be tightly controlled to achieve the precise position in which data is collected. Also, due to the autonomous requirements, large amounts of data will have to be transmitted between the nodes in order to perform the computations and data compression so that information sent to the ground can be minimized. The second requirement is multiple access. This allows the nodes to share the same medium, thus reducing the number of transceivers per satellite. Multicast and broadcast messaging will also be necessary. Using a MAC layer should allow for ad hoc networking, and, thus, the ability for nodes to be added or dropped without disrupting communications between other nodes. The third and last requirement for communications in a formation flying system is acknowledgements at the data link layer. This will improve reliability of transfer and will shorten the time that it takes for a node to respond to a corrupted frame. This paper presents the simulated performance analysis and comparison of two data link layer protocols in a formation flying context. The two data link layer protocols compared are IEEE 802.11a and W1394. Both use FTP at the application layer over TCP/IP, and both protocols fulfill the requirements stated earlier. We consider two configurations for the formation flying system. The difference between the configurations is the traffic load simulated within the formation. The performance of 802.11a and W1394 are compared for the two configurations. 20th AIAA International Communication Satellite Systems Conference and Exhibit 12-15 May 2002, Montreal, Quebec, Canada AIAA 2002-1960 Copyright © 2002 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved. 2 American Institute of Aeronautics and Astronautics PROTOCOLS The data link protocols used in these simulations are IEEE 802.11a and W1394, which is not yet a standard. The features of these protocols are summarized in Table I. Also included in this table, for reference purposes, is Bluetooth. Bluetooth is a low power, low cost, and low data rate (less than 1 Mbps) protocol and is another potential candidate as it evolves. However, in the research presented in this paper Bluetooth was not included, since the data rate is too low. IEEE 802.11 specifies the data link and physical layers on the protocol stack. Three different physical layers are supported: direct sequence spread spectrum (DSSS), frequency hopping spread spectrum (FHSS) and infrared. The IEEE 802.11b standard supports data rates of 1, 2, 5.5 and 11 Mbps, while IEEE802.11a supports data rates up to 54 Mbps. In the simulations reported in this paper, IEEE 802.11a (54 Mbps) was used. The medium access control (MAC) for 802.11a is based on CSMA/CA (Carrier Medium Access with Collision Avoidance). Table I Wireless Data Link Protocols Attributes W1394 IEEE 802.11 Bluetooth Installations (related to satellite communications or industrial control) Unknown 1. Used in Caltech's multi-vehicle test bed. 4 1. ABB is experimenting with using for mobile monitoring stations such as PDAs in industrial settings 5 2. Used in Caltech's multi-vehicle test bed 4 Space Qualified No No No Authentication Unknown No No Privacy Unknown Yes, 128-bit cipher key (however, shown to have security weaknesses ) Yes, 128-bit cipher key 7 Arbitration 1. Asynchronous – reliable transfer, no guarantee of bandwidth 2. Isochronous guarantee of bandwidth, no guarantee of reliable transfer 8 Carrier Sense Medium Access with Collision Avoidance (CSMA/CA) 9 1. Synchronous Connection Oriented (SCO) 2. Asynchronous Connectionless (ACL) using master/slave architecture 10 Deterministic Yes (Isochronous mode) 8 No Yes 10 Max Data Rate ~ 100 Mbps 11,12 54 Mbps (802.11a) 13 780 Kbps 10 Max Distance ~ 10 Meters 12 ~300 Meters (without amplification) 10 meters normally, 100 meters with amplification 10 Max # of Nodes (without hub) 63 8 2048 per access point, 20 max recommended 14 8 active nodes (1 master, 7 slaves) + 256 'parked' (not active) nodes 10 Application Layer Unspecified Unspecified Unspecified Transport Layer TCP, UDP , None TCP, UDP, Unspecified Specified, TCP/UDP compatible 10 Network Layer IP , None IP, Unspecified Specified, IP compatible 10 Data Link Layer Specified 8 Specified 9 Specified 10 Physical Layer Specified 8 Specified 9 Specified 10 Frequency band Unknown possibly IR, 2.4 or 5 GHz 2.4 or 5 GHz 9,13 2.4 GHz 10

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.