I am studying about Fourier series from book"Signals and Systems Laboratory with MATLAB"

I came across topic "Orthogonality of Complex Exponential Signals"

I am confused in case when m=k, will the two signals still orthogonal in this case ?(highlighted yellow)

enter image description here

  • 3
    $\begingroup$ When m=k, the two signals are the same signal. They cannot be orthogonal to themselves. $\endgroup$ – Juancho May 7 '19 at 18:16

Well, when $m=k$ the integral is: $$ \int_0^T e^{j(m-k)\Omega_0t} dt = \int_0^T e^{j \cdot 0 \cdot\Omega_0t} dt = \int_0^T dt = T $$

So as Juancho says in the comments, it's the same signal and so can't be orthogonal to itself.

| improve this answer | |

The text is a bit cumbersome, in that it states "things" before defining them. And the zero-signal (or vector) is considered orthogonal to every other vector. This is why the sentence "If $I=0$ for..." seems unnecessarily complicated to me. And the $m$ and $k$ indices introduced at the beginning come out of nowhere until we understand they are integer complex exponential indices. Moreover, introducing $x$ and $y$ at the very end is quite inconsistent with the previous $x_m$ and $x_k$ notations.

In a mood for conciseness and precision, let me rewrite the text (with missing hypotheses, like $\Omega_{0}\neq 0 $), hoping it will be more direct.

Let us consider in the following complex-valued continuous-time periodic signals with period $T>0$. The inner product of $x_1$ and $x_2$ is defined as:

$$I(x_1,x_2)=\int_{t_{0}}^{t_{0}+T} x_1(t) x_2^{*}(t) d t$$

where $x^{*}$ denotes the complex conjugate of $x$. Two signals are orthogonal if their inner product is zero. Suppose now that $(m,k)\in \mathbb{Z}^2$, $\Omega_{0}\neq 0 $ and define $x_{1}(t)=e^{j m \Omega_{0} t}$ and $x_{2}(t)=e^{j k \Omega_{0} t}$. Since: $$I(x_1,x_2)=\int_{0}^{T} e^{j m \Omega_{0} t_{e}-j k \Omega_{0} t} d t=\int_{0}^{T} e^{j(m-k) \Omega_{0} t} d t=\left\{\begin{array}{ll}{T,} & {k=m} \\ {0,} & {k \neq m}\end{array}\right.\,,$$ then the complex exponentials $e^{j m \Omega_{0} t}$ and $e^{j k \Omega_{0} t}$ are orthogonal if $m\neq k$.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.