What Is a Feedback Loop?

A feedback loop is a closed signal path in which a system's output is sampled and returned to an earlier stage of the same system, where it influences subsequent behavior. The structure appears in electrical circuits, mechanical systems, biological organisms, and software architectures, making the feedback loop one of the most fundamental organizing principles in systems engineering. Its key property is self-reference: the system's future inputs depend on its own past outputs, enabling behaviors such as self-correction, oscillation, amplification, and homeostasis that would be impossible in a purely open-loop arrangement.

The feedback loop is defined by its polarity. In a negative feedback loop, the returned signal opposes the current state, driving the system back toward a set point when it deviates. In a positive feedback loop, the returned signal reinforces the current state, accelerating change in whichever direction the system is already moving. Most regulatory systems in engineering rely on negative feedback for stability, while positive feedback is exploited selectively in latching circuits, oscillators, and bistable elements. The mathematical treatment of both polarities traces to Norbert Wiener's cybernetics work and the control theory developed simultaneously at MIT and Bell Laboratories in the 1940s and 1950s.

Loop Components and Signal Flow

A feedback loop consists of four functional elements: a plant (the system being controlled or observed), a sensor that measures the plant output, a comparator that forms the difference between the measured output and a reference input, and an actuator or controller that translates the error signal into a corrective action applied back to the plant. Signal flow diagrams and block diagrams represent these elements as interconnected functional blocks, with arrows indicating the direction of signal propagation. The loop is closed by the path from the plant output, through the sensor and comparator, through the controller, and back to the plant input. The IEEE Control Systems Society historical overview documents how the block diagram formalism was developed alongside feedback control theory to give engineers a systematic way to analyze and design these closed-loop signal paths.

Loop Gain and Dynamic Behavior

The behavior of a feedback loop over frequency is governed by its loop gain, the product of all gains around the closed path at a given frequency. High loop gain at low frequencies drives the output to match the reference accurately, suppressing disturbances and compensating for plant variations. As frequency rises, phase shifts accumulate in the plant and controller, and if the total phase reaches 180 degrees while the loop gain remains greater than unity, the loop becomes unstable and oscillates. The gain and phase margins, read from a Bode plot of the open-loop frequency response, quantify how much additional phase shift or gain the loop can tolerate before instability. These stability criteria, established by Hendrik Bode and Harry Nyquist, are the standard design tools for electronic, electromechanical, and process control feedback loops. The textbook Feedback Systems by Åström and Murray develops both the frequency-domain and state-space stability frameworks in a unified treatment.

Feedback Loops Beyond Engineering

The feedback loop concept extends beyond hardware systems into economics, ecology, and organizational behavior, where the same polarity distinctions apply. A price-supply feedback loop in a competitive market is negative: rising prices attract additional suppliers, which drives prices back down. Population-resource cycles in ecology create negative feedback that limits growth. The NIST Communications Technology Laboratory studies feedback loops in communication networks, where congestion control protocols implement negative feedback between traffic load and transmission rate to maintain network stability.

Applications

Feedback loops have applications in a wide range of disciplines, including:

• Closed-loop motor speed control and servo positioning
• Phase-locked loop frequency synthesis in communications hardware
• Thermostat-based building climate regulation
• Biological homeostasis including blood glucose and temperature regulation
• TCP congestion control in internet networking