ELEC6103 Chap.11 Multiple Access, Small Satellites and NGSO Systems
Multiple Access, Small Satellites and NGSO Systems
Sharing one transponder among many stations
A satellite earns its keep by serving many users at once. The simplest layer of sharing is already built into the payload, since separate transponders sit on separate frequencies. Inside one transponder the course names three ways of dividing capacity: by frequency, by time and by code.
Only the first two are worked numerically; code division is named as a method and the official course information for this offering does not state how its codes are constructed or detected, so it is not developed here.
Frequency division in practice
Under FDMA every station gets its own carrier frequency and keeps it.
All stations in the footprint receive every downlink carrier, and each receiver uses a bandpass filter to keep the one it wants. The lecture calls this the most common commercial form, largely because filtering is a mature technology.
Each channel needs its occupied bandwidth plus a guard band, so dividing the transponder bandwidth by the channel width gives the number of stations.
Time division and its fixed overhead
TDMA lets each station use the entire transponder in a short burst within a repeating frame.
Every burst carries the same preamble and guard times however little data it holds, so slow stations waste most of the frame on overhead.
In the tutorial a 54 MHz transponder serves 360 stations at 80 kbps by FDMA but only 32 by TDMA, and the gap closes as the rate rises.
A low-orbit handheld system, priced per minute
The longest tutorial applies the link chapters to a 50-satellite constellation at 750 km linking L band handsets to Ka band hubs. A wide-beam handset uplink barely fails to close, which forces a multi-beam redesign.
Dividing the total system cost by circuit-minutes sold over ten years gives about eight cents a minute, and the discussion shows why that figure rests on optimistic load assumptions.
New space and large constellations
Lecture 1 contrasts traditional geostationary platforms with satellites of a few hundred kilograms in lower orbits.
Starlink, with over 5,000 satellites launched by January 2024, illustrates the scale, relying on laser inter-satellite links so traffic need not land at a gateway beneath every satellite.
What this chapter covers
- 01
Three named methods of sharing a transponder
- 02
FDMA carriers, bandpass filters and fixed assignment
- 03
Guard bands and the FDMA station count
- 04
TDMA frames, preambles and guard times
- 05
Why FDMA wins for low-rate stations in the tutorial
- 06
A 50-satellite LEO handheld system and its link budgets
- 07
Cost per circuit-minute and its load assumptions
- 08
Traditional space against new space and large constellations
Worked example · free
FDMA station count on a 72 MHz transponder
- 2QPSK carries 2 bits per symbol, so a 128 kbps station needs 128 / 2 = 64 ksps.
- 2Channel width = 1.25 x 64 + 100 = 80 + 100 = 180 kHz, so 72,000 / 180 = 400 stations.
- 2At 512 kbps: 256 ksps, channel width 1.25 x 256 + 100 = 420 kHz, and 72,000 / 420 = 171.4, so 171 whole stations.
- 2Guard band share: 100 / 180 = 56 per cent of each 128 kbps channel, against 100 / 420 = 24 per cent at 512 kbps.
Key terms
- Multiple Access
- Any scheme that lets many earth stations share the capacity of one satellite or transponder; the course names frequency, time and code division as the three methods.
- Frequency Division Multiple Access
- Sharing in which each station is assigned its own carrier frequency and bandwidth within the transponder, with guard bands between channels and filters selecting the wanted carrier.
- Time Division Multiple Access
- Sharing in which each station transmits a burst at the full transponder rate during its own slot in a repeating frame, paying a preamble and guard times per burst.
- Cost per Circuit-Minute
- Total system cost divided by the circuit-minutes the system is assumed to sell over its life; in the LEO tutorial it comes to roughly 8 cents.
- Inter-Satellite Link
- A direct link between satellites in a constellation, carried by laser at about 100 Gbps in the lectures, so traffic can be relayed without a ground gateway under each satellite.
Multiple Access, Small Satellites and NGSO Systems FAQ
Why does TDMA serve so few slow stations in the tutorial?
Each burst carries the same 148-bit preamble plus a pair of guard times regardless of payload. A station sending 80 kbps contributes only ten data bits per 125 microsecond frame, so roughly 97 per cent of its slot is overhead. FDMA has a fixed guard band too, but at 150 kHz per channel that cost is far smaller in proportion, giving 360 stations against 32 on the same 54 MHz transponder.
What changed in the handheld LEO design after the first link budget?
With a single 120 degree beam the satellite antenna gain is only about 3.6 dB, and the handset uplink ends up near -0.3 dB C/N, so the link does not close. The redesign splits coverage into several narrower beams with higher gain. The Ka band hub links had tens of decibels to spare, so the power-limited handheld direction is what sets the beam count.
Is eight cents per minute a realistic price for that LEO service?
Only if the assumed traffic actually materialises. The figure divides about $4,318M of satellites, operations and overhead by 200 busy circuits on every working satellite for ten years. Only a few satellites cover any region at a time, phones lose signal indoors and interest accrues while subscribers are still being signed up, so real revenue per minute must be considerably higher to recover the investment.
Exam move
Rework the tutorial FDMA and TDMA counts with pencil until you can do each in under five minutes, writing the symbol rate, channel or burst size, and overhead as separate labelled lines. Then change one input at a time, such as the transponder bandwidth, roll-off or data rate, and predict which method gains before calculating.
For the LEO handheld system, list each link with its band, power and limiting factor, and redo the cost per circuit-minute with a lower load to see how sensitive it is. Keep code division to a single sentence naming it as the third method, and spend the saved time on the new space contrasts and Starlink figures from Lecture 1.
Working through Multiple Access, Small Satellites and NGSO Systems in ELEC6103? Sia is AskSia’s AI Electrical Engineering tutor — ask any ELEC6103 Multiple Access, Small Satellites and NGSO Systems question and get a clear, step-by-step explanation grounded in how ELEC6103 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.