| ||||||||||||||||||||||||||||||||||||||||||||||||||
Satellite Earth Stations |
No | Application | Terminal | Typical Circuit Speed |
1 | Calculations (Basic) | Teletype | 10 chrs/sec |
2 | Typewriter | 15 chrs/sec | |
3 | Online Chat | Application built display | 1.2 to 4.8 kbps |
4 | Video tape | VDU | 1.2 to 4.8 kbps |
5 | Web-based education | Data transfer | 9.6 kbps |
6 | Circuit Design CAD/CAM | Graphic Terminal | 56kbps |
While these transmission speeds are typical depending on the application, the duty cycle or the average time the circuit is busy in a given period, is extremely important. This factor viz. the peak to average ratio is important because, if a high speed fixed rate transmission channel is chosen for a high peak to average bursty data terminal, the channel would be used inefficiently. In such situations, the channel should be such that it transmits in bursts with a very short response time. In this sense, the choice would be for a packet satellite network as compared to circuit switched channels.
The few basic advantages of satellite packet networks may be listed as:
Costs are independent of distance of location
Possibility of point to multi-point or broadcast transmission facilities.
Provision of receive only facilities corresponding to 1/2 circuit costs.
Availability of end-to-end digital transmission with arbitrarily low error rates using VSATs [2].
Provision of asymmetric data rate channel unlike terrestrial telephone networks. This can result in savings in cases of question and answer sessions as in data collection systems.
Ease of network design and reconfiguration of VSATs to suit changing network requirements.
Ability to multiplex data together on a common channel independent of geographic locations by time, frequency and code division multiplexing unlike data multiplexing on terrestrial circuits, which is done by co-located multiplexers or concentrators.
A mesh architecture provides full interconnectivity with single hop links, and high G/T earth stations. Although higher C/I earth stations would permit distributed network functions with higher transmission bit rates and with lesser transmission delay (due to single hop), such a scheme is not suitable for VSAT network because of the very small size of the earth station. VSATS cannot communicate with each except through a central station called ‘Hub’ because of their low E.I.R.P. capabilities.
Thus full connectivity of VSATS requires using double hop links with Hub as a control station. Such a star network incorporates complicated network control schemes for earth stations [3] but permits small station VSATs to be located at urban centres, either or roof tops or in backyards. Since two frequencies are involved at the satellite transponder associated with VSAT to Hub direction (in bound) and Hub to VSAT direction (out bound), two different access schemes are associated with each of these two carriers.
Packet networks using terrestrial media have been in evolving stages for more than a decade. Packet switching which is a form of message switching as opposed to circuit switching is advantageous for high peak to average, bursty traffic as in computer communication [4]. Packet switching using satellites have special features:
Broadcast nature of satellite transmission as opposed to point to point transmission in terrestrial circuits
Reduced switching functions at nodes and
Higher flexibility and reliability as compared to terrestrial based packet transmission.
Substituting a satellite circuit for a terrestrial circuit in a packet network without changing equipment or software at the terminals can have major disadvantages all related to propagation delay [5]. These are:
1. The circuit throughput can be severely degraded,
2. Mechanism regulating flow or pacing of data can be interfered with,
3. Response time of devices can be lengthened; sometimes terminals may cease working.
To avoid these problems, the following control procedures will have to be followed for data transmission on satellite circuits:
1. Stop and wait ARQ should be avoided and protocols such as Binary Synchronous line control may be used.
2. Continuous ARQ with pull back as used in High Level Data Link Control (HDLC); Synchronous data Link Control (SDLC) is efficient only if a suitable frame sized is employed.
3. When a high bit rate for transmission is used with HDLC protocol, a high value of M (>127) as well as a low bit error rate channel is needed. In other words, the satellite link shall be engineered to give a low error rate performance.
4. Selective repeat ARQ can be employed provided proper link control equipment is used.
5. When pacing and flow control mechanisms are used for distributed processing; computer networks should be properly selected [6].
6. Polling of satellite circuits for interactive systems is to be avoided.
7. Although terrestrial protocols work well over satellite, they do not take advantage of broadcast capabilities of satellite links.
The service objectives of any satellite data network would be to provide for
Interactive terminal to computer working
Distributed resource sharing
File transfer
Question and answer sessions to and from a central source
Electronic Mail/Facsimile.
With the above service objectives in mind and the capability of satellite to broadcast information from a point to several points, various access schemes suitable for packet switched networks are described below. These accesses are suitable only for DATA and in a few special cases digitised voice or slow scan/freeze frame video.
The most prominent ones are:
Fixed Assignment Time Division Multiple Access (F – TDMA)
Random Access
i. Pure Aloha
ii. Slotted Aloha
iii. Implicit reservation
iv. Explicit reservation
Hybrid schemes
Code Division Multiple Access (CDMA)
In a fixed assignment TDMA, each frame is divided into slots of fixed time duration among stations of the network. The assignment of stations to slots is permanent similar to TDMA systems carrying digitised voice except F-TDMA does not have network synchronisation - Packets are sent asynchronously, with no frame sync. signals. TDMA itself is a flexible multiple access scheme and can carry digitised voice, DATA and video of widely different capacities from each station. Them is no intermodulation problems caused by multi-carrier working. Consequently transponder utilisation is the highest. A typical frame is of the order of 1msec or more. The only synchronisation required is that the burst from stations must arrive at satellite exactly in the allocated slots without overlap.
8.2.1 Pure Aloha. In the simplest form, also called pure or unslotted Aloha. Stations transmit packets randomly and packets from different stations may collide. Stations retransmit the packets until they are received correctly. To avoid repeated overlaps, time interval of packet transmission is randomised. Prof. Abramsan and others gave analysis of the Aloha channel throughput [4,7] in term of traffic offered as S = Ge-2G where S is the average no. of packets transmitted successfully and G is the average no. of packet attempted to be transmitted.
8.2.2 Slotted Aloha. It is seen that the maximum throughput of an unslotted Aloha channel is limited to 18% (Fig. 1) due to collision and to reduce the probability of such collision, time slots are introduced so that the transmission can begin only at the start of slots. This network discipline reduces the collision and hence increases the maximum throughput efficiency of the channel. In S-Aloha, each station has 2 queues – the new packet queue and the retransmit packet queue.
Only if the retransmit queen is empty, a new packet is sent. The analysis of a slotted Aloha channel shows that S = Ge-G and maximum channel throughput is 36%. It may however, be noted that the bands (in Fig. 1) apply to large uniform networks. It is quite common to see a kind of dynamic reservation in slotted Aloha. Reservation of slots are monitored by all stations and synchronised by maintaining tables showing outstanding transmission requirements. The Reservation table is used by the channel scheduler to assign future slots on demand-access Round Robin-Fashion.
8.2.3 Implicit Reservation (Reservation via S-Aloha). In this form of slot reservation, it is indicated only by the use of slot in a frame time, slots having high traffic rates have one or more slots by reservation. These stations are removed from contention from the remaining slots. Control is distributed at each station based on global information of the network [8]. When a station uses the slot successfully by contention in a particular frame, this slot is assigned to that station in each successive frame till it stops using it. The frame time must at least be equal to the lime of transmission of a single hop, otherwise there are instabilities. This scheme has a higher throughput than either S-Aloha or F-TDMA depending on traffic.
.
8.2.4 Explicit Reservation. This is a form of reservation scheme implemented on a TDMA system. A Network scheduler makes a distinct assignment of slots to users in a TDMA frame. The slots not claimed by original allotee may be re-assigned on a Round Robin bases to stations having traffic to send.
There are many other schemes essentially based an either TDMA or Aloha but implementing features like Priority Oriented Demand Assignment (PODA) or a contention based Priority Oriented Demand Assignment (C- PODA) etc. All the schemes however will generally take into account the following one or more features:
1. Efficient use of satellite Bandwidth
2. Satisfaction of multiple Delay constraints
3. Can incorporate multiple priority levels
4. Handle variable packet (message) lengths
5. Handle different transmission rates
6. To have fairness in allotting slots to nodes. That is to say that one node does not use the slots all the time.
7. Incorporate efficient message acknowledgement procedures [8].
8. Network is robust in operation.
A comparison of the various access schemes described so far is represented in Table 2 incorporating Delay vs. Normalised Throughput in a 3-station configuration.
Spread spectrum technology was initially applied to military and radio astronomy applications. Only a few years ago, spread spectrum technology could be used by VSATs with no complexities or cost penalties. In addition it is also claimed to offer certain advantages.
The general features of a CDMA system are:
1. All stations operate on the same transponder frequency using a larger bandwidth than needed for the data rate.
2. The network needs no time or frequency coordination.
3. Provide anti-Jam capabilities or protection against interference.
4. Provide for a graceful degradation of network performance as the number of simultaneous users increase
5. Low spectral density compared to conventional emissions
The mostly widely used CDMA technique for VSAT application is Direct sequence (US) or Pseudo-Noise (PN) sequence modulation using a chip sequence to represent I or 0 of data bits. Frequency hopping and chirp modulation techniques are not so common yet for VSAT application. While advantages offered by CDMA techniques are unquestionable [10] (like low power spectral density and interference resistance of VSATs), certain drawbacks of SSMA techniques can not be left without mention. These are:
1. Larger transponder bandwidth requirements
2. Due to imperfect code orthogonalities, expected simultaneous users may be much limited.
3. Results in a highly complex Central Earth Station for the star configuration.
4. VSAT technology using SSMA is presently available only for a narrow band segment to cater to low bit rate applications.
Satellite teleconferencing is technology used to send a one-way video broadcast from one site to many sites through the use of satellite equipment. This one-way video broadcast is made interactive through the use of telephones and fax machines. Satellite teleconferencing is a one-way video, two-way audio (1V-2A) experience where participants can see and hear the presenter, but cannot be seen by the presenter and can interact with the presenter only through the use of other audio media such as telephone or fax. Satellite teleconferencing should not be confused with "videoconferencing" which is a two-way video, two-way audio (2V-2A) technology in which all parties are able to see and hear each other in real-time.
satellite uplink equipment and a production studio are required to produce your own satellite teleconference
satellite downlink equipment is required to receive (downlink) satellite teleconference events
satellite programs are purchased from the program provider in the form of site licenses
satellite programs are made interactive through the use of phone/fax by participants
satellite programs can often be videotaped (if authorized by the provider) for later viewing
The producer of the teleconference leases satellite time from a satellite owner and uplinks its program to the satellite at the time of the broadcast (Fig.2). The broadcast can then be down-linked by sites with appropriate satellite equipment – these sites simply need to be given the satellite downlink coordinates in order to do so. Once the program is down-linked to a particular site, participants at that site can view the broadcast and there is usually time allotted during the broadcast for phone/fax questions from participants to the program presenters.
The comparison of various access techniques is provided in Table 2.
ACCESS | THROUGHPUT EFFICIENCY | AVERAGE RESPONSE TIME | COMMENTS |
Pre Assigned TDMA | 68% | 2.0sec or greater | Lowest complexity Excellent efficiency Excessive response time |
Slotted | 10% | 0.45sec | Low complexity Poor efficiency Moderate response |
Reservation TDMA | 50% | 1-2sec | High complexity Moderate efficiency Moderate response |
CDMA with Thresholding | 80% | 0.25sec | Complex Best combination of efficiency and response time |
Satellite teleconferencing has been very expensive until now. It would not be cost-effective for most distant learning centers to use uplinks to originate distance-education classes unless the centers were in a position to market classes over wide geographic areas. It is reasonable, however, for a learning center to use a downlink to receive commercial courses that are delivered through satellite channels. One example of an educational system that makes use of satellite communication is EMG (Educational Management Group). Using the access techniques described above, various overhead costs in satellite communication for distance learning may be diminished or eliminated. In addition, as per the requirements of different distance learning centers, various access techniques cited above may be used.
[1] Ananasso, F. Delli Pricoli, "The Role of Satellites in Personal Communications Services," IEEE Journal on Selected Areas in Communications," V 13, N 2, (February 1995), pp.180-195.
[2] Jun Sun and Eytan Modiano, "Capacity Provisioning and Failure Recovery for Low Earth Orbit Satellite Networks," International Journal on Satellite Communications, June, 2003, pp. 278-279.
[3] E.V. Hoversten and H. L. Van Trees, "International Broadcast Packet Satellite Services," ICCC Conference Proceedings, Kyoto, Japan, September 1978.
[4] N. Abramson, "Packet Switching with Satellites," AFIPS Conference Proceedings, NCC, 1973, pp. 695-702.
[5] I.M. Jacobs, et al, "Packet Satellite Network Design Issues," Proceedings, NTC, November 1979.
[6] D.B. Hoang and K.J. Pye,Computer Communication Networks — Lecture Notes, School of Electronic Engineering, La Trobe University, 1995 Edition, pp. 86-88.
[7] W.W. Chu, et al, "Experimental Results on the Packet Satellite Network," Proceedings of the NTC, November 1979.
[8] Eytan Modiano, "Scheduling Algorithms for Message Transmission Over a Satellite Broadcast System," MILCOM 97, Monterey, CA, November 1997, pp. 629-631.
[9] Eytan Modiano, "Resource allocation and congestion control in next generation satellite networks," GBN 2001, Anchorage, AK, April 2001.
[10] Wenhua Jiao, "A Voice/Data Integration MAC Protocol for Ku Band CDMA VSAT Networks", Orlando, Florida, USA, IEEE Vehicular Technology Conference, VTC 2003 FALL, Sep. 2003
Jivesh Govil | Jivesh Govil is a final year student at Netaji Subhas Institute of Technology (DU), India. He is actively involved in several projects in the field of Wireless Communication and Analog Research. He has attended several international conferences and participated in seminars at IITs and other places also. He has undergone industrial training in GSM/CDMA/IN Systems (MTNL). He was the only one to be selected for internship in Satellite Communications at INSAT Network Operation Control Centre (BSNL) in winter 2005. At such a young age, he has already 16 publications to his credit. He was selected amongst top 1% candidates by TATA Institute of Fundamental Research (TIFR) in the field of physics. He was awarded the President Gold Medal Award in high School for best performance in all the major subjects. He was selected for short course in “Digital Communication” at Massachusetts Institute of Technology (USA). He is an IEEE and IEE member. Institute address: Correspondence Address Ph:+91 11 27181067 |