Deriving the General Response of a System from Its Impulse Response | 从冲激响应推导系统的一般响应

📚 Deriving the General Response of a System from Its Impulse Response | 从冲激响应推导系统的一般响应

In signal processing and control theory, the impulse response completely characterizes a linear time-invariant (LTI) system. Once the impulse response is known, the response to any arbitrary input can be obtained by convolution. This article explains the mathematical derivation of that general response, step by step, with examples and connections to the IB Mathematics curriculum.

在信号处理与控制理论中，冲激响应完整地刻画了一个线性时不变（LTI）系统。一旦冲激响应已知，任意输入信号所对应的系统响应便可通过卷积获得。本文将逐步解释这一一般响应的数学推导过程，并给出示例及其与IB数学课程的联系。

1. What Is an Impulse Response? | 什么是冲激响应？

An impulse response, usually denoted h(t), is the output of a system when the input is a Dirac delta function δ(t). This delta function represents an idealized instantaneous pulse with unit area. The impulse response captures all the dynamic properties of the system, including stability, natural frequencies, and transient behavior.

冲激响应通常记作 h(t)，是指输入为狄拉克δ函数 δ(t) 时系统的输出。δ函数是一种理想化的瞬时脉冲，其面积为1。冲激响应包含了系统的全部动态特性，包括稳定性、固有频率和瞬态行为。

Mathematically, we write:

δ(t) → system → h(t)

For a discrete-time system, the impulse response is the output to a unit sample sequence δ[n], and is denoted h[n].

对于离散时间系统，冲激响应是对单位采样序列 δ[n] 的输出，记为 h[n]。

2. The Two Key Properties: Linearity and Time Invariance | 两个关键性质：线性与时不变性

To derive the general response from the impulse response, we rely on two assumptions about the system. First, the system must be linear: if input x₁(t) produces output y₁(t) and input x₂(t) produces output y₂(t), then the input ax₁(t) + bx₂(t) produces the output ay₁(t) + by₂(t).

要从冲激响应推导一般响应，我们依赖于关于系统的两个假设。第一，系统必须是线性的：若输入 x₁(t) 产生输出 y₁(t)，输入 x₂(t) 产生输出 y₂(t)，则输入 ax₁(t) + bx₂(t) 产生输出 ay₁(t) + by₂(t)。

Second, the system must be time-invariant: if the input is delayed by τ, then the output is delayed by the same amount. In symbols, if x(t) → y(t), then x(t − τ) → y(t − τ).

第二，系统必须是时不变的：若输入延迟 τ，则输出也延迟相同的时间。用符号表示，若 x(t) → y(t)，则 x(t − τ) → y(t − τ)。

Together, these two properties define an LTI system. For an LTI system, the impulse response is a complete fingerprint.

这两个性质合在一起定义了LTI系统。对于LTI系统，冲激响应是系统的完整指纹。

3. Representing an Arbitrary Input as a Sum of Impulses | 将任意输入表示为冲激的叠加

The central idea is to decompose any continuous input signal into an infinite sum of shifted and scaled delta functions. A single impulse at time τ with strength x(τ) dτ can be written as x(τ)δ(t − τ) dτ.

核心思想是将任意连续输入信号分解为无穷多个平移且缩放后的δ函数之和。在时刻 τ 处强度为 x(τ) dτ 的单个冲激可写为 x(τ)δ(t − τ) dτ。

Summing over all possible τ gives the sifting property of the delta function:

x(t) = ∫₋∞^∞ x(τ)δ(t − τ) dτ

This integral is not an ordinary multiplication; it is a continuous superposition of impulses. It is valid for any reasonably well-behaved signal x(t).

这个积分不是普通的乘法，而是冲激的连续叠加。它对任何性质良好的信号 x(t) 都成立。

In discrete time, the analogous decomposition is:

x[n] = Σₖ₌₋∞^∞ x[k]δ[n − k]

Here x[k] is the value of the signal at time k, and δ[n − k] is a unit sample shifted by k.

这里 x[k] 是信号在时刻 k 的值，δ[n − k] 是平移 k 后的单位采样序列。

4. Applying the System to Each Impulse | 将系统作用于每个冲激

Because the system is linear, the response to the superposition of inputs is the superposition of individual responses. The response to the single impulse δ(t − τ) is h(t − τ), due to time invariance.

由于系统是线性的，对输入叠加的响应等于各个响应之和。根据时不变性，对单个冲激 δ(t − τ) 的响应是 h(t − τ)。

Now scale and integrate: the input component x(τ)δ(t − τ) dτ produces the output component x(τ)h(t − τ) dτ.

现在进行缩放与积分：输入分量 x(τ)δ(t − τ) dτ 产生输出分量 x(τ)h(t − τ) dτ。

Summing all these infinitesimal contributions yields the convolution integral.

将所有无穷小贡献相加，就得到卷积积分。

5. The Convolution Integral: The General Response | 卷积积分：一般响应

Applying linearity and time invariance to the decomposition in Section 3, the output y(t) is:

将第3节的分解应用于线性和时不变性，输出 y(t) 为：

y(t) = ∫₋∞^∞ x(τ)h(t − τ) dτ = (x ∗ h)(t)

This is the convolution integral. It states that the response of an LTI system to any input x(t) is the convolution of the input with the impulse response h(t).

这就是卷积积分。它表明LTI系统对任意输入 x(t) 的响应是输入与冲激响应 h(t) 的卷积。

Equivalently, by a change of variable, the convolution is commutative:

等价地，通过变量替换，卷积满足交换律：

y(t) = ∫₋∞^∞ h(τ)x(t − τ) dτ

Both forms appear frequently in IB math and engineering courses. The first is often easier when h(t) is simple; the second is useful when x(t) is simple.

这两种形式在IB数学和工程课程中经常出现。当 h(t) 简单时第一种形式更方便；当 x(t) 简单时第二种形式更有用。

6. Step-by-Step Derivation Process | 逐步推导过程

Let us outline a practical step-by-step method for deriving the general response from an impulse response.

我们来概述从冲激响应推导一般响应的实用步骤。

Step 1: Write the input x(t) as an integral of shifted impulses using the sifting property.

第一步：利用筛选性质将输入 x(t) 写成平移冲激的积分。
Step 2: Identify the response to a single shifted impulse δ(t − τ) as h(t − τ) by time invariance.

第二步：根据时不变性，确定单个平移冲激 δ(t − τ) 的响应为 h(t − τ)。
Step 3: Multiply by the strength x(τ)dτ and sum over all τ, using linearity.

第三步：利用线性性质，乘以强度 x(τ)dτ 并对所有 τ 求和。
Step 4: Evaluate the resulting integral, possibly by splitting it into intervals where x or h has simple forms.

第四步：计算所得积分，必要时将积分区间分段，使 x 或 h 具有简单形式。

This procedure replaces the difficult problem of solving a differential equation with a direct integration, provided h(t) is known.

只要 h(t) 已知，这一过程就将求解微分方程的困难问题转化为直接积分。

7. Causality and the Limits of Integration | 因果性与积分限

For a causal system, the output cannot depend on future input. This implies h(t) = 0 for t < 0. Consequently, when t − τ < 0, h(t − τ) = 0, so the integration upper limit becomes t.

对于因果系统，输出不能依赖于未来输入。这意味着 h(t) 在 t < 0 时为零。因此，当 t − τ < 0 时 h(t − τ) = 0，积分上限变为 t。

If the input x(τ) is also zero for τ < 0, the lower limit becomes 0. The convolution simplifies to:

如果输入 x(τ) 在 τ < 0 时也为零，则积分下限变为0。卷积简化为：

y(t) = ∫₀ᵗ x(τ)h(t − τ) dτ

This is the standard form for causal LTI systems with causal inputs, and it is widely used in IB exam problems.

这是具有因果输入的因果LTI系统的标准形式，在IB考试题中广泛使用。

8. Example 1: Exponential Input and Exponential Impulse Response | 示例1：指数输入与指数冲激响应

Let h(t) = e⁻ᵃᵗ for t ≥ 0, and x(t) = e⁻ᵇᵗ for t ≥ 0, with a, b > 0. Both signals are causal. Then for t ≥ 0:

设 h(t) = e⁻ᵃᵗ（t ≥ 0），x(t) = e⁻ᵇᵗ（t ≥ 0），其中 a, b > 0。两个信号都是因果的。那么对于 t ≥ 0：

y(t) = ∫₀ᵗ e⁻ᵇτ e⁻ᵃ⁽ᵗ⁻τ⁾ dτ = e⁻ᵃᵗ ∫₀ᵗ e⁽ᵃ⁻ᵇ⁾τ dτ

Evaluating the integral gives two cases. If a ≠ b, then:

计算积分得到两种情况。若 a ≠ b，则：

y(t) = [e⁻ᵇᵗ − e⁻ᵃᵗ] / (a − b)

If a = b, then y(t) = t e⁻ᵃᵗ. This example illustrates the classic “resonance” case in differential equations.

若 a = b，则 y(t) = t e⁻ᵃᵗ。这个例子展示了微分方程中经典的”共振”情形。

9. Example 2: Rectangular Pulse Input | 示例2：矩形脉冲输入

Suppose h(t) = u(t), where u(t) is the unit step function. This system is an integrator. Let the input be a rectangular pulse: x(t) = 1 for 0 ≤ t ≤ T, and x(t) = 0 otherwise.

设 h(t) = u(t)，其中 u(t) 是单位阶跃函数。该系统是一个积分器。输入为矩形脉冲：x(t) 在 0 ≤ t ≤ T 时为1，其余为0。

The convolution integral must be split into intervals. For 0 ≤ t ≤ T:

卷积积分需要分段求解。当 0 ≤ t ≤ T 时：

y(t) = ∫₀ᵗ 1 · 1 dτ = t

For t > T:

当 t > T 时：

y(t) = ∫₀ᵀ 1 · 1 dτ = T

Thus the output rises linearly to T and then stays constant. This is exactly the expected behavior of an integrator.

因此输出线性上升到 T，然后保持不变。这正是积分器的预期行为。

10. Discrete-Time Convolution | 离散时间卷积

In discrete time, the convolution sum replaces the integral. For an LTI system with impulse response h[n], the response to input x[n] is:

在离散时间中，卷积和代替了积分。对于冲激响应为 h[n] 的LTI系统，对输入 x[n] 的响应为：

y[n] = Σₖ₌₋∞^∞ x[k]h[n − k]

This sum can be computed by flipping h, shifting it by n, multiplying pointwise, and summing. For finite-length signals, the computation is often shown as a table or a graphical sliding process.

该求和可通过将 h 翻转、平移 n、逐点相乘并求和来计算。对于有限长信号，计算常以表格或图形滑动过程呈现。

The discrete form is especially important for digital signal processing and for IB AI/HL students studying difference equations.

离散形式对数字信号处理尤为重要，也适合IB数学AI/HL中学习差分方程的学生。

11. Connection to the Laplace Transform | 与拉普拉斯变换的联系

Convolution in the time domain corresponds to multiplication in the Laplace domain. If X(s), H(s), and Y(s) are the Laplace transforms of x(t), h(t), and y(t), then:

时域中的卷积对应于拉普拉斯域中的乘法。若 X(s)、H(s)、Y(s) 分别是 x(t)、h(t)、y(t) 的拉普拉斯变换，则：

Y(s) = H(s) X(s)

Here H(s) is called the transfer function of the system. This relationship makes it easy to compute the response: transform, multiply, and inverse transform.

这里 H(s) 称为系统的传递函数。这一关系使计算响应变得容易：变换、相乘、逆变换。

In IB mathematics, this is a direct application of the convolution theorem, often covered in IB DP Mathematics Analysis and Approaches HL or Further Mathematics.

在IB数学中，这是卷积定理的直接应用，常见于IBDP数学分析与方法HL或进阶数学课程。

12. Why This Matters for Problem Solving | 为什么这对解题很重要

Knowing the impulse response allows us to find the response to any input without solving differential equations each time. This is a powerful reduction in complexity.

知道冲激响应后，我们无需每次都求解微分方程，就能找到对任意输入的响应。这极大地降低了复杂度。

For IB students, key exam skills include identifying systems as LTI, setting up the convolution integral correctly, choosing the right integration limits, and handling piecewise functions. Mastering these skills also builds intuition for transfer functions, stability, and frequency response.

对于IB学生，关键考试技能包括：判断系统是否为LTI、正确建立卷积积分、选择正确的积分限以及处理分段函数。掌握这些技能也有助于建立对传递函数、稳定性和频率响应的直觉。

In summary, the impulse response is not just one output: it is the key that unlocks the system’s behavior for every possible input.

总之，冲激响应不仅仅是一个输出：它是解锁系统对所有可能输入行为的钥匙。

Published by TutorHao | IB Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Deriving the General Response of a System from Its Impulse Response | 从冲激响应推导系统的一般响应