What Is Array Noise?

Array noise refers to the aggregate of unwanted signals and statistical disturbances that affect the performance of sensor arrays, including antenna arrays, acoustic arrays, seismic arrays, and phased-array radar systems. Unlike noise in a single-channel system, array noise encompasses the thermal and electronic noise generated within each individual element and its receiver chain, as well as correlated interference that can couple across elements through mutual coupling, shared electromagnetic environments, and spatially coherent noise sources. Understanding and managing array noise is central to the design of beamforming systems, where the goal is to separate signals of interest from spatially distributed interference while preserving signal-to-noise ratio (SNR) across the array output.

The characterization of array noise draws from both statistical signal processing and electromagnetic theory. Each array element captures signals from a range of directions simultaneously, which means that noise arriving from extended spatial regions (sky noise, ambient acoustic reverberation, or urban radio interference) has directional structure that interacts with the array's spatial filtering properties. The noise figure of an array system is therefore not a simple per-element quantity but depends on the array geometry, element coupling, beamforming weights, and the spatial distribution of the noise field.

Sources of Noise in Sensor Arrays

Array noise originates from several distinct physical mechanisms. Thermal noise (Johnson-Nyquist noise) is generated in each receiving element and its associated amplifier by the random thermal motion of charge carriers. For a low-noise amplifier (LNA) at the front end of each element, this contribution is characterized by the device's noise figure and its equivalent noise temperature. In antenna arrays, sky noise and ground thermal emission add to the antenna temperature, making the total input noise temperature dependent on the pointing direction of the array beam. Mutual coupling between array elements introduces a second category of noise: signals received by one element can leak into adjacent elements, effectively correlated noise whose level depends on element spacing and the impedance environment. In acoustic and seismic arrays, ambient noise from wind, waves, traffic, and machinery arrives as spatially correlated disturbances that the array may amplify if their direction coincides with the look direction.

Noise Figure and SNR in Array Systems

The noise figure of an active phased array system is defined as the ratio of the total output noise power to the output noise power that would result if only the input source noise were present. For arrays with per-element LNAs, the system noise figure is dominated by the first-stage amplifier when its gain is sufficiently high, because the LNA gain suppresses the contribution of downstream components per the Friis cascade formula. However, amplitude tapering applied to beamforming weights to reduce sidelobe levels reduces the taper efficiency and increases the effective noise figure by approximately the reciprocal of the taper efficiency, as shown in research on noise figure in active antenna arrays published in IEEE journals. The spatial distribution of noise also affects the output SNR: noise arriving from directions where the beam has high sidelobe response degrades the SNR more than the same noise power arriving from nulled directions. The URSI analysis of noise figure in beamforming systems formalizes the relationship between array weighting, taper efficiency, and effective noise temperature.

Noise Reduction Through Array Processing

Array processing exploits the spatial degrees of freedom available in a multi-element system to suppress noise and interference that arrive from directions different from the signal of interest. Conventional delay-and-sum beamforming provides a spatial gain (array gain) proportional to the number of elements for spatially uncorrelated noise, because summing N coherent signal contributions increases signal power by N squared while uncorrelated noise powers add as N. Adaptive beamforming algorithms such as the minimum variance distortionless response (MVDR) and linearly constrained minimum variance (LCMV) methods place spatial nulls in the direction of interference sources by optimizing the beamforming weight vector to minimize total output power subject to a constraint that the response toward the desired signal remains unity.

Applications

Array noise analysis and mitigation techniques have applications in many fields, including:

• Phased-array radar, where noise figure directly limits target detection range
• Radio telescope arrays and radio astronomy, where sky noise and antenna temperature must be carefully characterized
• Acoustic arrays for speech enhancement, noise cancellation, and room acoustics measurement
• Seismic arrays for earthquake monitoring and subsurface imaging
• Millimeter-wave and sub-terahertz imaging systems for security screening