6.3.2 POVM#

Prompts

What is a POVM, and how does it generalize projective measurement? Why can a POVM have more outcomes than the Hilbert space dimension?
Explain unambiguous state discrimination: how can a three-outcome POVM distinguish non-orthogonal states with certainty (when it gives a definite answer)?
State Naimark’s theorem. What does it mean physically that every POVM can be realized as a projective measurement on a larger system?
Give an example of a POVM that is not a projective measurement and explain what advantage it provides.

Lecture Notes#

Overview#

Projective measurements are elegant but restrictive: outcomes must be orthogonal, and measurement outcomes correspond to eigenvalues of a Hermitian operator. Yet in many practical scenarios—measuring non-orthogonal quantum states, detecting weak signals, or performing measurements on subsystems—we need more flexibility. Positive operator-valued measures (POVMs) generalize the measurement postulate by removing the requirement that measurement operators be orthogonal projectors, enabling a vast toolkit for quantum information processing and state discrimination.

POVM Definition#

A positive operator-valued measure (POVM) is a set of operators \(\{\hat{M}_1, \hat{M}_2, \ldots, \hat{M}_n\}\) on Hilbert space \(\mathcal{H}\) satisfying two conditions:

1. Positive semidefiniteness:

(136)#\[ \hat{M}_i \geq 0 \quad \text{for all } i \quad (\text{all eigenvalues} \geq 0) \]

2. Completeness:

(137)#\[ \sum_i \hat{M}_i = I \]

These conditions ensure that measurement probabilities are non-negative and sum to one.

Measurement with a POVM#

If a system in state \(\hat{\rho}\) is measured using POVM \(\{\hat{M}_i\}\):

Outcome probability:

(138)#\[ p_i = \text{Tr}(\hat{M}_i \hat{\rho}) \]

Post-measurement state:

(139)#\[ \hat{\rho}'_i = \frac{\sqrt{\hat{M}_i} \hat{\rho} \sqrt{\hat{M}_i}}{\text{Tr}(\hat{M}_i \hat{\rho})} \]

where \(\sqrt{\hat{M}_i}\) is the unique positive square root (defined via spectral decomposition: \(\hat{M}_i = \sum_j \lambda_{i,j} |v_{i,j}\rangle\langle v_{i,j}| \Rightarrow \sqrt{\hat{M}_i} = \sum_j \sqrt{\lambda_{i,j}} |v_{i,j}\rangle\langle v_{i,j}|\)).

Normalization check: \(\sum_i p_i = \sum_i \text{Tr}(\hat{M}_i \hat{\rho}) = \text{Tr}(I \hat{\rho}) = 1\) ✓

Discussion

A POVM can have more outcomes than the Hilbert space dimension (\(n > \dim(\mathcal{H})\)). Why is this impossible for projective measurements? What does this overcomplete structure enable?

POVM vs. Projective Measurement#

A projective measurement is a special case of a POVM where elements are orthogonal projectors \(\hat{M}_i = \hat{P}_i\) with:

\(\hat{P}_i^2 = \hat{P}_i\) (idempotent)
\(\hat{P}_i \hat{P}_j = \delta_{ij} \hat{P}_i\) (orthogonal)
\(\sum_i \hat{P}_i = I\) (complete)

Key differences:

Property	Projective Measurement	General POVM
Outcomes	Limited to \(d\) (Hilbert space dimension)	Can have \(n > d\) (overcomplete)
Orthogonality	\(\hat{M}_i \hat{M}_j = 0\) for \(i \neq j\)	Elements can overlap: \(\hat{M}_i \hat{M}_j \neq 0\)
Information	Binary outcome per qubit	More outcomes = more information per measurement
Rank	Rank-1 projectors	Can be rank-1 or higher
Physical meaning of outcome	Eigenvalue of measured observable	Index number (no direct physical meaning)

Information trade-off: General POVMs extract more information about the system by allowing non-orthogonal measurement elements, at the cost of less definite outcome meaning.

Unambiguous State Discrimination#

One of the most striking applications of POVMs: distinguishing non-orthogonal quantum states with certainty when a definite answer is given.

Projective measurement: No. Non-orthogonal states have overlapping eigenspaces, so any projective measurement will sometimes give wrong answers.

POVM solution: Yes, with a trade-off. Use a three-outcome POVM:

Outcome 1: “State is \(|\psi_1\rangle\)” (definite, no error)
Outcome 2: “State is \(|\psi_2\rangle\)” (definite, no error)
Outcome 3: “Inconclusive” (try again or use different strategy)

Construction: Unambiguous State Discrimination POVM

Define POVM elements:

\[ \hat{M}_1 = \frac{1}{1+|c|^2}(I - |\psi_2\rangle\langle\psi_2|) \]

\[ \hat{M}_2 = \frac{1}{1+|c|^2}(I - |\psi_1\rangle\langle\psi_1|) \]

\[ \hat{M}_3 = I - \hat{M}_1 - \hat{M}_2 \]

Key property: If the state is \(|\psi_1\rangle\):

\(\langle\psi_1|\hat{M}_2|\psi_1\rangle = 0\) (outcome 2 never occurs)
Outcomes 1 and 3 occur with probabilities \(1-|c|^2\) and \(|c|^2\) respectively

Similarly, if state is \(|\psi_2\rangle\): outcome 1 never occurs. This guarantees no false identifications.

Naimark’s Theorem: POVM as Extended Projective Measurement#

Every POVM on a Hilbert space can be realized as a projective measurement on an extended Hilbert space (original space plus auxiliary system).

Naimark’s Theorem

Every POVM \(\{\hat{M}_i\}\) on \(\mathcal{H}\) can be realized as a projective measurement on an extended space \(\mathcal{H} \otimes \mathcal{H}_\text{aux}\). Specifically, there exist:

An auxiliary (ancilla) Hilbert space \(\mathcal{H}_\text{aux}\)
Orthogonal projectors \(\{\hat{P}_i\}\) on \(\mathcal{H} \otimes \mathcal{H}_\text{aux}\)

such that:

(140)#\[ \hat{M}_i = \text{Tr}_\text{aux}(\hat{P}_i (\hat{\rho} \otimes |0\rangle\langle 0|_\text{aux})) \]

Physical picture: Entangle the system with an ancilla, perform a projective measurement on the combined system, and trace out (discard) the ancilla measurement outcome.

This theorem reveals a profound truth: POVMs are not fundamentally more general than projective measurements—they are equivalent to projective measurements on a larger system. The apparent “softness” of a POVM comes from not observing (or not having access to) the ancilla measurement outcome.

Examples#

Example: Six-State (SIC-POVM) on a Qubit

Define 6 POVM elements:

\[ \hat{M}_k = \frac{1}{3}|\phi_k\rangle\langle\phi_k| \quad (k=1,\ldots,6) \]

where \(|\phi_k\rangle\) are 6 equally-spaced directions on the Bloch sphere (vertices of a regular polyhedron).

Properties:

\(\sum_k \hat{M}_k = I\) (completeness verified by symmetry)
Informationally complete: Probabilities \(\{p_k\}\) uniquely determine any qubit state \(\hat{\rho}\)
6 outcomes (overcomplete) for a 2-dimensional space

Application: Quantum state tomography with minimal number of outcomes.

Example: Heterodyne Detection (Optical POVM)

In quantum optics, measuring quadrature amplitudes (amplitude and phase of a light field) is described by a continuous POVM:

\[ \hat{M}(\alpha) = \frac{1}{\pi}|\alpha\rangle\langle\alpha|_\text{coherent} \]

where \(|\alpha\rangle\) are coherent states indexed by complex amplitude \(\alpha \in \mathbb{C}\).

Properties:

Continuous outcome space (not discrete)
Coherent states are highly non-orthogonal: \(\langle\alpha|\beta\rangle = \mathrm{e}^{-|\alpha-\beta|^2/2} \neq 0\) for \(\alpha \neq \beta\)
Enables full quantum tomography of optical field states

Summary#

POVM: Set of positive semidefinite operators \(\{\hat{M}_i\}\) with \(\sum_i \hat{M}_i = I\); generalizes orthogonal projectors.
Measurement rule: Outcome \(i\) has probability \(p_i = \text{Tr}(\hat{M}_i \hat{\rho})\); post-measurement state involves \(\sqrt{\hat{M}_i}\).
Generality: POVM elements need not be orthogonal or rank-1; allows \(n > \dim(\mathcal{H})\) outcomes (overcomplete).
Key advantage: Enables unambiguous state discrimination for non-orthogonal states (using three outcomes: identify state 1, identify state 2, or inconclusive).
Naimark dilation: Every POVM on \(\mathcal{H}\) equals a projective measurement on \(\mathcal{H} \otimes \mathcal{H}_\text{aux}\) followed by tracing out the ancilla.
Examples: SIC-POVM for tomography (6 outcomes on qubit), heterodyne detection in quantum optics (continuous outcome space).

Homework#

1. Show that any projective measurement \(\{P_i\}\) (with \(P_i^2 = P_i\), \(P_i P_j = \delta_{ij}P_i\), \(\sum_i P_i = \mathbf{1}\)) is a special case of a POVM by setting \(E_i = P_i\). Verify the POVM conditions.

2. Consider a trine POVM on a qubit: \(E_k = \frac{2}{3}|\phi_k\rangle\langle\phi_k|\) for \(k=0,1,2\), where \(|\phi_0\rangle = |0\rangle\), \(|\phi_1\rangle = -\frac{1}{2}|0\rangle + \frac{\sqrt{3}}{2}|1\rangle\), \(|\phi_2\rangle = -\frac{1}{2}|0\rangle - \frac{\sqrt{3}}{2}|1\rangle\). Verify that \(\sum_{k=0}^2 E_k = \mathbf{1}\).

3. (Unambiguous state discrimination) Consider two non-orthogonal states \(|\psi_1\rangle\) and \(|\psi_2\rangle\) with \(\langle\psi_1|\psi_2\rangle = s \neq 0\). A POVM for unambiguous discrimination has three outcomes: “\(\psi_1\)”, “\(\psi_2\)”, and “?” (inconclusive). Explain why this task is impossible with a projective measurement if \(s \neq 0\), but possible with a POVM. What condition must the POVM elements satisfy to make discrimination unambiguous when a definite answer is given?

4. Construct an explicit POVM for unambiguous discrimination of \(|0\rangle\) and \(|+\rangle = \frac{1}{\sqrt{2}}(|0\rangle + |1\rangle)\). Find POVM elements \(E_1\) (identifies \(|0\rangle\)), \(E_2\) (identifies \(|+\rangle\)), and \(E_?\) (inconclusive) that satisfy all POVM conditions. What is the probability of an inconclusive result for each state?

5. (Naimark dilation) Any POVM \(\{E_i\}\) on \(\mathcal{H}\) can be realized as a projective measurement on a larger Hilbert space \(\mathcal{H} \otimes \mathcal{K}\): introduce an ancilla in state \(|0\rangle_\mathcal{K}\), perform a unitary \(U\) on \(\mathcal{H} \otimes \mathcal{K}\), then measure the ancilla. Show that if \(U(|\psi\rangle \otimes |0\rangle) = \sum_i (\sqrt{E_i}|\psi\rangle) \otimes |i\rangle_\mathcal{K}\), then the probability of outcome \(i\) is \(\mathrm{Tr}[E_i |\psi\rangle\langle\psi|]\).

6. An informationally complete POVM (IC-POVM) has enough elements to uniquely determine any density matrix \(\hat{\rho}\) from the measurement statistics \(\{p_i = \mathrm{Tr}[E_i \hat{\rho}]\}\). For a \(d\)-dimensional system, what is the minimum number of POVM elements needed for an IC-POVM? Justify your answer by counting the degrees of freedom in \(\hat{\rho}\).

7. For a POVM with elements \(\{E_i\}\), the probability of outcome \(i\) on mixed state \(\hat{\rho}\) is \(p_i = \mathrm{Tr}[E_i \hat{\rho}]\). Show that if \(\hat{\rho} = \sum_k q_k \hat{\rho}_k\) is a convex mixture, then \(p_i = \sum_k q_k \mathrm{Tr}[E_i \hat{\rho}_k]\). This shows POVM statistics are linear in the state.

8. Compare the information gain of a POVM vs. a projective measurement. For a qubit initially in an unknown state on the Bloch sphere: (a) A projective measurement gives binary outcomes. How much classical information (in bits) does it convey? (b) A 4-element symmetric IC-POVM (SIC-POVM) with \(E_k = \frac{1}{2}|\phi_k\rangle\langle\phi_k|\) for 4 tetrahedral states. Explain qualitatively why this conveys more information about \(\hat{\rho}\) per measurement than a projective measurement.