Quantum-Mechanics

Quantum gates, state representations, fidelity metrics, noise channels, and error-correcting codes for qubit-based computation.

Namespace: CSharpNumerics.Physics.Quantum

using CSharpNumerics.Physics.Quantum;
using CSharpNumerics.Physics.Quantum.NoiseModels;
using CSharpNumerics.Physics.Quantum.ErrorCorrection;

---

⚛️ Quantum Gates

Namespace: CSharpNumerics.Physics.Quantum

Quantum gates are unitary operators acting on one or more qubits. Each gate is represented by a ComplexMatrix and knows how to apply itself to a ComplexVectorN state vector.

All gates extend the abstract QuantumGate base class:

Class	Qubits	Matrix	Description
HadamardGate	1	$\frac{1}{\sqrt{2}}\begin{pmatrix}1&1\\1&-1\end{pmatrix}$	Creates equal superposition
PauliXGate	1	$\begin{pmatrix}0&1\\1&0\end{pmatrix}$	Quantum NOT — flips |0⟩ ↔ |1⟩
PauliYGate	1	$\begin{pmatrix}0&-i\\i&0\end{pmatrix}$	Bit + phase flip, $Y^2 = I$
PauliZGate	1	$\begin{pmatrix}1&0\\0&-1\end{pmatrix}$	Phase flip — |1⟩ → −|1⟩
SGate	1	$\begin{pmatrix}1&0\\0&i\end{pmatrix}$	Phase gate (π/2), $S^2 = Z$
TGate	1	$\begin{pmatrix}1&0\\0&e^{i\pi/4}\end{pmatrix}$	π/8 gate, $T^2 = S$
PhaseGate(θ)	1	$\begin{pmatrix}1&0\\0&e^{i\theta}\end{pmatrix}$	General phase gate, $P(\pi) = Z$
RxGate(θ)	1	$\begin{pmatrix}\cos\frac{\theta}{2}&-i\sin\frac{\theta}{2}\\-i\sin\frac{\theta}{2}&\cos\frac{\theta}{2}\end{pmatrix}$	Rotation about X-axis
RyGate(θ)	1	$\begin{pmatrix}\cos\frac{\theta}{2}&-\sin\frac{\theta}{2}\\\sin\frac{\theta}{2}&\cos\frac{\theta}{2}\end{pmatrix}$	Rotation about Y-axis
RzGate(θ)	1	$\begin{pmatrix}e^{-i\theta/2}&0\\0&e^{i\theta/2}\end{pmatrix}$	Rotation about Z-axis
CNOTGate	2	4×4 permutation	Flips target qubit when control is |1⟩
CZGate	2	$\text{diag}(1,1,1,-1)$	Phase flip on |11⟩
CPhaseGate(θ)	2	$\text{diag}(1,1,1,e^{i\theta})$	Controlled phase, $CP(\pi) = CZ$
SWAPGate	2	4×4 permutation	Swaps two qubit states
ToffoliGate	3	8×8 permutation	CCNOT — flips target when both controls are |1⟩
FredkinGate	3	8×8 permutation	CSWAP — swaps targets when control is |1⟩
PhaseOracle(n, states)	n	2ⁿ×2ⁿ diagonal	Flips phase of marked basis states: |w⟩ → −|w⟩
ControlledGate(U)	n+1	2ⁿ⁺¹×2ⁿ⁺¹ block	Applies U when control qubit is |1⟩
ModularMultiplyGate(a,N,n)	n	2ⁿ×2ⁿ permutation	|y⟩ → |ay mod N⟩, used in Shor's algorithm

QuantumGate (abstract)

public abstract class QuantumGate
{
    public abstract int QubitCount { get; }
    public abstract ComplexMatrix GetMatrix();
    public ComplexVectorN Apply(ComplexVectorN amplitudes, int[] qubitIndices, int totalQubits);
}

Apply uses the gate's unitary matrix to transform the relevant amplitudes in-place (tensor-product expansion) — works for arbitrary qubit counts and target indices.

Usage

Gates are consumed by QuantumInstruction and QuantumCircuit in the Engines Quantum section:

using CSharpNumerics.Physics.Quantum;
using CSharpNumerics.Engines.Quantum;

var circuit = new QuantumCircuit(2);
circuit.AddInstruction(new QuantumInstruction(new HadamardGate(), new List<int> { 0 }));
circuit.AddInstruction(new QuantumInstruction(new CNOTGate(), new List<int> { 0, 1 }));

var state = new QuantumSimulator().Run(circuit);   // Bell state (|00⟩ + |11⟩)/√2

---

🌐 BlochVector

Represents a single-qubit pure state on the Bloch sphere. Given $|\psi\rangle = \alpha|0\rangle + \beta|1\rangle$:

$x = 2\,\text{Re}(\alpha^*\beta), \quad y = 2\,\text{Im}(\alpha^*\beta), \quad z = |\alpha|^2 - |\beta|^2$

using CSharpNumerics.Physics.Quantum;
using CSharpNumerics.Engines.Quantum;

// From a circuit result
var circuit = new QuantumCircuit(1);
circuit.AddInstruction(new QuantumInstruction(new HadamardGate(), new List<int> { 0 }));
var state = new QuantumSimulator().Run(circuit);

BlochVector bloch = state.GetBlochVector();  // (1, 0, 0) — +X axis
double theta = bloch.Theta;                  // polar angle θ
double phi   = bloch.Phi;                    // azimuthal angle φ
double r     = bloch.Radius;                 // 1.0 for pure states
Vector v     = bloch.ToVector();             // 3D Vector for rendering

// Directly from amplitudes
var b = BlochVector.FromAmplitudes(
    new ComplexNumber(1, 0), new ComplexNumber(0, 0));  // |0⟩ → (0, 0, 1)

Property	Description
X, Y, Z	Cartesian Bloch coordinates
Theta	Polar angle $\theta \in [0, \pi]$ from +Z
Phi	Azimuthal angle $\varphi \in (-\pi, \pi]$
Radius	Vector length (1 for pure states)
ToVector()	Returns a 3D Vector for visualization

Canonical states on the sphere:

State	Bloch
$\|0\rangle$	$(0, 0, 1)$ — north pole
$\|1\rangle$	$(0, 0, -1)$ — south pole
$(\|0\rangle+\|1\rangle)/\sqrt{2}$	$(1, 0, 0)$ — +X
$(\|0\rangle-\|1\rangle)/\sqrt{2}$	$(-1, 0, 0)$ — −X
$(\|0\rangle+i\|1\rangle)/\sqrt{2}$	$(0, 1, 0)$ — +Y
$(\|0\rangle-i\|1\rangle)/\sqrt{2}$	$(0, -1, 0)$ — −Y

---

📏 QuantumFidelity

Static methods for computing the overlap between quantum states. Fidelity ranges from 0 (orthogonal) to 1 (identical).

State fidelity — $F = |\langle\psi|\phi\rangle|^2$

using CSharpNumerics.Physics.Quantum;
using CSharpNumerics.Engines.Quantum;

var sim = new QuantumSimulator();
var s0    = sim.Run(QuantumCircuitBuilder.New(1).Build());        // |0⟩
var sPlus = sim.Run(QuantumCircuitBuilder.New(1).H(0).Build());  // |+⟩
var s1    = sim.Run(QuantumCircuitBuilder.New(1).X(0).Build());  // |1⟩

double f1 = QuantumFidelity.Fidelity(s0, s0);     // 1.0  — identical
double f2 = QuantumFidelity.Fidelity(s0, sPlus);  // 0.5  — overlap
double f3 = QuantumFidelity.Fidelity(s0, s1);     // 0.0  — orthogonal

Vector fidelity — works directly on ComplexVectorN

double f = QuantumFidelity.Fidelity(psi, phi);   // |⟨ψ|φ⟩|²

Bloch fidelity — geometric formula for single-qubit states: $F = \frac{1}{2}(1 + \hat{n}_1 \cdot \hat{n}_2)$

var b1 = s1State.GetBlochVector();
var b2 = s2State.GetBlochVector();
double f = QuantumFidelity.BlochFidelity(b1, b2);
// Matches state fidelity for pure single-qubit states

---

🔊 Noise Models

The Physics.Quantum.NoiseModels namespace provides quantum noise channels using the Kraus operator formalism. Each channel implements INoiseChannel and satisfies the completeness relation $\sum E_k^\dagger E_k = I$.

Channel	Parameter	Kraus Operators	Physical Model
DepolarizingNoise(p)	$p \in [0,1]$	$E_0 = \sqrt{1 - 3p/4}\,I$, $E_{1,2,3} = \sqrt{p/4}\,\{X, Y, Z\}$	Random Pauli error — qubit replaced by maximally mixed state with probability $p$
DephasingNoise(p)	$p \in [0,1]$	$E_0 = \sqrt{1-p}\,I$, $E_1 = \sqrt{p}\,Z$	Phase-flip error — $T_2$ decoherence
AmplitudeDampingNoise(γ)	$\gamma \in [0,1]$	$E_0 = [[1,0],[0,\sqrt{1-\gamma}]]$, $E_1 = [[0,\sqrt{\gamma}],[0,0]]$	Energy dissipation — $\|1\rangle \to \|0\rangle$ decay ($T_1$)

Usage with NoisyQuantumSimulator (from CSharpNumerics.Engines.Quantum):

using CSharpNumerics.Physics.Quantum.NoiseModels;
using CSharpNumerics.Engines.Quantum;

var circuit = QuantumCircuitBuilder.New(2).H(0).CNOT(0, 1).Build();

// Ideal simulation
var ideal = new QuantumSimulator().Run(circuit);

// Noisy simulation — channels stacked, applied after each gate
var noisy = new NoisyQuantumSimulator(new Random(42))
    .WithNoise(new DepolarizingNoise(0.01))
    .WithNoise(new AmplitudeDampingNoise(0.005))
    .Run(circuit);

double fidelity = QuantumFidelity.Fidelity(ideal, noisy);

INoiseChannel interface

public interface INoiseChannel
{
    int QubitCount { get; }
    ComplexMatrix[] GetKrausOperators();
}

---

🛡️ Quantum Error Correction — Code Definitions

Namespace: CSharpNumerics.Physics.Quantum.ErrorCorrection

The ErrorCorrection sub-namespace defines quantum error-correcting codes as mathematical objects — stabilizers, correction maps, and logical operators. The simulation/orchestration layer lives in Engines.Quantum.ErrorCorrection.

IQuantumErrorCorrectionCode

Interface for any [[n, k, d]] stabilizer code:

Property / Method	Description
PhysicalQubits	Number of physical qubits (n)
LogicalQubits	Number of logical qubits (k)
Distance	Code distance (d)
SyndromeQubits	Number of ancilla qubits for syndrome extraction
GetStabilizers()	Returns stabilizer generators as (qubit, Pauli) lists
GetCorrectionMap()	Syndrome integer → corrective (qubit, Pauli) operations
GetLogicalX(i)	Logical X̄ operator for logical qubit i
GetLogicalZ(i)	Logical Z̄ operator for logical qubit i

BitFlipCode3 — [[3,1,1]]

Encodes $|\psi\rangle = \alpha|0\rangle + \beta|1\rangle$ into $\alpha|000\rangle + \beta|111\rangle$. Corrects any single-qubit X (bit-flip) error.

Stabilizers: $Z_0 Z_1$, $Z_1 Z_2$

Syndrome	Error	Correction
00	None	—
01	$X_0$	Apply X to qubit 0
10	$X_2$	Apply X to qubit 2
11	$X_1$	Apply X to qubit 1

Logical operators: $\bar{X} = X_0 X_1 X_2$, $\bar{Z} = Z_0$

PhaseFlipCode3 — [[3,1,1]]

Encodes $|\psi\rangle$ into $\alpha|{+}{+}{+}\rangle + \beta|{-}{-}{-}\rangle$. Corrects any single-qubit Z (phase-flip) error.

Stabilizers: $X_0 X_1$, $X_1 X_2$

Syndrome	Error	Correction
00	None	—
01	$Z_0$	Apply Z to qubit 0
10	$Z_2$	Apply Z to qubit 2
11	$Z_1$	Apply Z to qubit 1

Logical operators: $\bar{X} = X_0$, $\bar{Z} = Z_0 Z_1 Z_2$

using CSharpNumerics.Physics.Quantum.ErrorCorrection;

var code = new BitFlipCode3();
var stabs = code.GetStabilizers();       // [{(0,'Z'),(1,'Z')}, {(1,'Z'),(2,'Z')}]
var map   = code.GetCorrectionMap();     // 0→[], 1→[(0,'X')], 2→[(2,'X')], 3→[(1,'X')]
var logX  = code.GetLogicalX();          // [(0,'X'),(1,'X'),(2,'X')]

ShorCode9 — [[9,1,3]]

Shor's 9-qubit code — the first code to correct any single-qubit error (X, Z, or Y). Uses three blocks of three qubits: inner bit-flip repetition within blocks, outer phase-flip repetition across blocks.

$|0\rangle_L = \frac{(|000\rangle+|111\rangle)(|000\rangle+|111\rangle)(|000\rangle+|111\rangle)}{2\sqrt{2}}$ $|1\rangle_L = \frac{(|000\rangle-|111\rangle)(|000\rangle-|111\rangle)(|000\rangle-|111\rangle)}{2\sqrt{2}}$

8 stabilizer generators:

Generator	Operator	Type
$g_1 \ldots g_6$	$Z_i Z_{i+1}$ within each block	Bit-flip detection
$g_7$	$X_0 X_1 X_2 X_3 X_4 X_5$	Phase-flip detection (blocks 0–1)
$g_8$	$X_3 X_4 X_5 X_6 X_7 X_8$	Phase-flip detection (blocks 1–2)

Correction: The 8-bit syndrome (256 values) maps to at most one X correction + one Z correction. Each block's 2-bit sub-syndrome identifies bit-flip errors exactly as in BitFlipCode3. The 2-bit phase-flip sub-syndrome identifies which block suffered a Z error.

Logical operators: $\bar{X} = X_0 X_1 \ldots X_8$, $\bar{Z} = Z_0 Z_3 Z_6$

var shor = new ShorCode9();
var stabs = shor.GetStabilizers();  // 8 generators
var map   = shor.GetCorrectionMap(); // 256 syndrome entries

SteaneCode7 — [[7,1,3]]

The Steane code — the smallest CSS (Calderbank-Shor-Steane) code, correcting any single-qubit error using only 7 physical qubits. Built from the classical [7,4,3] Hamming code and its dual [7,3,4] code.

$|0\rangle_L = \frac{1}{\sqrt{8}} \sum_{x \in C_2} |x\rangle \qquad |1\rangle_L = \frac{1}{\sqrt{8}} \sum_{x \in C_2 + v} |x\rangle$

where $C_2$ is the [7,3,4] dual Hamming code and $v = (1110000)$.

6 stabilizer generators:

Generator	Operator	Type
$g_1$	$Z_3 Z_4 Z_5 Z_6$	X-error detection
$g_2$	$Z_1 Z_2 Z_5 Z_6$	X-error detection
$g_3$	$Z_0 Z_2 Z_4 Z_6$	X-error detection
$g_4$	$X_3 X_4 X_5 X_6$	Z-error detection
$g_5$	$X_1 X_2 X_5 X_6$	Z-error detection
$g_6$	$X_0 X_2 X_4 X_6$	Z-error detection

Correction: The 6-bit syndrome has two independent 3-bit halves — Z-stabilizer syndrome (bits 0–2) identifies X errors, X-stabilizer syndrome (bits 3–5) identifies Z errors, using the Hamming decoding map: syndrome $s \to$ qubit $j$ where column $j$ of the parity-check matrix equals the binary expansion of $s$.

Logical operators: $\bar{X} = X_0 X_1 \ldots X_6$, $\bar{Z} = Z_0 Z_1 \ldots Z_6$

var steane = new SteaneCode7();
var stabs  = steane.GetStabilizers();   // 6 generators (3 Z-type + 3 X-type)
var map    = steane.GetCorrectionMap();  // 64 syndrome entries