In any wave-based sensing system—whether using sound, radio, or light—engineers face a persistent physical conflict: the better you want to see small details, the less deeply you can see into a material. This constraint, known as the resolution-penetration tradeoff, is driven by the frequency-dependent nature of signal attenuation.
The Physics of Frequency and Fading
The intuition behind this tradeoff lies in the relationship between wavelength and energy loss. High-frequency waves have short wavelengths, which allow them to resolve tiny features. However, as Hiroshi Kanda and Kageyoshi Katakura noted in their early work on acoustic microscopy, these same high-frequency waves suffer from significant absorption when traveling through a medium.
In a coupling medium like water or air, higher frequencies cause the particles of the medium to oscillate more rapidly, converting more of the wave's energy into heat. This means that while a 1 GHz acoustic wave can resolve sub-micron structures, it may only travel a few micrometers before the signal becomes too weak to detect. Conversely, lower frequencies possess longer wavelengths that "ignore" small-scale obstacles and experience less internal friction, allowing them to penetrate deep into a sample at the cost of blurring fine details.
Mapping the Subsurface in Materials Science
In Scanning Acoustic Microscopy (SAM), this tradeoff dictates the entire architecture of the instrument. To characterize extremely thin "damaged layers" in machined silicon, Isao Ishikawa demonstrated that researchers must use the highest possible frequencies to capture surface acoustic wave changes. However, when the goal shifts to detecting internal defects in thick polymers, Yukio Ogura et al. argue that one must tune the frequency downward.
Lower frequencies enable "deep-tissue" tomography of bulk mechanical flaws, while higher frequencies are reserved for micro-scale crystalline morphologies near the surface. As Hisayoshi Hashimoto et al. found, maintaining signal integrity at 1 GHz requires minimizing the distance the wave travels through the coupling fluid, often necessitating specialized lenses and micromachining to keep the path length short.
Sensing Through Obstacles and Soil
The same principle applies to electromagnetic waves used in radar and geophysical exploration. In search-and-rescue operations, Yang Zhang et al. utilized 400 MHz ultra-wideband (UWB) radar to detect human breathing through 30 cm brick walls. At this low frequency, the signal can penetrate dense obstacles, but the spatial resolution is coarse.
In contrast, Ruofeng Liu et al. showed that millimeter-wave (mmWave) radar provides the high precision needed to track subtle finger movements in Virtual Reality, yet these signals are easily blocked by a simple wooden door. The challenge for classification, as Tyler Rust et al. describe, is that lower-frequency bands like the S-band can "see" through vegetation and soil, but the resulting images are inherently grainy, requiring advanced deep learning to reconstruct target details.
In civil engineering and archaeology, the tradeoff is managed by selecting specific antenna frequencies for Ground Penetrating Radar (GPR). Duhee Park et al. found that:
- High frequencies resolve small voids (e.g., 10 cm) within the first meter of a structure.
- Mid-range frequencies balance resolution and depth for general detection at 2–3 meters.
- Low frequencies reach deeper geological interfaces but lose the ability to distinguish small features from surrounding soil variations.
Medical Diagnostics and the "Cheating" Problem
In clinical settings, the tradeoff forces a choice between surface precision and deep-organ visibility. Nianzhou Yu et al. identified 18 MHz as an "optimal balance" for measuring melanoma thickness; it is high enough to resolve the tumor base but low enough to penetrate the dense stratum corneum of the skin.
However, relying on a single frequency can lead to diagnostic errors. Gautam Rajendrakumar Gare et al. discovered that AI models trained on low-frequency lung ultrasound often "cheat" by focusing on superficial fat and skin layers because the probe's settings prioritize the depth needed to see the lungs, thereby sacrificing the resolution of the pleural line where early injuries actually occur.
Strategies for Overcoming the Limit
Researchers have developed several methods to mitigate these physical constraints:
- Interference and Phase Analysis: By using an "interference mode" that leverages internal reflections, Isao Ishikawa boosted depth resolution fivefold without increasing frequency. Similarly, Jinsong Liu et al. used laser cavity dynamics to increase LiDAR resolution without needing a wider frequency sweep.
- Multi-Modal Integration: In archaeological sites like Saqqara, Ahmed El-khteeb and Walid M. Mabrouk combined GPR (high resolution, low depth) with Seismic Refraction and Electrical Resistivity (low resolution, high depth) to create a comprehensive subsurface map.
- Algorithmic Compensation: Felix Heide et al. used sparse convolutional coding to recover depth information in scattering media, effectively using computation to fill the gaps left by physical signal attenuation.
Ultimately, the resolution-penetration tradeoff is not a problem to be "solved" but a physical boundary to be navigated. Whether the medium is a titanium alloy, a brick wall, or human tissue, the choice of frequency remains the primary lever for determining what a system can see—and how deep that vision goes.
Key Foundational Work
Evaluation of materials by acoustic microscope. (1985) — contributed 3 concept elements to this direction.
Measurement of a damaged layer thickness with SAM. (1989) — contributed 1 concept elements to this direction.