Ozye˘ginUniversity¨ Spring 2020 CS480/CS580 Quantum Computing
Lecture 4 - Quantum Circuit Model, Entanglement, Superdense Coding 02.03.2020
Scribe: Eren Balatkan Lecturer: Ozlem¨ Salehi K¨oken
4 Quantum Circuit Model
In this section, we will be talking about the quantum circuit model.
4.1 1-Qubit Gates
Not (X) : Corresponds to a 180 rotation around X - Axis
0 1 Z |0i X = 1 0
X |0i = |1i Y |0i
X |0i X |1i = |0i
X(α |0i + β |1i) = β |0i + α |1i |1i
Y-Gate (Y) : Corresponds to a 180 rotation around Y - Axis
0 −i Z |0i Y = i 0
Y |0i = i |1i
Y |0i
X |0i Y |1i = −i |0i
Y (α |0i + β |1i) = −βi |0i + α |1i |1i
1 Z-Gate (Z) : Corresponds to a 180 rotation around Z - Axis
1 0 Z |0i Z = 0 −1
Z |0i = |0i Y |0i
X |0i Z |1i = − |1i
Z(α |0i + β |1i) = α |0i + −β |1i |1i
Rotation Gates
Let |Ψi be a quantum state and let µ be a unitary operator. The action of µ on |Ψi can be thought as a rotation on the Bloch sphere
−iθX/2 Rx(θ) = e
−iθY/2 Ry(θ) = e
−iθZ/2 Rz(θ) = e
If A2 = I, eiAx = cos(x)I + isin(x)A
θ θ cos −isin θ θ 2 2 Rx(θ) = cos I + isin X = 2 2 θ θ −isin cos 2 2 θ θ cos −sin θ θ 2 2 Ry(θ) = cos I + isin Y = 2 2 θ θ sin cos 2 2
θ θ R (θ) = e−iθZ/2 = cos I − Z z 2 2 θ θ cos 0 isin 0 2 2 = − θ θ 0 cos 0 −isin 2 2 θ θ cos − isin 0 2 2 = θ θ 0 cos + isin 2 2 e−iθ/2 0 = 0 eiθ/2
2 Example:
σ σ Let |Ψ| = cos |0i + eiϕsin |1i 2 2 σ −iθ/2 cos e 0 2 Rz(θ) = iθ/2 σ 0 e eiϕsin 2 σ e−iθ/2cos = 2 σ eiθ/2eiϕsin 2 σ σ = e−iθ/2cos |0i + eiθ/2eiϕsin |1i 2 2 σ σ = e−iθ/2(cos |0i + eiθeiϕsin |1i) 2 2 σ σ = e−iθ/2(cos |0i + ei(θ+ϕ)sin |1i) 2 2
Effect of Rz(θ) is to change the angle ϕ to ϕ + θ which is a rotation around z-axis
iα Theorem 1. Suppose µ is a 1-Qubit unitary gate. Then there exists α, β, γ and δ such that µ = e Rz(β)Ry(γ)Rz(δ)
Theorem 2. Any unitary µ can be expressed as µ = eiαAXBXC where ABC = I, A,B,C are unitary
S-Gate T-Gate
1 0 1 0 S = T = 0 i 0 eπi/4
S = Rz(π/2) T = Rz(π/4) S2 = Z T 2 = S
Hadamard Gate (H)
! 1 1 √1 √1 H = √1 = 2 2 2 1 −1 √1 − √1 2 2 ! ! √1 √1 1 √1 |+i = H |0i = 2 2 = 2 = √1 |0i + √1 |1i √1 − √1 0 √1 2 2 2 2 2 ! ! √1 √1 0 √1 |−i = H |1i = 2 2 = 2 = √1 |0i − √1 |1i √1 − √1 1 − √1 2 2 2 2 2
3 1 1 H |+i = H √ |0i + √ |1i 2 2 1 1 1 1 1 1 = √ √ |0i + √ |1i + √ √ |0i − √ |1i 2 2 2 2 2 2 1 1 1 1 = |0i + |1i + |0i − |1i 2 2 2 2 = |0i
H2 = I
(H ⊗ H)(|00i) = H⊗2 |00i 1 1 = √ (|0i + |1i) ⊗ √ (|0i + |1i) 2 2 1 1 1 1 = |00i + |01i + |10i + |11i 2 2 2 2
1 n H⊗n |0...0i = √ P2 −1 |xi where x is written in binary 2n x=0 1 1 1 1 H⊗3 |000i = √ P7 |xi = √ |000i + √ |001i + ... + √ |111i 8 x=0 8 8 8
Hadamard corresponds to a rotation of π around √1 , 0, √1 . When restricted to real amplitudes, Hadamard 2 2 is a reflection matrix in 2D plane over angle π/8
|0i
|+i
H
π/8 |1i
4 4.2 Multi-Qubit Gates
Controlled Not (CNOT)
1 0 0 0 CNOT |10i = |11i 0 1 0 0 CNOT = 0 0 0 1 CNOT |11i = |10i 0 0 1 0 CNOT |01i = |01i
CNOT |00i = |00i
Example:
Let |Ψi = √1 |00i + √1 |10i = √1 |0i + √1 |1i ⊗ |0i 2 2 2 2
CNOT (|Ψi = √1 |00i + √1 |11i) 2 2
Controlled Z-Gate (CZ)
1 0 0 0 CZ |00i = |00i 0 1 0 0 CZ = 0 0 1 0 CZ |01i = |01i 0 0 0 −1 CZ |10i = |10i
CZ |11i = − |11i
Toffoli Gate (CCX)
1 0 0 0 0 0 0 0 CCX |000i = |000i 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 CCX |101i = |101i 0 0 0 1 0 0 0 0 CCX = 0 0 0 0 1 0 0 0 CCX |110i = |111i 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 CCX |111i = |110i
5 Example:
C4NOT : Apply NOT to target if 4 control qubits are in state |1i ?
Input 1 Input 2 Input 3 Input 4
Ancilla 1
Ancilla 2
Target
Example:
Apply NOT to target if control qubit is on state |0i ?
Input 1 X X
Target
6 5 Basic Quantum Protocols
5.1 Entanglement
Suppose that we have a quantum system in |Ψi = |00i. Let’s apply the following operations.
Qubit 1 H
Qubit 2
1 1 (H ⊗ I)(|00i) = √ |0i + √ |1i ⊗ |0i 2 2 1 1 = √ |00i + √ |10i 2 2
CNOT √1 |00i + √1 |10i) = √1 |00i + √1 |11i 2 2 2 2
If we measure first qubit and observe |0i, then the second qubit collapses to |0i. Similarly, if we measure the second qubit and observe |1i, then the second qubit collapses to |1i. This is true even if the qubits are seperated from each other. This is called entanglement.
Bell States
•| Φ+i = √1 |00i + √1 |11i 2 2
•| Φ−i = √1 |00i − √1 |11i 2 2
•| Ψ+i = √1 |01i + √1 |10i 2 2
•| Ψ−i = √1 |01i − √1 |10i 2 2 The states above are known as the Bell states and they are entangled states. Entangled states can not be written as a tensor product of two subsystems.
7 5.2 Superdense Coding
Suppose that Alice wants to send Bob two classical bits of information. Alice will send only a single qubit to Bob to achieve this task. • Initially, Alice and Bob have an entangled pair of qubits in state √1 |00i + √1 |11i. 2 2 • Alice has the first qubit and Bob has the second qubit. Alice wants to send two bits of classical information, ab, to Bob. • If a=1, Alice applies Z-Gate to her qubit. • If b=1, Alice applies X and sends to Bob.
ab Operation Result ———– ———– ———– 00 - √1 |00i + √1 |11i 2 2 01 X √1 |10i + √1 |01i 2 2 10 Z √1 |00i − √1 |11i 2 2 11 XZ √1 |10i − √1 |01i 2 2
• Now Bob has both qubits. • Bob applies CNOT where first qubit is control and second qubit is target. • Bob applies Hadamard to first qubit • He makes the measuerment and observes quantum state |abi
1’st Case - |00i
1 1 |Ψi = CNOT √ |00i + √ |11i 2 2 1 1 = √ |00i + √ |10i 2 2
1 1 1 1 1 1 (H ⊗ I) |Ψi = √ √ |0i + √ |1i |0i + √ √ |0i − √ |1i |0i 2 2 2 2 2 2 1 1 1 1 = √ |00i + √ |10i + √ |00i − √ |10i = |00i 2 2 2 2 2’nd Case - |01i
1 1 |Ψi = CNOT √ |10i + √ |01i 2 2 1 1 = √ |11i + √ |01i 2 2
1 1 1 1 1 1 (H ⊗ I) |Ψi = √ √ |0i − √ |1i |1i + √ √ |0i + √ |1i |1i 2 2 2 2 2 2 1 1 1 1 = √ |01i − √ |11i + √ |01i − √ |11i = |01i 2 2 2 2
8