4.10 Signal Distortion

Signal distortion is any alteration of a transmitted signal that causes the received signal to differ from the intended transmitted signal in ways other than a simple scaling or time delay. While noise introduces random and unpredictable fluctuations into a signal, distortion introduces systematic, deterministic modifications that change the signal's waveform, frequency content, or amplitude relationships in ways that may degrade the accuracy of the information it carries. In communication systems, distortion and noise are both sources of error, but they arise from different mechanisms and require different countermeasures.

Distortion originates in the characteristics of the transmission medium, amplifiers, filters, and other circuit elements that a signal passes through. Unlike ideal linear time-invariant systems, which reproduce signals faithfully apart from amplitude scaling and phase shifting, real-world systems introduce various forms of distortion that depend on the signal's frequency, amplitude, or history.

Amplitude distortion occurs when a system responds differently to signals of different amplitudes, typically due to nonlinear elements. The relationship between input and output ceases to be proportional, causing the waveform shape to be altered and generating harmonic frequencies that were not present in the original signal. An amplifier driven into saturation produces amplitude distortion by compressing large-amplitude portions of the waveform.

Frequency distortion arises when a system does not transmit all frequency components of a signal with equal gain. Different frequencies are amplified or attenuated by different amounts, altering the spectral composition of the signal. A channel with a limited bandwidth that does not extend to the high-frequency components of the signal introduces frequency distortion by attenuating those components, effectively filtering out fine temporal details of the transmitted waveform.

Phase distortion occurs when different frequency components of a signal experience different phase shifts as they pass through a system. Even if all amplitudes are preserved, misaligned phases cause the temporal relationships between components to shift, changing the shape of the time-domain waveform. Phase distortion is particularly problematic for digital pulse transmission and for audio signals, where the relative timing of frequency components affects perceived quality.

Group delay distortion is closely related to phase distortion and occurs when the group delay of a channel, defined as the negative derivative of phase shift with respect to angular frequency, varies across the signal bandwidth:

When group delay is constant across the bandwidth, the signal envelope passes through the channel without distortion. When group delay varies, different frequency components of a pulse arrive at the receiver at different times, causing the pulse to spread out in time, a phenomenon called pulse dispersion. This is a critical concern in optical fiber communications, where chromatic dispersion causes different wavelength components to travel at slightly different speeds.

Intermodulation distortion is a nonlinear effect that occurs when two or more signals are simultaneously present in a nonlinear system. The nonlinearity causes the signals to mix, producing sum and difference frequency components at frequencies not present in the original signals. These intermodulation products can fall within the signal band and cause interference that cannot be filtered out without also removing the desired signal components.

In digital communication systems, the practical consequences of distortion manifest as intersymbol interference (ISI), where the distorted response to one transmitted symbol spreads into the time slots of adjacent symbols. A receiver attempting to decode each symbol independently will suffer from the contaminating energy of neighboring symbols. ISI is addressed through equalization techniques, where the receiver applies a filter that inverts or compensates for the distortion introduced by the channel. Linear equalizers, decision feedback equalizers, and maximum likelihood sequence estimators are commonly used approaches.

The effect of distortion on information capacity is analyzed differently from that of noise. Additive white Gaussian noise reduces the signal-to-noise ratio and thus directly reduces the achievable information rate according to the Shannon–Hartley theorem. Distortion, by contrast, alters the signal in ways that may be partially predictable and thus partially correctable by a receiver with knowledge of the channel's distortion characteristics. When the channel's transfer function is known, equalization can undo the distortion and restore the signal, recovering some or all of the lost information. When the channel is unknown or time-varying, adaptive equalization techniques must track and compensate for the changing distortion profile.

In the broader cybernetic context, distortion represents a systematic failure of a communication channel to faithfully reproduce the intended signal. Where noise makes individual messages unreliable in a probabilistic sense, distortion can introduce systematic biases that degrade all messages in a predictable way. Understanding and compensating for distortion is therefore a central concern in the design of any communication system that aims to transmit information accurately across a physical medium.