What Are Sampled Data Systems?

Sampled data systems are dynamical systems in which signals are processed not continuously but at discrete points in time, typically at periodic intervals governed by a clock or sampler. They sit at the intersection of continuous-time physical processes and digital computation, arising whenever a real-world signal is measured and acted upon by a digital controller, processor, or recording system. The class encompasses digital control systems, digital signal processing chains, data acquisition networks, and any hybrid system in which continuous-time plant dynamics interact with a discrete-time controller.

The theoretical framework for sampled data systems emerged from work by Claude Shannon, Ragazzini, Jury, and others in the late 1940s and 1950s, drawing on the sampling theorem and on the z-transform as the discrete-time counterpart to the Laplace transform. These tools allowed engineers to analyze stability, frequency response, and transient behavior in systems where control actions occur only at sampling instants, even when the plant itself evolves continuously between samples.

Discrete-Time Representation and the Sampling Theorem

When a continuous-time signal is sampled at a rate fs, the resulting sequence of values represents the original signal exactly only if the signal contains no frequency components above fs/2, a limit known as the Nyquist frequency. Violating this condition causes aliasing, in which high-frequency content folds into the sampled band and cannot be distinguished from genuine low-frequency content. In practice, anti-aliasing filters precede the sampler to suppress out-of-band energy. Once sampled, a continuous-time system described by differential equations is converted to an equivalent discrete-time model described by difference equations, with state-transition matrices computed from the matrix exponential of the continuous-time system matrix times the sampling period. The IEEE Xplore publication on stability analysis of sampled-data control systems illustrates how continuous-time and discrete-time perspectives must be combined to obtain complete stability results.

The Z-Transform and Frequency Analysis

The z-transform is the primary mathematical tool for analyzing sampled data systems in the frequency domain, playing a role analogous to the Laplace transform in continuous-time systems. The z-transform maps a discrete-time sequence into a function of the complex variable z, allowing difference equations to be manipulated as algebraic expressions and transfer functions to be defined for discrete-time systems. The frequency response of a sampled data system is obtained by evaluating the z-domain transfer function on the unit circle, where z equals the complex exponential of the normalized frequency. The mapping between the s-plane (continuous-time) and the z-plane (discrete-time) takes the left-half s-plane to the interior of the unit circle in the z-plane, so a stable continuous-time system maps to a stable discrete-time system under this correspondence. Lecture materials on discrete-time systems and the z-transform from the University of Swansea provide a clear pedagogical treatment of these mappings.

Stability and Digital Control Design

Stability in sampled data systems requires that all poles of the closed-loop discrete-time transfer function lie strictly within the unit circle in the z-plane. Sampling introduces additional dynamics: a zero-order hold used to reconstruct a continuous-time control signal from discrete samples contributes phase lag that was absent in the continuous-time design, and choosing too low a sampling rate can destabilize a loop that is stable in continuous time. The analysis of systems with periodically time-varying sampling rates, including the effects of jitter and non-uniform sampling, is addressed in IEEE Transactions research on sampled-data systems with varying sampling rates. Modern digital control design methods, including pole placement, linear-quadratic-Gaussian (LQG) synthesis, and model predictive control, all operate within the sampled data framework.

Applications

Sampled data systems have applications in a range of fields, including:

• Digital feedback control in robotics, aerospace, and industrial automation
• Digital signal processing chains for audio, radar, and communications
• Data acquisition and logging systems for scientific instrumentation
• Networked control systems with variable latency and packet loss
• Medical device control loops such as insulin delivery and ventilator management
• Embedded real-time control in automotive and power electronics systems