Skip to content
Omnikalc

Bit error rate calculator (PSK and QAM)

Pick the modulation, enter the Eb/N0 in dB, a target error rate and a packet length to get the bit error rate, the Eb/N0 you need, the margin and the packet error rate.

Updated on
Your inputs
dB

Value out of range

bits

Value out of range

Results

Bit errors per million bits

9.7362

Eb/N0 needed for the target
10.53 dB
Margin over the requirement
-0.93 dB
Below the target: errors exceed the goal
Packet error rate
0.969 %
Ideal spectral efficiency
2 bit/s/Hz
Matching Es/N0
12.61 dB

The bit error rate calculator for PSK and QAM gives the bit error rate of a modulation from the Eb/N0, the Eb/N0 needed for a target error rate, the margin and the packet error rate.

How the calculation works

  • Pick the modulation and enter the Eb/N0 in dB.
  • Pick the target bit error rate and enter the packet length.
  • The calculator gives the errors per million bits, the required Eb/N0, the margin, the packet error rate and the spectral efficiency.

Example

With QPSK at an Eb/N0 of 9.6 dB (a ratio of 9.12), the BER is Q(√18.2) = 9.7362 errors per million bits, about 10⁻⁵. To reach 10⁻⁶, 10.53 dB are needed, so the margin is -0.93 dB, negative. A 1,000-bit packet is lost 0.969 % of the time. With 16-QAM, more Eb/N0 is needed for the same rate, but 4 bits are sent per symbol.

Why it matters

The calculator helps students in telecommunications, radio engineers and designers of wireless or satellite links who need to know which Eb/N0 a modulation requires for a given quality.

Good to know

The results are theoretical for an additive white Gaussian noise channel with perfect synchronization. Real links lose 1 to 3 dB from implementation losses, and fading makes the error rate fall much more slowly with the Eb/N0. Forward error correction codes cut the required Eb/N0 by several decibels, which is the coding gain. Higher-order modulations carry more bits per symbol but need more power for the same error rate, the basic trade-off between power and bandwidth. At very low error rates, the calculation is limited by the numerical precision, and the output shows zero when the rate is smaller than about 10⁻¹⁵. Use the Es/N0 value when a data sheet specifies the symbol energy rather than the bit energy.

The SNR calculator converts signal-to-noise ratios, the Nyquist sampling calculator relates bandwidth and sampling, the noise figure calculator covers receivers and the normal distribution calculator gives tail probabilities.

Frequently asked questions

What is Eb/N0?

It is the energy per bit divided by the noise power density, the standard measure of signal quality in digital links that lets different modulations be compared fairly. It is given in decibels. Es/N0, the energy per symbol, is larger by 10 log10 of the bits per symbol.

Which formulas are used?

For BPSK and QPSK, BER = Q(√(2 Eb/N0)). For square M-QAM with Gray coding, BER ≈ 4/log2(M) × (1 − 1/√M) × Q(√(3 log2(M) Eb/N0 / (M − 1))). Q(x) is the tail of the normal distribution. These hold on a simple noise channel without fading or coding.

Why does the packet error rate matter?

A packet fails if any bit is wrong, so its error rate is 1 − (1 − BER)^L. A bit error rate of 10⁻⁵ looks small, but loses about 1 % of 1,000-bit packets, and 10 % of 10,000-bit ones, which is why error correction is used.

Was this calculator helpful?

Thanks for your feedback!

Results are estimates provided for information purposes. They do not replace professional advice.