Geometric parameters for the orthocross inverse
Abstract
The standard orthocross frame has diagonal d, upper entry a = (1-i)/2 and lower entry b = (1+i)/2. A geometric ratio and a nonzero denominator specify its proposed inverse entries.
Definition 1.1 (Upper frame entry).
Formalization. D5/S3/Quantum/Measurement/OrthocrossInverseFrame.upperEntry (✓ std3).
Source. Repository-derived.
Commentary.
The upper frame entry is defined by this equality.
Definition 1.2 (Lower frame entry).
Formalization. D5/S3/Quantum/Measurement/OrthocrossInverseFrame.lowerEntry (✓ std3).
Source. Repository-derived.
Commentary.
The lower frame entry is defined by this equality.
Definition 1.3 (Complex scale).
Formalization. D5/S3/Quantum/Measurement/OrthocrossInverseFrame.scale (✓ std3).
Source. Repository-derived.
Commentary.
The complex scale is defined by this equality.
Definition 1.4 (Geometric ratio).
Formalization. D5/S3/Quantum/Measurement/OrthocrossInverseFrame.ratio (✓ std3).
Source. Repository-derived.
Commentary.
The geometric ratio is defined by this equality.
Definition 1.5 (Geometric denominator).
Formalization. D5/S3/Quantum/Measurement/OrthocrossInverseFrame.denominator (✓ std3).
Source. Repository-derived.
Commentary.
The geometric denominator is defined by this equality.
Definition 1.6 (Upper inverse coefficient).
Formalization. D5/S3/Quantum/Measurement/OrthocrossInverseFrame.upperConstant (✓ std3).
Source. Repository-derived.
Commentary.
The upper inverse coefficient is defined by this equality.
Definition 1.7 (Diagonal inverse coefficient).
Formalization. D5/S3/Quantum/Measurement/OrthocrossInverseFrame.diagonalConstant (✓ std3).
Source. Repository-derived.
Commentary.
The diagonal inverse coefficient is defined by this equality.
Definition 1.8 (Hermitian triangular candidate).
Formalization. D5/S3/Quantum/Measurement/OrthocrossInverseFrame.candidate (✓ std3).
Source. Repository-derived.
Commentary.
For indices j,k in Fin d, the candidate has diagonal t, entry u q^(j-k+1) above the diagonal, and the conjugate of the corresponding upper entry below the diagonal. Integer exponents are used.
Theorem 1.9 (The ratio is nonzero).
Proof. Machine-checked in Lean as D5/S3/Quantum/Measurement/OrthocrossInverseFrame.ratio_ne_zero (✓ std3). ∎
Source. Repository-derived.
Commentary.
The scale has imaginary part one half, so it and its conjugate are nonzero.
Theorem 1.10 (The ratio is distinct from one).
Proof. Machine-checked in Lean as D5/S3/Quantum/Measurement/OrthocrossInverseFrame.ratio_ne_one (✓ std3). ∎
Source. Repository-derived.
Commentary.
The difference q-1 equals -i divided by the nonzero scale.
Theorem 1.11 (Unit modulus).
Proof. Machine-checked in Lean as D5/S3/Quantum/Measurement/OrthocrossInverseFrame.norm_ratio (✓ std3). ∎
Source. Repository-derived.
Commentary.
The numerator is the conjugate of L. Its norm equals the nonzero norm of L.
Theorem 1.12 (Conjugation inverts the ratio).
Proof. Machine-checked in Lean as D5/S3/Quantum/Measurement/OrthocrossInverseFrame.star_ratio (✓ std3). ∎
Source. Repository-derived.
Commentary.
Conjugating the quotient interchanges its numerator and denominator.
Theorem 1.13 (Nonzero denominator).
Proof. Machine-checked in Lean as D5/S3/Quantum/Measurement/OrthocrossInverseFrame.denominator_ne_zero (✓ std3). ∎
Source. Repository-derived.
Commentary.
A zero denominator would imply q^d = i. For d at least two, the telescoping bound |q^d-1| <= d |q-1|, together with |L|^2 = d^2-d+1/2, contradicts |i-1|^2 = 2. Dimension one is evaluated exactly.
Theorem 1.14 (Nonzero upper coefficient).
Proof. Machine-checked in Lean as D5/S3/Quantum/Measurement/OrthocrossInverseFrame.upperConstant_ne_zero (✓ std3). ∎
Source. Repository-derived.
Commentary.
All factors in the numerator and denominator of u are nonzero in positive dimension.
Theorem 1.15 (The candidate inverts the frame).
Proof. Machine-checked in Lean as D5/S3/Quantum/Measurement/OrthocrossInverseFrame.frame_mul_candidate (✓ std3). ∎
Source. Repository-derived.
Commentary.
Scale the candidate by L^2 delta. The geometric column sum is i L q^(d-k-1), which gives the first product row. Subtracting adjacent frame rows leaves only two entries, and the geometric recurrence gives the difference of the corresponding identity rows. Induction gives every row.
Theorem 1.16 (Closed inverse form).
Proof. Machine-checked in Lean as D5/S3/Quantum/Measurement/OrthocrossInverseFrame.inverse_frame_eq (✓ std3). ∎
Source. Repository-derived.
Commentary.
A right inverse of a square matrix equals its nonsingular inverse. Hermiticity identifies the lower entries with the conjugates of the upper entries.
Theorem 1.17 (Positive real diagonal).
Proof. Machine-checked in Lean as D5/S3/Quantum/Measurement/OrthocrossInverseFrame.diagonalConstant_pos (✓ std3). ∎
Source. Repository-derived.
Commentary.
The standard frame is positive definite, and so is its inverse. Every diagonal entry of the inverse is therefore a strictly positive real number.
Theorem 1.18 (Conjugate coefficient phase).
Proof. Machine-checked in Lean as D5/S3/Quantum/Measurement/OrthocrossInverseFrame.upperConstant_phase (✓ std3). ∎
Source. Repository-derived.
Commentary.
Conjugating delta gives -delta q^(-d), while conjugating L gives qL. The factors in the conjugate of u therefore cancel to a/(L^2 delta).
Theorem 1.19 (A short quotient for the diagonal ratio).
Proof. Machine-checked in Lean as D5/S3/Quantum/Measurement/OrthocrossInverseFrame.diagonal_div_upperConstant (✓ std3). ∎
Source. Repository-derived.
Commentary.
The identities t = 1/L + qu and qt - conjugate(u) = 1/L give t(1-q) = qu - conjugate(u). Substituting the coefficient phase yields the quotient.
Theorem 1.20 (Small Gaussian polynomials do not vanish).
Proof. Machine-checked in Lean as D5/S3/Quantum/Measurement/OrthocrossInverseFrame.polynomial_at_ratio_ne_zero (✓ std3). ∎
Source. Repository-derived.
Commentary.
The polynomial p has Gaussian-integer coefficients. Their Gaussian norm is the squared complex modulus. Write q=A/B with A=(d-1)-di and B=d-(d-1)i. An explicit Bezout identity makes A and B coprime. Scaling polynomial roots by B shows that a root at q would force B to divide the nonzero leading coefficient. Its norm would then be at least norm(B)=2d^2-2d+1, a contradiction.
References
- Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.candidate - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.denominator - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.denominator_ne_zero - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.diagonalConstant - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.diagonalConstant_pos - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.diagonal_div_upperConstant - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.frame_mul_candidate - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.inverse_frame_eq - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.lowerEntry - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.norm_ratio - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.polynomial_at_ratio_ne_zero - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.ratio - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.ratio_ne_one - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.ratio_ne_zero - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.scale - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.star_ratio - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.upperConstant - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.upperConstant_ne_zero - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.upperConstant_phase - Truth anchor:
D5/S3/Quantum/Measurement/OrthocrossInverseFrame.upperEntry - Dependency: D5/S3/AnalyticClosure/ComplexPowerDifference
- Dependency: D5/S3/Quantum/Measurement/OrthocrossGramHalfInteger