Preprint
Article

This version is not peer-reviewed.

Unitary States and Gates of Three-Qubit Quantum Systems

Submitted:

19 October 2026

Posted:

20 October 2026

You are already at the latest version

Abstract
Pure states of three-qubit quantum systems are represented by eight-dimensional complex vectors, whereas their unitary transformations are described by $8 \times 8$ complex matrices. Motivated by this correspondence, we extend our algebraic model for two-qubit systems to eight-dimensional spaces endowed with commutative and associative multiplication. To each eight-dimensional number we assign a matrix and prove that the resulting correspondence is a homomorphic bijection. We derive conditions under which the associated matrix is unitary and hence represents a three-qubit quantum gate. This leads to a specific algebraic notion of a unitary state: a normalized state whose multiplication matrix is unitary. We construct 16,384 basis spaces; within each fixed space, multiplication generates a commuting family of quantum gates. For any two unitary states in the class considered, we give an explicit transition-gate construction within one of these algebraic families. For real components, the unitary states in a representative basis space form a union of thirty-two four-dimensional manifolds in \(\mathbb{R}^{8}\). The construction therefore yields new structured classes of three-qubit unitary operators and invariant state sets, while its extension to larger multiqubit systems remains open.
Keywords: 
;  ;  ;  ;  ;  ;  

1. Introduction

Pure states of three-qubit quantum systems are represented by eight-dimensional vectors with complex components, whereas unitary operators are described by 8 × 8 complex matrices. It is therefore natural to regard the states of a three-qubit system as points in an eight-dimensional complex space. This approach extends our study of two-qubit systems [1], in which four-dimensional complex spaces were used to describe states jointly with the corresponding 4 × 4 matrices. The transition from four to eight components produces a qualitatively new algebraic structure: in the present work, 16,384 basis spaces are constructed, classes of 8 × 8 matrices are investigated, and the geometry of the set of unitary states is analyzed. Four- and eight-dimensional spaces with real components were previously considered in [2–5].
Consider a quantum system composed of three qubits. Its quantum state | φ 〉 is represented by a linear combination of eight basis vectors with complex coefficients [6,7]:
| φ 〉 = x 1 | 000 〉 + x 2 | 001 〉 + x 3 | 010 〉 + x 4 | 011 〉 + x 5 | 100 〉 + x 6 | 101 〉 + x 7 | 110 〉 + x 8 | 111 〉 ,
where x i ∈ C and ∑ i = 1 8 | x i | 2 = 1 . The vectors
| 000 〉 , | 001 〉 , | 010 〉 , | 011 〉 , | 100 〉 , | 101 〉 , | 110 〉 , | 111 〉
are referred to as basis states.
Any 8 × 8 unitary matrix can be regarded as a gate for a three-qubit quantum system. In practice, a quantum processor implements a basic set of primitive gates, while other gates are synthesized from them as quantum circuits. The available primitive gates depend on the technology used to build the physical quantum computer. More sophisticated gates are then constructed from these primitives. For example, the open-source Qiskit quantum-computing platform provides a large collection of gates, many of which are built from elementary gates [8]. At the hardware level, experimental physicists and engineers optimize elementary gates, whereas computer scientists seek to design efficient high-level gates [6].
We construct 8 × 8 unitary matrices for three-qubit quantum systems. Commonly used three-qubit gates include the Hadamard gate H 3 , the Toffoli (CCNOT) gate, and the Fredkin (CSWAP) gate. Their matrices are, respectively,
H 3 = 2 4 1 1 1 − 1 1 1 1 − 1 1 1 1 − 1 − 1 − 1 − 1 1 1 1 1 − 1 1 1 1 − 1 1 1 1 − 1 − 1 − 1 − 1 1 1 1 1 − 1 1 1 1 − 1 1 1 1 − 1 − 1 − 1 − 1 1 − 1 − 1 − 1 1 − 1 − 1 − 1 1 − 1 − 1 − 1 1 1 1 1 − 1 ,
C C N O T = 1 0 0 1 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 1 1 0 ,
C S W A P = 1 0 0 1 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 1 0 0 0 0 0 1 .
Three-qubit systems occupy a special place in quantum information theory because they are the simplest composite systems in which genuinely multipartite structures of states and transformations arise. Unlike the two-qubit case, pure three-qubit states do not admit a single analogue of the Schmidt decomposition; instead, their properties are described by several entanglement classes and sets of algebraic invariants. Enríquez, Delgado, and Życzkowski investigated random pure three-qubit states generated by the Haar measure on U(8), including the distributions of their canonical-decomposition parameters and polynomial invariants [9]. Pérez-Salinas et al. proposed a quantum circuit that transforms an arbitrary three-qubit state into canonical form by means of local parameterized unitary transformations and thereby enables its tripartite entanglement to be determined [10].
The geometric and information-theoretic properties of three-qubit states remain an active area of research. Kalaga et al. established relationships among mixedness, coherence, and entanglement for a family of three-qubit states [11]. Benedito and Sierra proposed a geometric representation of three-qubit entanglement classes and related their geometry to a tripartite entanglement measure and the Cayley hyperdeterminant [12]. Wang et al. studied the local-unitary equivalence of quantum states using tensor decompositions of unitary matrices and derived the corresponding equivalence criteria [13].
Another important research direction concerns the construction and implementation of multiqubit unitary operators. An arbitrary transformation of a three-qubit system is represented by a matrix in U(8); however, hardware implementation generally requires this matrix to be decomposed into a sequence of one- and two-qubit gates. Krol et al. developed an efficient implementation of quantum Shannon decomposition and showed that exploiting the internal structure of the input matrix can substantially reduce the resulting circuit [14]. Tang et al. proposed a fast implementation of a three-qubit CCZ gate using Rydberg atoms and applied it to three-qubit versions of the Deutsch–Jozsa and Grover algorithms [15].
Grigoryan et al. constructed a QR decomposition of arbitrary three-qubit unitary operations using the discrete signal-induced heap transform and obtained permutation-free circuits comprising 28 controlled rotations [16]. In a subsequent study, this approach was extended to arbitrary real and complex three- and four-qubit unitary matrices, yielding circuits that require no permutation gates [17].
Thus, existing studies address the classification and geometry of three-qubit states, criteria for local-unitary equivalence, decompositions of arbitrary matrices in U(8), and physical implementations of individual three-qubit gates. However, the construction of entire classes of three-qubit unitary operators by means of specially defined commutative and associative eight-dimensional algebras has received little attention. In the present work, states are identified with eight-dimensional numbers, and their associated matrices form algebraically defined families of quantum gates.
The remainder of this paper is organized as follows. Section 2 defines associative and commutative multiplication of eight-dimensional numbers with complex components and constructs the corresponding number spaces. Section 3 investigates the so-called basis spaces of eight-dimensional numbers and defines 16,384 such spaces. A matrix is then associated with every eight-dimensional number, and the properties of this mapping are studied in Section 4. The mapping between numbers and matrices is shown to be a homomorphic bijection. Section 5 derives conditions under which the matrix associated with an eight-dimensional number is a gate for a three-qubit quantum system. Section 6 introduces the concept of a unitary state of a three-qubit quantum system. It is shown that any unitary state can be transformed into any other unitary state in a single step, and the gate implementing this transition is constructed explicitly. Section 7 addresses the general form of unitary states and gates with real components in the basis spaces. The unitary states are shown to form a union of thirty-two four-dimensional manifolds in an eight-dimensional space. The final section discusses the results.

2. Spaces of Eight-Dimensional Numbers with Complex Components

In [2], spaces of four-dimensional numbers with complex components were considered, while [3,4] studied eight-dimensional numbers with real components equipped with associative and commutative addition and multiplication. Following these studies, consider an eight-dimensional number X = ( x 1 , x 2 , x 3 , x 4 , x 5 , x 6 , x 7 , x 8 ) with complex components x i ∈ C . Define addition of X and Y componentwise, i.e., X + Y = ( x 1 + y 1 , x 2 + y 2 , x 3 + y 3 , x 4 + y 4 , x 5 + y 5 , x 6 + y 6 , x 7 + y 7 , x 8 + y 8 ) . Define multiplication Z = X · Y as follows
z 1 = x 1 y 1 + γ 2 γ 3 x 2 y 2 + γ 2 γ 4 x 3 y 3 + γ 3 γ 4 x 4 y 4 + + γ 5 γ 8 x 5 y 5 + γ 2 γ 3 γ 8 γ 5 x 6 y 6 + γ 2 γ 4 γ 5 γ 8 γ 6 2 x 7 y 7 + γ 3 γ 4 γ 5 γ 8 γ 7 2 x 8 y 8 z 2 = x 2 y 1 + x 1 y 2 + γ 4 x 4 y 3 + γ 4 x 3 y 4 + + γ 8 x 6 y 5 + γ 8 x 5 y 6 + γ 4 γ 5 γ 8 γ 6 γ 7 x 8 y 7 + γ 4 γ 5 γ 8 γ 6 γ 7 x 7 y 8 z 3 = x 3 y 1 + γ 3 x 4 y 2 + x 1 y 3 + γ 3 x 2 y 4 + + γ 5 γ 8 γ 6 x 7 y 5 + γ 3 γ 8 γ 7 x 8 y 6 + γ 5 γ 8 γ 6 x 5 y 7 + γ 3 γ 8 γ 7 x 6 y 8 z 4 = x 4 y 1 + γ 2 x 3 y 2 + γ 2 x 2 y 3 + x 1 y 4 + + γ 5 γ 8 γ 7 x 8 y 5 + γ 2 γ 8 γ 6 x 7 y 6 + γ 2 γ 8 γ 6 x 6 y 7 + γ 5 γ 8 γ 7 x 5 y 8 z 5 = x 5 y 1 + γ 2 γ 3 γ 5 x 6 y 2 + γ 2 γ 4 γ 6 x 7 y 3 + γ 3 γ 4 γ 7 x 8 y 4 + + x 1 y 5 + γ 2 γ 3 γ 5 x 2 y 6 + γ 2 γ 4 γ 6 x 3 y 7 + γ 3 γ 4 γ 7 x 4 y 8 z 6 = x 6 y 1 + γ 5 x 5 y 2 + γ 5 γ 4 γ 7 x 8 y 3 + γ 5 γ 4 γ 6 x 7 y 4 + + γ 5 x 2 y 5 + x 1 y 6 + γ 5 γ 4 γ 6 x 4 y 7 + γ 5 γ 4 γ 7 x 3 y 8 z 7 = x 7 y 1 + γ 3 γ 6 γ 7 x 8 y 2 + γ 6 x 5 y 3 + γ 3 γ 6 γ 5 x 6 y 4 + + γ 6 x 3 y 5 + γ 3 γ 6 γ 5 x 4 y 6 + x 1 y 7 + γ 3 γ 6 γ 7 x 2 y 8 z 8 = x 8 y 1 + γ 2 γ 7 γ 6 x 7 y 2 + γ 2 γ 7 γ 5 x 6 y 3 + γ 7 x 5 y 4 + + γ 7 x 4 y 5 + γ 2 γ 7 γ 5 x 3 y 6 + γ 2 γ 7 γ 6 x 2 y 7 + x 1 y 8
where γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 are arbitrary nonzero complex numbers.
Theorem 1.
For any fixed generating tuple ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) , multiplication (1) is commutative and associative.
Proof. 
The proof follows by direct verification. □
A proof of this theorem for eight-dimensional numbers with real components is given in [4].
It follows that infinitely many distinct associative and commutative multiplications can be defined in the space of eight-dimensional numbers; each ordered tuple ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) determines a particular multiplication.
Introduce the basis numbers J k 1 = ( 0 , 0 , … , 0 , 1 , 0 , … , 0 ) and J k 2 = ( 0 , 0 , … 0 , i , 0 , … , 0 ) = i · J k 1 , where k ranges from 1 to 8 and only the kth component of a basis number is nonzero. The number ( a , 0 , 0 , 0 , 0 , 0 , 0 , 0 ) is called a complex number if a ∈ C and a real number if a ∈ R .
Lemma 1.
Let a be a real or complex number and let X = ( x 1 , x 2 , x 3 , x 4 , x 5 , x 6 , x 7 , x 8 ) be an eight-dimensional number with complex components. Then their product is well defined and
a · X = ( a x 1 , a x 2 , a x 3 , a x 4 , a x 5 , a x 6 , a x 7 , a x 8 ) .
Proof. 
Substitution of Y = ( a , 0 , 0 , 0 , 0 , 0 , 0 , 0 ) into definition (1) yields the required identity. □
This lemma immediately implies that every eight-dimensional number X = ( x 1 re + i x 1 im , x 2 re + i x 2 im , … , x 8 re + i x 8 im ) can be expanded in the basis numbers:
X = ∑ k = 1 8 x k r e J k 1 + ∑ k = 1 8 x k i m J k 2
Products of basis numbers are summarized in Table 1, which lists J k 1 J m 1 for k , m = 1 , … , 8 . The remaining products satisfy
J k 1 J m 2 = i · J k 1 J m 1 , J k 2 J m 2 = − J k 1 J m 1 .
As in the case of numbers with real components [4], we define a conjugate number. For this purpose, together with X = ( x 1 , x 2 , x 3 , x 4 , x 5 , x 6 , x 7 , x 8 ) , consider the following seven numbers:
X ( 5 , 6 , 7 , 8 ) = ( x 1 , x 2 , x 3 , x 4 , − x 5 , − x 6 , − x 7 , − x 8 ) X ( 3 , 4 , 5 , 6 ) = ( x 1 , x 2 , − x 3 , − x 4 , − x 5 , − x 6 , x 7 , x 8 ) X ( 2 , 4 , 5 , 7 ) = ( x 1 , − x 2 , x 3 , − x 4 , − x 5 , x 6 , − x 7 , x 8 ) X ( 2 , 3 , 5 , 8 ) = ( x 1 , − x 2 , − x 3 , x 4 , − x 5 , x 6 , x 7 , − x 8 ) X ( 3 , 4 , 7 , 8 ) = ( x 1 , x 2 , − x 3 , − x 4 , x 5 , x 6 , − x 7 , − x 8 ) X ( 2 , 4 , 6 , 8 ) = ( x 1 , − x 2 , x 3 , − x 4 , x 5 , − x 6 , x 7 , − x 8 ) X ( 2 , 3 , 6 , 7 ) = ( x 1 , − x 2 , − x 3 , x 4 , x 5 , − x 6 , − x 7 , x 8 )
Table 1. Products of basis numbers
Table 1. Products of basis numbers
J 1 1
J 2 1
J 3 1
J 4 1
J 5 1
J 6 1
J 7 1
J 8 1
J 1 1 J 1 1 J 2 1 J 3 1 J 4 1 J 5 1 J 6 1 J 7 1 J 8 1
J 2 1 J 2 1 γ 2 γ 3 J 1 1 γ 2 J 4 1 γ 3 J 3 1 γ 5 J 6 1 γ 2 γ 3 γ 5 J 5 1 γ 2 γ 7 γ 6 J 8 1 γ 3 γ 6 γ 7 J 7 1
J 3 1 J 3 1 γ 2 J 4 1 γ 2 γ 4 J 1 1 γ 4 J 2 1 γ 6 J 7 1 γ 2 γ 7 γ 5 J 8 1 γ 2 γ 4 γ 6 J 5 1 γ 5 γ 4 γ 7 J 6 1
J 4 1 J 4 1 γ 3 J 3 1 γ 4 J 2 1 γ 3 γ 4 J 1 1 γ 7 J 8 1 γ 3 γ 6 γ 5 J 7 1 γ 5 γ 4 γ 6 J 6 1 γ 3 γ 4 γ 7 J 5 1
J 5 1 J 5 1 γ 5 J 6 1 γ 6 J 7 1 γ 7 J 8 1 γ 5 γ 8 J 1 1 γ 8 J 2 1 γ 5 γ 8 γ 6 J 3 1 γ 5 γ 8 γ 7 J 4 1
J 6 1 J 6 1 γ 2 γ 3 γ 5 J 5 1 γ 2 γ 7 γ 5 J 8 1 γ 3 γ 6 γ 5 J 7 1 γ 8 J 2 1 γ 2 γ 3 γ 8 γ 5 J 1 1 γ 2 γ 8 γ 6 J 4 1 γ 3 γ 8 γ 7 J 3 1
J 7 1 J 7 1 γ 2 γ 7 γ 6 J 8 1 γ 2 γ 4 γ 6 J 5 1 γ 5 γ 4 γ 6 J 6 1 γ 5 γ 8 γ 6 J 3 1 γ 2 γ 8 γ 6 J 4 1 γ 2 γ 4 γ 5 γ 8 γ 6 2 J 1 1 γ 4 γ 5 γ 8 γ 6 γ 7 J 2 1
J 8 1 J 8 1 γ 3 γ 6 γ 7 J 7 1 γ 5 γ 4 γ 7 J 6 1 γ 3 γ 4 γ 7 J 5 1 γ 5 γ 8 γ 7 J 4 1 γ 3 γ 8 γ 7 J 3 1 γ 4 γ 5 γ 8 γ 6 γ 7 J 2 1 γ 3 γ 4 γ 5 γ 8 γ 7 2 J 1 1
We calculate the product
X · X ( 5 , 6 , 7 , 8 ) · X ( 3 , 4 , 5 , 6 ) · X ( 2 , 4 , 5 , 7 ) · X ( 2 , 3 , 5 , 8 ) · X ( 3 , 4 , 7 , 8 ) · X ( 2 , 4 , 6 , 8 ) · X ( 2 , 3 , 6 , 7 ) .
The computation was performed using a Python script for symbolic computation. The resulting eight-dimensional number is ( M ( X ) , 0 , 0 , 0 , 0 , 0 , 0 , 0 ) : all components except the first vanish, and the first component M ( X ) is a homogeneous polynomial of degree eight containing 835 terms. Thus,
X · X * = M r e ( X ) · J 1 1 + M i m ( X ) · J 1 2
where M re ( X ) and M im ( X ) are the real and imaginary parts of M ( X ) , respectively, and
X * = X ( 5 , 6 , 7 , 8 ) · X ( 3 , 4 , 5 , 6 ) · X ( 2 , 4 , 5 , 7 ) · X ( 2 , 3 , 5 , 8 ) · X ( 3 , 4 , 7 , 8 ) · X ( 2 , 4 , 6 , 8 ) · X ( 2 , 3 , 6 , 7 )
The eight-dimensional number X * is called the conjugate of X, whereas the eighth root M ( X ) 8 of the complex number M ( X ) is called the symplectic modulus of X. We denote this modulus by | X | . We now calculate the conjugates of the basis numbers.
J 1 1 * = J 1 1 , J 2 1 * = γ 2 3 γ 3 3 J 2 1 , J 3 1 * = γ 2 3 γ 4 3 J 3 1 , J 4 1 * = γ 3 3 γ 4 3 J 4 1 ,
J 5 1 * = γ 5 3 γ 8 3 J 5 1 , J 6 1 * = γ 2 3 γ 3 3 γ 8 3 γ 5 3 J 6 1 , J 7 1 * = γ 2 3 γ 4 3 γ 5 3 γ 8 3 γ 6 6 J 7 1 , J 8 1 * = γ 3 3 γ 4 3 γ 5 3 γ 8 3 γ 7 6 J 8 1 .
J 1 2 * = − J 1 2 , J 2 2 * = − γ 2 3 γ 3 3 J 2 2 , J 3 2 * = − γ 2 3 γ 4 3 J 3 2 , J 4 2 * = − γ 3 3 γ 4 3 J 4 2 ,
J 5 2 * = − γ 5 3 γ 8 3 J 5 2 , J 6 2 * = − γ 2 3 γ 3 3 γ 8 3 γ 5 3 J 6 2 , J 7 2 * = − γ 2 3 γ 4 3 γ 5 3 γ 8 3 γ 6 6 J 7 2 , J 8 2 * = − γ 3 3 γ 4 3 γ 5 3 γ 8 3 γ 7 6 J 8 2 .
Thus, in a certain sense, the basis numbers are self-conjugate.
Suppose that the symplectic modulus | X | = M ( X ) 8 of the eight-dimensional number X = ( x 1 , x 2 , x 3 , x 4 , x 5 , x 6 , x 7 , x 8 ) is nonzero. Then there exists a unique inverse X − 1 such that
X · X − 1 = X − 1 · X = J 1 1 .
Multiplying this identity by X * , we obtain
X − 1 · X · X * = X * , ( M ( X ) , 0 , 0 , 0 , 0 , 0 , 0 , 0 ) X − 1 = X * .
Multiplication by
M ( X ) ¯ | M ( X ) | 2 , 0 , 0 , 0 , 0 , 0 , 0 , 0
and the identity
M ( X ) ¯ | M ( X ) | 2 , 0 , 0 , 0 , 0 , 0 , 0 , 0 · ( M ( X ) , 0 , 0 , 0 , 0 , 0 , 0 , 0 ) = J 1 1
give
X − 1 = X * · M ( X ) ¯ | M ( X ) | 2 , 0 , 0 , 0 , 0 , 0 , 0 , 0 .
Therefore,
X − 1 = M ( X ) ¯ M ( X ) 2 X *
If the symplectic modulus of X is nonzero, division of eight-dimensional numbers is defined by
Y X = Y · X − 1 = M ( X ) ¯ | M ( X ) | 2 X * Y , M ( X ) = | X | 8 .
We have thus defined addition, subtraction, multiplication, and division for eight-dimensional numbers with complex components. Infinitely many multiplication laws are obtained as the seven generating numbers vary. We next construct basis spaces in which multiplication is specified explicitly and from which all admissible multiplications can be obtained. This produces a finite collection of basis spaces endowed with explicit associative and commutative addition and multiplication operations.

3. Basis Spaces of Eight-Dimensional Numbers

As in the four-dimensional complex case, we determine the basis spaces of eight-dimensional numbers. Introduce the following change of variables:
x ¯ 1 = x 1 x ¯ 2 = γ 2 γ 3 x 2 x ¯ 3 = γ 2 γ 4 x 3 x ¯ 4 = γ 3 γ 4 x 4 x ¯ 5 = γ 5 γ 8 x 5 x ¯ 6 = γ 2 γ 3 γ 8 γ 5 x 6 x ¯ 7 = γ 2 γ 4 γ 5 γ 8 γ 6 x 7 x ¯ 8 = γ 3 γ 4 γ 5 γ 8 γ 7 x 8
The inverse transformation is
x 1 = x ¯ 1 x 2 = 1 γ 2 γ 3 x ¯ 2 x 3 = 1 γ 2 γ 4 x ¯ 3 x 4 = 1 γ 3 γ 4 x ¯ 4 x 5 = 1 γ 5 γ 8 x ¯ 5 x 6 = γ 5 γ 2 γ 3 γ 8 x ¯ 6 x 7 = γ 6 γ 2 γ 4 γ 5 γ 8 x ¯ 7 x 8 = γ 7 γ 3 γ 4 γ 5 γ 8 x ¯ 8
Applying the change of variables to (1), with X ¯ = ( x ¯ 1 , x ¯ 2 , x ¯ 3 , x ¯ 4 , x ¯ 5 , x ¯ 6 , x ¯ 7 , x ¯ 8 ) and Y ¯ = ( y ¯ 1 , y ¯ 2 , y ¯ 3 , y ¯ 4 , y ¯ 5 , y ¯ 6 , y ¯ 7 , y ¯ 8 ) , we obtain
z 1 = x ¯ 1 · y ¯ 1 + x ¯ 2 · y ¯ 2 + x ¯ 3 · y ¯ 3 + x ¯ 4 · y ¯ 4 + + x ¯ 5 · y ¯ 5 + x ¯ 6 · y ¯ 6 + x ¯ 7 · y ¯ 7 + x ¯ 8 · y ¯ 8 z 2 = 1 γ 2 γ 3 x ¯ 2 · y ¯ 1 + 1 γ 2 γ 3 x ¯ 1 · y ¯ 2 + 1 γ 2 γ 3 x ¯ 4 · y ¯ 3 + 1 γ 2 γ 3 x ¯ 3 · y ¯ 4 + + 1 γ 2 γ 3 x ¯ 6 · y ¯ 5 + 1 γ 2 γ 3 x ¯ 5 · y ¯ 6 + 1 γ 2 γ 3 x ¯ 8 · y ¯ 7 + 1 γ 2 γ 3 x ¯ 7 · y ¯ 8 z 3 = 1 γ 2 γ 4 x ¯ 3 · y ¯ 1 + 1 γ 2 γ 4 x ¯ 4 · y ¯ 2 + 1 γ 2 γ 4 x ¯ 1 · y ¯ 3 + 1 γ 2 γ 4 x ¯ 2 · y ¯ 4 + + 1 γ 2 γ 4 x ¯ 7 · y ¯ 5 + 1 γ 2 γ 4 x ¯ 8 · y ¯ 6 + 1 γ 2 γ 4 x ¯ 5 · y ¯ 7 + 1 γ 2 γ 4 x ¯ 6 · y ¯ 8 z 4 = 1 γ 3 γ 4 x ¯ 4 · y ¯ 1 + 1 γ 3 γ 4 x ¯ 3 · y ¯ 2 + 1 γ 3 γ 4 x ¯ 2 · y ¯ 3 + 1 γ 3 γ 4 x ¯ 1 · y ¯ 4 + + 1 γ 3 γ 4 x ¯ 8 · y ¯ 5 + 1 γ 3 γ 4 x ¯ 7 · y ¯ 6 + 1 γ 3 γ 4 x ¯ 6 · y ¯ 7 + 1 γ 3 γ 4 x ¯ 5 · y ¯ 8 z 5 = 1 γ 5 γ 8 x ¯ 5 · y ¯ 1 + 1 γ 5 γ 8 x ¯ 6 · y ¯ 2 + 1 γ 5 γ 8 x ¯ 7 · y ¯ 3 + 1 γ 5 γ 8 x ¯ 8 · y ¯ 4 + + 1 γ 5 γ 8 x ¯ 1 · y ¯ 5 + 1 γ 5 γ 8 x ¯ 2 · y ¯ 6 + 1 γ 5 γ 8 x ¯ 3 · y ¯ 7 + 1 γ 5 γ 8 x ¯ 4 · y ¯ 8 z 6 = γ 5 γ 2 γ 3 γ 8 x ¯ 6 · y ¯ 1 + γ 5 γ 2 γ 3 γ 8 x ¯ 5 · y ¯ 2 + γ 5 γ 2 γ 3 γ 8 x ¯ 8 · y ¯ 3 + γ 5 γ 2 γ 3 γ 8 x ¯ 7 · y ¯ 4 + + γ 5 γ 2 γ 3 γ 8 x ¯ 2 · y ¯ 5 + γ 5 γ 2 γ 3 γ 8 x ¯ 1 · y ¯ 6 + γ 5 γ 2 γ 3 γ 8 x ¯ 4 · y ¯ 7 + γ 5 γ 2 γ 3 γ 8 x ¯ 3 · y ¯ 8 z 7 = γ 6 γ 2 γ 4 γ 5 γ 8 x ¯ 7 · y ¯ 1 + γ 6 γ 2 γ 4 γ 5 γ 8 x ¯ 8 · y ¯ 2 + γ 6 γ 2 γ 4 γ 5 γ 8 x ¯ 5 · y ¯ 3 + γ 6 γ 2 γ 4 γ 5 γ 8 x ¯ 6 · y ¯ 4 + + γ 6 γ 2 γ 4 γ 5 γ 8 x ¯ 3 · y ¯ 5 + γ 6 γ 2 γ 4 γ 5 γ 8 x ¯ 4 · y ¯ 6 + γ 6 γ 2 γ 4 γ 5 γ 8 x ¯ 1 · y ¯ 7 + γ 6 γ 2 γ 4 γ 5 γ 8 x ¯ 2 · y ¯ 8 z 8 = γ 7 γ 3 γ 4 γ 5 γ 8 x ¯ 8 · y ¯ 1 + γ 7 γ 3 γ 4 γ 5 γ 8 x ¯ 7 · y ¯ 2 + γ 7 γ 3 γ 4 γ 5 γ 8 x ¯ 6 · y ¯ 3 + γ 7 γ 3 γ 4 γ 5 γ 8 x ¯ 5 · y ¯ 4 + + γ 7 γ 3 γ 4 γ 5 γ 8 x ¯ 4 · y ¯ 5 + γ 7 γ 3 γ 4 γ 5 γ 8 x ¯ 3 · y ¯ 6 + γ 7 γ 3 γ 4 γ 5 γ 8 x ¯ 2 · y ¯ 7 + γ 7 γ 3 γ 4 γ 5 γ 8 x ¯ 1 · y ¯ 8
Using notation (7) for Z, these equalities can be written as
z ¯ 1 = x ¯ 1 · y ¯ 1 + x ¯ 2 · y ¯ 2 + x ¯ 3 · y ¯ 3 + x ¯ 4 · y ¯ 4 + x ¯ 5 · y ¯ 5 + x ¯ 6 · y ¯ 6 + x ¯ 7 · y ¯ 7 + x ¯ 8 · y ¯ 8 z ¯ 2 = x ¯ 2 · y ¯ 1 + x ¯ 1 · y ¯ 2 + x ¯ 4 · y ¯ 3 + x ¯ 3 · y ¯ 4 + x ¯ 6 · y ¯ 5 + x ¯ 5 · y ¯ 6 + x ¯ 8 · y ¯ 7 + x ¯ 7 · y ¯ 8 z ¯ 3 = x ¯ 3 · y ¯ 1 + x ¯ 4 · y ¯ 2 + x ¯ 1 · y ¯ 3 + x ¯ 2 · y ¯ 4 + x ¯ 7 · y ¯ 5 + x ¯ 8 · y ¯ 6 + x ¯ 5 · y ¯ 7 + x ¯ 6 · y ¯ 8 z ¯ 4 = x ¯ 4 · y ¯ 1 + x ¯ 3 · y ¯ 2 + x ¯ 2 · y ¯ 3 + x ¯ 1 · y ¯ 4 + x ¯ 8 · y ¯ 5 + x ¯ 7 · y ¯ 6 + x ¯ 6 · y ¯ 7 + x ¯ 5 · y ¯ 8 z ¯ 5 = x ¯ 5 · y ¯ 1 + x ¯ 6 · y ¯ 2 + x ¯ 7 · y ¯ 3 + x ¯ 8 · y ¯ 4 + x ¯ 1 · y ¯ 5 + x ¯ 2 · y ¯ 6 + x ¯ 3 · y ¯ 7 + x ¯ 4 · y ¯ 8 z ¯ 6 = x ¯ 6 · y ¯ 1 + x ¯ 5 · y ¯ 2 + x ¯ 8 · y ¯ 3 + x ¯ 7 · y ¯ 4 + x ¯ 2 · y ¯ 5 + x ¯ 1 · y ¯ 6 + x ¯ 4 · y ¯ 7 + x ¯ 3 · y ¯ 8 z ¯ 7 = x ¯ 7 · y ¯ 1 + x ¯ 8 · y ¯ 2 + x ¯ 5 · y ¯ 3 + x ¯ 6 · y ¯ 4 + x ¯ 3 · y ¯ 5 + x ¯ 4 · y ¯ 6 + x ¯ 1 · y ¯ 7 + x ¯ 2 · y ¯ 8 z ¯ 8 = x ¯ 8 · y ¯ 1 + x ¯ 7 · y ¯ 2 + x ¯ 6 · y ¯ 3 + x ¯ 5 · y ¯ 4 + x ¯ 4 · y ¯ 5 + x ¯ 3 · y ¯ 6 + x ¯ 2 · y ¯ 7 + x ¯ 1 · y ¯ 8
Equations (9) define a canonical multiplication of eight-dimensional numbers with complex components. Several canonical forms are possible. For example, the following change of variables
x ¯ 1 = x 1 x ¯ 2 = ( 1 + i ) γ 2 γ 3 2 x 2 x ¯ 3 = ( 1 + i ) γ 2 γ 4 2 x 3 x ¯ 4 = γ 3 γ 4 x 4 x ¯ 5 = γ 5 γ 8 x 5 x ¯ 6 = ( 1 + i ) γ 2 γ 3 γ 8 2 γ 5 x 6 x ¯ 7 = ( 1 + i ) γ 2 γ 4 γ 5 γ 8 2 γ 6 x 7 x ¯ 8 = γ 3 γ 4 γ 5 γ 8 γ 7 x 8
leads to another canonical form:
z ¯ 1 = x ¯ 1 · y ¯ 1 − i x ¯ 2 · y ¯ 2 − i x ¯ 3 · y ¯ 3 + x ¯ 4 · y ¯ 4 + x ¯ 5 · y ¯ 5 − i x ¯ 6 · y ¯ 6 − i x ¯ 7 · y ¯ 7 + x ¯ 8 · y ¯ 8 z ¯ 2 = x ¯ 2 · y ¯ 1 + x ¯ 1 · y ¯ 2 + x ¯ 4 · y ¯ 3 + x ¯ 3 · y ¯ 4 + x ¯ 6 · y ¯ 5 + x ¯ 5 · y ¯ 6 + x ¯ 8 · y ¯ 7 + x ¯ 7 · y ¯ 8 z ¯ 3 = x ¯ 3 · y ¯ 1 + x ¯ 4 · y ¯ 2 + x ¯ 1 · y ¯ 3 + x ¯ 2 · y ¯ 4 + x ¯ 7 · y ¯ 5 + x ¯ 8 · y ¯ 6 + x ¯ 5 · y ¯ 7 + x ¯ 6 · y ¯ 8 z ¯ 4 = x ¯ 4 · y ¯ 1 − i x ¯ 3 · y ¯ 2 − i x ¯ 2 · y ¯ 3 + x ¯ 1 · y ¯ 4 + x ¯ 8 · y ¯ 5 − i x ¯ 7 · y ¯ 6 − i x ¯ 6 · y ¯ 7 + x ¯ 5 · y ¯ 8 z ¯ 5 = x ¯ 5 · y ¯ 1 − i x ¯ 6 · y ¯ 2 − i x ¯ 7 · y ¯ 3 + x ¯ 8 · y ¯ 4 + x ¯ 1 · y ¯ 5 − i x ¯ 2 · y ¯ 6 − i x ¯ 3 · y ¯ 7 + x ¯ 4 · y ¯ 8 z ¯ 6 = x ¯ 6 · y ¯ 1 + x ¯ 5 · y ¯ 2 + x ¯ 8 · y ¯ 3 + x ¯ 7 · y ¯ 4 + x ¯ 2 · y ¯ 5 + x ¯ 1 · y ¯ 6 + x ¯ 4 · y ¯ 7 + x ¯ 3 · y ¯ 8 z ¯ 7 = x ¯ 7 · y ¯ 1 + x ¯ 8 · y ¯ 2 + x ¯ 5 · y ¯ 3 + x ¯ 6 · y ¯ 4 + x ¯ 3 · y ¯ 5 + x ¯ 4 · y ¯ 6 + x ¯ 1 · y ¯ 7 + x ¯ 2 · y ¯ 8 z ¯ 8 = x ¯ 8 · y ¯ 1 − i x ¯ 7 · y ¯ 2 + x ¯ 6 · y ¯ 3 − i x ¯ 5 · y ¯ 4 + x ¯ 4 · y ¯ 5 − i x ¯ 3 · y ¯ 6 − i x ¯ 2 · y ¯ 7 + x ¯ 1 · y ¯ 8
These transformations indicate that the basis multiplications are those for which every generating number γ k , k = 2 , 3 , … , 8 , takes one of the values 1 , − 1 , i , − i . Hence, there are 4 7 = 16384 such multiplications. We denote the corresponding basis spaces by M ( j 2 j 3 j 4 j 5 j 6 j 7 j 8 ) , where j k ∈ { 1 , 2 , 3 , 4 } . Here j k = 1 , 2 , 3 , 4 corresponds to γ k = 1 , − 1 , i , − i , respectively. For example, multiplication in the space M ( 1111111 ) is defined by
z 1 = x 1 y 1 + x 2 y 2 + x 3 y 3 + x 4 y 4 + x 5 y 5 + x 6 y 6 + x 7 y 7 + x 8 y 8 z 2 = x 2 y 1 + x 1 y 2 + x 4 y 3 + x 3 y 4 + x 6 y 5 + x 5 y 6 + x 8 y 7 + x 7 y 8 z 3 = x 3 y 1 + x 4 y 2 + x 1 y 3 + x 2 y 4 + x 7 y 5 + x 8 y 6 + x 5 y 7 + x 6 y 8 z 4 = x 4 y 1 + x 3 y 2 + x 2 y 3 + x 1 y 4 + x 8 y 5 + x 7 y 6 + x 6 y 7 + x 5 y 8 z 5 = x 5 y 1 + x 6 y 2 + x 7 y 3 + x 8 y 4 + x 1 y 5 + x 2 y 6 + x 3 y 7 + x 4 y 8 z 6 = x 6 y 1 + x 5 y 2 + x 8 y 3 + x 7 y 4 + x 2 y 5 + x 1 y 6 + x 4 y 7 + x 3 y 8 z 7 = x 7 y 1 + x 8 y 2 + x 5 y 3 + x 6 y 4 + x 3 y 5 + x 4 y 6 + x 1 y 7 + x 2 y 8 z 8 = x 8 y 1 + x 7 y 2 + x 6 y 3 + x 5 y 4 + x 4 y 5 + x 3 y 6 + x 2 y 7 + x 1 y 8
whereas multiplication in M ( 1234321 ) is defined by
z 1 = x 1 y 1 − x 2 y 2 + i x 3 y 3 − i x 4 y 4 − i x 5 y 5 − i x 6 y 6 − x 7 y 7 − x 8 y 8 z 2 = x 2 y 1 + x 1 y 2 + i x 4 y 3 + i x 3 y 4 + x 6 y 5 + x 5 y 6 + i x 8 y 7 + i x 7 y 8 z 3 = x 3 y 1 − x 4 y 2 + x 1 y 3 − x 2 y 4 − x 7 y 5 + x 8 y 6 − x 5 y 7 + x 6 y 8 z 4 = x 4 y 1 + x 3 y 2 + x 2 y 3 + x 1 y 4 + i x 8 y 5 − i x 7 y 6 − i x 6 y 7 + i x 5 y 8 z 5 = x 5 y 1 − i x 6 y 2 + x 7 y 3 + i x 8 y 4 + x 1 y 5 − i x 2 y 6 + x 3 y 7 + i x 4 y 8 z 6 = x 6 y 1 − i x 5 y 2 − x 8 y 3 − i x 7 y 4 − i x 2 y 5 + x 1 y 6 − i x 4 y 7 − x 3 y 8 z 7 = x 7 y 1 + i x 8 y 2 + i x 5 y 3 + x 6 y 4 + i x 3 y 5 + x 4 y 6 + x 1 y 7 + i x 2 y 8 z 8 = x 8 y 1 + i x 7 y 2 − i x 6 y 3 − x 5 y 4 − x 4 y 5 − i x 3 y 6 + i x 2 y 7 + x 1 y 8
Thus, all multiplication laws and their corresponding spaces are ordered from M ( 1111111 ) to M ( 4444444 ) . For completeness, multiplication in M ( 4444444 ) is
z 1 = x 1 y 1 − x 2 y 2 − x 3 y 3 − x 4 y 4 − x 5 y 5 − x 6 y 6 − x 7 y 7 − x 8 y 8 z 2 = x 2 y 1 + x 1 y 2 − i x 4 y 3 − i x 3 y 4 − i x 6 y 5 − i x 5 y 6 − i x 8 y 7 − i x 7 y 8 z 3 = x 3 y 1 − i x 4 y 2 + x 1 y 3 − i x 2 y 4 − i x 7 y 5 − i x 8 y 6 − i x 5 y 7 − i x 6 y 8 z 4 = x 4 y 1 − i x 3 y 2 − i x 2 y 3 + x 1 y 4 − i x 8 y 5 − i x 7 y 6 − i x 6 y 7 − i x 5 y 8 z 5 = x 5 y 1 − i x 6 y 2 − i x 7 y 3 − i x 8 y 4 + x 1 y 5 − i x 2 y 6 − i x 3 y 7 − i x 4 y 8 z 6 = x 6 y 1 − i x 5 y 2 − i x 8 y 3 − i x 7 y 4 − i x 2 y 5 + x 1 y 6 − i x 4 y 7 − i x 3 y 8 z 7 = x 7 y 1 − i x 8 y 2 − i x 5 y 3 − i x 6 y 4 − i x 3 y 5 − i x 4 y 6 + x 1 y 7 − i x 2 y 8 z 8 = x 8 y 1 − i x 7 y 2 − i x 6 y 3 − i x 5 y 4 − i x 4 y 5 − i x 3 y 6 − i x 2 y 7 + x 1 y 8
Each basis multiplication defines a vector space of eight-dimensional numbers with complex components over C . Any other admissible multiplication can be reduced, by a change of variables analogous to (7) or (10), to one of the basis multiplications. For every basis space M ( j 2 j 3 j 4 j 5 j 6 j 7 j 8 ) , the products of basis numbers are obtained from Table 1 by substituting the values determined by j k , k = 2 , 3 , … , 8 . Table 2 gives the products for M ( 1234321 ) .
Table 2. Products of basis numbers in M ( 1234321 ) .
Table 2. Products of basis numbers in M ( 1234321 ) .
J 1 1
J 2 1
J 3 1
J 4 1
J 5 1
J 6 1
J 7 1
J 8 1
J 1 1 J 1 1 J 2 1 J 3 1 J 4 1 J 5 1 J 6 1 J 7 1 J 8 1
J 2 1 J 2 1 − J 1 1 J 4 1 − J 3 1 − i J 6 1 − i J 5 1 i J 8 1 i J 7 1
J 3 1 J 3 1 − J 4 1 i J 1 1 i J 2 1 i J 7 1 − i J 8 1 J 5 1 − J 6 1
J 4 1 J 4 1 − J 3 1 i J 2 1 − i J 1 1 − J 8 1 J 7 1 − i J 6 1 i J 5 1
J 5 1 J 5 1 − i J 6 1 i J 7 1 − J 8 1 − i J 1 1 J 2 1 − J 3 1 i J 4 1
J 6 1 J 6 1 − i J 5 1 − i J 8 1 J 7 1 − i J 2 1 − i J 1 1 − i J 4 1 J 3 1
J 7 1 J 7 1 i J 8 1 J 5 1 − i J 6 1 − J 3 1 − i J 4 1 − J 1 1 i J 2 1
J 8 1 J 8 1 i J 7 1 − J 6 1 i J 5 1 i J 4 1 J 3 1 i J 2 1 − J 1 1
The products J k 2 J m 2 follow the same pattern as Table 2 with every sign reversed, whereas the products J k 2 J m 1 are obtained by multiplying each entry of Table 2 by i. The complete 16 × 16 multiplication table exhibits a remarkable symmetry: of its 16 diagonal entries, four are J 1 1 , four are − J 1 1 , four are J 1 2 , and four are − J 1 2 . The same property holds in every space M ( j 2 j 3 j 4 j 5 j 6 j 7 j 8 ) . This symmetry distinguishes these spaces from the octonions [5,6], in which one basis element is real and the remaining elements are imaginary. The symmetric role of the basis numbers makes the proposed spaces a natural extension of one-, two- (complex), and four-dimensional algebras.

4. Matrix Spaces Associated with Eight-Dimensional Numbers

Introduce the mapping A : X ↦ A ( X ) , which assigns the following matrix to every eight-dimensional number X:
x 1 γ 2 γ 3 x 2 x 2 x 1 γ 2 γ 4 x 3 γ 3 γ 4 x 4 γ 4 x 4 γ 4 x 3 x 3 γ 3 x 4 x 4 γ 2 x 3 x 1 γ 3 x 2 γ 2 x 2 x 1 γ 5 γ 8 x 5 γ 2 γ 3 γ 8 γ 5 x 6 γ 8 x 6 γ 8 x 5 γ 2 γ 4 γ 5 γ 8 γ 6 2 x 7 γ 3 γ 4 γ 5 γ 8 γ 7 2 x 8 γ 4 γ 5 γ 8 γ 6 γ 7 x 8 γ 4 γ 5 γ 8 γ 6 γ 7 x 7 γ 5 γ 8 γ 6 x 7 γ 3 γ 8 γ 7 x 8 γ 5 γ 8 γ 7 x 8 γ 2 γ 8 γ 6 x 7 γ 5 γ 8 γ 6 x 5 γ 3 γ 8 γ 7 x 6 γ 2 γ 8 γ 6 x 6 γ 5 γ 8 γ 7 x 5 x 5 γ 2 γ 3 γ 5 x 6 x 6 γ 5 x 5 γ 2 γ 4 γ 6 x 7 γ 3 γ 4 γ 7 x 8 γ 5 γ 4 γ 7 x 8 γ 5 γ 4 γ 6 x 7 x 7 γ 3 γ 6 γ 7 x 8 x 8 γ 2 γ 7 γ 6 x 7 γ 6 x 5 γ 3 γ 6 γ 5 x 6 γ 2 γ 7 γ 5 x 6 γ 7 x 5 x 1 γ 2 γ 3 γ 5 x 2 γ 5 x 2 x 1 γ 2 γ 4 γ 6 x 3 γ 3 γ 4 γ 7 x 4 γ 5 γ 4 γ 6 x 4 γ 5 γ 4 γ 7 x 3 γ 6 x 3 γ 3 γ 6 γ 5 x 4 γ 7 x 4 γ 2 γ 7 γ 5 x 3 x 1 γ 3 γ 6 γ 7 x 2 γ 2 γ 7 γ 6 x 2 x 1
where ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) is the fixed generating tuple of the eight-dimensional space. Every entry of A ( X ) is determined by the coordinates of the corresponding eight-dimensional number X.
The mapping A is bijective. Indeed, distinct numbers X and Y correspond to distinct matrices, and every matrix of the stated form is associated with a unique eight-dimensional number with complex components.
Theorem 2.
The set of all matrices of the form (14) is closed under matrix addition, subtraction, multiplication, and scalar multiplication. The inverse of any nonsingular matrix also has the form (14).
Proof. 
The assertion follows by direct verification. □
The product of two eight-dimensional numbers X and Y can be written in matrix–vector form. On the left-hand side, multiplication is understood in the sense of (1), whereas on the right-hand side it denotes multiplication of a matrix by a vector:
X · Y = A ( X ) Y .
The significance of this matrix becomes clear when its determinant is evaluated.
We denote the matrices (14) associated with the basis spaces by A ( j 2 j 3 j 4 j 5 j 6 j 7 j 8 ) , where j k ∈ { 1 , 2 , 3 , 4 } , in accordance with the notation for the corresponding generating numbers. For example, A ( 1234321 ) has the form
x 1 − x 2 x 2 x 1 i x 3 − i x 4 i x 4 i x 3 x 3 − x 4 x 4 x 3 x 1 − x 2 x 2 x 1 − i x 5 − i x 6 x 6 x 5 − x 7 − x 8 i x 8 i x 7 − x 7 x 8 i x 8 − i x 7 − x 5 x 6 − i x 6 i x 5 x 5 − i x 6 x 6 − i x 5 x 7 i x 8 − x 8 − i x 7 x 7 i x 8 x 8 i x 7 i x 5 x 6 − i x 6 − x 5 x 1 − i x 2 − i x 2 x 1 x 3 i x 4 − i x 4 − x 3 i x 3 x 4 − x 4 − i x 3 x 1 i x 2 i x 2 x 1
Theorem 3.
Let γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 be nonzero complex numbers, and consider the eight-dimensional space generated by them. Then, for arbitrary X , Y ∈ C 8 , the mapping A = A ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) has the following properties:
1) 
A ( X ± Y ) = A ( X ) ± A ( Y ) ;
2) 
A ( c X ) = c A ( X ) for every scalar c ∈ C ;
3) 
A ( X Y ) = A ( X ) A ( Y ) ;
4) 
A X − 1 = A ( X ) − 1 ;
5) 
det A ( X ) = M ( X ) , where M ( X ) is defined by (4);
6) 
det A ( X ) ± A ( Y ) = M ( X ± Y ) , where M ( X ± Y ) is defined by (4);
7) 
det A ( X ) A ( Y ) = M ( X Y ) , where M ( X Y ) is defined by (4);
8) 
det A ( X ) − 1 = M X − 1 , provided that the symplectic modulus of X is nonzero.
Proof. 
We prove the assertion for A ( 1234321 ) ; the proofs for the remaining spaces are analogous. The first two properties are immediate. To prove the third property, note that A ( X Y ) has the form (15), with each x i replaced by the corresponding z i , where the z i are defined by (12). Let B = A ( X ) A ( Y ) . Its entries are calculated below.
b 11 = x 1 y 1 − x 2 y 2 + i x 3 y 3 − i x 4 y 4 − i x 5 y 5 − i x 6 y 6 − x 7 y 7 − x 8 y 8 = z 1 ,
b 12 = − x 1 y 2 − x 2 y 1 − i x 3 y 4 − i x 4 y 3 − x 5 y 6 − x 6 y 5 − i x 7 y 8 − i x 8 y 7 = − z 2 ,
b 13 = i x 1 y 3 − i x 2 y 4 + i x 3 y 1 − i x 4 y 2 − i x 5 y 7 + i x 6 y 8 − i x 7 y 5 + i x 8 y 6 = i z 3 ,
b 14 = − i x 1 y 4 − i x 2 y 3 − i x 3 y 2 − i x 4 y 1 + x 5 y 8 − x 6 y 7 − x 7 y 6 + x 8 y 5 = − i z 4 ,
b 15 = − i x 1 y 5 − x 2 y 6 − i x 3 y 7 + x 4 y 8 − i x 5 y 1 − x 6 y 2 − i x 7 y 3 + x 8 y 4 = − i z 5 ,
b 16 = − i x 1 y 6 − x 2 y 5 + i x 3 y 8 − x 4 y 7 − x 5 y 2 − i x 6 y 1 − x 7 y 4 + i x 8 y 3 = − i z 6 ,
b 17 = − x 1 y 7 − i x 2 y 8 − i x 3 y 5 − x 4 y 6 − i x 5 y 3 − x 6 y 4 − x 7 y 1 − i x 8 y 2 = − z 7 ,
b 18 = − x 1 y 8 − i x 2 y 7 + i x 3 y 6 + x 4 y 5 + x 5 y 4 + i x 6 y 3 − i x 7 y 2 − x 8 y 1 = − z 8 .
Thus, the first row of B coincides with the first row of A(XY). The remaining rows are verified in the same way, proving property 3. For property 4, use (6), property 3, and identity (4) to obtain
A X − 1 = M ( X ) ¯ | M ( X ) | 2 A ( X * ) .
A ( X − 1 ) A ( X ) = M ( X ) ¯ | M ( X ) | 2 A ( X * ) A ( X ) = M ( X ) ¯ | M ( X ) | 2 A ( X X * ) = M ( X ) ¯ | M ( X ) | 2 M re ( X ) A ( J 1 1 ) + M im ( X ) A ( J 1 2 ) = M ( X ) ¯ | M ( X ) | 2 M ( X ) A ( J 1 1 ) = E .
where E is the identity matrix. This proves property 4. To prove property 5, we calculate the determinant of matrix (15) using the following change of variables:
y 1 = x 1 , y 2 = i x 2 , y 3 = i x 3 , y 4 = i i x 4 , y 5 = i i x 5 , y 6 = − i x 6 , y 7 = − i x 7 , y 8 = − i x 8
In the new variables y k , matrix (15) takes the form
y 1 i y 2 y 2 i y 1 i y 3 i i y 4 i y 4 i i y 3 y 3 i i y 4 i y 4 i i y 3 i y 1 i y 2 y 2 i y 1 i i y 5 i i y 6 − i y 6 i − i y 5 i i y 7 i y 8 y 8 y 7 i y 7 − i y 8 y 8 − y 7 i y 5 i − i y 6 i − y 6 i y 5 i y 5 i i − y 6 i y 6 i i − y 5 i y 7 i y 8 − y 8 i − y 7 y 7 i y 8 y 8 i y 7 y 5 i y 6 i i − y 6 i − y 5 i i y 1 − y 2 − y 2 y 1 y 3 i y 4 i − y 4 i − y 3 i i y 3 − i y 4 1 i y 4 1 − i y 3 y 1 y 2 y 2 y 1
Multiply the second and fourth rows of matrix (17) by i and subtract them from the first and third rows, respectively; add the sixth row to the fifth, and subtract the eighth row from the seventh.
r e s i z e b o x w i d t h 418.45818 p t y 1 − y 2 i y 2 − i y 1 y 2 i y 1 i y 3 − i y 4 i i y 4 − i i y 3 i y 4 i i y 3 y 3 i − y 4 i i y 4 i − i y 3 i y 4 i i y 3 i y 1 − y 2 i y 2 − i y 1 y 2 i y 1 i i y 5 + i i y 6 i i y 6 + i i y 5 − i y 6 i − i y 5 i i y 7 − i y 8 i y 8 − i y 7 y 8 y 7 i y 7 − i y 8 − i y 8 + i y 7 y 8 − y 7 i y 5 i + i y 6 i − i y 6 i − i y 5 i − y 6 i y 5 i y 5 i i + y 6 i i − y 6 i − y 5 i y 6 i i i i y 5 y 7 i − y 8 i y 8 − y 7 − y 8 i − y 7 y 7 i − y 8 i y 8 − y 7 y 8 i y 7 y 5 i + y 6 i y 6 i i + y 5 i i − y 6 i − y 5 i i y 1 − y 2 − y 2 + y 1 − y 2 y 1 y 3 i − y 4 i y 4 i − y 3 i − y 4 i − y 3 i i y 3 − i y 4 − i y 4 + i y 3 i y 4 − i y 3 y 1 − y 2 y 2 − y 1 y 2 y 1
Next, multiply the first and third columns of the resulting matrix by i and add them to the second and fourth columns, respectively; subtract the fifth column from the sixth, and add the seventh column to the eighth. Interchange rows 2 and 5 and rows 4 and 7, and then perform the analogous interchanges of columns. The resulting block matrix is
A = A 1 0 A 2 A 3 ,
in which the upper-right block is the zero matrix and the remaining blocks have the form
A 1 = y 1 − y 2 i i y 5 + i i y 6 y 5 i i + y 6 i i y 1 − y 2 i y 3 − i y 4 i y 7 − i y 8 y 7 i − y 8 i y 3 i − y 4 i y 3 i − y 4 i i y 7 − i y 8 y 7 i − y 8 i i y 3 − i y 4 y 1 − y 2 i y 5 i + i y 6 i y 5 i + y 6 i y 1 − y 2 ,
A 2 = y 2 i − i y 6 i y 6 i i − y 2 i y 4 i y 8 − y 8 i − y 4 i y 4 i i y 8 y 8 i i y 4 y 2 i − y 6 i − y 6 i y 2 ,
A 3 = y 1 + y 2 i y 6 − i y 5 − y 5 i + y 6 i y 1 + y 2 i y 3 + i y 4 y 7 + y 8 − y 7 − y 8 − y 3 i − y 4 i y 3 i + y 4 i − y 7 − y 8 y 7 + y 8 − i y 3 − i y 4 y 1 + y 2 y 5 − y 6 i i y 5 − i y 6 y 1 + y 2 .
Therefore, det A = det A 1 det A 3 . Evaluating the determinants of these four-dimensional matrices gives
det A = ( y 1 − y 2 − y 5 − y 6 ) 2 − ( y 3 − y 4 − y 7 + y 8 ) 2 × ( y 1 − y 2 + y 5 + y 6 ) 2 − ( y 3 − y 4 + y 7 − y 8 ) 2 × ( y 1 + y 2 − y 5 + y 6 ) 2 − ( y 3 + y 4 − y 7 − y 8 ) 2 × ( y 1 + y 2 + y 5 − y 6 ) 2 − ( y 3 + y 4 + y 7 + y 8 ) 2 .
Returning to the original variables x k yields
det A = ( x 1 − i x 2 − i i x 5 + i x 6 ) 2 − ( i x 3 − i i x 4 + i x 7 − i x 8 ) 2 × ( x 1 − i x 2 + i i x 5 − i x 6 ) 2 − ( i x 3 − i i x 4 − i x 7 + i x 8 ) 2 × ( x 1 + i x 2 − i i x 5 − i x 6 ) 2 − ( i x 3 + i i x 4 + i x 7 + i x 8 ) 2 × ( x 1 + i x 2 + i i x 5 + i x 6 ) 2 − ( i x 3 + i i x 4 − i x 7 − i x 8 ) 2 .
Direct evaluation of product (4) now confirms that det A = M ( X ) . This calculation can be performed either manually or with symbolic-computation software. Property 5 is proved, and the remaining assertions follow from the properties already established. □
Thus, for every fixed tuple ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) , the bijection A ( j 2 j 3 j 4 j 5 j 6 j 7 j 8 ) identifies the space of eight-dimensional numbers with complex components with the corresponding space of matrices of the form (14). It preserves the arithmetic operations and is therefore a homomorphism with respect to both addition and multiplication. We have consequently constructed 16,384 matrix spaces, each isomorphic to its corresponding space of eight-dimensional complex numbers. The invertible elements of each matrix space form an Abelian group under multiplication, while the full space is an Abelian group under addition.

5. Gates of Three-Qubit Quantum Systems

Quantum computation consists of the successive application of unitary operators U, called quantum gates, to the state of a quantum system. A unitary operator acting on an n-qubit system is represented by a 2 n × 2 n matrix. Thus, a gate for a three-qubit system is represented by an 8 × 8 matrix U:
U = x 11 x 12 x 21 x 22 x 13 x 14 x 23 x 24 x 31 x 32 x 41 x 42 x 33 x 34 x 43 x 44 x 15 x 16 x 25 x 26 x 17 x 18 x 27 x 28 x 35 x 36 x 45 x 46 x 37 x 38 x 47 x 48 x 51 x 52 x 61 x 62 x 53 x 54 x 63 x 64 x 71 x 72 x 81 x 82 x 73 x 74 x 83 x 84 x 55 x 56 x 65 x 66 x 57 x 58 x 67 x 68 x 75 x 76 x 85 x 86 x 77 x 78 x 87 x 88
where x i j ∈ C . If U is unitary, its Hermitian conjugate U * is also unitary; therefore,
U * = x ¯ 11 x ¯ 21 x ¯ 12 x ¯ 22 x ¯ 31 x ¯ 41 x ¯ 32 x ¯ 42 x ¯ 13 x ¯ 23 x ¯ 14 x ¯ 24 x ¯ 33 x ¯ 43 x ¯ 34 x ¯ 44 x ¯ 51 x ¯ 61 x ¯ 52 x ¯ 62 x ¯ 71 x ¯ 81 x ¯ 72 x ¯ 82 x ¯ 53 x ¯ 63 x ¯ 54 x ¯ 64 x ¯ 73 x ¯ 83 x ¯ 74 x ¯ 84 x ¯ 15 x ¯ 25 x ¯ 16 x ¯ 26 x ¯ 35 x ¯ 45 x ¯ 36 x ¯ 46 x ¯ 17 x ¯ 27 x ¯ 18 x ¯ 28 x ¯ 37 x ¯ 47 x ¯ 38 x ¯ 48 x ¯ 55 x ¯ 65 x ¯ 56 x ¯ 66 x ¯ 75 x ¯ 85 x ¯ 76 x ¯ 86 x ¯ 57 x ¯ 67 x ¯ 58 x ¯ 68 x ¯ 77 x ¯ 87 x ¯ 78 x ¯ 88
is likewise a gate for a three-qubit quantum system. By definition, U · U * = E , where E is the 8 × 8 identity matrix.
In the preceding section, we constructed spaces of eight-dimensional matrices whose entries are determined by the components of an eight-dimensional number. We now determine the conditions under which these matrices are quantum gates for three-qubit systems.
Let X = ( x 1 , x 2 , x 3 , x 4 , x 5 , x 6 , x 7 , x 8 ) ∈ C 8 represent the state | φ 〉 of a three-qubit quantum system, i.e.,
| φ 〉 = x 1 | 000 〉 + x 2 | 001 〉 + x 3 | 010 〉 + x 4 | 011 〉 + x 5 | 100 〉 + x 6 | 101 〉 + x 7 | 110 〉 + x 8 | 111 〉
To every such state | φ 〉 , we can associate one of the 16,384 matrices of the form (14), with a generating tuple ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) ∈ { 1 , − 1 , i , − i } 7 . Since | φ 〉 is normalized, its components satisfy
ρ 0 x 1 , x 2 , x 3 , x 4 , x 5 , x 6 , x 7 , x 8 ≡ ∑ k = 1 8 x k 2 = 1
Theorem 4.
Let X = ( x 1 , x 2 , x 3 , x 4 , x 5 , x 6 , x 7 , x 8 ) ∈ C 8 be a state of a three-qubit quantum system, and let A ( X ) be its associated matrix in a basis space whose generating numbers satisfy γ k ∈ { 1 , − 1 , i , − i } . Suppose that the components of X satisfy the following conditions
ρ 1 ≡ x 1 x ¯ 2 + γ 2 γ 3 x 2 x ¯ 1 + γ 2 x 3 x ¯ 4 + γ 3 x 4 x ¯ 3 + + γ 5 x 5 x ¯ 6 + γ 2 γ 3 γ 5 x 6 x ¯ 5 + γ 2 γ 7 γ 6 x 7 x ¯ 8 + γ 3 γ 6 γ 7 x 8 x ¯ 7 = 0 ρ 2 ≡ x 1 x ¯ 3 + γ 2 x 2 x ¯ 4 + γ 2 γ 4 x 3 x ¯ 1 + γ 4 x 4 x ¯ 2 + + γ 6 x 5 x ¯ 7 + γ 2 γ 7 γ 5 x 6 x ¯ 8 + γ 2 γ 4 γ 6 x 7 x ¯ 5 + γ 4 γ 5 γ 7 x 8 x ¯ 6 = 0 ρ 3 ≡ x 1 x ¯ 4 + γ 3 x 2 x ¯ 3 + γ 4 x 3 x ¯ 2 + γ 3 γ 4 x 4 x ¯ 1 + + γ 7 x 5 x ¯ 8 + γ 3 γ 6 γ 5 x 6 x ¯ 7 + γ 4 γ 5 γ 6 x 7 x ¯ 6 + γ 3 γ 4 γ 7 x 8 x ¯ 5 = 0 ρ 4 ≡ x 1 x ¯ 5 + γ 5 x 2 x ¯ 6 + γ 6 x 3 x ¯ 7 + γ 7 x 4 x ¯ 8 + + γ 5 γ 8 x 5 x ¯ 1 + γ 8 x 6 x ¯ 2 + γ 5 γ 8 γ 6 x 7 x ¯ 3 + γ 5 γ 8 γ 7 x 8 x ¯ 4 = 0 ρ 5 ≡ x 1 x ¯ 6 + γ 2 γ 3 γ 5 x 2 x ¯ 5 + γ 2 γ 7 γ 5 x 3 x ¯ 8 + γ 3 γ 6 γ 5 x 4 x ¯ 7 + + γ 8 x 5 x ¯ 2 + γ 2 γ 3 γ 8 γ 5 x 6 x ¯ 1 + γ 2 γ 8 γ 6 x 7 x ¯ 4 + γ 3 γ 8 γ 7 x 8 x ¯ 3 = 0 ρ 6 ≡ x 1 x ¯ 7 + γ 2 γ 7 γ 6 x 2 x ¯ 8 + γ 2 γ 4 γ 6 x 3 x ¯ 5 + γ 4 γ 5 γ 6 x 4 x ¯ 6 + + γ 5 γ 8 γ 6 x 5 x ¯ 3 + γ 2 γ 8 γ 6 x 6 x ¯ 4 + γ 2 γ 4 γ 5 γ 8 γ 6 2 x 7 x ¯ 1 + γ 4 γ 5 γ 8 γ 6 γ 7 x 8 x ¯ 2 = 0 ρ 7 ≡ x 1 x ¯ 8 + γ 3 γ 6 γ 7 x 2 x ¯ 7 + γ 4 γ 5 γ 7 x 3 x ¯ 6 + γ 3 γ 4 γ 7 x 4 x ¯ 5 + + γ 5 γ 8 γ 7 x 5 x ¯ 4 + γ 3 γ 8 γ 7 x 6 x ¯ 3 + γ 4 γ 5 γ 8 γ 6 γ 7 x 7 x ¯ 2 + γ 3 γ 4 γ 5 γ 8 γ 7 2 x 8 x ¯ 1 = 0
where x ¯ j denotes the complex conjugate of x j . Then A ( X ) is a three-qubit quantum gate.
Proof. 
Multiply A ( X ) by its Hermitian conjugate A * ( X ) :
A ( X ) A * ( X ) = ρ 0 ρ 1 1 γ 2 γ 3 ρ 1 ρ 0 ρ 2 ρ 3 1 γ 3 ρ 3 1 γ 2 ρ 2 1 γ 2 γ 4 ρ 2 1 γ 4 ρ 3 1 γ 3 γ 4 ρ 3 1 γ 4 ρ 2 ρ 0 1 γ 2 ρ 1 1 γ 3 ρ 1 ρ 0 ρ 4 ρ 5 γ 5 γ 2 γ 3 ρ 5 1 γ 5 ρ 4 ρ 6 ρ 7 γ 7 γ 3 γ 6 ρ 7 γ 6 γ 2 γ 7 ρ 6 γ 6 γ 2 γ 4 ρ 6 γ 7 γ 4 γ 5 ρ 7 γ 7 γ 3 γ 4 ρ 7 γ 6 γ 4 γ 5 ρ 6 1 γ 6 ρ 4 γ 5 γ 2 γ 7 ρ 5 γ 5 γ 3 γ 6 ρ 5 1 γ 7 ρ 4 1 γ 5 γ 8 ρ 4 1 γ 8 ρ 5 γ 5 γ 2 γ 3 γ 8 ρ 5 1 γ 8 ρ 4 γ 6 γ 5 γ 8 ρ 6 γ 7 γ 5 γ 8 ρ 7 γ 7 γ 3 γ 8 ρ 7 γ 6 γ 2 γ 8 ρ 6 γ 6 2 γ 2 γ 4 γ 5 γ 8 ρ 6 γ 6 γ 7 γ 4 γ 5 γ 8 ρ 7 γ 7 2 γ 3 γ 4 γ 5 γ 8 ρ 7 γ 6 γ 7 γ 4 γ 5 γ 8 ρ 6 γ 6 γ 5 γ 8 ρ 4 γ 6 γ 2 γ 8 ρ 5 γ 7 γ 3 γ 8 ρ 5 γ 7 γ 5 γ 8 ρ 4 ρ 0 1 γ 5 ρ 1 γ 5 γ 2 γ 3 ρ 1 ρ 0 1 γ 6 ρ 2 1 γ 7 ρ 3 γ 5 γ 3 γ 6 ρ 3 γ 5 γ 2 γ 7 ρ 2 γ 6 γ 2 γ 4 ρ 2 γ 6 γ 4 γ 5 ρ 3 γ 7 γ 3 γ 4 ρ 3 γ 7 γ 4 γ 5 ρ 2 ρ 0 γ 6 γ 2 γ 7 ρ 1 γ 7 γ 3 γ 6 ρ 1 ρ 0
Here we used | γ i | = 1 and γ ¯ i = 1 / γ i , i = 2 , … , 8 , where γ ¯ i denotes the complex conjugate of γ i . Under conditions (22), A ( X ) A * ( X ) = E , which proves the theorem. □
We have therefore obtained 16,384 commuting families of gates for three-qubit quantum systems. For example, in the space M(1234321), conditions (22) take the form
ρ 1 = x 1 x ¯ 2 − x 2 x ¯ 1 + x 3 x ¯ 4 − x 4 x ¯ 3 − i x 5 x ¯ 6 − i x 6 x ¯ 5 + i x 7 x ¯ 8 + i x 8 x ¯ 7 = 0 ρ 2 = x 1 x ¯ 3 + x 2 x ¯ 4 + i x 3 x ¯ 1 + i x 4 x ¯ 2 + i x 5 x ¯ 7 − i x 6 x ¯ 8 + x 7 x ¯ 5 − x 8 x ¯ 6 = 0 ρ 3 = x 1 x ¯ 4 − x 2 x ¯ 3 + i x 3 x ¯ 2 − i x 4 x ¯ 1 − x 5 x ¯ 8 + x 6 x ¯ 7 − i x 7 x ¯ 6 + i x 8 x ¯ 5 = 0 ρ 4 = x 1 x ¯ 5 − i x 2 x ¯ 6 + i x 3 x ¯ 7 − x 4 x ¯ 8 − i x 5 x ¯ 1 + x 6 x ¯ 2 − x 7 x ¯ 3 + i x 8 x ¯ 4 = 0 ρ 5 = x 1 x ¯ 6 − i x 2 x ¯ 5 − i x 3 x ¯ 8 + x 4 x ¯ 7 + x 5 x ¯ 2 − i x 6 x ¯ 1 − i x 7 x ¯ 4 + x 8 x ¯ 3 = 0 ρ 6 = x 1 x ¯ 7 + i x 2 x ¯ 8 + x 3 x ¯ 5 − i x 4 x ¯ 6 − x 5 x ¯ 3 − i x 6 x ¯ 4 − x 7 x ¯ 1 + i x 8 x ¯ 2 = 0 ρ 7 = x 1 x ¯ 8 − i x 2 x ¯ 7 − x 3 x ¯ 6 + i x 4 x ¯ 5 + i x 5 x ¯ 4 + x 6 x ¯ 3 + i x 7 x ¯ 2 − x 8 x ¯ 1 = 0
Conditions (22) for all other gate families are obtained analogously by substituting the corresponding values of the generating numbers. If the state has real components, these conditions reduce to
ρ 1 ≡ 1 + γ 2 γ 3 x 1 x 2 + γ 5 x 5 x 6 + γ 2 + γ 3 x 3 x 4 + γ 6 γ 7 x 7 x 8 = 0 ρ 2 ≡ 1 + γ 2 γ 4 x 1 x 3 + γ 6 x 5 x 7 + γ 2 + γ 4 x 2 x 4 + γ 5 γ 7 x 6 x 8 = 0 ρ 3 ≡ ( 1 + γ 3 γ 4 ) ( x 1 x 4 + γ 7 x 5 x 8 ) + ( γ 3 + γ 4 ) ( x 2 x 3 + γ 5 γ 6 x 6 x 7 ) = 0 ρ 4 ≡ 1 + γ 5 γ 8 x 1 x 5 + γ 6 x 3 x 7 + γ 7 x 4 x 8 + γ 5 + γ 8 x 2 x 6 = 0 ρ 5 ≡ ( 1 + γ 2 γ 3 γ 5 γ 8 ) x 1 x 6 + ( γ 2 γ 3 γ 5 + γ 8 ) x 2 x 5 + + γ 7 ( γ 2 γ 5 + γ 3 γ 8 ) x 3 x 8 + γ 6 ( γ 3 γ 5 + γ 2 γ 8 ) x 4 x 7 = 0 ρ 6 ≡ ( 1 + γ 2 γ 4 γ 5 γ 8 ) x 1 x 7 + γ 6 γ 7 ( γ 2 + γ 4 γ 5 γ 8 ) x 2 x 8 + + γ 6 ( γ 2 γ 4 + γ 5 γ 8 ) x 3 x 5 + γ 6 ( γ 4 γ 5 + γ 2 γ 8 ) x 4 x 6 = 0 ρ 7 ≡ ( 1 + γ 3 γ 4 γ 5 γ 8 ) x 1 x 8 + γ 6 γ 7 ( γ 3 + γ 4 γ 5 γ 8 ) x 2 x 7 + + γ 7 ( γ 4 γ 5 + γ 3 γ 8 ) x 3 x 6 + γ 7 ( γ 3 γ 4 + γ 5 γ 8 ) x 4 x 5 = 0
Three-qubit gates with real components are considered in greater detail below, where their general form is derived.
The system of equations (21) and (22) is always consistent.
Lemma 2.
All basis numbers J k 1 and J k 2 , k = 1 , 2 , … , 8 , are solutions of system (21), (22).
Proof. 
The proof is immediate. □
Thus, every eight-dimensional number has an associated matrix of the form (14), which can be reduced to one of the basis spaces by a transformation of type (7) or (10). All matrices within a fixed basis space commute under multiplication, and the space is closed under associative addition and multiplication. When conditions (21) and (22) hold, the corresponding matrices are gates for three-qubit quantum systems. This construction provides infinitely many such gates explicitly. Gates from different basis families need not commute, whereas all gates belonging to the same basis family commute.

6. Unitary States of Three-Qubit Quantum Systems

Definition 1.
A state X = ( x 1 , x 2 , x 3 , x 4 , x 5 , x 6 , x 7 , x 8 ) ∈ C 8 of a three-qubit quantum system is called a unitary state if there exists a generating tuple ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) ∈ C 7 , with γ k ∈ { 1 , − 1 , i , − i } , such that the components of X satisfy conditions (21) and (22).
Terminological note. Throughout this paper, the term unitary state is used exclusively in a specific algebraic sense. It denotes a normalized three-qubit state whose multiplication matrix, constructed in one of the basis spaces considered here, is unitary. It does not mean that unitarity is attributed to a quantum state in the conventional sense; ordinarily, the term unitary applies to operators.
Equivalently, X is unitary if there exists a basis space M ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) in which its associated matrix A ( X ) is a three-qubit quantum gate.
Remark 1.
The restriction γ k ∈ { 1 , − 1 , i , − i } in Definition 1 is imposed only to confine the discussion to the basis spaces; it is not essential to the general construction.
By Lemma 2, every basis state of the quantum system is a unitary state in every basis space.
Definition 2.
The following states are called quasi-basis states:
X 1 = ( x 1 , 0 , 0 , 0 , 0 , 0 , 0 , 0 ) T , X 2 = ( 0 , x 2 , 0 , 0 , 0 , 0 , 0 , 0 ) T , X 3 = ( 0 , 0 , x 3 , 0 , 0 , 0 , 0 , 0 ) T , X 4 = ( 0 , 0 , 0 , x 4 , 0 , 0 , 0 , 0 ) T , X 5 = ( 0 , 0 , 0 , 0 , x 5 , 0 , 0 , 0 ) T , X 6 = ( 0 , 0 , 0 , 0 , 0 , x 6 , 0 , 0 ) T , X 7 = ( 0 , 0 , 0 , 0 , 0 , 0 , x 7 , 0 ) T , X 8 = ( 0 , 0 , 0 , 0 , 0 , 0 , 0 , x 8 ) T , x j ∈ C , | x j | = 1 .
Clearly, every quasi-basis state satisfies (22) and is therefore a unitary state.
Every unitary state X in a fixed basis space M ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) determines a unitary operator A ( X ) , which we call the gate of the unitary state X. Let Y be an arbitrary three-qubit state. Applying A ( X ) to Y gives A ( X ) Y = X · Y ; thus, gate application is equivalent to multiplication of the eight-dimensional numbers X and Y in that basis space. If Y is also unitary in the same space, the resulting state is unitary there. The following theorem describes what happens when Y is not unitary.
Theorem 5.
The gate associated with a unitary state in a basis space M ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) maps every unitary state to a unitary state and every nonunitary state to a nonunitary state within that space.
Proof. 
For simplicity, we prove the theorem for M ( 1234321 ) ; the remaining cases are analogous. Let X = ( x 1 , x 2 , x 3 , x 4 , x 5 , x 6 , x 7 , x 8 ) ∈ C 8 be a unitary state in M ( 1234321 ) , and let A ( X ) be its gate of the form (15), with the components of X satisfying (22). For an arbitrary three-qubit state Y, the gate A ( X ) produces the state
A ( X ) Y = x 1 − x 2 x 2 x 1 i x 3 − i x 4 i x 4 i x 3 x 3 − x 4 x 4 x 3 x 1 − x 2 x 2 x 1 − i x 5 − i x 6 x 6 x 5 − x 7 − x 8 i x 8 i x 7 − x 7 x 8 i x 8 − i x 7 − x 5 x 6 − i x 6 i x 5 x 5 − i x 6 x 6 − i x 5 x 7 i x 8 − x 8 − i x 7 x 7 i x 8 x 8 i x 7 i x 5 x 6 − i x 6 − x 5 x 1 − i x 2 − i x 2 x 1 x 3 i x 4 − i x 4 − x 3 i x 3 x 4 − x 4 − i x 3 x 1 i x 2 i x 2 x 1 y 1 y 2 y 3 y 4 y 5 y 6 y 7 y 8
= x 1 y 1 − x 2 y 2 + i x 3 y 3 − i x 4 y 4 − i x 5 y 5 − i x 6 y 6 − x 7 y 7 − x 8 y 8 x 2 y 1 + x 1 y 2 + i x 4 y 3 + i x 3 y 4 + x 6 y 5 + x 5 y 6 + i x 8 y 7 + i x 7 y 8 x 3 y 1 − x 4 y 2 + x 1 y 3 − x 2 y 4 − x 7 y 5 + x 8 y 6 − x 5 y 7 + x 6 y 8 x 4 y 1 + x 3 y 2 + x 2 y 3 + x 1 y 4 + i x 8 y 5 − i x 7 y 6 − i x 6 y 7 + i x 5 y 8 x 5 y 1 − i x 6 y 2 + x 7 y 3 + i x 8 y 4 + x 1 y 5 − i x 2 y 6 + x 3 y 7 + i x 4 y 8 x 6 y 1 − i x 5 y 2 − x 8 y 3 − i x 7 y 4 − i x 2 y 5 + x 1 y 6 − i x 4 y 7 − x 3 y 8 x 7 y 1 + i x 8 y 2 + i x 5 y 3 + x 6 y 4 + i x 3 y 5 + x 4 y 6 + x 1 y 7 + i x 2 y 8 x 8 y 1 + i x 7 y 2 − i x 6 y 3 − x 5 y 4 − x 4 y 5 − i x 3 y 6 + i x 2 y 7 + x 1 y 8 .
To verify that the resulting state is unitary, evaluate ρ k ( A ( X ) Y ) using (23). Substitution of the components of A ( X ) Y into the first expression in (23), followed by collection of like terms, gives
ρ 1 ( A ( X ) Y ) = ρ 0 ( X ) ρ 1 ( Y ) + ρ 1 ( X ) ρ 0 ( Y ) − i ρ 2 ( X ) ρ 3 ( Y ) − i ρ 3 ( X ) ρ 2 ( Y ) + ρ 4 ( X ) ρ 5 ( Y ) + ρ 5 ( X ) ρ 4 ( Y ) − i ρ 6 ( X ) ρ 7 ( Y ) − i ρ 7 ( X ) ρ 6 ( Y ) .
Because X is unitary, condition (23) gives ρ 1 ( A ( X ) Y ) = ρ 1 ( Y ) . The remaining quantities are obtained analogously: ρ i ( A ( X ) Y ) = ρ i ( Y ) , i = 2 , 3 , … , 7 . Hence, if Y is unitary in M ( 1234321 ) , then A ( X ) Y is also unitary there; conversely, if Y is nonunitary in that space, then A ( X ) Y is nonunitary as well. □
Remark 2.
In this theorem, nonunitarity means nonunitarity relative to the basis space under consideration. A state may be nonunitary in one basis space but unitary in another.
An important property of unitary states of two-qubit systems [1] is that any unitary state can be transformed into any other unitary state in a single step, with an explicit gate implementing the transformation. Three-qubit unitary states possess an analogous property.
Theorem 6.
Let X ∈ M ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) be a unitary state of a three-qubit quantum system. For every quasi-basis state X k , k = 1 , 2 , … , 8 , there exists a gate A ( Y ) such that A ( Y ) X = X k , for some Y ∈ M ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) .
Proof. 
Set Y = X k X − 1 . Since X is unitary, its inverse exists and belongs to the same basis space. The quasi-basis state X k also belongs to that space; hence so does Y. Multiplication by X gives Y X = X k . Because Y X = A ( Y ) X , the required gate is A ( Y ) = A ( X k X − 1 ) . □
Theorem 6 states that, within a fixed basis space M ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) , any unitary state can be mapped in one step to any quasi-basis state and hence, in particular, to any basis state.
Corollary 1.
Let X , Y ∈ M ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) be two unitary states. Then there exists a gate A ( Z ) that maps X to Y.
Proof. 
By Theorem 6, there are gates A ( X ˜ ) and A ( Y ˜ ) that map X and Y, respectively, to the same quasi-basis state X k . Hence A − 1 ( Y ˜ ) A ( X ˜ ) maps X to Y. □
Theorem 7.
Let X and Y be two unitary states of a three-qubit quantum system. Then there exists a gate that maps X to Y.
Proof. 
Suppose that X belongs to M ( γ 2 , γ 3 , γ 4 , γ 5 , γ 6 , γ 7 , γ 8 ) , whereas Y belongs to M ( δ 2 , δ 3 , δ 4 , δ 5 , δ 6 , δ 7 , δ 8 ) . By Lemma 2, the basis state X 1 = ( 1 , 0 , 0 , 0 , 0 , 0 , 0 , 0 ) T belongs to both spaces. Theorem 6 therefore provides a gate A ( Z 1 ) mapping X to X 1 and a gate A ( Z 2 ) mapping Y to X 1 . Consequently, A − 1 ( Z 2 ) A ( Z 1 ) maps X to Y. □
Thus, the proposed definition distinguishes unitary and nonunitary states. All quasi-basis states, and consequently all basis states, are unitary. The general existence of a unitary operator mapping one normalized vector to another is well known. The specific content of Theorem 7 is constructive: the required gate is expressed explicitly through multiplication and inversion of eight-dimensional numbers and belongs to one of the algebraic families constructed above. The choice of a quasi-basis state yields infinitely many such explicit representations.

7. General Form of Unitary States with Real Components

Consider unitary states of three-qubit quantum systems with real components. As shown above, such states satisfy (21) and (24). Assume that the generating numbers are also real, so that each takes the value + 1 or − 1 . As proved in [4], there are 112 basis spaces in which the symplectic modulus and spectral norm can be defined. In these spaces, conditions (24) reduce to three equations. For example, in M ( 1111211 ) , where γ 2 = γ 3 = γ 4 = γ 5 = γ 7 = γ 8 = 1 and γ 6 = − 1 , they take the form
x 1 x 2 + x 3 x 4 + x 5 x 6 + x 7 x 8 = 0 x 1 x 3 + x 2 x 4 + x 5 x 7 + x 6 x 8 = 0 x 1 x 4 + x 2 x 3 + x 5 x 8 + x 6 x 7 = 0
Analogous conditions can be written for the other 111 spaces. It was proved in [4] that, together with (21), these conditions are equivalent to orthogonality of A ( X ) ; the corresponding eight-dimensional numbers were denoted by N 0 .
We now derive the general form of three-qubit unitary states with real components. Rewrite system (21), (25) as follows:
x 1 2 + x 2 2 + x 3 2 + x 4 2 = 1 − x 5 2 − x 6 2 − x 7 2 − x 8 2 x 1 x 2 + x 3 x 4 = − x 5 x 6 − x 7 x 8 x 1 x 3 + x 2 x 4 = − x 5 x 7 − x 6 x 8 x 1 x 4 + x 2 x 3 = − x 5 x 8 − x 6 x 7
Regard this as a system in the unknowns x 1 , x 2 , x 3 , x 4 . Its left-hand side resembles the square of the four-dimensional number ( x 1 , x 2 , x 3 , x 4 ) . We therefore consider the general system
x 1 2 + x 2 2 + x 3 2 + x 4 2 = a 1 2 x 1 x 2 + 2 x 3 x 4 = a 2 2 x 1 x 3 + 2 x 2 x 4 = a 3 2 x 1 x 4 + 2 x 2 x 3 = a 4
Subtract the second equation from the first and the fourth from the third; then add the first two equations and the last two equations. System (27) becomes
( x 1 − x 2 ) 2 + ( x 3 − x 4 ) 2 = a 1 − a 2 x 1 − x 2 x 3 − x 4 = 1 2 a 3 − 1 2 a 4 ( x 1 + x 2 ) 2 + ( x 3 + x 4 ) 2 = a 1 + a 2 x 1 + x 2 x 3 + x 4 = 1 2 a 3 + 1 2 a 4
Introduce the notation
y 1 = x 1 − x 2 , y 2 = x 3 − x 4 , y 3 = x 1 + x 2 , y 4 = x 3 + x 4
Then system (28) takes the form
( y 1 ) 2 + ( y 2 ) 2 = a 1 − a 2 2 y 1 y 2 = a 3 − a 4 ( y 3 ) 2 + ( y 4 ) 2 = a 1 + a 2 2 y 3 y 4 = a 3 + a 4 .
Transform this system into the following equivalent form:
( y 1 + y 2 ) 2 = a 1 − a 2 + a 3 − a 4 ( y 1 − y 2 ) 2 = a 1 − a 2 − a 3 + a 4 ( y 3 + y 4 ) 2 = a 1 + a 2 + a 3 + a 4 ( y 3 − y 4 ) 2 = a 1 + a 2 − a 3 − a 4
The system separates into two independent subsystems:
( y 1 + y 2 ) 2 = a 1 − a 2 + a 3 − a 4 ( y 1 − y 2 ) 2 = a 1 − a 2 − a 3 + a 4 ,
( y 3 + y 4 ) 2 = a 1 + a 2 + a 3 + a 4 ( y 3 − y 4 ) 2 = a 1 + a 2 − a 3 − a 4 ,
each of which has four solutions, written as
y 1 , y 2 = b 1 + b 2 2 , b 1 − b 2 2 , b 1 − b 2 2 , b 1 + b 2 2 , − b 1 + b 2 2 , − b 1 − b 2 2 , − b 1 − b 2 2 , − b 1 + b 2 2 ,
where
b 1 = a 1 − a 2 + a 3 − a 4 , b 2 = a 1 − a 2 − a 3 + a 4
b 3 = a 1 + a 2 + a 3 + a 4 , b 4 = a 1 + a 2 − a 3 − a 4
Four analogous solutions are obtained for ( y 3 , y 4 ) in terms of b 3 and b 4 . Hence, system (30) has 16 possible solutions. Substitution into (29) gives the 16 solutions of the original system (27), listed in Table 3.
Table 3. Solutions of system (27)
Table 3. Solutions of system (27)
№
x 1
x 2
x 3
x 4
1 b 1 + b 2 + b 3 + b 4 4 − b 1 − b 2 + b 3 + b 4 4 b 1 − b 2 + b 3 − b 4 4 − b 1 + b 2 + b 3 − b 4 4
2 b 1 + b 2 + b 3 − b 4 4 − b 1 − b 2 + b 3 − b 4 4 b 1 − b 2 + b 3 + b 4 4 − b 1 + b 2 + b 3 + b 4 4
3 b 1 + b 2 − b 3 − b 4 4 − b 1 − b 2 − b 3 − b 4 4 b 1 − b 2 − b 3 + b 4 4 − b 1 + b 2 − b 3 + b 4 4
4 b 1 + b 2 − b 3 + b 4 4 − b 1 − b 2 − b 3 + b 4 4 b 1 − b 2 − b 3 − b 4 4 − b 1 + b 2 − b 3 − b 4 4
5 b 1 − b 2 + b 3 + b 4 4 − b 1 + b 2 + b 3 + b 4 4 b 1 + b 2 + b 3 − b 4 4 − b 1 − b 2 + b 3 − b 4 4
6 b 1 − b 2 + b 3 − b 4 4 − b 1 + b 2 + b 3 − b 4 4 b 1 + b 2 + b 3 + b 4 4 − b 1 − b 2 + b 3 + b 4 4
7 b 1 − b 2 − b 3 − b 4 4 − b 1 + b 2 − b 3 − b 4 4 b 1 + b 2 − b 3 + b 4 4 − b 1 − b 2 − b 3 + b 4 4
8 b 1 − b 2 − b 3 + b 4 4 − b 1 + b 2 − b 3 + b 4 4 b 1 + b 2 − b 3 − b 4 4 − b 1 − b 2 − b 3 − b 4 4
9 − b 1 − b 2 + b 3 + b 4 4 b 1 + b 2 + b 3 + b 4 4 − b 1 + b 2 + b 3 − b 4 4 b 1 − b 2 + b 3 − b 4 4
10 − b 1 − b 2 + b 3 − b 4 4 b 1 + b 2 + b 3 − b 4 4 − b 1 + b 2 + b 3 + b 4 4 b 1 − b 2 + b 3 + b 4 4
11 − b 1 − b 2 − b 3 − b 4 4 b 1 + b 2 − b 3 − b 4 4 − b 1 + b 2 − b 3 + b 4 4 b 1 − b 2 − b 3 + b 4 4
12 − b 1 − b 2 − b 3 + b 4 4 b 1 + b 2 − b 3 + b 4 4 − b 1 + b 2 − b 3 − b 4 4 b 1 − b 2 − b 3 − b 4 4
13 − b 1 + b 2 + b 3 + b 4 4 b 1 − b 2 + b 3 + b 4 4 − b 1 − b 2 + b 3 − b 4 4 b 1 + b 2 + b 3 − b 4 4
14 − b 1 + b 2 + b 3 − b 4 4 b 1 − b 2 + b 3 − b 4 4 − b 1 − b 2 + b 3 + b 4 4 b 1 + b 2 + b 3 + b 4 4
15 − b 1 + b 2 − b 3 − b 4 4 b 1 − b 2 − b 3 − b 4 4 − b 1 − b 2 − b 3 + b 4 4 b 1 + b 2 − b 3 + b 4 4
16 − b 1 + b 2 − b 3 + b 4 4 b 1 − b 2 − b 3 + b 4 4 − b 1 − b 2 − b 3 − b 4 4 b 1 + b 2 − b 3 − b 4 4
Express b i , i = 1 , 2 , 3 , 4 , in terms of x 5 , x 6 , x 7 , x 8 :
b 1 = 1 − x 5 − x 6 + x 7 − x 8 2 b 2 = 1 − x 5 − x 6 − x 7 + x 8 2 b 3 = 1 − x 5 + x 6 + x 7 + x 8 2 b 4 = 1 − x 5 + x 6 − x 7 − x 8 2
Consequently, if the following conditions hold,
x 5 + x 6 + x 7 + x 8 2 ≤ 1 x 5 + x 6 − x 7 − x 8 2 ≤ 1 x 5 − x 6 + x 7 − x 8 2 ≤ 1 x 5 − x 6 − x 7 + x 8 2 ≤ 1
then system (26) has 16 distinct solutions ( x 1 , x 2 , x 3 , x 4 , x 5 , x 6 , x 7 , x 8 ) . The variables x 5 , x 6 , x 7 , x 8 are arbitrary subject to (34), whereas x 1 , x 2 , x 3 , x 4 are listed in Table 3, with b i , i = 1 , 2 , 3 , 4 , defined by (33). Each solution is a four-dimensional manifold in the eight-dimensional space, and every point of the manifold is a unitary state of the three-qubit quantum system.
There are 16 additional manifolds consisting of unitary states. To obtain them, rewrite system (26) as
x 5 2 + x 6 2 + x 7 2 + x 8 2 = 1 − x 1 2 − x 2 2 − x 3 2 − x 4 2 x 5 x 6 + x 7 x 8 = − x 1 x 2 − x 3 x 4 x 5 x 7 + x 6 x 8 = − x 1 x 3 − x 2 x 4 x 5 x 8 + x 6 x 7 = − x 1 x 4 − x 2 x 3
which has the form (27). Solving it for x 5 , x 6 , x 7 , x 8 yields the 16 solutions listed in Table 3, with
b 1 = 1 − x 1 − x 2 + x 3 − x 4 2 b 2 = 1 − x 1 − x 2 − x 3 + x 4 2 b 3 = 1 − x 1 + x 2 + x 3 + x 4 2 b 4 = 1 − x 1 + x 2 − x 3 − x 4 2
while the parameters x 1 , x 2 , x 3 , x 4 satisfy
x 1 + x 2 + x 3 + x 4 2 ≤ 1 x 1 + x 2 − x 3 − x 4 2 ≤ 1 x 1 − x 2 + x 3 − x 4 2 ≤ 1 x 1 − x 2 − x 3 + x 4 2 ≤ 1
Thus, the unitary states with real components form thirty-two four-dimensional manifolds in the eight-dimensional space. We summarize this result in the following theorem.
Theorem 8.
In M ( 1111211 ) , unitary states with real components form 32 four-dimensional manifolds in R 8 . For the first 16, the variables x 5 , x 6 , x 7 , x 8 satisfy (34), whereas x 1 , x 2 , x 3 , x 4 are specified in Table 3 with the quantities b i , i = 1 , 2 , 3 , 4 , defined by (33). For the other 16, the variables x 1 , x 2 , x 3 , x 4 satisfy (37), whereas x 5 , x 6 , x 7 , x 8 are specified in Table 3 with b i defined by (36).
An analogous theorem can be established for every other eight-dimensional space with real components.
Example 1.
Choose ( 0 , 0 , 0 , 0 ) ∈ R 4 , which satisfies (34) and (37). In M ( 1111211 ) , this point generates 32 unitary states of the three-qubit quantum system. Sixteen are ± J k 1 , k = 1 , 2 , … , 8 ; the remaining sixteen have the form
± 1 2 ( 1 , − 1 , 1 , 1 , 0 , 0 , 0 , 0 ) , ± 1 2 ( 1 , − 1 , − 1 , − 1 , 0 , 0 , 0 , 0 ) , ± 1 2 ( 1 , 1 , 1 , − 1 , 0 , 0 , 0 , 0 ) , ± 1 2 ( 1 , 1 , − 1 , 1 , 0 , 0 , 0 , 0 ) , ± 1 2 ( 0 , 0 , 0 , 0 , 1 , − 1 , 1 , 1 ) , ± 1 2 ( 0 , 0 , 0 , 0 , 1 , − 1 , − 1 , − 1 ) , ± 1 2 ( 0 , 0 , 0 , 0 , 1 , 1 , 1 , − 1 ) , ± 1 2 ( 0 , 0 , 0 , 0 , 1 , 1 , − 1 , 1 ) .

8. Discussion

We constructed spaces of eight-dimensional numbers with complex components, extending the four-dimensional construction for two-qubit systems [1] and earlier studies of eight-dimensional real spaces [3,4]. The correspondence between an eight-dimensional number and its matrix is a homomorphic bijection. Consequently, multiplication of numbers simultaneously represents composition of the corresponding operators.
An explicit algebraic construction of a transition gate was obtained for the introduced class of states. Unlike the general existence statement for an arbitrary matrix in U(8), the operator here is constructed directly from eight-dimensional numbers and belongs to a specifically defined matrix family. Entire classes of three-qubit gates can therefore be generated within a unified formalism.
For real components, the set of unitary states in M(1111211) is represented as a union of thirty-two four-dimensional manifolds. This description reveals the geometry of the distinguished invariant set and the discrete sign transformations connecting its components. The same construction applies to the other basis spaces.
The symmetry aspects of the construction appear in several forms. In every fixed eight-dimensional space, associativity and commutativity of multiplication generate a commuting matrix family; the set of unitary states is invariant under the action of the corresponding gates; and sign and permutation relations among the basis elements provide a discrete structure for the family of 16,384 spaces.
Several limitations should be noted. Gates belonging to one fixed commutative space commute and therefore do not by themselves constitute a universal quantum gate set. The present work also does not address optimal decompositions of the constructed matrices into physically implementable one- and two-qubit operations. Comparison of the obtained classes with the full group U(8), their relationships to the Hadamard, CCNOT, and CSWAP gates, and estimates of circuit complexity remain topics for future research.
The authors gratefully acknowledge financial support from the Ministry of Science and Higher Education of the Republic of Kazakhstan (Grant No. AP23489825).

References

  1. Gabbassov, M.B.; Kuanov, T.Z.; Yermagambetov, T.K.; Tuleuov, B.I. Four-Dimensional Spaces of Complex Numbers and Unitary States of Two-Qubit Quantum Systems. Symmetry 2025, 17, 1789. [Google Scholar] [CrossRef]
  2. Abenov, M.M.; Gabbassov, M.B.; Kuanov, T.Z.; Tuleuov, B.I. Anisotropic Four-Dimensional Spaces of Real Numbers. Symmetry 2025, 17, 795. [Google Scholar] [CrossRef]
  3. Gabbassov, M.B.; Kuanov, T.Z.; Tuleuov, B.I.; Suleimenov, K.M.; Yermagambetov, T.K. Associative and Commutative Algebra of Eight-Dimensional Real Numbers. Conference Proceedings «Automorphisms and Identities of Algebras», 8 – 12 September, 2025; p. P. 10. [Google Scholar]
  4. Gabbassov, M.B.; Kuanov, T.Z.; Suleimenov, K.M.; Tuleuov, B.I. Anisotropic Spaces of Eight-Dimensional Numbers with Real Components. Preprints.org. [CrossRef]
  5. Abenov, M.M. Four-Dimensional Mathematics: Methods and Applications; Al-Farabi Kazakh National University Press: Almaty, Kazakhstan, 2019; 175 p. (In Russian) [Google Scholar]
  6. Sutor, Robert S. Dancing with Qubits. How quantum computing works and how it can change the world. Packt Publ. 2021, 492p. [Google Scholar]
  7. Hidary, J.D. Quantum Computing: An Applied Approach; DMK Press: Moscow, Russia, 2021; 370 p, (Russian edition). [Google Scholar]
  8. Qiskit.org. Qiskit: An Open-source Framework for Quantum Computing. Available online: https://qiskit.org/documentation/.
  9. Enríquez, M.; Delgado, F.; Życzkowski, K. Entanglement of Three-Qubit Random Pure States. Entropy 2018, 20, 745. [Google Scholar] [CrossRef] [PubMed]
  10. Pérez-Salinas, A.; García-Martín, D.; Bravo-Prieto, C.; Latorre, J.I. Measuring the Tangle of Three-Qubit States. Entropy 2020, 22, 436. [Google Scholar] [CrossRef] [PubMed]
  11. Kalaga, J.K.; Leoński, W.; Szczęśniak, R.; Peřina, J., Jr. Mixedness, Coherence and Entanglement in a Family of Three-Qubit States. Entropy 2022, 24, 324. [Google Scholar] [CrossRef] [PubMed]
  12. Benedito, A.; Sierra, G. Visualizing Three-Qubit Entanglement. Entropy 2025, 27, 800. [Google Scholar] [CrossRef] [PubMed]
  13. Wang, J.; Liu, X.; Xu, L.; Li, M.; Li, L.; Shen, S. Local Unitary Equivalence of Quantum States Based on the Tensor Decompositions of Unitary Matrices. Entropy 2023, 25, 1139. [Google Scholar] [CrossRef] [PubMed]
  14. Krol, A.M.; Sarkar, A.; Ashraf, I.; Al-Ars, Z.; Bertels, K. Efficient Decomposition of Unitary Matrices in Quantum Circuit Compilers. Appl. Sci. 2022, 12, 759. [Google Scholar] [CrossRef]
  15. Tang, S.; Yang, C.; Li, D.; Shao, X. Implementation of Quantum Algorithms via Fast Three-Rydberg-Atom CCZ Gates. Entropy 2022, 24, 1371. [Google Scholar] [CrossRef] [PubMed]
  16. Grigoryan, A.M.; Gomez, A.; Espinoza, I.; Agaian, S.S. Signal-Induced Heap Transform-Based QR-Decomposition and Quantum Circuit for Implementing 3-Qubit Operations. Information 2025, 16, 466. [Google Scholar] [CrossRef]
  17. Grigoryan, A.M. New Permutation-Free Quantum Circuits for Implementing 3- and 4-Qubit Unitary Operations. Information 2025, 16, 621. [Google Scholar] [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.