Signals and Systems – Symmetric Impulse Response of Linear-Phase System

Distortion-less Transmission

When a signal is transmitted through a system and there is a change in the shape of the signal, it called the distortion. If the output of the system is an exact replica of the input signal, then the transmission of the signal through the system is called distortionless transmission.

Linear Phase System

For distortionless transmission through a system, there should not be any phase distortion, i.e., the phase of the system should be linear. For the linear phase system, the impulse response of the system is symmetrical about the delay time $\mathit{(t_{d})}$.

Proof

For a linear phase system, we have,

$$\mathrm{ \mathit{H\left ( \omega \right )\mathrm{=}\left |H\left ( \omega \right ) \right |e^{-j\omega t_{d}}}}$$

The impulse response of such a system can be obtained by finding the inverse Fourier transform, i.e.,

$$\mathrm{ \mathit{h\left ( t \right )\mathrm{=}F^{-\mathrm{1}}\left [H\left ( \omega \right ) \right ]\mathrm{=}F\left [ \left |H\left ( \omega \right ) \right |e^{-j\omega t_{d}} \right ]}}$$

Therefore, from the definition of inverse Fourier transform, we have,

$$\mathrm{ \mathit{x\left ( t \right )\mathrm{=}F^{-\mathrm{1}}\left [x\left ( \omega \right ) \right ]\mathrm{=}\frac{\mathrm{1}}{\mathrm{2}\pi }\int_{-\infty }^{\infty }X\left ( \omega \right )e^{j\omega t}d\omega }}$$

$$\mathrm{\Rightarrow \mathit{h\left ( t \right )\mathrm{=}F^{-\mathrm{1}}\left [H\left ( \omega \right ) \right ]\mathrm{=}\frac{\mathrm{1}}{\mathrm{2}\pi }\int_{-\infty }^{\infty }\left [ \left |H\left ( \omega \right ) \right |e^{-j\omega t_{d}} \right ]e^{j\omega t}d\omega }}$$

$$\mathrm{\Rightarrow \mathit{h\left ( t \right )\mathrm{=}\frac{\mathrm{1}}{\mathrm{2}\pi }\int_{-\infty }^{\infty }\left |H\left ( \omega \right ) \right |e^{j\omega \left ( t-t_{d} \right )}d\omega }}$$

$$\mathrm{\Rightarrow \mathit{h\left ( t \right )\mathrm{=}\frac{\mathrm{1}}{\mathrm{2}\pi }\left [ \int_{-\infty }^{\mathrm{0} }\left |H\left ( \omega \right ) \right |e^{j\omega \left ( t-t_{d} \right )}d\omega \mathrm{+} \int_{\mathrm{0} }^{\infty }\left |H\left ( \omega \right ) \right |e^{j\omega \left ( t-t_{d} \right )}d\omega \right ]}} $$

$$\mathrm{\Rightarrow \mathit{h\left ( t \right )\mathrm{=}\frac{\mathrm{1}}{\mathrm{2}\pi }\left [ \int_{\mathrm{0} }^{\infty }\left |H\left ( \omega \right ) \right |e^{-j\omega \left ( t-t_{d} \right )}d\omega \mathrm{+} \int_{\mathrm{0} }^{\infty }\left |H\left ( \omega \right ) \right |e^{j\omega \left ( t-t_{d} \right )}d\omega \right ]}} $$

$$\mathrm{\Rightarrow \mathit{h\left ( t \right )\mathrm{=}\frac{\mathrm{1}}{\mathrm{2}\pi }\int_{\mathrm{0} }^{\infty }\left |H\left ( \omega \right ) \right |\left [ e^{j\omega \left ( t-t_{d} \right )} \mathrm{+} e^{-j\omega \left ( t-t_{d} \right )} \right ]d\omega}}$$

$$\mathrm{\mathit{\because \left ( \frac{e^{j\omega \left ( t-t_{d} \right )}\: \mathrm{\mathrm{+}} \: e^{-j\omega \left ( t-t_{d} \right )}}{\mathrm{2}}\right )\mathrm{=}\cos \omega \left ( t-t_{d} \right )}} $$

$$\mathrm{\mathit{\therefore h\left ( t \right )\mathrm{=}\frac{\mathrm{1}}{\pi }\int_{\mathrm{0}}^{\infty }\left | H\left ( \omega \right ) \right |\cos \omega \left ( t-t_{d} \right )d\omega}} $$

$$\mathrm{\mathit{\Rightarrow h\left ( t_{d}\:\mathrm{+} \:t \right )\mathrm{=}\frac{\mathrm{1}}{\pi }\int_{\mathrm{0}}^{\infty }\left | H\left ( \omega \right ) \right |\cos \omega t\; d\omega}}$$

And,

$$\mathrm{\mathit{h\left ( t_{d}-t \right )\mathrm{=}\frac{\mathrm{1}}{\pi }\int_{\mathrm{0}}^{\infty }\left | H\left ( \omega \right ) \right |\cos \omega t\; d\omega}}$$

Therefore, we have,

$$\mathrm{\mathit{h\left ( t_{d}\mathrm{+}t \right )\mathrm{=}h\left ( t_{d}-t \right )}} $$

Hence, this proves that for a linear phase system, the impulse response ℎ(𝑡) of the system is symmetrical about the delay time (𝑡_𝑑) and it is non-causal, i.e., non-zero for 𝑡 < 0.

The value of impulse response ℎ(𝑡) of the linear-phase filter or system is maximum at delay time, i.e., at 𝑡 = 𝑡_𝑑 and is given by,

$$\mathrm{\mathit{h\left (t \right )|_{max}\mathrm{=}h\left ( t_{d} \right )\mathrm{=}\frac{\mathrm{1}}{\pi }\int_{\mathrm{0}}^{\infty }\left | H\left ( \omega \right ) \right |d\omega }}$$

Manish Kumar Saini

Updated on 17-Dec-2021 07:31:04

• Related Questions & Answers
• Signals and Systems – Response of Linear Time Invariant (LTI) System
• Signals and Systems: Linear and Non-Linear Systems
• Signals and Systems – What is a Linear System?
• Signals and Systems: Linear Time-Invariant Systems
• Signals and Systems – Filter Characteristics of Linear Systems
• Signals and Systems – Transfer Function of Linear Time Invariant (LTI) System
• Signals and Systems – Properties of Linear Time-Invariant (LTI) Systems
• Signals and Systems: Stable and Unstable System
• Signals and Systems: Static and Dynamic System
• Signals and Systems: Multiplication of Signals
• Signals and Systems: Classification of Systems
• Signals and Systems: Amplitude Scaling of Signals
• Signals and Systems – Classification of Signals
• Signals and Systems – Causal and Non-Causal System
• Signals and Systems: Addition and Subtraction of Signals