Published: 2026-05-24 A phase shifter is a two-port microwave or radio frequency (RF) component that introduces a controllable change in the phase of an electromagnetic signal propagating through it, ideally without significantly altering the signal's amplitude. Formally, if a signal enters the device as V_in = A·cos(ωt), the output takes the form V_out = A'·cos(ωt − φ), where φ is the imposed phase shift and A' is the (ideally unchanged) output amplitude. The phase shift φ may be fixed or variable, and may be commanded electronically, magnetically, or mechanically depending on the device technology. Phase shifters are fundamental building blocks in phased array antenna systems, where they enable electronic beam steering by progressively shifting the phase of the signal fed to (or received from) each radiating element. They also appear in beamforming networks, phase discriminators, vector modulators, instrumentation, and various signal-processing architectures. In a phased array of N elements, the far-field beam direction θ₀ is related to the progressive inter-element phase shift Δφ by: $$\Delta\varphi = \frac{2\pi d \sin\theta_0}{\lambda}$$ where d is the element spacing and λ is the free-space wavelength. By changing Δφ in real time, the beam can be steered electronically without physical movement of the antenna — a capability critical to modern radar, electronic warfare, satellite communications, and 5G/6G wireless systems. Phase shifters are broadly classified as digital or analog based on how the phase is controlled. Digital phase shifters provide a discrete set of phase states. A phase shifter of n bits offers 2ⁿ equally spaced phase states spanning 360°. For example, a 5-bit phase shifter provides 32 states at 11.25° increments. Digital designs are implemented by cascading individual bit cells (180°, 90°, 45°, 22.5°, 11.25°, etc.), each of which is switched between two states. The command interface is a simple binary word, making digital phase shifters straightforward to integrate with digital beam-steering computers — a decisive advantage in modern active electronically scanned arrays (AESAs). PIN diode and MEMS phase shifters are inherently digital in their most common implementations. Analog (or continuously variable) phase shifters provide a smooth, continuous range of phase adjustment — in principle offering infinite resolution. Varactor-diode-based and ferrite-based designs are naturally suited to analog control, where a bias voltage or current sets the phase to any value within a defined range (commonly 0°–360° or more). Analog phase shifters eliminate quantization error entirely and are valuable in applications such as adaptive beamforming, phase-locked loops, and precision instrumentation. However, they require accurate, stable, and often linear analog control signals, and they can be more susceptible to drift, noise, and temperature-dependent variations than their digital counterparts. In practice, the choice between digital and analog is driven by the system architecture. Most large-scale phased arrays use digital phase shifters for their ease of control and integration, while analog designs find niche roles where continuous adjustment or ultra-fine resolution is required. Insertion loss — the attenuation of the signal as it passes through the phase shifter — is a critical performance metric, particularly in large arrays where many phase shifters are cascaded or where the cumulative loss in the transmit/receive chain directly impacts system sensitivity and effective radiated power. Contributors to insertion loss include: - Resistive losses in semiconductor switches (ON-state resistance of PIN diodes, finite isolation in the OFF state), MEMS contact resistance, or conductor losses in ferrite waveguide structures. - Mismatch losses due to imperfect return loss in one or more phase states. A phase shifter may be well matched in one state but poorly matched in another, causing state-dependent loss variations. - Dielectric losses in substrates and packaging materials. - Radiative or coupling losses in structures with discontinuities (e.g., switched-line bends). The typical insertion loss ranges by technology are: | Technology | Typical Insertion Loss | Key Loss Mechanism | |---|---|---| | Ferrite (waveguide) | 0.5–1.5 dB | Conductor/dielectric in waveguide | | MEMS | 0.5–2.0 dB | Contact resistance, substrate | | PIN diode | 2–5 dB (per bit) | Diode ON-resistance | | Varactor diode | 1–3 dB | Junction resistance, mismatch | In an AESA with N elements, a phase shifter with 1 dB of excess loss (relative to an ideal device) degrades the array's receive noise figure by approximately 1 dB and reduces the effective isotropic radiated power (EIRP) on transmit by roughly 1 dB — a direct and often unacceptable performance hit in radar sensitivity or communications link budget. This drives relentless engineering effort to minimize phase shifter loss, particularly at millimeter-wave frequencies where losses rise steeply. Designers also pay close attention to insertion loss variation across phase states (sometimes called amplitude taper or RMS amplitude error). If all phase states do not have identical insertion loss, the resulting amplitude errors across the aperture further degrade the radiation pattern — increasing sidelobe levels and reducing gain independently of the quantization effects described above.