# What is the matrix representation for $n$-qubit gates?

Let's say I have more than one qbits $$|0\rangle|1\rangle$$ and I want to perform a $$H$$ on both of them. I know the matrix representation for the Hadamard on a single qbit is

$$\frac{1}{\sqrt{2}}\begin{bmatrix}1 & 1\\1 & -1\end{bmatrix}$$

If I represent the qbits with the vector $$\begin{bmatrix} 0 \\ 1 \\ 0 \\ 0\end{bmatrix}$$ I think that the representation for a two qbit Hadamard is the tensor $$H\otimes H$$ giving

$$\frac{1}{2}\begin{bmatrix}1 & 1 & 1 & 1\\1 & -1 & 1 & -1 \\ 1 & 1 & -1 & -1 \\ 1 & -1 & -1 & 1 \end{bmatrix}$$

and so

$$\frac{1}{2}\begin{bmatrix}1 & 1 & 1 & 1\\1 & -1 & 1 & -1 \\ 1 & 1 & -1 & -1 \\ 1 & -1 & -1 & 1 \end{bmatrix} \begin{bmatrix} 0 \\ 1 \\ 0 \\ 0\end{bmatrix} = \frac{1}{2} \begin{bmatrix} 1 \\ -1 \\ 1 \\ -1 \end{bmatrix}$$

which feels correct as

\begin{align}\frac{1}{\sqrt{2}}\begin{bmatrix}1 & 1\\1 & -1\end{bmatrix} \begin{bmatrix} 1 \\ 0 \end{bmatrix} \otimes \frac{1}{\sqrt{2}}\begin{bmatrix}1 & 1\\1 & -1\end{bmatrix} \begin{bmatrix} 0 \\ 1 \end{bmatrix} &= \\ \frac{1}{\sqrt{2}} \begin{bmatrix} 1 \\ 1 \end{bmatrix} \otimes \frac{1}{\sqrt{2}} \begin{bmatrix} 1 \\ -1 \end{bmatrix} &= \\ \frac{1}{2} \begin{bmatrix} 1 \\ -1 \\ 1 \\ -1 \end{bmatrix} \end{align}

But is this actually correct? And how does it (or is it possible to?) generalize to different gate compositions like $$H \otimes CNOT$$? Does it scale to $$n$$ qbits?

• Just to note $$H \otimes CNOT = \frac{1}{\sqrt{2}}\begin{bmatrix} CNOT & CNOT \\ CNOT & -CNOT\end{bmatrix}$$ And of course, this works generally. Commented Nov 14, 2019 at 14:54
• @MartinVesely yes, that makes sense, and that fits my observations informally. I guess my question was about the formal generalization -- it appears, as given in the answers below, that the Kronecker product gives the proper generalization. thanks for commenting -- this relation is handy on its own
– 1ijk
Commented Nov 18, 2019 at 0:51

In general, given two matrices $$A$$ and $$B$$ of dimensions $$n_1\times n_2$$ and $$m_1\times m_2$$, respectively, their tensor product $$A\otimes B$$ can be represented using the Kronecker product as $$(A\otimes B)_{n_1 m_1,n_2m_2}=A_{n_1,n_2}B_{m_1, m_2}.$$ The indices on the left hand side are a standard way to enumerate the integers from $$1$$ to $$n_1 m_1$$ and from $$1$$ to $$n_2 m_2$$. This is what you already observed in the case of $$H\otimes H$$, where two $$2\times 2$$ matrices become one $$4\times 4$$ matrix, whose elements are the product of elements of the two copies of $$H$$.
Note that a column-vector can be considered as a matrix with the size $$n \times 1$$, so the Kronecker product rule also applies.