Pulsed Polarimetry: Techniques & Applications

• Pulsed polarimetry is a technique where the state of polarization is captured in a single shot through time-resolved measurements, avoiding sequential sampling of static sources.
• It employs methods like single-shot Stokes polarimetry using digital micromirror devices and dual-comb Jones-matrix spectroscopy for rapid, high-fidelity polarization reconstruction.
• Its wide-ranging applications include astrophysical pulsar observations, distributed plasma diagnostics in fibers and THz regimes, and advanced searches in fundamental physics such as dark matter.

In the literature considered here, pulsed polarimetry denotes a family of polarimetric measurement strategies in which the state of polarization is resolved in a single shot, as a function of pulse phase, or across explicitly time-dependent signals rather than through sequential measurements that assume a static source. Its implementations span Stokes polarimetry of dynamic optical beams, Jones-matrix spectroscopic polarimetry based on polarization-multiplexed pulse sequences, phase-resolved and timing-resolved astrophysical measurements, distributed fiber and THz diagnostics in magnetized plasmas, and cavity-enhanced searches for weak oscillatory signals (Trippe, 2014, Bo-Zhao et al., 2019, Ingram, 2022).

1. Mathematical descriptors and measurement primitives

Polarization is a fundamental property of light, tied to the temporal behavior of the electric-field vector. For a monochromatic wave propagating in the $z$-direction, the field components may be written as

$$E_x(t) = E_{x0}\cos(\omega t - \phi_1),\qquad E_y(t) = E_{y0}\cos(\omega t - \phi_2),$$

with amplitudes, phases, and frequency defining the waveform (Trippe, 2014).

For observational work, the standard macroscopic description is the Stokes vector. The Stokes parameters are

$$I = \langle E_x^2 \rangle + \langle E_y^2 \rangle,\quad Q = \langle E_x^2 \rangle - \langle E_y^2 \rangle,\quad U = 2\langle E_x E_y \cos \delta \rangle,\quad V = 2\langle E_x E_y \sin \delta \rangle,$$

where $\delta = \phi_2 - \phi_1$ and the averaging time is much larger than the wave period. Common derived quantities are the total, linear, and circular degrees of polarization,

$$m_P = \frac{I_P}{I},\qquad m_L = \frac{\sqrt{Q^2 + U^2}}{I},\qquad m_C = \frac{V}{I},$$

and the polarization angle

$$\chi = \frac{1}{2}\tan^{-1}\!\left(\frac{U}{Q}\right)$$

(Trippe, 2014).

In time-resolved and phase-resolved work, these quantities are evaluated in successive time bins or pulse-phase intervals. In pulsar measurements, for example, the signal is folded on the pulse period and the Stokes parameters are computed for each phase bin, enabling direct tracking of $Q(\phi)$, $U(\phi)$, $V(\phi)$, the linear-polarization fraction, the circular-polarization fraction, and the position-angle swing (Trippe, 2014).

A complementary formalism is Jones calculus. In Jones-matrix spectroscopic polarimetry, the polarization state is represented by a Jones vector,

$$J = \begin{bmatrix} E_{x0} e^{i\phi_x}\ E_{y0} e^{i\phi_y} \end{bmatrix},$$

or in normalized form through an amplitude ratio $$E_x(t) = E_{x0}\cos(\omega t - \phi_1),\qquad E_y(t) = E_{y0}\cos(\omega t - \phi_2),$$0 and phase difference $$E_x(t) = E_{x0}\cos(\omega t - \phi_1),\qquad E_y(t) = E_{y0}\cos(\omega t - \phi_2),$$1, while a sample is represented by a $$E_x(t) = E_{x0}\cos(\omega t - \phi_1),\qquad E_y(t) = E_{y0}\cos(\omega t - \phi_2),$$2 Jones matrix $$E_x(t) = E_{x0}\cos(\omega t - \phi_1),\qquad E_y(t) = E_{y0}\cos(\omega t - \phi_2),$$3 with $$E_x(t) = E_{x0}\cos(\omega t - \phi_1),\qquad E_y(t) = E_{y0}\cos(\omega t - \phi_2),$$4. This permits direct reconstruction of the full complex polarization response as a function of wavelength rather than only the Stokes observables at a single setting (Koresawa et al., 2023).

2. Single-shot Stokes polarimetry for dynamic and pulsed beams

Stokes polarimetry reconstructs the state of polarization from four intensity measurements. In the conventional implementation, four intensity projections are acquired sequentially, typically corresponding to total intensity, horizontal, diagonal, and right-circular projections. Because all intensities must be measured on the same beam, sequential acquisition limits the method to beams with static state of polarization (Bo-Zhao et al., 2019).

A single-shot alternative was demonstrated using a polarization-insensitive Digital Micromirror Device. The key element is a multiplexed digital hologram that superposes four diffraction gratings at distinct spatial frequencies. When the input beam passes through this hologram, it is split into four identical copies, each directed along a separate path. Each path contains polarization optics configured to analyze one of the required intensity projections, $$E_x(t) = E_{x0}\cos(\omega t - \phi_1),\qquad E_y(t) = E_{y0}\cos(\omega t - \phi_2),$$5, $$E_x(t) = E_{x0}\cos(\omega t - \phi_1),\qquad E_y(t) = E_{y0}\cos(\omega t - \phi_2),$$6, $$E_x(t) = E_{x0}\cos(\omega t - \phi_1),\qquad E_y(t) = E_{y0}\cos(\omega t - \phi_2),$$7, and $$E_x(t) = E_{x0}\cos(\omega t - \phi_1),\qquad E_y(t) = E_{y0}\cos(\omega t - \phi_2),$$8, and the four beams are directed onto distinct, non-overlapping regions of a CCD detector for simultaneous acquisition (Bo-Zhao et al., 2019).

The corresponding Stokes reconstruction is

$$E_x(t) = E_{x0}\cos(\omega t - \phi_1),\qquad E_y(t) = E_{y0}\cos(\omega t - \phi_2),$$9

Because the splitting is performed in the wavefront domain rather than by amplitude division, each copy retains the coherence and polarization characteristics of the original beam, avoiding beam mismatch and energy imbalance associated with beam-splitter architectures (Bo-Zhao et al., 2019).

The DMD grants independent and rapid ($$I = \langle E_x^2 \rangle + \langle E_y^2 \rangle,\quad Q = \langle E_x^2 \rangle - \langle E_y^2 \rangle,\quad U = 2\langle E_x E_y \cos \delta \rangle,\quad V = 2\langle E_x E_y \sin \delta \rangle,$$0 kHz) manipulation of each beam, while the demonstrated real-time measurement speed was up to $$I = \langle E_x^2 \rangle + \langle E_y^2 \rangle,\quad Q = \langle E_x^2 \rangle - \langle E_y^2 \rangle,\quad U = 2\langle E_x E_y \cos \delta \rangle,\quad V = 2\langle E_x E_y \sin \delta \rangle,$$1 Hz, limited by the CCD camera. The reported reconstructions had high fidelity for both scalar and vector polarization states, and the single-shot acquisition effectively “freezes” the state of polarization at one instant. The method was explicitly identified as relevant to pulsed polarimetry, including characterizing the polarization of single ultrafast pulses, and to structured-light applications requiring real-time polarimetry (Bo-Zhao et al., 2019).

3. Spectroscopic multiplexing and cavity ring-down variants

Pulsed polarimetry extends naturally into spectroscopic and cavity-enhanced regimes. In Jones-matrix dual-comb spectroscopic polarimetry, dual-comb spectroscopy is combined with polarization control pulse sequences (PCPS) to recover the wavelength-dependent Jones matrix of a sample without mechanical polarization modulation. A femtosecond optical frequency comb pulse is split into two parts; one remains $$I = \langle E_x^2 \rangle + \langle E_y^2 \rangle,\quad Q = \langle E_x^2 \rangle - \langle E_y^2 \rangle,\quad U = 2\langle E_x E_y \cos \delta \rangle,\quad V = 2\langle E_x E_y \sin \delta \rangle,$$2-polarized, the other is set to $$I = \langle E_x^2 \rangle + \langle E_y^2 \rangle,\quad Q = \langle E_x^2 \rangle - \langle E_y^2 \rangle,\quad U = 2\langle E_x E_y \cos \delta \rangle,\quad V = 2\langle E_x E_y \sin \delta \rangle,$$3-polarization and delayed in time. After recombination and rotation, the result is a polarization-multiplexed, time-multiplexed sequence delivered to the sample (Koresawa et al., 2023).

Detection uses two phase-locked optical frequency combs with slightly different repetition rates. The sample comb generates the PCPS, the local comb provides interferometric detection, and the output interferogram is separated into $$I = \langle E_x^2 \rangle + \langle E_y^2 \rangle,\quad Q = \langle E_x^2 \rangle - \langle E_y^2 \rangle,\quad U = 2\langle E_x E_y \cos \delta \rangle,\quad V = 2\langle E_x E_y \sin \delta \rangle,$$4- and $$I = \langle E_x^2 \rangle + \langle E_y^2 \rangle,\quad Q = \langle E_x^2 \rangle - \langle E_y^2 \rangle,\quad U = 2\langle E_x E_y \cos \delta \rangle,\quad V = 2\langle E_x E_y \sin \delta \rangle,$$5-polarization components at a PBS and measured by photodiodes. Fourier analysis of the temporally separated interferograms yields amplitude and phase spectra for both polarizations and for each pulse in the sequence, from which all four Jones-matrix elements are reconstructed at each optical frequency (Koresawa et al., 2023).

This architecture eliminates mechanical polarization modulation, provides fast (sub-ms/shot), broadband (multi-THz) acquisition, and enables full polarization multiplexing with absolute phase measurement. It therefore occupies a distinct position within pulsed polarimetry: the time structure of the incident field is itself the multiplexing mechanism for polarimetric inversion (Koresawa et al., 2023).

A different branch is cavity ring-down polarimetry. In pulsed-based CRDP, short laser pulses are stored in a high-finesse cavity, and intracavity birefringence or dichroism produces time-resolved polarization modulations in the transmitted decay. A general fitting model is

$$I = \langle E_x^2 \rangle + \langle E_y^2 \rangle,\quad Q = \langle E_x^2 \rangle - \langle E_y^2 \rangle,\quad U = 2\langle E_x E_y \cos \delta \rangle,\quad V = 2\langle E_x E_y \sin \delta \rangle,$$6

where $$I = \langle E_x^2 \rangle + \langle E_y^2 \rangle,\quad Q = \langle E_x^2 \rangle - \langle E_y^2 \rangle,\quad U = 2\langle E_x E_y \cos \delta \rangle,\quad V = 2\langle E_x E_y \sin \delta \rangle,$$7 is the polarization-rotation or beat frequency. For circular birefringence induced by a Faraday bias, the mode splitting obeys

$$I = \langle E_x^2 \rangle + \langle E_y^2 \rangle,\quad Q = \langle E_x^2 \rangle - \langle E_y^2 \rangle,\quad U = 2\langle E_x E_y \cos \delta \rangle,\quad V = 2\langle E_x E_y \sin \delta \rangle,$$8

(2002.04538).

Continuous-Wave Cavity Ring-Down Polarimetry extends this lineage by employing rapidly-pulsed single-frequency CW laser sources. The CW approach was described as extending current cavity-based spectropolarimetric techniques while allowing gains in spectral resolution, signal intensity and data acquisition rate compared to traditional pulsed-based cavity ring-down polarimetry. In the demonstrated modality, the observable is the polarization-rotation frequency during a ring-down event generated by large intracavity polarization anisotropies. The analysis showed that the frequency-based methodology alleviates the requirement for high finesse cavities and renders the method particularly suitable for robust portable polarimetric instrumentations (2002.04538).

4. Pulse-phase polarimetry in pulsars and accreting neutron stars

In astrophysics, pulsed polarimetry is most directly associated with periodic sources such as pulsars. At radio wavelengths, the field amplitudes and phases can be recorded directly, and the Stokes parameters are built from cross- and auto-correlations of orthogonally oriented dipoles. More generally, the observational procedure is to fold the signal on the pulse period, compute the Stokes parameters for each phase bin, and track the phase evolution of the linear-polarization angle, the linear-polarization fraction, and the circular-polarization fraction (Trippe, 2014).

Phase- and frequency-resolved polarimetry of millisecond pulsars has recently been formalized through a frequency-dependent template framework for all four Stokes parameters, $$I = \langle E_x^2 \rangle + \langle E_y^2 \rangle,\quad Q = \langle E_x^2 \rangle - \langle E_y^2 \rangle,\quad U = 2\langle E_x E_y \cos \delta \rangle,\quad V = 2\langle E_x E_y \sin \delta \rangle,$$9, $\delta = \phi_2 - \phi_1$0, $\delta = \phi_2 - \phi_1$1, and $\delta = \phi_2 - \phi_1$2. Using wideband ($\delta = \phi_2 - \phi_1$3–$\delta = \phi_2 - \phi_1$4 GHz) observations with the Parkes Murriyang UWL receiver, the method constructs two-dimensional portraits in phase and frequency and applies principal component analysis to each Stokes parameter separately. These templates can be used in matrix template matching, and the reported timing measurement uncertainties were reduced up to $\delta = \phi_2 - \phi_1$5–$\delta = \phi_2 - \phi_1$6 for Parkes Pulsar Timing Array millisecond pulsars (Curyło et al., 10 Dec 2025).

High-energy implementations require more elaborate response modeling. For the Crab pulsar with POLAR, the analysis used phase subdivision, off-pulse background subtraction, and a joint-fitting method in which modulation curves from multiple datasets were numerically de-rotated by an angle $\delta = \phi_2 - \phi_1$7 to a common reference frame before stacking and fitting. Detector response and efficiency for each incident direction were simulated using Geant4, and the phase-resolved polarization results for the pulsed emission were $\delta = \phi_2 - \phi_1$8 and $\delta = \phi_2 - \phi_1$9 for the averaged pulsed signal, $$m_P = \frac{I_P}{I},\qquad m_L = \frac{\sqrt{Q^2 + U^2}}{I},\qquad m_C = \frac{V}{I},$$0 and $$m_P = \frac{I_P}{I},\qquad m_L = \frac{\sqrt{Q^2 + U^2}}{I},\qquad m_C = \frac{V}{I},$$1 for P1, and $$m_P = \frac{I_P}{I},\qquad m_L = \frac{\sqrt{Q^2 + U^2}}{I},\qquad m_C = \frac{V}{I},$$2 and $$m_P = \frac{I_P}{I},\qquad m_L = \frac{\sqrt{Q^2 + U^2}}{I},\qquad m_C = \frac{V}{I},$$3 for P2 (Li et al., 2021).

IXPE measurements of the Crab in the $$m_P = \frac{I_P}{I},\qquad m_L = \frac{\sqrt{Q^2 + U^2}}{I},\qquad m_C = \frac{V}{I},$$4–$$m_P = \frac{I_P}{I},\qquad m_L = \frac{\sqrt{Q^2 + U^2}}{I},\qquad m_C = \frac{V}{I},$$5 keV band yielded a markedly different soft X-ray picture. The total pulsar pulsed emission was reported as unpolarised, while significant polarisation up to $$m_P = \frac{I_P}{I},\qquad m_L = \frac{\sqrt{Q^2 + U^2}}{I},\qquad m_C = \frac{V}{I},$$6 was detected only in the core of the main peak. For the phase bin $$m_P = \frac{I_P}{I},\qquad m_L = \frac{\sqrt{Q^2 + U^2}}{I},\qquad m_C = \frac{V}{I},$$7–$$m_P = \frac{I_P}{I},\qquad m_L = \frac{\sqrt{Q^2 + U^2}}{I},\qquad m_C = \frac{V}{I},$$8, the reported OP-subtracted values were $$m_P = \frac{I_P}{I},\qquad m_L = \frac{\sqrt{Q^2 + U^2}}{I},\qquad m_C = \frac{V}{I},$$9, $$\chi = \frac{1}{2}\tan^{-1}\!\left(\frac{U}{Q}\right)$$0, a polarisation degree of $$\chi = \frac{1}{2}\tan^{-1}\!\left(\frac{U}{Q}\right)$$1, and a polarisation angle of $$\chi = \frac{1}{2}\tan^{-1}\!\left(\frac{U}{Q}\right)$$2 (Bucciantini et al., 2022).

A further case is the accreting pulsar 4U 1626–67. IXPE measurements in the $$\chi = \frac{1}{2}\tan^{-1}\!\left(\frac{U}{Q}\right)$$3–$$\chi = \frac{1}{2}\tan^{-1}\!\left(\frac{U}{Q}\right)$$4 keV band gave an upper limit on the pulse-averaged linear polarization of $$\chi = \frac{1}{2}\tan^{-1}\!\left(\frac{U}{Q}\right)$$5 at $$\chi = \frac{1}{2}\tan^{-1}\!\left(\frac{U}{Q}\right)$$6 confidence. No significant polarized flux was detected in pulse-phase intervals when the bandpass was subdivided by energy, but spectropolarimetric modeling over the full bandpass in pulse phase intervals yielded a marginal detection of polarization of the power-law spectral component at the $$\chi = \frac{1}{2}\tan^{-1}\!\left(\frac{U}{Q}\right)$$7 level at $$\chi = \frac{1}{2}\tan^{-1}\!\left(\frac{U}{Q}\right)$$8 confidence (Marshall et al., 2022).

Source	Measurement	Result
Crab pulsar with POLAR (Li et al., 2021)	Phase-resolved pulsed emission	Averaged: $$\chi = \frac{1}{2}\tan^{-1}\!\left(\frac{U}{Q}\right)$$9, $Q(\phi)$0; P1: $Q(\phi)$1, $Q(\phi)$2; P2: $Q(\phi)$3, $Q(\phi)$4
Crab pulsar with IXPE (Bucciantini et al., 2022)	Soft X-ray phase-resolved pulsed emission	Total pulsed emission unpolarised; P1 core: $Q(\phi)$5, $Q(\phi)$6
4U 1626–67 with IXPE (Marshall et al., 2022)	Pulse-averaged and pulse-phase spectropolarimetry	$Q(\phi)$7 at $Q(\phi)$8 confidence; power-law component $Q(\phi)$9 at $U(\phi)$0 confidence
PPTA millisecond pulsars (Curyło et al., 10 Dec 2025)	Polarimetric template timing	Timing measurement uncertainties reduced up to $U(\phi)$1–$U(\phi)$2

These measurements underscore an important observational point: phase-resolved polarimetry can reveal sharp, localized polarized components even when the pulse-averaged signal is consistent with zero. In the Crab IXPE study, the near-zero average pulsed polarisation was reported as contrasting with most existing models, while in 4U 1626–67 the low net phase-averaged polarization was discussed as consistent with some geometries because swings in the electric-vector position angle can average out the net signal (Bucciantini et al., 2022, Marshall et al., 2022).

5. X-ray polarimetry-timing beyond coherent pulsations

Pulsed polarimetry in X-rays is not limited to strictly periodic sources. X-ray polarimetry-timing was defined as the characterization of rapid variability in the X-ray polarization degree and angle, with the goal of recovering causal information about compact-object environments. Theoretical expectations include pulse-phase polarization swings from rotating neutron-star hotspots, QED vacuum birefringence in magnetars, lagged polarization changes associated with propagating accretion-rate fluctuations, polarization reverberation from reflection, and polarization-angle oscillations synchronized with low-frequency quasi-periodic oscillations in Lense–Thirring precession models (Ingram, 2022).

In this regime, the most direct observables are the Stokes parameters estimated from per-photon modulation angles: $U(\phi)$3 with

$U(\phi)$4

For strictly periodic signals, pulse-phase stacking is the standard method; the achievable number of phase bins is set by source strength, polarization degree, background, and total exposure. The usual detection threshold is summarized by the minimum detectable polarization,

$U(\phi)$5

(Ingram, 2022).

For stochastic variability, direct time binning of $U(\phi)$6 and $U(\phi)$7 is generally impractical because of Poisson noise. A Fourier alternative was introduced specifically for fast stochastic X-ray polarimetry-timing. Instead of forming $U(\phi)$8 and $U(\phi)$9 directly in short bins, photons are sorted into light curves according to measured modulation angle $V(\phi)$0, and each angle-bin light curve is cross-correlated in the Fourier domain with a reference time series: $V(\phi)$1 The modulation function for the counts in each angle bin is

$V(\phi)$2

This cross-spectral approach was described as enabling statistically robust detection of stochastic polarisation variability for arbitrarily short variability timescales and as analogous to spectral-timing methods already used in X-ray astronomy (Ingram et al., 2017).

The distinction from classical phase-folding is fundamental. Phase-folding works for coherent pulsations; it fails for quasi-periodic oscillations whose phase drifts unpredictably. The Fourier method was proposed as a route to detect the quasi-periodic swings in polarisation angle predicted by Lense–Thirring precession of the inner accretion flow, contingent on a mean polarisation degree greater than $V(\phi)$3–$V(\phi)$4 (Ingram et al., 2017). This establishes a broader definition of pulsed polarimetry in which the “pulse” can be a periodic phase structure, a statistically reconstructed oscillation, or a transient modulation recovered in the frequency domain.

6. Distributed pulsed polarimetry in fibers and fusion plasmas

In plasma diagnostics, pulsed polarimetry has acquired a strongly spatial meaning. Fiber optic pulsed polarimetry was described as a LIDAR-like fiber sensing technique that uses a backscatter enhanced single mode backscatter-tailored optical fiber to measure distributed magnetic fields. The fiber contains a series of wavelength resonant reflection fiber Bragg gratings written uniformly along its length, so that a short polarized pulse launched into the fiber produces time-delayed reflections at known positions. The local state of polarization of each reflected pulse is modified by the Faraday effect, whose strength is set by the fiber’s Verdet constant (Smith, 27 Oct 2025).

The accumulated Faraday rotation is

$V(\phi)$5

and the local field can be inferred from the incremental rotation between adjacent reflectors. The reported performance for the backscatter-tailored optical fiber was high repetition rates of $V(\phi)$6 MHz, spatial resolution of $V(\phi)$7–$V(\phi)$8 cm, magnetic-field accuracy $V(\phi)$9, and temporal response in the ns range (Smith, 27 Oct 2025).

This implementation raises a characteristic inversion problem: multipathing in the fiber produces third-order reflections that contaminate the LIDAR signal. The contamination was analyzed algorithmically for uniform and flat reflection designs, and the combinatorics of higher-order paths were related to Narayana numbers and the Catalan series. Thus, pulsed polarimetry in this setting is not only a measurement technique but also a problem in path-ensemble modeling and reflector design (Smith, 27 Oct 2025).

THz pulsed polarimetry for magnetic fusion devices extends the same logic into free-space plasma probing. With intense pulsed THz sources generated by optical rectification in crystalline lithium niobate, non-perturbative measurements of the internal local magnetic field become possible. Pulsed Polarimetry was described as resolving the sightline magnetic field and density, analogous to RADAR measurements of the spatial distribution of precipitation. The relevant magneto-optic effects are Faraday rotation, which probes the parallel field, and the Cotton-Mouton effect, which produces ellipticity from the perpendicular field (Smith, 19 Sep 2025).

The line-integrated Faraday response is

$$J = \begin{bmatrix} E_{x0} e^{i\phi_x}\ E_{y0} e^{i\phi_y} \end{bmatrix},$$0

and, with added spectroscopy and a heterodyne receiver, collective Thomson backscatter can be used to recover the local ion spectral density function, the bulk ion temperature, and confined fast ion velocity distributions. The method was presented as diagnostically agile because the sightline can be steered without loss of alignment, permitting temporal and spatial real-time feedback across the poloidal plane (Smith, 19 Sep 2025).

7. Fundamental-physics searches and conceptual scope

Pulsed polarimetry also appears in searches for weak coherent signals beyond standard laboratory and astrophysical settings. In proposed dark-matter polarimetry experiments, a thick birefringent solid inside a Fabry–Pérot cavity is used to search for scalar field dark matter through oscillatory variations in the solid’s thickness and refractive index, which modulate the relative phase shift

$$J = \begin{bmatrix} E_{x0} e^{i\phi_x}\ E_{y0} e^{i\phi_y} \end{bmatrix},$$1

A second configuration places two quarter-wave plates inside the cavity instead of the birefringent solid and targets axion-like particles through an oscillatory rotation of the polarization plane (Ejlli et al., 2022).

Both schemes use a heterodyne readout based on a photoelastic modulator and demodulation at the PEM frequency. The cavity acts as a band-pass filter and enhances the effective interaction length by the buildup factor $$J = \begin{bmatrix} E_{x0} e^{i\phi_x}\ E_{y0} e^{i\phi_y} \end{bmatrix},$$2. The signals are oscillatory at the Compton frequency of the dark-matter field and were described as narrow-band monochromatic features in the polarimeter output. The same work noted that this is not strictly “pulsed” in a classical sense, but that the detection is analogous to pulsed polarimetry because a time-dependent, oscillatory signal is sought with strong rejection of DC and slow drifts (Ejlli et al., 2022).

A further extension is cross-correlation of twin polarimeters. Because the dark-matter signal is expected to be coherent over long spatial scales, two nearby polarimeters should observe the same signal but uncorrelated local noise. Cross-correlation was therefore proposed as a route to integrate beyond a single dark-matter coherence time and improve sensitivity by suppressing uncorrelated noise (Ejlli et al., 2022).

This range of examples suggests that pulsed polarimetry is best understood not as a single instrumental class but as a common measurement principle: polarization information is encoded in temporal structure and then recovered through simultaneous acquisition, phase folding, Fourier-domain estimation, heterodyne demodulation, or time-of-flight inversion. A plausible implication is that future developments will continue to favor architectures that replace mechanical modulation with multiplexing, model-based calibration, and direct recovery of all available Stokes or Jones information across time, frequency, and space.