What Is Thermal Noise?

Thermal noise is the random voltage or current fluctuation generated by the thermal agitation of charge carriers in any electrical conductor at a temperature above absolute zero. Because electrons in a conductor are in constant random motion driven by thermal energy, they produce a fluctuating current even in the absence of any applied voltage. The effect is ubiquitous: every resistor, every conductor, and every impedance element generates thermal noise, and the noise power depends only on temperature and bandwidth, not on the composition or geometry of the conductor. This makes thermal noise a fundamental physical limit that cannot be eliminated, only reduced by cooling or bandwidth restriction.

Thermal noise was first measured experimentally by John B. Johnson at Bell Laboratories in 1926 and subsequently explained theoretically by Harry Nyquist using statistical mechanics, which is why the phenomenon is also known as Johnson-Nyquist noise. Nyquist showed that the available noise power from a resistor of resistance R at temperature T over bandwidth Δf is proportional to kBTΔf, where kB is the Boltzmann constant. The spectral distribution is flat across all frequencies within the range where quantum mechanical effects can be ignored, earning thermal noise the designation of white noise.

Physical Origin and the Johnson-Nyquist Formula

The physical origin of thermal noise lies in the equipartition theorem of statistical mechanics: at thermal equilibrium, each degree of freedom in the system carries an average energy of kBT/2. Electrons in a conductor occupy such degrees of freedom, producing random current fluctuations whose spectral density is determined by the resistance and temperature. The mean-square open-circuit noise voltage across a resistor is given by vn² = 4RkBTΔf, a result derived by Nyquist in 1928. NIST research on Johnson noise thermometry exploits this relationship in reverse, using precision measurements of thermal noise voltage to determine absolute temperature with uncertainties below 1 millikelvin, because the noise depends only on fundamental constants and temperature.

Thermal Noise in Electronic Circuits

In electronic circuits, thermal noise sets a lower bound on signal detectability. Every resistive element in a signal path contributes noise that adds in quadrature with the signal, so the signal-to-noise ratio degrades as the signal propagates through lossy stages. As analyzed by Cadence's electronics resource on Johnson-Nyquist noise in circuits, both temperature and resistance value directly control the noise power: a resistor at 290 K (standard room temperature) generates a noise voltage spectral density of approximately 4 nV/√Hz per kiloohm of resistance. Low-noise amplifier design therefore prioritizes using low-resistance input stages and cooling the first amplification element to reduce thermal noise at the point where the signal is weakest.

Conductors themselves contribute thermal noise in proportion to their resistive component. Even an ideal inductor or capacitor, which are purely reactive, generates no thermal noise; the noise comes entirely from the resistive losses present in real-world components. This distinction guides the selection of low-loss components in sensitive receiver front ends, precision measurement instruments, and radio astronomy receivers.

Measurement and Mitigation

Thermal noise is characterized by its noise figure, noise temperature, or noise power spectral density, depending on the application. In RF and microwave systems, the noise figure of a receiver chain aggregates the thermal noise contributions of each stage according to the Friis formula. Practical mitigation strategies include cooling amplifier front ends with thermoelectric or cryogenic coolers, restricting measurement bandwidth to the minimum needed for the signal, and using impedance matching to maximize signal power relative to noise. Electronics Notes' treatment of RF thermal noise explains how the concept of noise temperature unifies these analysis approaches across different system types.

Applications

Thermal noise considerations arise in a range of fields, including:

• Radio astronomy receivers requiring cryogenic cooling to approach the noise floor set by physical temperature
• Precision resistive sensors and analog measurement instrumentation
• Low-noise amplifier design in cellular and satellite communication receivers
• Johnson noise thermometry for high-accuracy absolute temperature measurement
• Quantum computing, where thermal noise drives decoherence in superconducting qubits