---
title: Optical Power Beaming
url: https://www.emergentmind.com/topics/optical-power-beaming-opb
type: topic
---

# Optical Power Beaming

Searching arXiv for recent papers on optical power beaming, resonant beam communication, and lunar OPB to ground the article in current literature.
Optical power beaming (OPB) is the transfer of energy by directing optical radiation—typically laser or resonant optical beams—to a remote receiver that converts the incident optical power into electrical power, commonly through photovoltaic or photonic power-conversion devices. In current research, OPB is closely coupled to free-space optics, simultaneous wireless information and power transfer (SWIPT), power-over-fiber, and cislunar or lunar-surface energy infrastructure. Recent work spans meter-scale eye-safe indoor links using multi-segment GaAs photonic power converters (PPCs), resonant-beam systems with intrinsic self-alignment and fail-safe interruption, watt-level fiber-mediated delivery using hollow-core photonic crystal fibers, and long-range lunar architectures in which orbital constellations, EMLP-2 platforms, or phased-array transmitters support persistent power delivery under stringent diffraction, pointing, and environmental constraints [2510.06205] [1809.11091] [2207.02324] [2508.10855].

## 1. Definition and system archetypes

OPB denotes wireless power transmission in which the carrier is optical rather than microwave or wired current. Across the cited literature, the canonical OPB chain consists of a laser or resonant optical source, beam-shaping or steering optics, a propagation channel, an optical receiver aperture or direct-conversion surface, and a photovoltaic or photonic converter that produces usable electrical output [1809.11091] [2402.16320]. In some systems, the same optical beam also carries communication data, yielding SWIPT functionality through direct modulation or OFDM-based signaling [2510.06205] [1809.11091].

Several architectural families recur in the literature. Free-space direct laser beaming uses a conventional transmitter and remote photovoltaic receiver, often modeled with Gaussian-beam propagation and geometric capture terms [2405.00034] [2412.14083]. Resonant-beam communication (RBCom) places the transmitter and receiver inside a shared free-space optical resonator closed by retroreflectors, so that mobility and self-alignment are built into the cavity physics; a beam splitter extracts a fraction of the intra-cavity field for a photovoltaic receiver and data front-end [1809.11091] [2108.00004]. Power-over-fiber variants replace the free-space channel with guided delivery through hollow-core photonic crystal fibers, preserving high optical intensity with minimal light–glass overlap [2207.02324]. Space-based OPB extends the same principles to lunar-surface and cislunar links, where aperture size, slant range, constellation geometry, line-of-sight continuity, pointing jitter, and dust losses dominate system performance [2504.11300] [2402.16320] [2508.10855].

A persistent theme is that OPB is not solely a link-budget problem. Device physics at the receiver, particularly capacitance, responsivity, photovoltaic efficiency, thermal behavior, and mismatch under nonuniform illumination, are equally decisive. This is especially explicit in multi-segment GaAs PPCs, where segmentation alters both bandwidth and alignment sensitivity [2510.06205].

## 2. Optical and electro-optical principles

Most OPB analyses use Gaussian-beam propagation or Friis-like optical link formulations. In free-space lunar and terrestrial models, the beam radius evolves as
$$
w(z)=w_0\sqrt{1+\Bigl(\tfrac{\lambda z}{\pi w_0^2}\Bigr)^2},
$$
or, when beam quality is included,
$$
z_R=\frac{\pi w_0^2}{M^2\lambda}, \qquad \theta = \frac{M^2\lambda}{\pi w_0},
$$
with harvested power depending on the fraction of beam power intercepted by the receiver aperture [2504.11300] [2412.14083]. In cislunar phased-array analysis, effective aperture enlargement reduces divergence according to
$$
\theta_{\rm tx} = M^2\,\frac{2\lambda}{\pi D_{\rm eff}},
$$
which directly improves received power and end-to-end efficiency at long range [2508.10855].

At the receiver, OPB performance depends on the optical-to-electrical conversion chain. In the multi-segment GaAs PPC work, junction capacitance per subcell follows
$$
C_j \propto \epsilon\cdot A/d,
$$
and for \(N\) equal subcells in series,
$$
C_{\rm tot} \approx C_j/N,
$$
so the RC-limited electrical bandwidth becomes
$$
B = \frac{1}{2\pi\cdot R_{\rm load}\cdot C_{\rm tot}}.
$$
This formalizes the central OPB receiver trade-off: larger active area raises photocurrent, but also raises capacitance and lowers bandwidth; segmentation preserves total light-collecting area while lowering total capacitance [2510.06205].

The same work defines optical-to-electrical conversion efficiency as
$$
\eta = P_e/P_{\rm opt},
$$
with
$$
P_e = V\cdot I \approx R\cdot P_{\rm opt}
$$
in the short-circuit approximation, hence
$$
\eta = R.
$$
Although this expression is specific to the presentation in that study, it captures the paper’s device-centric view that responsivity and harvested power are directly entangled with link-level communication performance [2510.06205].

In resonant-beam systems, the defining condition is intra-cavity oscillation. Steady operation requires the round-trip gain to exceed or equal loss:
$$
G_{\rm rt} = R_1\,R_2\,e^{2\,g\,L} \ge 1.
$$
Once established, the resonant beam supports both power extraction and data modulation. A simplified received-power model is
$$
P_r(L) = P_t\,T_{\rm sys}\,e^{-\alpha L},
$$
while charging power and data throughput depend on cavity loss, extraction ratio, photovoltaic efficiency, and the electrical transfer response of the communication-and-energy-harvesting front-end [1809.11091].

## 3. Receiver technologies and integrated power–data reception

A major direction in contemporary OPB research is the use of photovoltaic or photonic conversion hardware not merely as energy harvesters but also as high-speed optical receivers. The most explicit example is the GaAs-based multi-segment PPC platform introduced for simultaneous energy harvesting and optical wireless communication [2510.06205].

That device divides each PPC chip into \(N = 2, 4\) or \(6\) circular GaAs subcells with diameters \(d = 1, 1.5\) or \(2.08\) mm. The subcells are electrically isolated by etched trenches in a semi-insulating GaAs substrate; the trenches are filled with polyimide and overlaid with metal bridges in a “pizza-configuration” to connect all subcells in series. The active junction is a 3.65 µm GaAs pn-junction grown by MOVPE on GaAs, with front and back passivation by GaInP layers, including a 400 nm front field layer transparent at 850 nm. Post-epitaxy processing includes photolithography, selective wet etching, dielectric passivation, metal evaporation, and anti-reflection coating [2510.06205].

The technical rationale is precise. GaAs-based PPCs provide six times greater electron mobility than silicon- or cadmium telluride-based cells, enabling faster data detection and improved power efficiency, but their bandwidth is constrained by junction capacitance, which increases with active area [2510.06205]. Segmentation reduces capacitance while maintaining light collection, so the device can act as both energy harvester and data detector in a single optical front-end.

Experimentally, these PPCs were used in an eye-safe 1.5 m optical wireless link employing OFDM with adaptive bit and power loading. The reported world-record metrics were a data rate of \(R_{\rm total} = 3.8\) Gbps for the 6-segment, \(d = 2.08\) mm device; electrical bandwidth \(B \approx 0.96\) GHz; and peak optical-to-electrical power conversion efficiency \(\eta \approx 39.7\%\) for the 2-segment, \(d = 2.08\) mm cell [2510.06205]. The abstract further states that the system converts 39.7% of optical power from a beam of 2.3 mW and that the achieved 3.8 Gbps is four times higher than prior works [2510.06205].

The same paper also quantifies the cost of segmentation. As the number of segments increases, the subcell area shrinks, so uniform illumination becomes more critical. Non-uniform beam profiles or slight misalignments create current mismatch in the series-connected subcells. The mismatch metric \(I_{\rm mp}/I_{\rm sc}\) drops from approximately 99% for 2-segment devices to approximately 66% for 6-segment devices, and the power-conversion efficiency falls correspondingly from approximately 39.7% to approximately 15.1% [2510.06205]. This establishes a concrete OPB design tension between bandwidth maximization and alignment tolerance.

Other receiver modalities appear in related OPB systems. RBCom uses a photovoltaic panel plus an AC-coupling network comprising capacitor \(C_0\), inductor \(L_0\), and resistor \(R_C\) to separate the DC charging path from the AC communication path [1809.11091]. Long-range resonant-beam SWIPT variants split the intra-cavity output by a ratio \(\mu\), sending a fraction to a photovoltaic array and the remainder to an APD, so power and data can be traded directly through the optical split [2108.00004]. In power-over-fiber, an InGaAs-based JDSU PPC-9LW receiver optimized for 1300–1550 nm achieved peak conversion efficiency of approximately 16.7% at an optimum load of 82 \(\Omega\), while under a 500 \(\Omega\) camera load it delivered approximately 0.035 W from 0.75 W incident optical power, corresponding to approximately 4.7% efficiency in that operating condition [2207.02324].

## 4. Modulation, SWIPT, and control

The data-bearing branch of OPB research is dominated by OFDM-based SWIPT formulations. In the GaAs multi-segment PPC link, a DCO-OFDM frame with M-QAM symbols is generated by an AWG and upconverted onto an 847 nm VCSEL biased at 1.78 V and 6 mA. At the receiver, the AC component is extracted via a bias-tee, sampled by a 10 GHz oscilloscope, and demodulated in Matlab using channel estimation, equalization, and adaptive bit/power loading [2510.06205]. Per-subcarrier SNR is estimated as
$$
{\rm SNR}_k = |H_k|^2P_k/N_0,
$$
and the bit loading is
$$
b_k = \log_2(1 + {\rm SNR}_k/\Gamma),
$$
yielding the total data rate
$$
R_{\rm total} = \sum_k b_k\cdot \Delta f.
$$
This places OPB squarely within the standard adaptive multicarrier communications framework [2510.06205].

RBCom uses a closely related OFDM abstraction. The AC photocurrent per subcarrier is modeled through a frequency-selective photovoltaic front-end \(H_{\rm ph}(\omega)\), with subcarrier SNR
$$
{\rm SNR}_k = \frac{|H_{\rm ph}(\omega_k)|^2(\gamma^2 P_{\rm sub})}{N_0 B_{\rm sub}},
$$
and total capacity
$$
C = \sum_{k=0}^{N-1} B_{\rm sub}\log_2(1+{\rm SNR}_k).
$$
Under typical parameters \(L_0 = 10\) nH, \(R_C = 140\,\Omega\), and \(P_{\rm laser}\approx 200\) mW, the paper reports bandwidth of approximately 200 MHz and total capacity of approximately 1.76 Gbit/s [1809.11091]. A related long-range resonant-beam SWIPT system reports numerical results of 0–9 W electrical power and 18 bit/s/Hz spectral efficiency over 20 m distance [2108.00004].

Beyond waveform design, OPB has also prompted work on adaptive control under channel distortion. For atmospheric power beaming with a fiber-array laser transmitter, one study considers a phased telescope array with \(N_{\rm sa}=19\) subapertures and \(K=57\) control channels for piston and tip-tilt compensation over a 5 km horizontal path at \(\lambda = 1.064\) µm [2204.05227]. The baseline control law is stochastic parallel gradient descent (SPGD),
$$
u^{\,k}_{t+1}
=u^{\,k}_t
+\frac{\gamma_t}{\sigma_t^k}\,\delta J_t\,\delta u_t^k,
$$
while the proposed self-learning controller uses a DNN driven by target-plane photovoltaic-array sensor data, current metric \(J_t\), previous control vector, and a short history window of size \(N_{\rm ws}=4\). The network contains time-distributed 2D convolution and max-pooling layers, a stateful GRU layer of 10 K units, a dense layer of 6 K units, and a linear output layer [2204.05227]. Numerical experiments show that in training mode the AI controller exceeds SPGD by approximately 5–7% in average \(J\) versus wind speed, while inference mode lags by up to approximately 10% except at very low wind [2204.05227]. This suggests that OPB increasingly intersects adaptive optics and learning-based control rather than remaining a static optical link problem.

Low-rate control reuse of the power beam itself is demonstrated in mobile robotics. The Phaser system directs a 915 nm narrow beam to moving robots, using the photovoltaic cell for power harvesting and a separate zero-bias photodiode for FSK reception. It delivers optical power densities of over 110 mW/cm\(^2\) and error-free data at multi-meter ranges, with on-board decoding drawing 0.3 mA, described as 97% less current than Bluetooth Low Energy [2504.17865]. While this platform is specialized, it exemplifies a broader OPB trend: co-design of power transfer, steering, and communication on a common optical carrier.

## 5. Propagation environments and channel impairments

OPB performance is dominated by the propagation environment once range increases. On Earth, atmospheric attenuation, turbulence, beam wander, and scattering constrain both received power and coupling efficiency. In the fiber-array atmospheric-beaming model, the transmission factor is
$$
T_{\rm atm} = e^{-\alpha L},
$$
and the total optical efficiency is
$$
\eta_{\rm opt} = P_r/P_t = T_{\rm atm}\,\eta_c,
$$
where \(\eta_c\) captures footprint mismatch, spread, and wander on the photovoltaic target [2204.05227]. High-energy terrestrial infrared OPB adds thermal blooming, turbulence, fog, smoke, rain, and safety exclusion zones to the practical design envelope [2405.00034].

In guided OPB, channel impairment shifts from open-air disturbance to fiber architecture and power handling. The hollow-core photonic crystal fiber study uses a single-ring tubular-lattice inhibited-coupling HCPCF with \(N=8\) untouching silica capillaries, wall thickness \(t = 1.05\) µm, and core diameter \(D_{\rm co} = 35\) µm. Two low-loss bands occur in the 700–1700 nm window, namely 770–940 nm and 1150–1590 nm. At the operating wavelength \(\lambda_0 = 1480\) nm, the attenuation is 35.3 dB/km, corresponding to a linear loss coefficient of approximately \(8.1\times 10^{-3}\,{\rm m}^{-1}\), while the dielectric overlap is approximately \(10^{-5}\), meaning less than 0.001% of power resides in silica [2207.02324]. The HCPCF safely transmitted 3 W input stably for over an hour, delivering 1.31 W without damage or power fluctuations of \(\sigma \approx 0.01\) W [2207.02324]. This is relevant to OPB because it addresses the power-ceiling limitations of conventional solid-core power-over-fiber systems.

Lunar OPB introduces a different impairment regime. Two recent studies focus on lofted lunar dust (LLD) and suspended regolith. One models OPB attenuation using the T-matrix method and Gaussian beam theory, finding that LLD significantly attenuates ground-to-ground transmission in illuminated regions, making OPB more suitable in darker areas such as permanently shadowed regions or during the lunar night [2412.14083]. The other introduces a detailed diffraction-plus-scattering model with altitude-dependent complex refractive index derived from particle density [2507.13982].

The latter gives especially explicit quantitative degradation. For a 1 kW, 1064 nm beam with \(\omega_0 = 5\) cm and a \(0.5\times 0.5\) m rover panel at \(h_p = 2\) m, efficiency \(\eta = P_r/P_0\) falls from 92.4% at 5 km and 50.4% at 50 km in dust-free conditions to 81.8% at 5 km, 12.5% at 30 km, and 3.7% at 50 km with 175 nm dust and source height \(h_0=2\) m. Raising the source to 12 m improves performance to 91.0% at 5 km, 32.7% at 30 km, and 25.0% at 50 km. With 250 nm particles, the viable transmission range drops below 30 km at 6% efficiency, and 50 km performance becomes negligible at 2 m source height and only 0.2% at 12 m [2507.13982]. The paper further notes that the beam profile shifts because the lower edge experiences stronger attenuation, producing a small upward centroid shift [2507.13982].

These results complicate a common assumption that the lunar surface is effectively a vacuum optical channel. The literature instead indicates that OPB system elevation, dust size distribution, and operational timing are mission-critical parameters on the Moon [2412.14083] [2507.13982].

## 6. Applications across terrestrial, robotic, and lunar systems

Current OPB applications span short-range embedded links, remote terrestrial power, mobile robotics, spacecraft charging, and lunar infrastructure.

For terrestrial short-range SWIPT, the multi-segment GaAs PPC system is presented as a solution for off-grid backhaul for future communication networks such as 6th generation cellular. The combination of greater than 1 GHz bandwidth and tens of percent power-conversion efficiency over meter-scale links suggests use in lightweight, rapidly deployable backhaul for 6G small cells or remote IoT clusters without wired power [2510.06205]. This suggests that OPB may occupy a niche where optical front-haul, energy autonomy, and compact receiver hardware must be integrated.

For mobile and safe SWIPT, RBCom targets 6G scenarios in which high-rate data and power are simultaneously desired. Numerical results show more than 40 mW charging power and 1.6 Gbit/s channel capacity with OFDM in one formulation [1809.11091]. The longer-range resonant-beam extension reports simultaneous delivery of 0–9 W electrical power and 18 bit/s/Hz over 20 m [2108.00004]. The self-terminating cavity behavior under obstruction makes these systems particularly relevant where mobility and beam safety are central design requirements [1809.11091].

Power-over-fiber broadens OPB to guided delivery in harsh or remote settings. The HCPCF demonstration activated a representative camera circuit using a watt-level continuous-wave laser beam delivered through a 6 m hollow-core fiber. The authors argue that hollow-core fibers are eligible candidates for next-generation power-over-fiber devices potentially able to lift the power restrictions of current solid-core systems [2207.02324]. This is not free-space OPB in the narrowest sense, but it is part of the same photonic energy-transfer continuum.

For robotic systems, Phaser fully powered gram-scale battery-free robots, simultaneously controlling them to navigate around obstacles and along paths. It achieved average robot speed of 1 cm/s, an 82% improvement over 5.5 mm/s under sunlight illumination in prior work, while maintaining zero BER up to 5 m and under diverse lighting [2504.17865]. The platform shows how OPB can be embedded in cyber-physical systems rather than restricted to fixed infrastructure.

Space and lunar applications are a rapidly expanding domain. “Continuous Power Beaming to Lunar Far Side from EMLP-2 Halo Orbit” finds that an equidistant triple-satellite scheme on an EMLP-2 halo orbit with semi-major axis \(A_z=15{,}000\) km provides full surface-coverage percentage for the lunar far side and is essential for continuous wireless power transmission [2402.16320]. With \(P_T=1\) kW, \(\lambda = 1064\) nm, \(d_R = 1\) m, \(\eta_T = 51\%\), and \(\eta_H = 50.8\%\), the stable L2 satellite case gives a median harvested power near 41.6 W, while for the revolving halo case the probability that \(P_H \le 41.6\) W is approximately 0.99 because of longer and varying slant ranges together with looser pointing accuracy [2402.16320].

For the lunar south pole, “Multi-Orbiter Continuous Lunar Beaming” studies multiple low-lunar-orbit satellites beaming to a circular solar-array receiver of diameter \(d_r = 2\) m. A 40-satellite quadruple-plane constellation yields 100% line-of-sight coverage over 27.3 days, while the average system efficiencies for single, 30-satellite, and 40-satellite schemes are 2.84%, 32.33%, and 33.29%, respectively, for a tracking panel, and 0.97%, 18.33%, and 20.44%, respectively, for a fixed panel [2504.11300]. Tracking yields an extra approximately 56% average power over the fixed case at \(M=40\), specifically 332.9 W versus 204.4 W [2504.11300].

More generally, the phased-array cislunar framework emphasizes that large effective apertures can produce orders-of-magnitude increases in delivered surface power under equivalent orbital and power conditions [2508.10855]. A representative example at \(R=10^6\) m with \(D_{\rm sub}=0.1\) m, \(N=100\), \(D_{\rm eff}=1\) m, \(D_r=1\) m, \(P_t=1\) kW, \(\eta_{\rm misc}=0.9\), and \(\eta_{\rm conv}=0.5\) yields \(P_r\approx 620\) W and \(P_{\rm elec}\approx 310\) W [2508.10855]. This suggests that aperture synthesis, rather than only higher laser power, is likely to dominate future long-range OPB scalability.

## 7. Trade-offs, misconceptions, and open technical questions

A recurring misconception is that OPB can be assessed by transmitter power and distance alone. The recent literature shows that the decisive trade space is multidimensional. Receiver segmentation raises bandwidth but reduces alignment margin [2510.06205]. Resonant extraction ratio and mirror reflectivity influence both charging power and data capacity [1809.11091]. In lunar networks, constellation multiplicity improves availability but increases deployment complexity [2504.11300] [2402.16320]. High source elevation can substantially mitigate dust-induced loss, but this adds structural mass and operational complexity [2507.13982].

Another misconception is that photovoltaic receivers are intrinsically too slow for useful data detection. The GaAs PPC results directly contradict that view by demonstrating gigahertz-class electrical bandwidth and a 3.8 Gbps world-record data rate in an eye-safe 1.5 m optical wireless link [2510.06205]. A more precise statement is that conventional large-area photovoltaic devices are often bandwidth-limited by capacitance, but receiver architecture and segmentation can alter that limitation materially.

Safety is also technically nuanced. Near-IR OPB is frequently associated with severe eye hazards, and some systems indeed require exclusion zones or airborne platforms; in the 1075 nm high-energy terrestrial study, the maximum permissible exposure for near-IR is cited as approximately 10 W/m\(^2\), and practical deployment requires management of beam pointing, atmospheric effects, and safety zones [2405.00034]. By contrast, RBCom embeds a fail-safe safety mechanism: an obstacle in the beam path increases intra-cavity diffraction loss, and once the loss exceeds round-trip gain, laser oscillation self-terminates in nanoseconds and re-establishes automatically when the obstacle is removed [1809.11091]. Meter-scale PPC communication links are likewise described as eye-safe and verified against IEC 60825-1 MPE limits for extended sources at 850 nm [2510.06205].

Open technical questions remain across scales. For terrestrial high-power OPB, the 1075 nm Yb-doped fiber-laser study reports that 20 kW illumination of a 0.6 m\(^2\) silicon solar panel can produce 3000 W at panel temperature 550 K, while a hybrid PV–TEG module can raise total efficiency to 0.43 and scale to approximately 4000 W output for the modeled panel [2405.00034]. This suggests a route to multi-kilowatt OPB, but only under demanding thermal-management and safety assumptions. For space systems, the PaddleSat concept examines hundreds-of-meters spacecraft-to-spacecraft charging with a 980 nm VCSEL array, 0.15 m optics, and approximately 10.5% total conversion chain, yielding approximately 18.4 W delivered from a 175 W electrical laser budget when the full beam is captured [2512.11629]. This indicates that formation flying and relative pointing may become as central as laser efficiency in spacecraft OPB.

The combined literature suggests that OPB is transitioning from proof-of-concept beam delivery to an integrated systems discipline. Device engineering, adaptive control, orbital geometry, safety architecture, thermal design, and environmental scattering are all first-order variables. A plausible implication is that future OPB systems will not converge on a single canonical architecture; instead, resonant self-aligning links, segmented high-speed PPC receivers, hollow-core guided channels, and phased-array cislunar transmitters are likely to persist as distinct solutions matched to different range, safety, and power-density regimes [1809.11091] [2510.06205] [2207.02324] [2508.10855].

Source: https://www.emergentmind.com/topics/optical-power-beaming-opb