Electron bunch profile reconstruction based on phase-constrained iterative algorithm

Abstract

The phase retrieval problem occurs in a number of areas in physics and is the subject of continuing investigation. The one-dimensional case, e.g., the reconstruction of the temporal profile of a charged particle bunch, is particularly challenging and important for particle accelerators. Accurate knowledge of the longitudinal (time) profile of the bunch is important in the context of linear colliders, wakefield accelerators and for the next generation of light sources, including x-ray SASE FELs. Frequently applied methods, e.g., minimal phase retrieval or other iterative algorithms, are reliable if the Blaschke phase contribution is negligible. This, however, is neither known a priori nor can it be assumed to apply to an arbitrary bunch profile. We present a novel approach which gives reproducible, most-probable and stable reconstructions for bunch profiles (both artificial and experimental) that would otherwise remain unresolved by the existing techniques.

Diagnostics for plasma-based electron accelerators

“…III.D.1). Finally, researchers transformed transition radiation (TR), a beam diagnostic of long standing for conventional accelerators, into an effective diagnostic of few-fs LWFA bunches by developing multi-octave, high-dynamic-range, single-shot spectrometers (Heigoldt et al, 2015;Lundh et al, 2011), and new algorithms for recovering bunch temporal profiles from TR spectral intensity measurements (Bakkali Taheri et al, 2016). TR spectroscopy complements timedomain MO and TD methods by characterizing τ b outside the accelerator without probe pulses (Sec.…”

“…Hence the inverse problem of reconstructing ρ (z) from CTR spectra is ill-posed, analogous to determining the structure of a molecule from a diffraction pattern generated by coherent x-rays. Iterative algorithms -see (Bajlekov et al, 2013;Bakkali Taheri et al, 2016;Heigoldt et al, 2015) and references therein -are used to reconstruct ρ (z) -subject to physical constraints -from spectral intensity measurements alone. Briefly, most reconstruction algorithms are variants of the following approach: one starts with an initial guess ρ (0) (z), calculates its complex Fourier transform |G(k)| exp[iψ(k)], then replaces the amplitude |G(k)| with the experimental |F (k)|.…”

Plasma-based particle sources

“…Specialized spectrometers record COTR spectral intensity from ultraviolet to long-wave-infrared [45], a range that spans wavelengths close to the dimensions of the e-bunch itself, in which the key structural information resides. Since intensity measurements lose phase information, iterative algorithms [46] guided by physical constraints and independent measurements reconstruct the longitudinal profile from recorded spectra. Recently high-resolution imaging of COTR from a generating foil to a multi-spectral detector array has augmented longitudinal with transverse structural information, enabling the first 3D spatial reconstructions of LWFA e-bunches [47].…”

Beam impedance minimization for accelerator beamline insertion devices

“…−iω s ′ t dt, and in the case of a train of Gaussian bunches, the spectrum will have a set of peaks, 17 which will appear at ω s ′ , i.e., a quasi-discrete spectrum. The peaks will have the amplitudes defined by the envelope shown [Fig.…”