Spectral Multiplexing

• Spectral multiplexing is the process of dividing a broad spectral bandwidth into multiple independent frequency channels, enabling high-capacity data transmission.
• It employs techniques like wavelength division, orthogonal modulation, and holographic encoding to create discrete, non-overlapping channels in various physical systems.
• Applications span optical communications, quantum memories, hyperspectral imaging, and on-chip photonics, driving advances in both classical and quantum technologies.

Spectral multiplexing is the process of encoding, transmitting, or processing multiple independent channels of information—optical, electrical, or quantum—simultaneously in distinct frequency (spectral) modes within a single physical system or device. It is fundamental to high-capacity communications, advanced photonic instrumentation, and scalable quantum information science. Methods differ dramatically across classical communications, computational optics, and quantum networks, but all leverage discrete spectral degrees of freedom to expand throughput, efficiency, or scalability beyond what time- or spatial-multiplexing alone can achieve.

1. Fundamental Principles and Physical Mechanisms

Spectral multiplexing exploits the division of a broad spectral bandwidth into multiple orthogonal channels, each defined by a central frequency and bandwidth. In classical systems, these channels may be realized via narrowband filters, orthogonal modulation formats, or holographic encoding in the Fourier domain of temporal signals. In quantum and photonic devices, spectral channels correspond to individually addressable frequency bins of emitters, light fields, or collective spin excitations, made available via inhomogeneous broadening, engineered mode structures, or cavity resonance splitting.

Distinct spectral channels can be established in various ways:

• Optical communications: Division into wavelength bands (WDM), orthogonal subcarriers (OFDM), or artificial parallel channels via pulse-shaping and meta-multiplexing (Ji et al., 2017, Hamamreh et al., 2020, Khodaei et al., 2021).
• Single-photon/quantum memories: Discrete frequency bins in atomic ensembles or crystals, with spectral filtering, frequency-bin manipulation, and programmable mapping of input and recalled states (Sinclair et al., 2013, Yu et al., 2021, Ulanowski et al., 2023).
• Computational imaging/instrumentation: Spectrally resolved channels via diffractive optics, holographic gratings, or spectrometer arrays for parallel sensing or imaging (Deng et al., 2018, Alessio et al., 2017, Kulkov et al., 6 Sep 2025, Tolila et al., 2024).
• Spectral islands: Domain-engineered nonlinear crystals supporting multiple non-overlapping, factorable biphoton wavefunctions for quadratic scaling in two-photon entanglement delivery (Shapiro et al., 19 Jul 2025).

The underlying mechanism always combines spectral orthogonality with a detection, routing, or encoding protocol that preserves channel independence and allows for recombination or selective readout.

2. Core Methodologies

2.1 Classical and Communication Protocols

Traditional frequency-division multiplexing (FDM) and advanced strategies such as orthogonal frequency-division multiplexing with subcarrier power modulation (OFDM-SPM) or meta-multiplexing deliberately exploit the spectral domain for massive increases in throughput (Hamamreh et al., 2020, Ji et al., 2017). In meta-multiplexing, overlapping modulated streams in time/frequency are decoded into "artificial" parallel subchannels, effectively linearizing the classical SNR-to-capacity scaling up to the occupied signal bandwidth (Ji et al., 2017).

Holographic spectral multiplexing (HSM) encodes high-entropy data symbols into ultrashort laser pulses in the temporal Fourier domain, harnessing the spatial distribution of frequency components within a 2D hologram to form distinct logical channels—realizing optical MIMO without explicit subchannelization (Khodaei et al., 2021).

Stacked volume-phase holographic gratings (VPHGs) and Fourier-domain spectral basis multiplexing enable simultaneous acquisition of multiple spectra or compressed hyperspectral data from dynamic scenes (Alessio et al., 2017, Deng et al., 2018). The use of stacked, spectrally selective diffractive elements allows the readout of multiple bands at once, directly boosting spectral coverage and resolution within a single exposure.

2.2 Quantum and Photonics Protocols

Quantum systems exploit spectral multiplexing to circumvent probabilistic limits of single-mode sources or memories. Protocols based on atomic frequency combs (AFC) store or retrieve photons in hundreds or thousands of spectral bins, with feed-forward control selecting the recalled bin via phase modulation (Sinclair et al., 2013). Spectral multiplexing in heralded single-photon sources boosts indistinguishable photon emission rates linearly with the number of bins, while preserving single-photon purity (as measured by $g^{(2)}(0)$) and indistinguishability (Puigibert et al., 2017, Yu et al., 2021).

The zero-added-loss multiplexing (ZALM) regime realizes heralded quantum entanglement across many spectral bins by feed-forward frequency shifting—all while incurring no loss penalty for the surviving photon (Chapman et al., 19 Dec 2025, Shapiro et al., 19 Jul 2025). Entangled-pair generation rates can scale as $N^2$ with cross-bin heralding, in contrast to conventional $N$-scaling.

Material platforms play a crucial role: rare-earth-doped crystals (e.g., Er:YSO, Er:Eu:YSO) provide ensembles of narrow, addressable spectral lines within inhomogeneous broadening, allowing hundreds of channels with lifetime-limited optical coherence (Ulanowski et al., 2021, Ulanowski et al., 2023).

3. Information-Throughput, Fidelity, and Scalability

3.1 Capacity and Spectral Efficiency

In classical multiplexing, spectral efficiency $\eta$ (bits/s/Hz) is a central metric. Methods such as meta-multiplexing have experimentally achieved $\eta \sim 81.7$ bits/s/Hz (Ji et al., 2017). The Shannon capacity for $Q$ orthogonal channels is given by:

$$\eta = \sum_{i=1}^Q \log_2 \left(1 + \text{SINR}_i\right)$$

where each channel's signal-to-interference-plus-noise ratio (SINR) includes interchannel crosstalk terms (Cagliero et al., 2016). Approaches leveraging artificial subchannels, as in meta-multiplexing or HSM, can surpass conventional $\log_2(1+\text{SNR})$ scaling within a fixed bandwidth, limited only by the pulse shape and channel decoding complexity (Ji et al., 2017). OFDM-SPM doubles per-subcarrier spectral efficiency by encoding bits both in constellation and power levels (Hamamreh et al., 2020).

In quantum applications, multiplexing increases entanglement or single-photon rates as:

• Heralded source: rate $\propto$ number of spectral bins $N$ for SMUX (Puigibert et al., 2017, Yu et al., 2021).
• ZALM: entanglement delivery rate $\propto N^2$ using phase-matched spectral islands and cross-bin heralding (Shapiro et al., 19 Jul 2025).
• AFC memories: mode capacity $N \sim B/\Delta$ is set by total bandwidth $B$ and spectral spacing $\Delta$ (Sinclair et al., 2013).

3.2 Fidelity and Crosstalk

Ensuring orthogonality and minimal crosstalk between spectral channels is critical. In rare-earth architectures, channel fidelity is quantified by corrected $g^{(2)}(0)$ measurements, with $g_{\rm corr}^{(2)}(0)<0.15$ demonstrating negligible channel overlap (Ulanowski et al., 2023). For cavity-based sources, the achieved single-channel linewidths ($<0.2$ MHz), Purcell factors ($>35$), and optical coherence times ($T_2 \approx 0.6$ ms with dynamical decoupling) guarantee high-fidelity spectral multiplexing (Ulanowski et al., 2021, Ulanowski et al., 2023).

In quantum repeater architectures, cross-channel isolation ensures that heralded entanglement remains deterministic and robust to loss and multipair emission, as demonstrated by conditional Bell-state fidelities of 0.99 at telecom wavelengths in both low and high-loss regimes (Shapiro et al., 19 Jul 2025).

4. Experimental Systems and Instrumentation Architectures

A diversity of platforms have been realized:

• Single-photon spectrometers: Linear SPAD arrays (e.g., LinoSPAD2) provide 40 pm spectral and 40 ps temporal resolution across 100 channels, enabling massively parallel Hanbury Brown–Twiss interference and scaling two-photon throughput by orders of magnitude (Kulkov et al., 6 Sep 2025, Tolila et al., 2024).
• Stacked VPHG devices: Directly installed in telescopes (e.g., GTC-OSIRIS), these enable multi-band, high-efficiency astronomical spectroscopy in a single exposure (Alessio et al., 2017).
• VIPA-based spectrometers: Provide GHz-resolution channelization, with cross-talk $<$–25 dB and up to 17% efficiency in fiber-coupled platforms for quantum repeater applications (Chakraborty et al., 2022).
• Fabry–Pérot resonators with rare-earth membranes: Achieve up to 360 individually addressable spectral channels, combining sub-MHz spectral diffusion, lifetime-limited $T_2$ times, and massive co-doping-enabled channelization (Ulanowski et al., 2023).
• Lensless multi-distance imaging and Fourier spectral multiplexing: Allow simultaneous recovery of multi-wavelength complex fields or hyperspectral datacubes with few intensity frames and tractable computational postprocessing (You et al., 2024, Deng et al., 2018).

Different architectures offer trade-offs in complexity, scalability, and per-channel SNR. For interferometric sensing, spectral multiplexing reduces integration time or increases sensitivity as $\sqrt{N}$, as demonstrated in both laboratory and field settings (Tolila et al., 2024, Kulkov et al., 6 Sep 2025).

5. Applications Across Disciplines

Spectral multiplexing underpins high-performance systems in diverse fields:

• Optical communications: Drives multi-Tbit/s dense wavelength-division multiplexing (DWDM) systems and advanced optical MIMO schemes (Khodaei et al., 2021, 0903.2471).
• Quantum networks and repeaters: Enables scalable distribution of entanglement via frequency-multiplexed sources, memories, and Bell-state analyses, essential for quantum repeaters and all-photonic quantum computing (Sinclair et al., 2013, Chakraborty et al., 2022, Chapman et al., 19 Dec 2025).
• Photon-starved astronomy: Massively parallel spectral HBT correlation and multiplexed intensity interferometry increase precision, reduce observation time, and access fainter targets (Kulkov et al., 6 Sep 2025, Tolila et al., 2024).
• Snapshot hyperspectral imaging: Advances real-time material analysis, dynamic scene monitoring, and high-throughput microscopy (Deng et al., 2018, Alessio et al., 2017).
• On-chip photonic circuits: Spectral-line-shape multiplexing, achieved through chiral metasurfaces and photonic spin–orbit interaction, enables programmable filtering, switching, and sensing on scalable platforms (Cheng et al., 2023).

6. Contemporary Challenges and Prospects

A central challenge is the engineering of device architectures with minimal loss, low crosstalk, and rapid, programmable access to $O(10^2$–$10^4)$ spectral modes. At the component level, this includes the development of high-finesse, low-insertion-loss frequency shifters, tunable microcavity filters, and on-chip gratings. For quantum network applications, further progress in spectral dynamic range, electro-optic response (sub-dB, $V_\pi<0.5$ V), and memory lifetimes is pivotal (Ulanowski et al., 2023, Ulanowski et al., 2021, Chapman et al., 19 Dec 2025).

In all domains—and especially for quantum repeater architectures—the convergence of spectral and temporal multiplexing, precise feed-forward control, and hybrid integration will extend capacity, scalability, and fault tolerance.

7. Key Theoretical Underpinnings and Capacity Analysis

• The Shannon capacity for spectral-multiplexed systems scales as a sum over independent channel capacities, but only when sufficiently isolated spectral modes are available (Cagliero et al., 2016).
• Artificial parallelization via meta-multiplexing or HSM can linearize SNR-to-capacity scaling under suitable waveform and decoding constraints (Ji et al., 2017, Khodaei et al., 2021).
• In quantum regimes, the scaling of deterministic source or entanglement rates depends critically on the effective number of non-overlapping, programmable frequency bins and the protocol for feed-forward channel addressing (Shapiro et al., 19 Jul 2025, Chapman et al., 19 Dec 2025).
• In high-loss or finite-fidelity scenarios, channel crosstalk, multipair emission, and storage inefficiency set fundamental bounds on usable spectral-multiplexed operation (Ulanowski et al., 2021, Sinclair et al., 2013, Shapiro et al., 19 Jul 2025).

Spectral multiplexing thus serves as a cornerstone technique at the intersection of photonic engineering, quantum information science, and optical communications. It is advancing the practical limits of channel capacity, measurement fidelity, and parallelization in both classical and quantum architectures (Sinclair et al., 2013, Ulanowski et al., 2021, Tolila et al., 2024, Shapiro et al., 19 Jul 2025, Alessio et al., 2017, Kulkov et al., 6 Sep 2025, Ji et al., 2017, Hamamreh et al., 2020).