Quantum Metrology & the Heisenberg Limit
Entanglement-enhanced sensing and its limits.
Measure a phase with independent probes and your error falls as : the standard quantum limit. Entangle the probes and the bound becomes : the Heisenberg limit. This lesson proves both bounds from the quantum Fisher information, shows how GHZ and NOON states reach the second, and explains why noise usually takes most of the advantage away.
What you’ll learn
- Define the classical and quantum Fisher information and state the quantum Cramér–Rao bound.
- Derive the standard quantum limit for product probes and prove the Heisenberg limit for arbitrary probes.
- Show that GHZ and NOON states attain the Heisenberg limit, and explain the phase-ambiguity problem and its resolution.
- Prove that uncorrelated Markovian dephasing removes the entanglement advantage in frequency estimation, and describe when error correction helps.
Fisher information and the Cramér–Rao bound
A probe state depends on an unknown phase . Measuring it gives outcome with probability , and you repeat times. Any unbiased estimator obeys the Cramér–Rao bound
and maximum-likelihood estimation attains it as . is the classical Fisher information of the chosen measurement.
The Mach–Zehnder (or Ramsey) fringe.
Use . One photon (or atom) carries exactly one unit of Fisher information, at every working point.
Standard quantum limit versus Heisenberg limit
Each of qubit probes picks up the phase independently.
Variances add for product states: the standard quantum limit (shot-noise limit).
Up to a global phase. The two branches are eigenvectors of with eigenvalues , so the relative phase is .
The Heisenberg limit: a quadratic gain in Fisher information.
The circuit below prepares a 3-qubit GHZ state, applies to each qubit, disentangles, and interferes qubit 0. What is the probability (in percent) that qubit 0 reads ?
Show the answer
100 %
The GHZ branches acquire relative phase , so qubit 0 ends in : . A single probe with the same gives only . The fringe runs three times faster: super-resolution.
Reaching the limit in practice
A GHZ state measures only modulo , so on its own it cannot tell from . The fix is the same as in phase estimation: combine probes (or repeated applications) with and read the digits of from coarse to fine. Higgins et al. (2007) realized Heisenberg-limited phase estimation this way with single photons passed repeatedly through the phase, without entanglement, using adaptive measurements.
With photons, the analogue of GHZ is the NOON state across the two arms, whose relative phase is . Small- experiments showed super-resolved fringes (Mitchell, Lundeen & Steinberg 2004; Walther et al. 2004) and sensitivity beyond the standard quantum limit (Nagata et al. 2007).
| Strategy | Scaling of | Comment |
|---|---|---|
| independent probes | (SQL) | robust; the classical benchmark |
| GHZ / NOON | (HL) | fragile: one lost particle destroys the phase information |
| spin- or light-squeezed states | between SQL and HL | robust and practical; used in clocks and gravitational-wave detectors |
| sequential / phase estimation | in the number of uses | no entanglement; needs coherence over long total times |
Noise and the limits of the limit
Estimate a frequency ; each run lasts and dephasing at rate shrinks the fringe contrast.
Each qubit gives Fisher information at the best working point, with contrast .
The GHZ coherence decays times faster, , exactly cancelling its -fold faster phase.
With optimized interrogation times the two strategies give identical precision, : back to . More generally, for uncorrelated Markovian noise the asymptotic gain over the SQL is at most a constant factor (Escher, de Matos Filho & Davidovich 2011; Demkowicz-Dobrzański, Kołodyński & Guţă 2012).
Under independent Markovian dephasing, with the interrogation time optimized separately for each strategy, how does a GHZ state of atoms compare with unentangled atoms for frequency estimation?
- GHZ gives scaling, a quadratic improvement
- GHZ is times better
- They achieve the same precision; the faster phase accumulation is exactly offset by the -fold faster decoherence
- GHZ is useless because it decoheres instantly
Show the answer
C. They achieve the same precision; the faster phase accumulation is exactly offset by the -fold faster decoherence
Both optimize to . The GHZ state accumulates phase times faster but loses coherence times faster, so its optimal run is times shorter and the gain disappears.
Key takeaways
- Precision is bounded by , with for pure states; one interferometer probe has .
- Product probes give (SQL, ); any probes obey (Heisenberg limit, ), saturated by GHZ and NOON states.
- The phase ambiguity is resolved by multi-scale (phase-estimation-like) strategies, and sequential schemes reach without entanglement.
- Uncorrelated Markovian noise reduces the gain to a constant factor; error correction restores only when the signal lies outside the noise's Lindblad span.
Check yourself
In the interactive lesson these are scored with your confidence, and what you miss comes back for review.
unentangled qubit probes, a single repetition (), ideal measurement. What is the smallest achievable ?
Show the answer
B.
Product probes give , so . The Heisenberg limit with entangled probes would be .
An -qubit GHZ state evolves under on each qubit. What relative phase separates its two branches?
Show the answer
C.
and are eigenvectors of with eigenvalues , so their relative phase is , which gives .
Why can't a single GHZ-type measurement estimate a completely unknown at the Heisenberg limit?
- GHZ states cannot be prepared for
- The Heisenberg limit applies only to photons
- The quantum Fisher information of a GHZ state is only
- It determines only modulo , so is ambiguous up to multiples of ; a ladder of values or adaptive scheme is needed
Show the answer
D. It determines only modulo , so is ambiguous up to multiples of ; a ladder of values or adaptive scheme is needed
The fringe repeats times over . Resolving which period you are in requires coarser measurements as well, exactly as phase estimation combines
When can quantum error correction restore Heisenberg scaling in the presence of Markovian noise?
- When the signal Hamiltonian is not in the Lindblad span of the noise (HNLS)
- Always, given enough ancilla qubits
- Never: noise always forces
- Only for photonic sensors
Show the answer
A. When the signal Hamiltonian is not in the Lindblad span of the noise (HNLS)
If the signal generator lies outside the span of the noise operators (and their products), a code can correct the noise while leaving the signal intact, giving scaling. Otherwise any code that removes the noise also removes the signal.
Further reading
- S. L. Braunstein & C. M. Caves, "Statistical distance and the geometry of quantum states", Phys. Rev. Lett. 72, 3439 (1994).
- V. Giovannetti, S. Lloyd & L. Maccone, "Quantum metrology", Phys. Rev. Lett. 96, 010401 (2006).
- S. F. Huelga, C. Macchiavello, T. Pellizzari, A. K. Ekert, M. B. Plenio & J. I. Cirac, "Improvement of frequency standards with quantum entanglement", Phys. Rev. Lett. 79, 3865 (1997).
- B. L. Higgins, D. W. Berry, S. D. Bartlett, H. M. Wiseman & G. J. Pryde, "Entanglement-free Heisenberg-limited phase estimation", Nature 450, 393 (2007).
- L. Pezzè, A. Smerzi, M. K. Oberthaler, R. Schmied & P. Treutlein, "Quantum metrology with nonclassical states of atomic ensembles", Rev. Mod. Phys. 90, 035005 (2018).