Fact-checked by Grok 2 weeks ago

Infinite impulse response

An infinite impulse response (IIR) filter is a type of in characterized by an of theoretically infinite duration, arising from its recursive nature where the output at any time depends not only on current and past inputs but also on past outputs. These filters are governed by linear constant-coefficient difference equations that incorporate , distinguishing them from (FIR) filters, which lack such and have finite-duration responses. IIR filters exhibit properties analogous to classical analog filters, enabling efficient approximation of continuous-time systems in discrete domains. IIR filters are implemented via recursive algorithms, typically represented by the general form y = \sum_{k=0}^{M} b_k x[n-k] - \sum_{k=1}^{N} a_k y[n-k], where x is the input, y is the output, b_k are coefficients, and a_k are coefficients. In the z-domain, their is a H(z) = \frac{\sum_{k=0}^{M} b_k z^{-k}}{1 + \sum_{k=1}^{N} a_k z^{-k}}, with poles inside the unit circle ensuring for causal systems. Design methods for IIR filters often involve bilinear transformation or to convert analog prototypes, such as Butterworth or , into digital equivalents. A key advantage of IIR filters over filters is their computational , as they achieve comparable selectivity with significantly lower filter orders—often requiring fewer multiplications and additions per sample—making them suitable for resource-constrained environments like systems. However, this comes with drawbacks, including the risk of if poles lie outside the unit circle and inherent non-linear phase responses that can distort signal waveforms, unlike the linear phase achievable with filters. Fixed-point implementations further introduce challenges like quantization and , necessitating careful and . IIR filters find widespread applications in tasks requiring sharp transitions or low-latency processing, including audio equalization, noise suppression in biomedical signals (e.g., ECG filtering), enhancement, and in . In audio applications, they enable equalizers for real-time frequency adjustments, while in , they support . Their origins trace back to the of analog filter designs in the mid-20th century, evolving with advances in hardware to become staples in modern systems like and wireless communications.

Fundamentals

Definition and Characteristics

An infinite impulse response (IIR) filter is a whose output depends on both the current and past input samples as well as past output samples, due to its recursive structure incorporating . This mechanism allows IIR filters to achieve sharp frequency responses with fewer coefficients compared to non-recursive filters, making them computationally efficient for applications like audio processing and control systems. IIR filters originated in the and as emerged, with early efforts focused on approximating the behavior of continuous-time analog filters to leverage the advantages of digital implementation, such as programmability and stability in certain conditions. Key theoretical foundations and design techniques for IIR filters were established in the seminal 1975 textbook by and Ronald W. Schafer, which formalized recursive filter structures and their properties. The defining characteristic of an IIR filter is its h, which remains non-zero for all n \geq 0 in theory, extending infinitely due to the recursive that propagates the effect of the initial indefinitely. Although practical implementations may exhibit decay in the response over time owing to finite precision or constraints, the theoretical infinite duration distinguishes IIR filters from those with finite-duration responses. The general of an IIR is depicted in a featuring a path from the input x with coefficients b_k (for k = 0, 1, \dots, M) that contribute to the zeros of the system, and a path from the output y with coefficients a_k (for k = 1, 2, \dots, N) that introduce the poles. This direct form realization sums the delayed input and output terms scaled by these coefficients to produce the current output, enabling the recursive computation essential to the 's operation.

Relation to Finite Impulse Response Filters

Finite impulse response (FIR) filters are non-recursive structures that compute their output based solely on the current and past input samples, resulting in a finite-duration impulse response that becomes and remains zero after a finite number of samples. In contrast, infinite impulse response (IIR) filters incorporate from previous output samples, making their computation recursive and leading to an impulse response that theoretically extends indefinitely. This recursive nature allows IIR filters to emulate the behavior of analog filters more closely, enabling sharper frequency selectivity with fewer coefficients compared to FIR filters, which often require higher orders to achieve similar characteristics. One key distinction lies in memory efficiency: IIR filters typically demand fewer parameters for equivalent performance in applications requiring steep transitions, as their pole-based structure provides analog-like efficiency in approximating ideal responses. However, this comes at the cost of potential due to the loop, where improper pole placement can cause unbounded outputs, unlike FIR filters, which are inherently regardless of coefficients. Regarding , IIR filters generally exhibit nonlinear phase, introducing group delay variations that can distort signal timing in phase-sensitive applications such as audio processing or data transmission. FIR filters, by design, can achieve —preserving shape and enabling constant group delay—which makes them preferable in scenarios like image processing or high-fidelity audio where must be minimized. Computationally, IIR filters offer advantages in lower-order implementations for achieving steep , reducing the number of multiplications and additions per sample, though the recursive structure may complicate implementation if is not ensured. This trade-off positions IIR filters as suitable for resource-constrained environments, while FIR filters excel in -critical or phase-linear applications despite higher computational demands.

Mathematical Foundation

Difference Equation

The time-domain behavior of an infinite impulse response (IIR) filter is described by a linear constant-coefficient difference that relates the output signal y to the input signal x through both and terms. The general form of this equation for a causal IIR is y = \sum_{k=0}^{M} b_k x[n-k] - \sum_{k=1}^{N} a_k y[n-k], where b_k (for k = 0, 1, \dots, M) are the feedforward coefficients determining the influence of current and past inputs, and a_k (for k = 1, 2, \dots, N) are the feedback coefficients governing the recursive dependence on past outputs. The of the is defined as the maximum of M and N, which indicates the highest delay involved in the and thus the complexity of the system. is inherent in this formulation, as the output y depends only on the current input x and previous values of both input and output, assuming initial rest conditions where all signals are zero for n < 0. This assumption simplifies analysis by ensuring the response starts from rest upon input application. To generate the impulse response h, the input is set to the unit impulse x = \delta, where \delta{{grok:render&&&type=render_inline_citation&&&citation_id=0&&&citation_type=wikipedia}} = 1 and \delta = 0 otherwise. Substituting into the difference equation yields h{{grok:render&&&type=render_inline_citation&&&citation_id=0&&&citation_type=wikipedia}} = b_0, and for n \geq 1, h = \sum_{k=1}^{\min(M,n)} b_k \delta[n-k] - \sum_{k=1}^{N} a_k h[n-k], resulting in a recursive computation that produces non-zero terms indefinitely due to the feedback, hence the "infinite" duration. In general, the total output response consists of the zero-state response, which is the output due to the input with zero initial conditions (y = 0 for n < 0), and the zero-input response, which arises from non-zero initial conditions with no input (x = 0). Initial conditions affect the transient behavior; under initial rest, the zero-input response is zero, isolating the zero-state component for standard filter analysis.

Z-Transform and Transfer Function

The z-transform provides a frequency-domain representation for discrete-time signals and systems, analogous to the in continuous-time analysis. For a discrete-time signal x, the bilateral z-transform is defined as X(z) = \sum_{n=-\infty}^{\infty} x z^{-n}, where z is a complex variable.https://www.dsprelated.com/freebooks/filters/Z_Transform.html For causal systems and signals that begin at n=0, the unilateral z-transform is employed, given by X(z) = \sum_{n=0}^{\infty} x z^{-n}. $$$$https://www.dsprelated.com/freebooks/filters/Z_Transform.html$$ This unilateral form is particularly relevant for [infinite impulse response](/page/Infinite_impulse_response) (IIR) filters, which are typically causal linear time-invariant (LTI) systems. To obtain the transfer function of an IIR filter, the z-transform is applied to its underlying difference equation, which describes the recursive relationship between input $x$ and output $y$. Assuming zero initial conditions for causality, the z-transform of the general difference equation y = \sum_{k=0}^{M} b_k x[n-k] - \sum_{k=1}^{N} a_k y[n-k] yields $Y(z) = \sum_{k=0}^{M} b_k z^{-k} X(z) - \sum_{k=1}^{N} a_k z^{-k} Y(z)$.$$https://www.dsprelated.com/freebooks/filters/Transfer_Function_Analysis.html$$ Rearranging terms gives Y(z) \left(1 + \sum_{k=1}^{N} a_k z^{-k}\right) = \left(\sum_{k=0}^{M} b_k z^{-k}\right) X(z), so the transfer function $H(z) = Y(z)/X(z)$ is H(z) = \frac{\sum_{k=0}^{M} b_k z^{-k}}{1 + \sum_{k=1}^{N} a_k z^{-k}}.https://www.dsprelated.com/freebooks/filters/Transfer_Function_Analysis.html This expression represents the system's frequency-domain behavior, with the numerator capturing the feedforward coefficients and the denominator the feedback terms characteristic of IIR filters. The transfer function takes the rational form H(z) = B(z)/A(z), where B(z) = \sum_{k=0}^{M} b_k z^{-k} is the numerator polynomial of degree M and A(z) = 1 + \sum_{k=1}^{N} a_k z^{-k} is the denominator polynomial of degree N, typically with N \geq M for proper filters.https://www.dsprelated.com/freebooks/filters/Transfer_Function_Analysis.html The impulse response h of the system is the inverse z-transform of H(z), which, for IIR filters, extends infinitely due to the feedback, hence the name infinite impulse response.https://www.dsprelated.com/freebooks/filters/Z_Transform.html

Analysis and Properties

Poles, Zeros, and Frequency Response

In infinite impulse response (IIR) filters, the transfer function H(z) = \frac{B(z)}{A(z)} is characterized by its poles and zeros, which are the roots of the denominator and numerator polynomials, respectively. Zeros are the values of z where B(z) = 0, resulting in H(z) = 0, and their locations in the z-plane directly influence the nulls in the frequency response. A zero located on the unit circle at z = e^{j\omega_0} produces a complete null (zero gain) at frequency \omega = \omega_0, while zeros inside or outside the unit circle create partial attenuation at nearby frequencies, with the depth of the null increasing as the zero approaches the unit circle. This placement allows designers to target specific frequencies for suppression, such as in where zeros are positioned to cancel unwanted tones. Poles are the values of z where A(z) = 0, causing |H(z)| to approach infinity, and they govern the resonance peaks and decay rates of the filter's response. The radial distance of a pole from the origin determines the decay rate of the corresponding exponential component in the impulse response, with poles closer to the unit circle yielding slower decay and higher Q-factor resonances. Clustering poles near the unit circle enhances sharp frequency transitions, such as steep roll-offs in bandpass or high-pass filters, by amplifying the response at the pole's angular frequency while maintaining overall filter selectivity. The frequency response of an IIR filter is obtained by evaluating H(z) on the unit circle, setting z = e^{j\omega} for |z| = 1, which yields H(e^{j\omega}). The magnitude |H(e^{j\omega})| describes the gain as a function of normalized frequency \omega (ranging from -\pi to \pi), revealing passband and stopband characteristics shaped by the proximity of poles and zeros to points on the unit circle. The phase response is given by \arg(H(e^{j\omega})), which exhibits nonlinear behavior due to the pole-zero configuration, often leading to phase distortion in IIR designs. For example, a low-pass IIR filter can be realized with a pair of complex-conjugate poles placed inside the unit circle near z = 1 (corresponding to low frequency \omega \approx 0), which boosts low-frequency gain and provides smooth attenuation at higher frequencies. Zeros may be positioned near z = -1 to further sharpen the cutoff, resulting in a magnitude response that closely approximates an ideal low-pass shape with minimal order.

Stability Conditions

Stability in infinite impulse response (IIR) filters is typically assessed using the bounded-input bounded-output (BIBO) criterion, which requires that every bounded input sequence produces a bounded output sequence. For linear time-invariant (LTI) discrete-time systems described by linear constant-coefficient difference equations, BIBO stability holds if and only if the impulse response h is absolutely summable, i.e., \sum_{n=-\infty}^{\infty} |h| < \infty. For causal IIR filters with rational transfer functions H(z) = \frac{B(z)}{A(z)}, where A(z) is the denominator polynomial, BIBO stability is equivalent to all poles of H(z) lying strictly inside the unit circle in the z-plane, meaning |p_i| < 1 for every pole p_i. This condition ensures that the region of convergence of H(z) includes the unit circle |z| = 1, allowing the frequency response to be well-defined and the impulse response to decay to zero. If any pole lies on or outside the unit circle (|p_i| \geq 1), the system is unstable, as the impulse response contains terms that do not decay or grow exponentially, leading to unbounded output growth for bounded inputs. To verify that all roots of the denominator polynomial A(z) = a_0 + a_1 z + \cdots + a_N z^N (with a_N > 0) lie inside the unit circle without explicitly solving for the roots, the Jury stability test provides a tabular algebraic criterion. This method, a simplification of the for real coefficients, constructs a table iteratively from the polynomial coefficients and checks specific conditions on the table entries. The Jury table begins with two rows: the first row contains the coefficients [a_0, a_1, \dots, a_N], and the second row is the reverse [a_N, a_{N-1}, \dots, a_0]. Subsequent rows are generated by computing elements b_k = -\frac{1}{a_0} \det \begin{vmatrix} a_0 & a_{N-k} \\ a_N & a_k \end{vmatrix} for the third row, and continuing similarly with decreasing polynomial degrees until a single element remains. Necessary conditions for stability include A(1) > 0, (-1)^N A(-1) > 0, and |a_0| < a_N. The full set requires that the absolute values of the first-column elements strictly decrease: |b_0| < |a_0|, and so on for subsequent rows. For low-order filters, the test simplifies significantly. For a first-order polynomial A(z) = a_0 + a_1 z (with a_1 > 0), requires |a_0| < a_1. For a second-order polynomial A(z) = a_0 + a_1 z + a_2 z^2 (with a_2 > 0), requires |a_0| < a_2, a_0 + a_1 + a_2 > 0, and a_0 - a_1 + a_2 > 0. The Jury table is:
Rowz^0z^1z^2
1a_0a_1a_2
2a_2a_1a_0
3b_0b_1
where b_0 = -\frac{1}{a_0} (a_0^2 - a_2^2) = \frac{a_2^2 - a_0^2}{a_0} and b_1 = -\frac{1}{a_0} \det \begin{vmatrix} a_0 & a_1 \\ a_2 & a_1 \end{vmatrix} = \frac{a_1 (a_2 - a_0)}{a_0}. These conditions ensure no roots exceed the unit circle magnitude. Alternative methods to verify pole locations include the Schur-Cohn algorithm, which recursively tests the for roots inside the unit circle using inner determinants, and direct numerical root-finding techniques applied to A(z) = 0. The Schur-Cohn approach is more general but computationally intensive for higher orders, while root-finding is straightforward with modern tools but less algebraic in nature.

Design Approaches

Impulse Invariance Transformation

The transformation is a design method for infinite (IIR) filters that preserves the time-domain of an analog prototype filter at the sampling instants. Specifically, the transfer function H_d(z) is constructed such that its satisfies h_d = T h_a(nT), where T is the sampling period and h_a(t) is the continuous-time of the analog filter H_a(s). This approach ensures that the mimics the transient behavior of the analog filter in the sampled domain, making it particularly useful for applications where time-domain characteristics, such as , are critical. The design process begins with the analog H_a(s), which is expressed in partial fraction form as H_a(s) = \sum_{k=1}^N \frac{A_k}{s - p_k}, where p_k are the poles and A_k the corresponding residues. The yields h_a(t) = \sum_{k=1}^N A_k e^{p_k t} u(t), and sampling gives h_d = T \sum_{k=1}^N A_k e^{p_k n T} u. Taking the results in H_d(z) = T \sum_{k=1}^N \frac{A_k}{1 - e^{p_k T} z^{-1}}, effectively each analog p_k to a digital at e^{p_k T}. This method inherently introduces because the digital is given by the infinite H_d(e^{j \omega}) = \sum_{k=-\infty}^{\infty} H_a \left( j \frac{\omega + 2\pi k}{T} \right), where \omega is the normalized digital in radians per sample. This folds higher-frequency components of H_a(j \Omega) into the |\omega| \leq \pi. To mitigate this, high sampling rates are used where contributions from |k| \geq 1 are negligible. The impulse invariance method is best suited for designing low-pass filters from analog s, where the analog filter's bandwidth is well below the to minimize distortion. For instance, consider designing a sixth-order low-pass from a Butterworth analog with a of 2 rad/s and sampling rate of 10 Hz (T = 0.1 s). The analog Butterworth is normalized and expanded into partial fractions, then transformed via impulse invariance to yield H_d(z) with poles at e^{p_k T}, resulting in a digital filter that closely matches the analog magnitude response in the while exhibiting controlled in the due to the low-pass nature. This example demonstrates the method's effectiveness for filters where the sampling rate exceeds twice the highest frequency of interest, ensuring the aliased tails from the infinite sum have minimal impact on the response.

Step Invariance Transformation

The step invariance transformation is a design technique for infinite impulse response (IIR) digital filters that preserves the step response characteristics of an analog . In this method, the unit step response of the digital is required to match the sampled values of the analog filter's unit step response at sampling instants nT, where T is the sampling period. This approach models the input to the analog filter as a (ZOH), approximating the continuous input as piecewise constant steps of duration T, which is particularly relevant for digital-to-analog conversion scenarios in control systems. The derivation begins with the analog transfer function H_a(s). The unit step response in the continuous domain is s_a(t) = \mathcal{L}^{-1} \left\{ \frac{H_a(s)}{s} \right\}, obtained via the inverse Laplace transform. Sampling this yields the desired digital step response s_d = s_a(nT) for n = 0, 1, 2, \dots. Since the z-transform of the unit step input u is \frac{1}{1 - z^{-1}}, the z-transform of the digital step response is \frac{H_d(z)}{1 - z^{-1}} = Z \left\{ s_d \right\}, leading to the digital transfer function H_d(z) = (1 - z^{-1}) \, Z \left\{ s_d \right\} = (1 - z^{-1}) \, Z \left\{ \mathcal{L}^{-1} \left\{ \frac{H_a(s)}{s} \right\} \bigg|_{t = nT} \right\}. This form uses the explicit step response computation via partial fraction expansion for practical implementation. This adjustment from impulse invariance incorporates integration to account for the cumulative nature of the step input. Compared to the impulse invariance method, step invariance reduces aliasing effects, particularly for low-frequency components, by emphasizing the integrated response, making it suitable for low-pass and band-pass filters in control applications where low sampling rates relative to the filter's frequency (e.g., f_s / f_p \approx 4) are common. It provides accurate magnitude responses in these cases and avoids issues with non-zero high-frequency gain seen in impulse invariance. However, it remains susceptible to some , especially for high-pass or band-reject filters, and introduces errors near resonant frequencies; consequently, it is less commonly applied in audio and , where the bilinear transformation is preferred for its aliasing-free frequency mapping.

Bilinear Transformation

The bilinear transformation is a widely used method for designing infinite impulse response (IIR) s by converting an analog into a discrete-time equivalent, ensuring a one-to-one that avoids issues inherent in other techniques. This approach substitutes the continuous-time s with a function of the discrete-time z, specifically s = \frac{2}{T} \frac{1 - z^{-1}}{1 + z^{-1}}, where T is the sampling period. The transformation derives from the approximation of the in the continuous-time system, providing a rational that preserves the filter's frequency-selective properties while adapting to the discrete domain. As a conformal mapping, the bilinear transformation compresses the entire infinite s-plane onto the unit disk in the z-plane, mapping the imaginary axis of the s-plane (corresponding to analog frequencies) onto the unit circle in the z-plane exactly once. This property ensures that stable analog filters, with poles in the left-half s-plane, result in stable digital filters with poles inside the unit circle (|z| < 1). Consequently, the method maintains stability without requiring additional checks, making it suitable for high-order designs. The design process begins with prewarping the critical frequencies to account for the nonlinear mapping. For a desired digital cutoff \omega_c (in radians per sample), the corresponding analog frequency \Omega_c is computed as \Omega_c = \frac{2}{T} \tan\left(\frac{\omega_c T}{2}\right), ensuring that and edges align accurately after transformation. Next, an analog prototype filter H_a(s) is designed using classical methods, such as Butterworth or Chebyshev approximations, based on the prewarped specifications to meet the required and cutoff. Finally, s in H_a(s) is replaced by the bilinear expression to obtain the digital transfer function H_d(z). A key characteristic is the frequency warping effect, where the analog frequency \Omega maps nonlinearly to the digital frequency \omega via \Omega = \frac{2}{T} \tan\left(\frac{\omega}{2}\right). This compression distorts higher frequencies more severely than lower ones, potentially shifting filter edges if not prewarped; for instance, frequencies near the Nyquist limit (\omega = \pi) are "squeezed" toward \omega = \pi/2. Prewarping mitigates this for specified bands, preserving the magnitude response shape within the warped scale. The advantages of the bilinear transformation include its immunity to , as the mapping is bijective and does not replicate frequencies beyond the , unlike methods. It is particularly effective for low-pass and band-pass filters using Butterworth or Chebyshev prototypes, where the is acceptable, and has become a standard in toolboxes for its computational simplicity and reliability.

Realization and Implementation

Direct Form Structures

Direct form structures implement infinite impulse response (IIR) filters by translating the difference equation into a network of delay elements (z^{-1}), multipliers, and adders, enabling iterative computation of the output sequence y from the input x. These structures directly realize the H(z) = \frac{\sum_{k=0}^{M} b_k z^{-k}}{1 + \sum_{k=1}^{N} a_k z^{-k}}, where the numerator coefficients b_k correspond to zeros and the denominator coefficients a_k to poles.

Direct Form I

The Direct Form I realization consists of separate and sections, with the feedforward part implementing the non-recursive (FIR-like) summation and the feedback part handling the recursive terms. It uses two distinct delay lines: one for the input signal with M delays to provide the terms x[n-k] for k = 0 to M, and another for the output signal with N delays to provide y[n-k] for k = 1 to N, resulting in a total of M + N delays for an Nth-order filter (assuming M ≈ N in typical designs). The output is computed iteratively according to the difference equation: \begin{align} y &= \sum_{k=0}^{M} b_k x[n - k] - \sum_{k=1}^{N} a_k y[n - k]. \end{align} In the block diagram, the input x enters the delay , where each tap is scaled by b_k and to form an intermediate signal, which is then added to the from the output delay scaled by -a_k; the resulting y feeds back into its own delay line. This separation avoids intermediate signal growth issues in but requires more memory than alternative forms.

Direct Form II

The Direct Form II realization optimizes hardware usage by employing a transposed structure with a single shared delay line for both feedforward and feedback operations, reducing the total number of delays to N for an Nth-order filter. This implements the poles first via on an intermediate signal w, followed by the zeros via scaling, making it particularly efficient for implementations where delay elements dominate resource costs. The iterative computation introduces w as: \begin{align} w &= x - \sum_{k=1}^{N} a_k w[n - k], \\ y &= \sum_{k=0}^{M} b_k w[n - k]. \end{align} The signal flow graph shows x entering a summer that subtracts the terms from delayed versions of w (scaled by a_k), producing w which then branches to the shared N-delay chain; taps from this chain are scaled by b_k and summed to yield y. This structure minimizes state variables but can lead to larger intermediate values at the summer, requiring careful scaling. Direct form structures, especially in their paths, are sensitive to quantization, where errors in the a_k can be amplified, particularly for low-order poles near the unit circle, potentially shifting pole locations and degrading performance.

Lattice and Other Forms

The form represents an orthogonal structure for realizing infinite impulse response (IIR) , characterized by reflection coefficients k_i that parameterize the filter sections. This form is particularly suited for all-pole , such as those in , where the coefficients k_i (also known as partial correlation or PARCOR coefficients) are bounded by |k_i| < 1 to ensure . For general pole-zero IIR , the structure is derived using the Gray-Markel , a two-step process that first transforms the denominator into an all-pole and then incorporates the numerator via tapped sections, providing a modular ladder-like realization. This structure exhibits robustness to quantization effects, as the orthogonal nature of the lattice minimizes error propagation in finite-word-length implementations, making it advantageous for fixed-point digital signal processing hardware. The low sensitivity to coefficient perturbations is especially beneficial for high-order designs, where small changes in parameters can otherwise lead to significant frequency response deviations. Conversion from direct form realizations to lattice form employs a recursion akin to the Levinson-Durbin algorithm, iteratively computing the k_i from the polynomial coefficients while preserving the transfer function. Additionally, the modular design facilitates adaptation in IIR filters, as updating individual k_i allows efficient reconfiguration without recomputing the entire structure, which is valuable in applications like speech processing and echo cancellation. Beyond lattice forms, other canonical realizations for IIR filters include and structures, which decompose the into or products of lower-order sections, typically second-order (biquad) blocks. In the form, the overall is expressed as a of second-order sections, each handling a pair of poles, which decouples the pole interactions and reduces sensitivity to rounding errors compared to forms. The form, conversely, implements the filter as a product of second-order sections in series, offering similar decoupling benefits while allowing independent scaling of sections to optimize and minimize overflow risks in . These forms enhance by localizing errors within individual sections, making them suitable for high-order filters in systems where pole-zero proximity can amplify quantization noise. Decomposed forms like and generally require a higher number of multipliers than direct or forms, particularly for low-order filters, increasing and hardware demands. For instance, an Nth-order (N even) or IIR implemented with N/2 second-order sections may need approximately 2.5N multipliers, versus 2N + 1 for direct form II or Gray–Markel (assuming M = N), though the trade-off favors these forms in scenarios prioritizing stability over minimal arithmetic operations.

Practical Aspects

Example: Second-Order IIR Filter

To illustrate the design and implementation of a second-order infinite impulse response (IIR) , consider the approximation of an analog second-order Butterworth low-pass filter using the . The analog prototype is H_a(s) = \frac{1}{s^2 + \sqrt{2}\, s + 1}, normalized such that the is \omega_c = 1 rad/s. This prototype provides a maximally flat magnitude response in the , as established in the seminal work on Butterworth filters. The bilinear transform is applied with sampling period T = 1 s (corresponding to a normalized sampling frequency f_s = 1 Hz). Prewarping adjusts the analog cutoff frequency using \Omega_c = \frac{2}{T} \tan\left(\frac{\omega_c T}{2}\right) \approx 1.093 rad/s to compensate for the nonlinear frequency mapping inherent in the transform, ensuring the digital filter's cutoff aligns with the desired specification of \omega_c = 1 rad/sample. The prewarped analog prototype is H_a(s) = \frac{\Omega_c^2}{s^2 + \sqrt{2} \Omega_c s + \Omega_c^2}. Substituting s = 2 \frac{1 - z^{-1}}{1 + z^{-1}} into the prewarped analog prototype yields the digital transfer function H(z) = \frac{0.1442 (1 + 2 z^{-1} + z^{-2})}{1 - 0.6772 z^{-1} + 0.2538 z^{-2}}. The numerator form $1 + 2 z^{-1} + z^{-2} = (1 + z^{-1})^2 arises naturally from the bilinear substitution for low-pass filters, scaled by 0.1442 to achieve unity DC gain. The filter is implemented using the Direct Form II structure, which uses only two delay elements for efficiency. The corresponding difference equation is y = 0.1442 x + 0.2884 x[n-1] + 0.1442 x[n-2] + 0.6772 y[n-1] - 0.2538 y[n-2], where the coefficients are b_0 = b_2 = 0.1442, b_1 = 0.2884, a_1 = -0.6772, and a_2 = 0.2538. The h is obtained recursively from the difference equation with input x = \delta (the unit impulse) and initial conditions y[-1] = y[-2] = 0. The response exhibits an initial peak followed by damped oscillations, decaying exponentially due to the locations inside the unit circle, and can be plotted to visualize the infinite duration characteristic of IIR filters. Verification of the frequency response involves evaluating |H(e^{j\omega})| along the unit circle. The magnitude response rolls off smoothly from 0 at , reaching -3 precisely at the prewarped cutoff frequency, consistent with the Butterworth approximation. Stability is confirmed by locating the poles as the roots of the denominator equation z^2 - 0.6772 z + 0.2538 = 0, yielding complex conjugate poles with magnitude \sqrt{0.2538} \approx 0.504 < 1, ensuring all poles lie within the unit circle.

Advantages, Limitations, and Applications

Infinite impulse response (IIR) filters offer significant advantages in due to their recursive structure, which enables high selectivity with a low filter order. By placing poles near the unit circle in the z-plane, IIR filters can achieve sharp frequency cutoffs and steep transition bands that would require much higher orders in (FIR) equivalents, making them ideal for applications demanding precise frequency shaping without excessive computational resources. Additionally, IIR filters are computationally efficient for processing, as they typically require fewer multiplications and additions per output sample compared to FIR filters of similar performance; for instance, a second-order IIR often suffices where an FIR might need dozens of taps, reducing both complexity and power consumption in systems. Despite these benefits, IIR filters have notable limitations that can impact their reliability in certain scenarios. Their mechanism introduces the potential for if poles lie outside the unit circle or if quantization errors accumulate, a absent in non-recursive filters. Furthermore, IIR filters generally exhibit nonlinear responses, leading to distortion where different components experience varying delays, which can degrade in applications sensitive to preservation, such as audio or . In fixed-point processors, IIR filters are also highly sensitive to precision; finite-word-length effects can amplify errors through , potentially shifting locations and causing or degraded performance. IIR filters find widespread applications across diverse fields leveraging their efficiency and selectivity. In audio processing, they are commonly used for equalization and , enabling real-time enhancement of sound signals with minimal . Echo cancellation in employs adaptive IIR structures to model acoustic paths and suppress reverberations effectively. Biomedical signal processing benefits from IIR filters in tasks like ECG filtering, where they remove baseline wander and power-line while preserving diagnostic features. In control systems, IIR filters provide robust compensation for and performance in real-time loops. Post-2020 has extended IIR concepts to neural networks, where differentiable IIR filters serve as building blocks for efficient, learnable in models, such as audio synthesis and upsampling. Emerging trends include hybrid IIR-FIR designs for adaptive filtering, which combine IIR's sharp response with FIR's to optimize performance in 5G for channel equalization and mitigation.

References

  1. [1]
    [PDF] INTRODUCTION TO DIGITAL FILTERS - Physics 123/253
    An FIR filter is one whose impulse response is of finite duration. An IIR filter is one whose impulse response theoretically continues for ever because the ...
  2. [2]
    [PDF] INFINITE-IMPULSE RESPONSE DIGITAL FILTERS Classical analog ...
    The impulse response h(n) is infinite in length. 2. A system described by this type of difference equation is called an IIR (Infinite Impulse Response) filter, ...
  3. [3]
    Infinite Impulse Response Filters | part of Digital Signal Processing
    Digital filters with an infinite impulse response (IIR), or recursive filters, have properties similar to those of analog filters.
  4. [4]
    [PDF] Lecture 28: IIR Filters
    An autoregressive filter is also known as an infinite impulse response (IIR) filter, because h[n] is infinitely long (never ends). A difference equation with ...Missing: definition | Show results with:definition
  5. [5]
    10. IIR Filters - DSP First
    Overview: In this chapter the class of infinite-impulse-response (IIR) filters is introduced. These digital filters involve feedback terms because past ...
  6. [6]
    [PDF] 1 IIR Filters
    Jul 22, 2002 · Impulse Response: The impulse response sequence of an IIR filter has infinite support, which is why we use the name IIR. (i) The terms of the ...<|control11|><|separator|>
  7. [7]
    [PDF] On Fast FIR Filters Implemented as Tail-Canceling IIR Filters
    Infinite impulse response (IIR) recursive linear digital filters are widely used because of their low computational cost and low storage overhead requirements.
  8. [8]
    [PDF] Filter Design and Implementation
    The primary advantage of IIR filters over FIR filters is that they typically meet a given set of specifications with a much lower filter order than a ...
  9. [9]
    Lab 3: Implementation of IIR filters - Patrick Schaumont
    Infinite Impulse Response Filter Implementation · This filter is created using five second-order sections ( MWSPT_NSEC ). · The NUM array contains the numerator ...
  10. [10]
    [PDF] Lecture 7 - IIR Filters - Colorado State University
    For an IIR filter to satisfy the linear phase condition, both the poles and zeros would need to have mirror images outside the unit circle of the z-plane.
  11. [11]
    [PDF] Fixed-point-IIR-filter challenges - spinlab
    Challenges include unique design issues, finite-word-length effects like register overflow and arithmetic errors, and the need to control register overflow.
  12. [12]
    INTRODUCTION TO DIGITAL FILTERS WITH AUDIO APPLICATIONS
    INTRODUCTION TO DIGITAL FILTERS WITH AUDIO APPLICATIONS · Frequency Response · Amplitude Response · Phase Response · Polar Form of the Frequency Response.
  13. [13]
    [PDF] Introduction to Digital Filters - UCSB ECE
    • Main disadvantages of digital filters over analog filters. 1. Speed limitation: The maximum bandwidth of signals that digital filters can handle, in real time ...
  14. [14]
    [PDF] Pros and cons 3. IIR digital filters 4. FIR digital filters 5.
    Digital filters have the following advantages. 1. Programmable (filter characteristics easily changed). 2. Reliable and repeatable. 3. Free from component ...
  15. [15]
    [PDF] Parametric IIR Filtering on an FPGA - Ashkan Ashrafi
    A useful application for a parametrically adjustable IIR filter is an audio equalizer. Audio equalizers, often used in music applications, boost or attenuate ...
  16. [16]
    [PDF] Intro to DSP.pptx - Washington
    FIR filters have several advantages that make them more desirable than IIR filters for certain design applications: – FIR can be designed to have linear ...
  17. [17]
    The Application of the IIR Filters Based on FPGA in the DTV Field
    This paper presents the design and realization of IIR filter based on FPGA. It firstly introduces the application of IIR filter in digital television ...
  18. [18]
    3 Infinite Impulse Response (IIR) Filters - Microchip Online docs
    Principle. Infinite Impulse Response (IIR) filters are feedback-based filters, i.e., the previous output plays a role in the current output.
  19. [19]
    [PDF] IIR vs. FIR - MIT OpenCourseWare
    Historically, digital IIR filters have been derived from their analog counterparts. There are several common types of analog filters: Butterworth which have ...<|control11|><|separator|>
  20. [20]
    Digital Signal Processing - Alan V. Oppenheim, Ronald W. Schafer
    Digital Signal Processing. Front Cover. Alan V. Oppenheim, Ronald W. Schafer. Prentice-Hall, 1975 - Digital electronics - 585 pages ... Export Citation, BiBTeX ...
  21. [21]
    [PDF] Chapter 5 – Design of IIR Filters
    In most cases a recursive filter has an impulse response which theoretically continues forever. It is therefore referred to as an infinite impulse response ( ...
  22. [22]
    [PDF] Ch. 8: IIR Filters • Difference equation • System function
    Dec 8, 2002 · IIR filters are recursive systems that depend on both current and past inputs and outputs, and have an infinitely long impulse response.Missing: infinite | Show results with:infinite
  23. [23]
  24. [24]
    [PDF] Chapter 3 - Introduction to Digital Filters
    Conversely, IIR filters require fewer coefficients than FIR filters for a sharp cut-off frequency response, and analogue filters can only be modelled using IIR ...<|control11|><|separator|>
  25. [25]
    Difference Equation - Stanford CCRMA
    Recursive filters are also called infinite-impulse-response (IIR) filters. ... difference equation may be referred to as an explicit finite difference scheme.
  26. [26]
    [PDF] 2.161 Signal Processing: Continuous and Discrete
    The Design of IIR Filters. An IIR filter is characterized by a recursive difference equation. N. M yn = akyn−k + bkfn−k k=1 k=0 and a rational transfer ...
  27. [27]
    Pole-Zero Analysis - Stanford CCRMA
    Pole-zero analysis specifies digital filters using poles and zeros. Zeros are where the transfer function is 0, and poles are where it approaches infinity.Missing: explanation | Show results with:explanation
  28. [28]
    Pole-Zero plot - Theory/Equations
    A pole-zero plot relates the frequency and Z-domains. The frequency response is obtained by evaluating the transfer function at z=e^jw, where z lies on the ...
  29. [29]
    [PDF] Discrete-Time Signals and Systems - Higher Education | Pearson
    2.2 DISCRETE-TIME SYSTEMS. A discrete-time system is defined mathematically as a transformation or operator that maps an input sequence with values x[n] into ...
  30. [30]
    Stability Revisited | Introduction to Digital Filters - DSPRelated.com
    This is because the transfer function is the z transform of the impulse ... $ H(z)$ moves outside the unit circle, and the impulse response has an ...<|control11|><|separator|>
  31. [31]
    [PDF] A Simplified Stability Criterion for Linear Discrete Systems - DTIC
    A recent investigation by this author has shown that the evaluation of the Schur-Cohn determinants can be simplified considerably by making use of the real ...Missing: IIR filter BIBO
  32. [32]
  33. [33]
    Impulse Invariant Method - Stanford CCRMA
    Performing the inverse Laplace transform on the partial fraction expansion we obtain the impulse response in terms of the system poles and residues:.
  34. [34]
    [PDF] Lecture 14 Design of IIR digital filters, part 1 - MIT OpenCourseWare
    The second transformation discussed is the use of impulse invariance, corresponding to obtaining the discrete-time unit sample response by sampling the analog ...
  35. [35]
    impinvar - Impulse invariance method for analog-to-digital filter ...
    Impulse invariance introduces a gain of 1 / f s to the digital filter. Multiply the analog impulse response by this gain to enable meaningful comparison.
  36. [36]
    [PDF] Lecture 24: Butterworth filters - MIT OpenCourseWare
    In this lecture, we illustrate the design of a discrete-time filter through the use of the impulse-invariant design procedure applied to a Butterworth filter.
  37. [37]
    [PDF] Derivation of Recursive Digital Filters by the Step-Invariant ... - DTIC
    The realization step is the process of converting the transfer function into a filter network using inter- connected unit delays, adders and multipliers. In the ...Missing: IIR | Show results with:IIR
  38. [38]
    [PDF] Design of Digital Filters - UTK-EECS
    digital filter be a sampled version of the impulse response of the analog filter. Step invariant design makes the step response of the digital filter be a ...
  39. [39]
    Bilinear transformation method for analog-to-digital filter conversion
    It transforms analog filters, designed using classical filter design techniques, into their discrete equivalents. The bilinear transformation maps the s-plane ...
  40. [40]
    [PDF] Digital Signal Processing IIR Filter Design via Bilinear Transform
    The basic procedure for IIR filter design via bilinear transform is: 1. Determine the CT filter class: 1.1 Butterworth. 1.2 Chebychev Type I or Type II. 1.3 ...<|control11|><|separator|>
  41. [41]
    [PDF] Lecture 8 - IIR Filters (II) - Colorado State University
    The corresponding discrete-time filter has a transfer function given by. H(z) = 1/2. 1 - e−(a−jb)t0 z−1. +. 1/2. 1 - e−(a+jb)t0 z−1. ,. (7) or. H(z) = B(z). A(z).
  42. [42]
    [PDF] Infinite Impulse Response (IIR) Digital Filters (III) Bilinear mapping
    Use Bilinear Transformation method and assume a 3 dB cut off frequency of ... Design a low-pass digital filter operating at the rate of 20 kHz. And ...
  43. [43]
    [PDF] Signal Flow Graphs IIR Filter Structures - MIT OpenCourseWare
    Note that the feedforward section determines the zeros of the transfer function, while the feedback section gives the poles. Interchanging the order of the ...
  44. [44]
    The Four Direct Forms | Introduction to Digital Filters
    The DF-I structure has the following properties: It can be regarded as a two-zero filter section followed in series by a two-pole filter section.
  45. [45]
    [PDF] Effect of Coefficient Quantization on IIR Filters - spinlab
    4. IIR filter response is often quite sensitive to denominator coefficient quantization. In fact, denominator coefficient quantization can cause an IIR filter ...
  46. [46]
    High performance IIR filter implementation on FPGA
    Jan 6, 2021 · However, the cascade and parallel realizations of IIR filters are more robust than the direct and canonic realization filters, and they have ...
  47. [47]
  48. [48]
    None
    ### Summary of Second-Order Butterworth IIR Filter Example with Bilinear Transform
  49. [49]
    3.6 Designing a digital filter in the frequency domain | OpenLearn
    Digital filters are designed in the frequency domain by modifying the input signal's spectrum, using software to generate a difference equation. There are two ...
  50. [50]
    [PDF] Filter-based fading channel modeling - Amir Alimohammad
    As the coefficients are quantized in any fixed-point implementation, the resulting numerical error is fed back in the IIR filter, possibly causing instability.<|separator|>
  51. [51]
    [PDF] IIR Multiple Notch Filter Design for Power Line Interference Removal
    As we design this multiple notch filter for two practical applications i.e. ECG and hum in audio system and speech recording up to five notches. We further ...
  52. [52]
    [PDF] Adaptive Filters Theory And Applications
    Common applications of adaptive filters include noise cancellation in audio signals, echo cancellation in telecommunications, system identification, and ...
  53. [53]
    [PDF] Practical Signals Theory With Matlab Applications Practical Signals ...
    Response (IIR) filters can achieve sharper cutoffs with lower order but may exhibit ... control systems. (feedback control). Q6: Are there ... are relevant (audio, ...
  54. [54]
    [PDF] Differentiable IIR filters for machine learning applications - DAFX
    Sep 8, 2020 · In this paper we present an approach to using traditional digital IIR filter structures inside deep-learning networks trained using back-.
  55. [55]
    Neural IIR Filter Field for HRTF Upsampling and Personalization
    Feb 27, 2024 · IIR filters mimic the modal nature of HRTFs, thus needing fewer coefficients to approximate them well compared to FIR filters. We find that our ...
  56. [56]
    FIR-IIR adaptive filters hybrid combination - Semantic Scholar
    Mar 27, 2014 · A hybrid combination of FIR and IIR adaptive filters (AFs) via a supervisor that senses which one is performing best is proposed, ...
  57. [57]
    Comparative Performance Analysis of IIR and FIR Filters for 5G ...
    The proposed hybrid scheme consists of a combination of precoding and non-linear companding techniques (NLCTs). A comparative analysis of the performance of ...