Satellite Transponder Bandwidth and Carrier Planning

Satellite transponder bandwidth is a payload parameter, not a ready-made throughput figure. A usable carrier plan must fit an approved frequency window, satisfy the transponder or payload power/loading model, close its link and interference budgets, comply with regulatory and operator masks, and deliver the contracted service rate at a declared interface.

The word bandwidth is especially dangerous here. Transponder output bandwidth, assigned frequency band, necessary bandwidth, occupied bandwidth, raised-cosine support, carrier spacing, leased MHz, receiver noise bandwidth, and user throughput describe different quantities. They cannot be added, divided, or compared until their definitions and reference planes match.

Carrier Bandwidth Definitions | Carrier Spacing | Link Budget

Quick Answer: Keep Two Resource Ledgers

A frequency fit does not prove a power fit, and neither proves throughput.

accepted payload plan
  = regulatory and operator frequency authority
  + carrier spans, centers, gaps, and edge clearances
  + payload gain, total loading, and per-carrier output plan
  + uplink/downlink C/(N+I) and interference closure
  + waveform, framing, and service-throughput calculation
  + controlled line-up and witnessed acceptance

Reject any capacity table that supplies only 36 MHz, a peak MODCOD, or saturated beam EIRP. Each omits at least one release gate.

Define the Payload Boundary First

Recommendation ITU-R S.1328-5 separates parameters that informal transponder explanations often combine. Its system-characteristics templates record, among other items:

• transparent or remodulating transponder type;
• uplink and downlink occupied bandwidth per carrier;
• transponder output bandwidth;
• access and modulation type;
• per-carrier uplink and downlink EIRP;
• transparent-transponder transmission gain and automatic level-control range;
• internal and external C/I terms; and
• long- and short-term C/(N+I) requirements.

That Recommendation is intended for FSS sharing studies, not as a lease specification. Its value here is the parameter separation: output bandwidth alone cannot replace the carrier, power, interference, and performance fields.

S.1328-5 describes a transparent transponder as a bent-pipe frequency translation and a remodulating transponder as one that demodulates to baseband. Do not assume every payload is a fixed analogue channel with one input filter, frequency converter, and dedicated HPA. The operator document must state how channels, beams, polarizations, amplifiers, gain control, digital processing, and resource pools map to the capacity being offered.

Use transponder on this page as an operator-identified payload path or channel. For flexible or processed payloads, the relevant commercial and engineering resource may instead be a channel, beam, polarization, gateway/user-link pair, capacity pool, or transponder equivalent.

Eight Bandwidth Terms That Are Not Interchangeable

In the ITU occupied-bandwidth definition, the frequencies below the lower limit and above the upper limit each contain a specified percentage β/2 of total mean power. With the default ITU percentage, 99% of total mean power lies between the limits. This is a measured-emission concept. It is not automatically Rs(1+α), a lease width, or the span from the first carrier edge to the last.

For a multi-carrier emission, internal gaps may lie inside the composite frequency span even though they do not contain carrier power. Calling “sum of carrier bandwidths plus guards” occupied bandwidth therefore obscures the ITU definition. Use required plan span for frequency geometry unless an applicable document defines another term.

Build the Frequency Ledger

For carrier i, use one declared planning-span definition. For a raised-cosine screening calculation:

B_i = Rs_i × (1 + alpha_i)

For N carriers inside an approved window W, with left and right clearances E_L and E_R, and internal edge-to-edge gaps G_i:

B_required = E_L + sum(B_i) + sum(G_i) + E_R

B_slack = W - B_required

For adjacent carriers i and i+1:

center spacing = B_i/2 + G_i + B_(i+1)/2

These equations prove only geometric fit. The operator-approved clearances and gaps must include the applicable emission masks, frequency tolerance, Doppler where relevant, transponder filtering and translation error, carrier power imbalance, nonlinear spectral regrowth, receiver selectivity, and required adjacent-carrier margin.

Reproducible three-carrier example

Assume only for the arithmetic:

• operator-approved window W = 34.000 MHz;
• Carrier A: Rs = 10 Msymbol/s, α = 0.20;
• Carrier B: Rs = 8 Msymbol/s, α = 0.20;
• Carrier C: Rs = 6 Msymbol/s, α = 0.10;
• approved internal gaps G_AB = 0.600 MHz, G_BC = 0.500 MHz;
• minimum edge clearances E_L = E_R = 0.700 MHz.

The waveform planning spans are:

B_A = 10(1.20) = 12.000 MHz
B_B =  8(1.20) =  9.600 MHz
B_C =  6(1.10) =  6.600 MHz

B_required = 0.700 + 12.000 + 0.600 + 9.600 + 0.500 + 6.600 + 0.700
           = 30.700 MHz

B_slack = 34.000 - 30.700
        = 3.300 MHz

If the unused 3.300 MHz is split equally between the two edges, each actual edge clearance becomes 2.350 MHz. Relative to the lower edge of the approved window:

The final right clearance is 34.000 - 31.650 = 2.350 MHz, so the arithmetic closes. No absolute RF frequency is implied. The example does not prove that the gaps, symbol rates, filters, payload, power plan, equipment, or emissions are acceptable on a real satellite.

Build the Power and Loading Ledger Separately

Frequency span and mean carrier power are separate axes. A carrier can fit in MHz and still fail because of payload drive, EIRP/PSD, nonlinearity, C/I, cross-polarization, adjacent-satellite limits, or receive threshold.

At one common reference plane, mean carrier powers add linearly:

P_total,linear = sum(10^(P_i,dB / 10))

P_total,dB = 10 log10(P_total,linear)

If three carrier mean powers are 0 dB, -3 dB, and -6 dB relative to the same reference:

linear sum = 1 + 10^(-3/10) + 10^(-6/10)
           = 1 + 0.501187 + 0.251189
           = 1.752376

total = 10 log10(1.752376)
      = 2.4363 dB relative to the strongest carrier

Likewise, four equal mean-power carriers total 10 log10(4) = 6.0206 dB above one carrier. Do not add dB values directly, and do not mistake this arithmetic for a back-off setting. The actual payload operating point depends on the amplifier/processor, gain state, ALC, waveform peaks, number and levels of carriers, linearization or predistortion, and operator loading rules.

Record at least:

The Satellite Link Budget page owns the end-to-end C/N calculation. This page owns the check that every carrier in the payload plan uses a consistent operator-approved power reference.

Back-Off Is Not a Universal Percentage

IBO and OBO are differences from declared input- and output-saturation references. They are not generic deductions that can be copied from another transponder.

A nonlinear payload can create gain compression, AM/AM and AM/PM distortion, spectral regrowth, and multi-carrier intermodulation. The required operating point changes with:

• single- or multi-carrier waveform and peak statistics;
• carrier count, relative levels, spacing, and modulation;
• TWTA, SSPA, linearized amplifier, digital processor, or shared-amplifier architecture;
• predistortion, crest-factor reduction, equalization, and ALC state;
• passband amplitude/group-delay behavior and memory effects; and
• the required C/(N+I), spectral mask, and out-of-band performance.

Therefore a statement such as “multi-carrier OBO is 2–5 dB” is not an engineering input. Use the operator payload model or measured transfer characteristic, simulate the actual loading, and verify it during line-up. A single carrier is not automatically allowed to operate at saturation, and multiple carriers are not automatically assigned equal power.

Spacing and Emission Compliance Are Different Tests

Recommendation ITU-R SM.1541-7 addresses unwanted emissions in the out-of-band domain, including single- and multi-carrier cases and space-service masks. Recommendation ITU-R SM.328-12 supplies the spectrum and bandwidth terminology. Neither turns a generic guard-band percentage into an approved carrier plan.

For every carrier pair, evaluate:

• modulator output mask and measured spectrum;
• center-frequency accuracy, drift, and translation tolerance;
• payload passband, ripple, group delay, and filtering;
• relative carrier PSD and nonlinear regrowth/intermodulation;
• receive filtering and demodulator adjacent-channel behavior;
• adjacent transponder, polarization, beam, and satellite conditions; and
• regulatory, coordination, operator, and lease edge rules.

Lower roll-off narrows the ideal raised-cosine support for the same symbol rate. It does not by itself prove smaller approved gaps, additional billable capacity, or acceptable interference. Detailed geometry remains in Carrier Spacing.

Convert Carrier Configuration to Rate—Not MHz to “Internet Speed”

For a standard that publishes spectral efficiency normalized to symbol rate:

R_standard = Rs × eta_Rs

eta_RC = eta_Rs / (1 + alpha)

ETSI EN 302 307-1 Table 13 lists 1.487473 bit/symbol for DVB-S2 QPSK 3/4 with a normal FECFRAME and no pilots. At Rs = 10 Msymbol/s:

R_standard = 10 × 1.487473
           = 14.87473 Mbit/s

With α = 0.20, the raised-cosine planning span is 12.000 MHz, so the efficiency normalized to that span is:

eta_RC = 14.87473 / 12.000
       = 1.23956 bit/s/Hz of B_RC

This is a standard-layer example, not IP throughput and not total transponder capacity. Frame length, pilots, input mode, BBFRAME data field, padding, encapsulation, packet headers, idle/dummy frames, ACM distribution, packet loss/retransmission, contention, and the measured service interface can change the delivered rate.

Never multiply transponder output bandwidth by a MODCOD label without declaring:

1. which carrier symbol rates fit the approved plan;
2. the exact standard table, frame, pilots, and roll-off;
3. which MODCODs close the link and for what percentage of time;
4. power, interference, and payload nonlinear losses;
5. service and encapsulation overhead; and
6. whether the result is one carrier, one beam, one polarization, or an aggregate reused system figure.

A Transponder Is Not the Whole Satellite Capacity

The same frequency may be reused across beams or orthogonal polarizations under a coordinated payload design. Conversely, channels can share amplifiers, gateways, feeder links, power, processing, or interference limits. A bandwidth figure must therefore carry its scope.

Do not multiply a channel width by a beam count and call the result usable capacity unless reuse, power, gateway, interference, scheduling, waveform, and traffic constraints are all included.

Operator Evidence Package

Before configuring a modem or transmitting, obtain a dated package for the named satellite and payload path:

• approved transmission plan or booking reference;
• satellite, transponder/channel, beam, polarization, and gateway/path identifiers;
• exact uplink/downlink center frequencies and inversion/translation rules;
• authorized carrier span, symbol rate, roll-off, waveform, FEC/MODCOD, pilots, and emission mask;
• edge clearances, adjacent carriers, frequency tolerance, PSD and EIRP limits;
• payload gain/ALC state, loading, per-carrier output allocation, and back-off definitions;
• link and interference budgets with clear-sky/faded operating states;
• terminal model, approvals, pointing/cross-polar results, and control interlocks;
• service-rate definitions, contention/traffic policy, monitoring point, and SLA; and
• line-up procedure, NOC contacts, stop-transmission authority, rollback, and change record.

Eutelsat's public space-segment access material is one operator example: its line-up information requires a valid transmission plan, the correct satellite/transponder, transmit frequency or carrier ID, polarization, and operating parameters such as EIRP, bit rate, and FEC. Apply the actual procedure of the selected operator; do not treat one operator's document as universal.

Controlled Line-Up and Acceptance

Before radiation

1. Confirm regulatory authorization and the operator-approved plan revision.
2. Validate frequency conversion, spectrum inversion, symbol rate, roll-off, waveform, and polarization offline.
3. Load power limits, mute/fail-safe behavior, remote control, and escalation contacts.
4. Verify instrument calibration, reference planes, RBW/VBW/detector settings, and measurement uncertainty.

During line-up

1. Follow NOC instructions and begin at the authorized test condition.
2. Verify satellite/channel identity, pointing, cross-polar performance, center frequency, and carrier ID.
3. Measure carrier spectrum, EIRP/PSD, frequency error, C/N or C/(N+I), and adjacent effects at the declared points.
4. Increase drive or enable traffic only as instructed; stop immediately on operator command or abnormal interference.

After activation

1. Confirm user-interface throughput separately from DVB and RF rates.
2. Exercise ACM/power-control states, traffic loading, alarms, mute, restoration, and rollback where applicable.
3. Preserve the final plan, payload state, modem configuration, measurements, screenshots/traces, exceptions, approver, and date.

Common Mistakes

FAQ

What is satellite transponder bandwidth?

Recommendation ITU-R S.1328-5 uses transponder output bandwidth as a space-station parameter. For design or procurement, add the exact payload path, frequency limits, beam, polarization, direction, configuration, gain/loading state, and date. The number alone does not state an approved carrier window or throughput.

Is a 36 MHz transponder exactly 36 MHz usable?

Do not assume so. 36 MHz may be a nominal channel label, physical parameter, lease unit, or transponder equivalent. Request the exact operator-approved frequency window, edge rules, payload response, power allocation, and commercial definition. There is no defensible universal 5–10% deduction.

How many carriers fit in one transponder?

First calculate the frequency geometry using compatible carrier-span definitions, approved gaps, and edge clearances. Then validate total and per-carrier power, nonlinear distortion, interference, link margin, equipment support, and operator acceptance. A carrier count from MHz alone is incomplete.

How does bandwidth become throughput?

Calculate each carrier from its symbol rate and the exact waveform/frame spectral-efficiency definition. Then apply pilots, framing, encapsulation, padding/idle behavior, ACM distribution, traffic policy, and service-interface measurement. There is no fixed Mbps value for a transponder width.

Why is transponder back-off required?

Back-off can be part of controlling payload nonlinearity, distortion, spectral regrowth, and intermodulation. The required value depends on the actual payload and loading. Use the operator model and measured line-up rather than a generic dB range.

Can one carrier fill the whole transponder?

Only if the approved plan, payload passband, power/loading model, waveform, mask, link/interference budget, equipment, and lease allow it. A single-carrier configuration is not automatically authorized to use every nominal MHz or the saturated output condition.

What is a transponder equivalent?

It is a commercial or planning normalization, often expressed as a bandwidth-equivalent quantity. It need not map one-to-one to a physical filter or amplifier. The contract must define bandwidth, power, beam/polarization, time, pre-emption, and performance rights.

Summary

Transponder bandwidth planning is a four-stage proof:

1. define the exact payload and regulatory boundary;
2. fit the carriers inside the approved frequency window;
3. close the payload power/loading, link, and interference cases; and
4. verify waveform-to-service throughput during a controlled operator line-up.

The useful result is not “a 36 MHz transponder carries X Mbps.” It is a dated carrier and power plan whose definitions, calculations, operator approvals, measurements, and service boundary can all be audited.

Related Articles

• Symbol Rate and Roll-Off — Raised-cosine span, regulatory occupied bandwidth, spacing, and throughput definitions
• Satellite Carrier Spacing — Center-frequency geometry and adjacent-carrier acceptance inputs
• Satellite Link Budget Calculation — Per-carrier EIRP, G/T, C/N, interference, and margin
• C/N, C/N0, Eb/N0 and Es/N0 — Ratio definitions and conversions at declared bandwidth reference planes
• Carrier-in-Carrier — Frequency-overlap geometry, power limits, product support, and ROI boundaries
• Satellite Network Topology — End-to-end paths, hop count, control, and failure boundaries