UHF RFID Antenna Gain Reference: dBi, dBd, dBiC and dBiL Converted

Four units describe the same antenna: dBi, dBd, dBiC and dBiL. A European reader is rated in e.r.p., a US one in e.i.r.p. Get one conversion right and you land exactly on the legal ceiling with every available decibel of read range working for you.

This page converts all four, both directions, with the arithmetic shown. It carries the formula the harmonised standard actually prints — P_C = P_erp − G_IC + 5,15 + C_L — and the axial-ratio correction that gives the exact dBiC-to-dBiL offset where the familiar “dBiC minus 3” rule gives a close approximation. Every row has an anchor, so a forum answer can link to one line. Every figure traces to a document cited at the end.

The conversion table, as text

Two reference frames matter. e.i.r.p. (equivalent isotropically radiated power) is referenced to an isotropic radiator. e.r.p. (effective radiated power) is referenced to a half-wave dipole, and is 2.15 dB lower for the same real transmission. Europe, and India, regulate in e.r.p.; the United States regulates conducted power plus antenna gain, which lands you in e.i.r.p.

Start from the conducted power at the reader port, P_C in dBm, and subtract total cable loss C_L in dB. Then apply the row for whichever unit your datasheet uses.

Gain quoted in	To e.r.p. (dBm)	To e.i.r.p. (dBm)
dBi	PC − CL + G − 2.15	PC − CL + G
dBd	PC − CL + G (unchanged)	PC − CL + G + 2.15
dBiL	PC − CL + G − 2.15	PC − CL + G
dBiC	PC − CL + G − 5.15	PC − CL + G − 3

The dipole constant is 2.15, and here is why

The directivity of a lossless half-wave dipole is 1.641 — antenna-theory.com states it as “1.64 (2.15 dB)”. In decibels: 10 × log₁₀(1.641) = 2.151, which rounds to 2.15 dB. That is the whole derivation. The harmonised standard’s own constant, 5,15, is 2,15 + 3 — so the standard itself settles the rounding question.

dBm to milliwatts, printed in full

dBm	mW	W
20	100.0	0.100
23	199.5	0.200
27	501.2	0.501
28	631.0	0.631
29	794.3	0.794
30	1 000.0	1.000
31	1 258.9	1.259
32	1 584.9	1.585
33	1 995.3	1.995
34	2 511.9	2.512
35	3 162.3	3.162
36	3 981.1	3.981

Note the rounding convention everyone uses: 2 W e.r.p. is written as 33 dBm, although 33 dBm is strictly 1 995.3 mW, and 2 000 mW is 33.01 dBm. The standard writes “2 W e.r.p. (33 dBm e.r.p.)” and that is the figure to design to.

What each unit is referenced to

Every gain figure is a ratio against some reference antenna. Name the reference and the unit stops being ambiguous.

• dBi — gain relative to an isotropic radiator, a theoretical point source radiating equally in all directions.
• dBd — gain relative to a half-wave dipole. Because the dipole already has 2.15 dBi of directivity, G[dBi] = G[dBd] + 2.15.
• dBiL — linear gain, referenced to a linearly polarised isotropic source.
• dBiC — circular gain, referenced to a circularly polarised isotropic source.

Nikitin and Rao put the last pair most cleanly: linear gain “is referenced to a linear isotropic source and measured in dBil, while circular gain is referenced to a circularly polarized isotropic source and measured in dBic.”

The 2.15 dB that sits between e.r.p. and e.i.r.p.

EN 302 208 defines effective radiated power as “the product of the power supplied to the antenna and its gain relative to a half wave dipole in the direction of maximum gain” (clause 4.3.3.2). India’s 2021 rules carry the same half-wave-dipole reference at rule 2(1)(c), defining e.r.p. as the power supplied to an antenna times its gain in a given direction relative to a half-wave dipole. So the dipole frame is legally load-bearing in both jurisdictions, and every compliance figure has to be stated against it.

Run the ceilings through the conversion and the apparent similarity resolves into three distinct numbers:

• EU and India lower band, 2 W e.r.p. (33 dBm e.r.p.) = 35.15 dBm e.i.r.p. = 3.273 W e.i.r.p.
• EU upper band, 4 W e.r.p. (36 dBm e.r.p.) = 38.15 dBm e.i.r.p. = 6.531 W e.i.r.p.
• US 47 CFR 15.247, 1 W conducted with a 6 dBi antenna = 36 dBm e.i.r.p. = 33.85 dBm e.r.p. = 2.427 W e.r.p.

Two ceilings both labelled “4 W” — the EU upper band and the US figure — differ by 2.15 dB because one is e.r.p. and the other e.i.r.p. A circularly polarised panel is specified in dBiC because that is what it radiates into a circular reference; quoting that same number as dBi overstates it by 3 dB, and against a dipole by 5.15 dB.

The formula in the harmonised standard: 5,15 dB

Most gain-conversion references stop at the rules of thumb. The harmonised standard prints an actual formula, and testing laboratories use it. From ETSI EN 302 208 V3.4.1 (2023-12), clause 5.5.3.2.3, Step 4, formula (1):

P_C = P_erp − G_IC + 5,15 + C_L   dBm

The standard defines its terms directly beneath the formula:

• P_C = interrogator conducted transmit power in dBm;
• G_IC = antenna gain of a circular antenna in dBic;
• C_L = total cable loss in dB.

Note the comma: ETSI writes decimals in the European style, so 5,15 means 5.15. That constant is 2.15 (dipole-to-isotropic) plus 3 (circular-to-linear), folded into one number so a test house applies a single constant. The lead-in sentence is explicit about direction — the formula calculates “the allowed conducted power with a circularly polarized antenna.” You give it a legal e.r.p. target and it returns the reader port setting.

Worked at the 2 W ceiling

Assume the lower-band ceiling of 33 dBm e.r.p., 0.5 dB of total cable loss, and an antenna whose beam-width qualifies it for the 2 W step (the beam-width ladder is the next section).

• 9 dBiC panel: P_C = 33 − 9 + 5.15 + 0.5 = 29.65 dBm (923 mW conducted)
• 5 dBiC panel: P_C = 33 − 5 + 5.15 + 0.5 = 33.65 dBm (2 317 mW conducted)
• 4 dBiC panel: P_C = 33 − 4 + 5.15 + 0.5 = 34.65 dBm (2 917 mW conducted)

The buying conclusion falls straight out of those three lines. A 9 dBiC panel reaches the legal ceiling with the reader port backed off to under 1 W conducted. A 4 dBiC panel asks for almost 3 W at the port to reach the same ceiling, which is why the panel is the component that sets the budget. The antenna, not the reader port, is usually what caps your e.r.p. Specify the panel first, then set the reader to suit it.

Because India’s 2021 rules name EN 302 208 as the reference standard for the RFID entry, the same formula governs deployments there. Indian sites run the identical four channels at the identical 2 W e.r.p., so the arithmetic above transfers without change — useful when you are commissioning warehouse portals across both regions from one configuration sheet.

Beamwidth sets the legal power ceiling

This is the clause integrators most often look for after the fact. In EN 302 208, how much e.r.p. you are permitted is gated by the beamwidth of the antenna, not only by the power figure. A wide-beam panel sits on a lower step of the ladder than a narrow one.

Clause 4.3.4.2 defines it precisely: “The beam-width of an antenna is the angle between the two half-power (-3 dB) points of the main lobe, when referenced to the peak effective radiated power of the main lobe.”

Clause 4.3.4.3 then sets the ladder. Lower band, 865 MHz to 868 MHz:

e.r.p.	Beam-width limit
≤ 500 mW	No restriction
> 500 mW to ≤ 1 000 mW	≤ 180°
> 1 000 mW to 2 000 mW	≤ 90°

Upper band, 915 MHz to 921 MHz, the same ladder shifted one step:

e.r.p.	Beam-width limit
≤ 1 000 mW	No restriction
> 1 000 mW to ≤ 2 000 mW	≤ 180°
> 2 000 mW to 4 000 mW	≤ 90°

Applied to beam-widths you can actually order

• A 105° panel exceeds 90°, so in the lower band it sits on the 1 W step. At 0.5 dB cable loss and 5 dBiC, that is P_C = 30 − 5 + 5.15 + 0.5 = 30.65 dBm.
• An 80° panel clears 90° and reaches the 2 W step.
• A 70° panel clears it comfortably.

So a coverage decision — wide beam for a forgiving read zone, narrow beam for a tight one — is simultaneously a power-budget decision with a 3 dB swing attached. Clause 5.5.4.2, Step 7, specifies that the measured figure is the horizontal beam-width, taken by rotating the antenna to the two points where the signal drops 3 dB. Where a datasheet quotes beamwidth in one plane, that is the plane the conformance test reads.

The exact dBiC-to-dBiL offset, by axial ratio

The flat “subtract 3 dB” rule holds exactly at an axial ratio of 0 dB, which describes perfect circular polarisation. Real antennas sit above that, and the true offset shrinks as the axial ratio grows.

Nikitin and Rao give the exact relation at equation (3) of their updated manuscript:

G[dBic] = G[dBil] + 3 + 20 log₁₀((1 + 10^(−A/20)) / 2)

where A is the axial ratio in dB. Their own note on the rule of thumb is that the 3 dB figure holds only for perfect circular polarisation, which practical antennas rarely achieve. Evaluating equation (3) for realistic axial ratios shows the gap is consistently narrower than 3 dB.

Evaluating that expression gives the real offset between dBiC and dBiL:

Axial ratio A	dBiC − dBiL offset
0 dB	3.000 dB
0.5 dB	2.754 dB
1 dB	2.514 dB
1.5 dB	2.282 dB
2 dB	2.057 dB
2.5 dB	1.840 dB
3 dB	1.629 dB
4 dB	1.228 dB

What that is worth in range

• A 9 dBiC panel with 1 dB axial ratio is 6.49 dBiL, where the flat rule gives 6.0 dBiL. That is 0.486 dB recovered, a range factor of 1.058×.
• A 4 dBiC panel with 2 dB axial ratio is 1.94 dBiL, where the flat rule gives 1.0 dBiL. That is 0.943 dB recovered, a range factor of 1.115×.

Free-space range scales with the square root of power, so a dB difference d becomes a range factor of 10^(d/20). Eleven per cent of range is the difference between a pallet reading reliably at the far edge of a portal and reading it intermittently.

One expert note, because the two numbers genuinely differ and both are right. EN 302 208’s 5,15 constant assumes perfect circular polarisation by design — a fixed legal constant has to be deterministic and conservative so that any two test houses reach the same figure. Equation (3) describes the physics. Use 5,15 for the compliance calculation and equation (3) for the link budget. They are answering different questions.

The axial-ratio unit trap

Axial ratio appears on datasheets in two incompatible forms, and the symbols look similar enough to swap by accident.

• In decibels — “2 dB typical”, “1 dB typical”. This is the quantity equation (3) expects.
• As a voltage ratio — “1.5:1”. This is a voltage ratio, and it converts to 3.52 dB.

Convert it and the gap is large: 20 log₁₀(1.5) = 3.52 dB. Run both through equation (3):

Datasheet text	Axial ratio in dB	dBiC − dBiL offset
“1.5 dB”	1.5 dB	2.28 dB
“1.5:1”	3.52 dB	1.42 dB

Reading a voltage ratio as though it were decibels credits your link budget with 0.87 dB more than the antenna actually delivers. Before using any axial-ratio figure, check whether the datasheet printed a unit after the number. If it printed a colon, convert first.

A sourced comparison of real shipping panels

Every figure below is taken from the manufacturer’s own datasheet, with the revision code quoted, rather than from a reseller listing. Where a listing and the datasheet differ, the datasheet revision is the figure to put in a specification.

Part	Band(s)	Gain	3 dB beamwidth	VSWR	Axial ratio	Front-to-back	Revision
Times-7 A6031	865–868 / 902–928 MHz	4 dBiC typical	80° both planes	1.4 typical	2 dB at boresight	−18 dB	v2.7
Times-7 A5020CP	865–868 (ETSI) / 902–928 (FCC)	5 dBiC typical	105° both planes	1.4 typical	2 dB typical	−10 dB typical	V2-03/24
Laird S9028PCR	902–928 MHz	9 dBic	70° azimuth	1.3:1 max	1 dB typical	20 dB	ANT-DS-...-0515

Three observations an integrator can act on.

Band coverage decides which rulebook applies before any arithmetic starts. Both Times-7 parts state 865–868 MHz, which covers the ETSI lower band and all four of the channels India permits at 2 W e.r.p. — centred at 865.7, 866.3, 866.9 and 867.5 MHz — as well as 902–928 MHz. The Laird S9028PCR datasheet states 902–928 MHz, so it is an FCC-band part and 47 CFR 15.247 governs it. Read the band line first, then pick the formula.

The S9028PCR sheet gives azimuth beamwidth only. Since EN 302 208 clause 5.5.4.2 measures the horizontal beam-width, that is the figure a lower-band conformance test addresses — and where you need the elevation figure for a coverage model, request it from the manufacturer rather than inferring it.

Polarisation handedness belongs in the specification. All three are right-hand circular in the variants quoted; the S9028 family also ships left-hand. Matching handedness across a portal keeps the full polarisation efficiency available to the link.

Worked conversions you can re-run

Each example states its inputs and then uses exactly those inputs. Substitute your own figures and the method holds.

Example 1 — a linear antenna, both frames

Inputs: 6 dBi linear antenna, 30 dBm conducted at the reader port, 1 dB total cable loss.

1. Power at the antenna: 30 − 1 = 29 dBm
2. e.i.r.p. = 29 + 6 = 35 dBm e.i.r.p. = 3.162 W
3. e.r.p. = 35 − 2.15 = 32.85 dBm e.r.p. = 1.928 W

That sits just under the 2 W e.r.p. ceiling, with 0.15 dB of headroom — thinner than most cable-loss estimates, so measure the cable rather than assuming it.

Example 2 — a circular panel via formula (1)

Inputs: a lower-band 9 dBiC panel, target of 33 dBm e.r.p., 0.5 dB cable loss.

1. P_C = P_erp − G_IC + 5,15 + C_L
2. P_C = 33 − 9 + 5.15 + 0.5 = 29.65 dBm = 923 mW conducted
3. Beamwidth check: the 2 W step requires ≤ 90°, so specify the panel at 90° or narrower.

Example 3 — the reverse direction

Inputs: the same lower-band 9 dBiC panel and 0.5 dB cable, reader port left at a default 30 dBm. Rearranging formula (1) gives P_erp = P_C + G_IC − 5.15 − C_L.

1. P_erp = 30 + 9 − 5.15 − 0.5 = 33.35 dBm e.r.p. = 2.163 W
2. That is above the 2 W (33 dBm) ceiling. Back the port off to 29.65 dBm to land on it.

High-gain panels are where this reverse check earns its keep: the reader ships at a default that a 9 dBiC panel turns into an over-limit transmission. Always run the reverse calculation after fitting a new antenna.

Example 4 — one 30 dBm port, each panel against its own applicable ceiling

Inputs: 30 dBm conducted at the reader port, 0.5 dB total cable loss.

Lower band, 865–868 MHz. e.r.p. = 30 + G − 5.15 − 0.5, and the clause 4.3.4.3 beam-width ladder decides which ceiling applies to each part.

Panel	Gain	3 dB beamwidth	Resulting e.r.p.	Applicable ceiling	Against it
A6031	4 dBiC	80°	28.35 dBm (684 mW)	2 000 mW (≤ 90° step)	34%
A5020CP	5 dBiC	105°	29.35 dBm (861 mW)	1 000 mW (> 90°)	86%

One extra decibel of panel gain plus a wider beam moves the A5020CP from roughly a third of its budget to 86% of it. The beam-width ladder, not the gain figure alone, is doing most of that work.

FCC band, 902–928 MHz. The S9028PCR datasheet states 902–928 MHz, so 47 CFR § 15.247 governs rather than EN 302 208. Paragraph (b)(3) permits 1 W (30 dBm) conducted for digitally modulated systems; paragraph (b)(4) requires the conducted power to be reduced, dB for dB, by the amount the antenna’s directional gain exceeds 6 dBi. Taking the datasheet’s 9 dBiC figure as the directional gain — the conservative reading — the reduction is 3 dB.

Panel	Gain	Permitted conducted power	Port setting at 0.5 dB cable	Resulting e.i.r.p.
S9028PCR	9 dBiC	30 − 2.5 = 27.5 dBm (562 mW)	27.5 dBm	36 dBm (3.981 W)

So the same 30 dBm default lands differently in each band: in the lower band it is comfortably inside the ladder for an 80° panel, and in 902–928 MHz the reduction is set by the directional gain net of cable loss: 9 dBiC less 0.5 dB of cable is 8.5 dBi effective, which exceeds 6 dBi by 2.5 dB. That is why we specify the antenna and the destination market before the reader on every portal we quote as an RFID hardware manufacturer and exporter, and why our UHF reader models are supplied with ETSI-band and FCC-band configurations set per order.

Primary documents behind this page

The documents behind every figure on this page, with the citation details as printed on each one.

• ETSI EN 302 208 V3.4.1 (2023-12) — Radio Frequency Identification Equipment operating in the band 865 MHz to 868 MHz with power levels up to 2 W and in the band 915 MHz to 921 MHz with power levels up to 4 W. Clauses used: 4.3.3.2, 4.3.3.3.1, 4.3.4.2, 4.3.4.3, 5.5.3.2.3, 5.5.4.2.
• P. V. Nikitin and K. V. S. Rao, “Antennas and Propagation in UHF RFID Systems”. The copy cited here states on its first page that it is “an updated version of the manuscript published in IEEE RFID 2008 conference proceedings (Las Vegas, NV, April 16-17, 2008)”, so the equation numbering used above — equation (3), the gain relation — is that of the updated manuscript.
• G.S.R. 853(E), Ministry of Communications, New Delhi, the 10th December 2021 — Use of Low Power Equipment in the Frequency Band 865-868 MHz for Short Range Devices (Exemption from Licence) Rules, 2021, made in supersession of the 2005 rules. Rule 2(1)(c) and the RFID entry naming EN 302 208.
• 47 CFR § 15.247 — paragraph (b)(3), 1 W conducted for digitally modulated systems in 902–928 MHz; paragraph (b)(4), dB-for-dB conducted power reduction above 6 dBi directional gain.
• antenna-theory.com, half-wave dipole page — the directivity figure of 1.64 (2.15 dB) used in the dipole-constant derivation.
• Manufacturer datasheets: Times-7 A6031 v2.7; Times-7 A5020CP V2-03/24; Laird S9028PCL/S9028PCR, document code ANT-DS-S9028PCL S9028PCR 0515.

Further reading cited by Nikitin and Rao

Nikitin and Rao cite three sources for the gain relation at equation (3), as references [42–44]. The bibliographic details below are reproduced as printed in that paper’s own bibliography, which is where they were read:

• B. Toh et al. [42], “Understanding and measuring circular polarization”, IEEE Trans. on Education, vol. 46, no. 3, Aug. 2003, pp. 313–318.
• T. Milligan [43], “Polarization loss in a link budget when using measured circular-polarization gains of antennas”, IEEE Antennas and Propagation Magazine, vol. 38, no. 1, Feb. 1996, pp. 56–58.
• Chelton Microwave [44], “Gain correction factor to convert from dBiL to dBiC”, technical note.

For completeness: the same paper’s reference [41], IEEE Standard 149-1979, is cited at its equation (1), for the mutual polarisation efficiency expression.

Frequently asked questions

How do I convert dBiC to dBi?

For the legal and datasheet-comparison case, subtract 3 dB: a 9 dBiC panel is treated as 6 dBi equivalent linear gain. To reach e.r.p. instead, subtract 5.15 dB in total (3 for circular-to-linear, 2.15 for isotropic-to-dipole). For a physical link budget, use the axial-ratio-corrected figure from Nikitin and Rao's equation (3) rather than the flat 3 dB.

Is dBiC minus 3 dB the right conversion?

It is exact at an axial ratio of 0 dB, meaning perfect circular polarisation, and real antennas sit above that value. At an axial ratio of 1 dB the real offset is 2.514 dB; at 2 dB it is 2.057 dB. Use the flat 3 dB for the EN 302 208 compliance calculation, which deliberately assumes ideal circular polarisation, and the corrected value from equation (3) for range prediction.

What is the 5.15 dB constant in EN 302 208?

It is the constant in formula (1) of clause 5.5.3.2.3, Step 4, written by ETSI as 5,15 in the European decimal style. The formula is P_C = P_erp - G_IC + 5,15 + C_L dBm, where P_C is the interrogator conducted transmit power, G_IC the circular antenna gain in dBic and C_L the total cable loss. The constant is 2.15 (dipole to isotropic) plus 3 (circular to linear) combined into one figure.

How do I turn an antenna gain figure into a legal e.r.p. setting?

Start from the conducted power at the reader port, subtract total cable loss, add the antenna gain, then apply the unit offset: subtract 2.15 dB for dBi or dBiL, subtract 5.15 dB for dBiC, or subtract nothing for dBd. Compare the result against your regional ceiling, which is 33 dBm e.r.p. in the ETSI lower band and in India, subject to the beam-width ladder. If you exceed it, rearrange formula (1) to find the conducted setting that lands exactly on the limit.

What is the difference between dBiL and dBiC?

Both are referenced to an isotropic source, but to differently polarised ones. dBiL is referenced to a linearly polarised isotropic source and dBiC to a circularly polarised one. The gap between the two figures for the same antenna is 3 dB at perfect circular polarisation, and narrows as axial ratio rises: 2.514 dB at 1 dB axial ratio, 2.057 dB at 2 dB.

Why is a dipole 2.15 dBi rather than 2.14 dBi?

The directivity of a lossless half-wave dipole is 1.641, published by antenna-theory.com as 1.64 (2.15 dB), and 10 x log10(1.641) = 2.151, which rounds to 2.15 dB. The harmonised standard settles the question in practice: EN 302 208 uses a combined constant of 5,15 dB, which is 2,15 plus the 3 dB circular-to-linear term.

Does antenna beamwidth affect how much power I am allowed to transmit?

Yes, and this is the clause most often looked up after the fact. EN 302 208 clause 4.3.4.3 ties the permitted e.r.p. to beamwidth. In the lower band there is no beamwidth restriction up to 500 mW e.r.p., a 180 degree limit above 500 mW up to 1 000 mW, and a 90 degree limit above 1 000 mW up to 2 000 mW. A 105 degree panel therefore sits on the 1 W step, while 80 degree and 70 degree panels reach the 2 W step.

Does the same gain arithmetic apply to a US deployment?

The gain conversion does, but the ceiling it runs against changes. In 902-928 MHz, 47 CFR 15.247(b)(3) sets 1 W conducted for digitally modulated systems, and (b)(4) reduces that conducted figure dB for dB where directional gain exceeds 6 dBi. So a US calculation caps conducted power at the antenna rather than radiated e.r.p., and the EN 302 208 beam-width ladder has no counterpart there. Read the band line on the antenna datasheet first, then pick the rulebook.

Sources

• ETSI EN 302 208 V3.4.1 (2023-12) - RFID equipment, 865-868 MHz up to 2 W and 915-921 MHz up to 4 W
• Nikitin & Rao, Antennas and Propagation in UHF RFID Systems - updated manuscript of the paper published in IEEE RFID 2008 proceedings (Las Vegas, NV, April 16-17, 2008); equation (3) and references [42-44]
• G.S.R. 853(E), 10 December 2021 - Low Power Equipment 865-868 MHz Short Range Devices (Exemption from Licence) Rules, 2021; rule 2(1)(c)
• 47 CFR 15.247 - Operation within the bands 902-928 MHz, 2400-2483.5 MHz, and 5725-5850 MHz; paragraphs (b)(3) and (b)(4)
• Laird S9028PCL / S9028PCR Circular Polarity RFID Panel Antenna datasheet (ANT-DS-S9028PCL S9028PCR 0515)
• Times-7 A6031 Circular Polarized UHF Antenna datasheet v2.7
• Times-7 A5020CP Circular Polarized Antenna datasheet V2-03/24
• Antenna Theory - Half-Wave Dipole Antenna: directivity 1.64 (2.15 dB)