Mutually Unbiased Bases
Maximally complementary measurements and where they come from.
Know a qubit's spin along perfectly and you know nothing about it along . Mutually unbiased bases make that complementarity exact in every dimension, and whether a complete set of them exists in dimension six is one of the most stubborn open problems in quantum information.
What you’ll learn
- Define mutually unbiased bases (MUBs) and interpret unbiasedness operationally.
- Prove that at most MUBs exist in .
- Construct complete sets of MUBs in odd prime dimensions with quadratic Gauss sums, and for qubits by partitioning the Pauli operators.
- Derive the MUB tomography formula and the identity .
- Describe the open existence problem in dimensions that are not prime powers, starting with .
Unbiasedness
Operationally: prepare any vector of one basis, measure in the other, and every outcome is equally likely. Perfect knowledge of one measurement's outcome means total ignorance of the other's.
For a qubit, , which equals exactly when the two Bloch axes are perpendicular. The , and eigenbases are therefore mutually unbiased, and there can be no fourth, because has only three mutually orthogonal axes.
In any dimension the Fourier basis is unbiased to the computational basis: . More generally, a basis is unbiased to the computational basis exactly when the unitary whose columns are its vectors has all entries of modulus : a rescaled complex Hadamard matrix.
At most d + 1 bases
Map each basis vector to a traceless Hermitian operator. is the real vector space of traceless Hermitian matrices, with the Hilbert–Schmidt inner product .
Within one basis the Gram matrix ( the all-ones matrix) has rank , so the operators span a -dimensional subspace.
For vectors from two different bases, unbiasedness is exactly Hilbert–Schmidt orthogonality.
MUBs give mutually orthogonal -dimensional subspaces of a -dimensional space.
A complete set, with , therefore splits the traceless operators into orthogonal pieces, one per basis. That is exactly what makes complete sets ideal for tomography, as you will see in a moment.
Constructions
Ivanović (1981) first constructed complete sets in prime dimensions. In the now-standard quadratic-phase form: for an odd prime , let and take the computational basis together with the bases made of the vectors
Compute the overlap.
Same , different : a full sum of -th roots of unity, which is . So each gives an orthonormal basis.
Different : complete the square, which is possible because is invertible modulo an odd prime. The sum becomes a phase times a quadratic Gauss sum, whose modulus is .
Every has entries of modulus , so it is also unbiased to the computational basis. That makes MUBs: a complete set.
For prime powers , replace by the finite field with elements (Wootters & Fields 1989). For qubits there is an elegant stabilizer construction: partition the non-identity Pauli operators into classes of mutually commuting operators. The common eigenbases of the classes are mutually unbiased (Bandyopadhyay, Boykin, Roychowdhury & Vatan 2002).
| Class | Commuting Paulis | Common eigenbasis |
|---|---|---|
| 1 | Computational (product) | |
| 2 | (product) | |
| 3 | (product) | |
| 4 | Entangled | |
| 5 | Entangled |
Tomography and 2-designs
Within each basis the 's form a Parseval frame for their subspace (their Gram matrix is a projector), and the mutually orthogonal subspaces fill all of . So this expansion is exact.
Since , collecting terms gives a linear inversion formula: measured probabilities in, density matrix out.
Take the Hilbert–Schmidt norm of the same expansion: .
For a qubit, turns the formula into with , and the identity reads .
A qubit gives , and . Using the identity above, what is its purity ?
Show the answer
B.
, so . Check: and .
Wootters and Fields showed that, among measurements in orthonormal bases, mutually unbiased ones minimize the statistical error of this reconstruction in a natural averaged sense. The identity above is the fingerprint of a deeper fact: the vectors of a complete set form a complex projective 2-design, so averaging over them reproduces the uniform (Haar) average. Applying Jensen's inequality to of the identity also gives an entropic uncertainty relation for complete sets: for pure states, .
Open problem: dimension six
Tensor products of complete sets show that with contains at least MUBs, so every dimension has at least 3. Whether more exist when is not a prime power is unknown, even for , where only three are known. Extensive analytical and numerical searches suggest that no fourth exists, but there is no proof.
| Dimension | Guaranteed by tensor products | Complete set ( bases) known? |
|---|---|---|
| 2, 3, 4, 5, 7, 8, 9 | Yes: prime powers | |
| 6 | 3 | Open; numerical evidence suggests not even a fourth basis exists |
| 10 | 3 | Open |
| 12 | 4 | Open |
Key takeaways
- Bases are mutually unbiased if : a vector of one basis gives uniformly random outcomes in the other.
- At most MUBs exist, by dimension counting in the space of traceless Hermitian operators. Complete sets exist in every prime-power dimension (Gauss sums, finite fields, Pauli partitions).
- A complete set gives exact linear tomography, , and forms a 2-design: .
- In other dimensions only MUBs are guaranteed, and whether has more than 3 is open.
Check yourself
In the interactive lesson these are scored with your confidence, and what you miss comes back for review.
What is the maximum number of mutually unbiased bases in ?
Show the answer
C.
The bound is , and since 5 is prime the Gauss-sum construction achieves it.
Which pair of qubit bases is mutually unbiased?
- The basis and the basis
- The basis and the eigenbasis of the Hadamard gate
- The basis and the basis
- The basis and the eigenbasis of
Show the answer
A. The basis and the basis
Mutually unbiased qubit bases correspond to perpendicular Bloch axes, and .
In the proof of the bound, what does mutual unbiasedness turn into?
- Linear independence of the basis vectors
- Commutation of the projectors
- Equality of the traces of the projectors
- Hilbert–Schmidt orthogonality of the traceless parts and
Show the answer
D. Hilbert–Schmidt orthogonality of the traceless parts and
, which vanishes exactly for unbiased pairs. Each basis then occupies its own -dimensional slice of a -dimensional space.
What is known about mutually unbiased bases in ?
- A complete set of 7 has been constructed
- Three are known; whether a fourth exists is open, and numerical evidence suggests it does not
- It has been proven that exactly 3 exist
- None exist, because 6 is not a prime power
Show the answer
B. Three are known; whether a fourth exists is open, and numerical evidence suggests it does not
The tensor-product construction gives . Extensive searches have failed to find a fourth, but there is no proof.
Further reading
- I. D. Ivanović, "Geometrical description of quantal state determination", Journal of Physics A 14 (1981).
- W. K. Wootters & B. D. Fields, "Optimal state-determination by mutually unbiased measurements", Annals of Physics 191 (1989).
- S. Bandyopadhyay, P. O. Boykin, V. Roychowdhury & F. Vatan, "A new proof for the existence of mutually unbiased bases", Algorithmica 34 (2002).
- T. Durt, B.-G. Englert, I. Bengtsson & K. Życzkowski, "On mutually unbiased bases", International Journal of Quantum Information 8 (2010).