In the transition toward 6G and Terahertz (THz) communications, researchers are grappling with a fundamental physical mismatch. To achieve massive data rates, we need two things: extremely wide bandwidths and extremely large antenna arrays. However, when these two are combined using traditional hardware, the wireless beam physically "unravels."
This phenomenon, known as the beam split or beam squint effect, occurs because traditional phase shifters are frequency-independent. As Zhaolin Wang, Xidong Mu, and Yuanwei Liu explain in Near-Field Communications: What Will Be Different?, the phase shift required to steer a beam in a specific direction depends on the signal's frequency. When a single phase shift is applied across a wide bandwidth, the different frequency components (subcarriers) end up pointing in different directions, resembling a "THz rainbow."
The Physics of the Beam Split
To understand why beams split, consider a phased array trying to steer a signal toward a user. To point the beam, the array must create a specific phase gradient across its elements. The catch is that a "phase shift" is actually a proxy for a "time delay." For a signal to arrive at a distant point simultaneously from all antennas, the signals from the further antennas must be sent slightly earlier.
Traditional hardware uses phase shifters, which provide a constant phase offset regardless of frequency. But because phase is a function of wavelength, a fixed phase shift represents different time delays for different frequencies. In wideband systems, the high-frequency edge of the signal and the low-frequency edge "see" different effective steering angles. As Linglong Dai and Jingbo Tan note in Delay-Phase Precoding for Wideband THz Massive MIMO, this causes massive array gain loss, as only a small fraction of the signal's energy actually hits the target.
True Time Delay as the Solution
The solution is to replace or augment phase shifters with True-Time Delayers (TTDs). A TTD introduces an actual temporal delay ($\tau$) rather than a phase offset. Because the phase shift $\phi$ created by a time delay is linearly proportional to frequency ($\phi = 2\pi f \tau$), a TTD automatically adjusts the phase for every subcarrier in the signal.
By integrating a small number of TTDs into the beamforming architecture—a method Linglong Dai and colleagues call Delay-Phase Precoding (DPP)—the system can compensate for spatial dispersion. The TTDs handle the coarse, frequency-dependent timing required to align the "rainbow," while traditional phase shifters provide the fine-grained control needed for high-resolution steering.
From Beamsteering to Beamfocusing
As antenna arrays grow larger (Extremely Large-Scale Antenna Arrays, or ELAAs), the "near-field" region expands. In the far-field, we treat waves as flat planes and focus on beamsteering (direction). In the near-field, waves are spherical, allowing for beamfocusing (a specific point in 3D space).
This shift transforms the wireless channel. In Near-Field Communications: A Tutorial Review, Zhaolin Wang and Jiaqi Xu describe how beamfocusing acts like a "spotlight" rather than a "flashlight." However, the beam split effect is even more destructive here: without TTDs, different frequencies focus at different distances and coordinates.
Research into TTD configurations—whether parallel, serial, or hybrid—aims to minimize hardware complexity. As explored in TTD Configurations for Near-Field Beamforming: Parallel, Serial, or Hybrid?, serial TTD architectures can significantly reduce the maximum delay required, making it feasible to implement these systems on-chip for 6G devices.
Embracing the Squint
While most research focuses on mitigating beam squint, some researchers are turning this "bug" into a feature. In Embracing Beam-Squint Effects for Wideband LEO Satellite Communications, Juha Park and Wonjae Shin propose using the rainbow effect to provide 3D coverage for satellites. By intentionally spreading frequencies across a footprint, a satellite can serve an entire area simultaneously without the need for complex beam-hopping.
Similarly, in Integrated Sensing and Communications (ISAC), Hongliang Luo and Feifei Gao demonstrate that controllable beam squint can be used for environmental mapping. By observing how different frequencies reflect off objects, the system can perform spatial sensing and high-speed communication using the same wideband waveform.
Summary of the Direction
The core of this research area is the realization that for wideband, high-frequency arrays, time is a more robust control dimension than phase. By using True-Time Delay architectures, researchers are:
- Correcting the beam split effect to maintain high array gain.
- Enabling high-precision beamfocusing in the near-field.
- Exploiting frequency-dependent dispersion for novel sensing and satellite coverage applications.