Sensing-Assisted OFDM with Index Modulation: A Waveform for Optical Wireless Integrated Communication and Localisation
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Electrical Engineering. The day in electrical engineering: signals, systems and the hardware that carries them.
Rosa: We're looking at the paper: "Sensing-Assisted OFDM with Index Modulation".
Dev: Here's the gist of the paper. Sensing-assisted OFDM with index modulation (SA-OFDM-IM) proposes a novel waveform for joint communication and localisation over optical wireless systems, enabling low-complexity detection of sensing and communication subcarriers while minimizing error rates and localization errors Waveform Architecture The SA-OFDM-IM waveform employs a multilayer frequency-domain architecture that partitions each layer’s admissible subcarriers into disjoint communication and distributed sensing-pilot subsets, applying group-wise index modulation (IM) within the communication subset The first layer coincides with the odd PF subcarriers, as in ACOOFDM, while successive layers occupy progressively sparser, higher-order decimations Within each layer, admissible subcarriers are partitioned into Gl disjoint groups of size gl, adaptively selected as gl = max(g ∈ 16, 8, 4: g Ul), so that Gl = Ul/gl The PF subcarrier mapping nl,i,r resolves group position igl + r into an absolute bin index For each tuple ν, Rl is partitioned into disjoint communication, sensing-pilot, and null index subsets Cν, Sν, and Zν of cardinalities kl, pl, and zl = gl − kl − pl respectively Receiver Processing Chain The receiver evaluates the per-tone energy metric Γj,ν(r) = Yj,m
nl,i,r: 2 for r ∈ Rl Because the sensing pilots are allocated strictly higher per-subcarrier energy than the communication symbols by construction, the imposed power-domain separation renders the sensing and communication subcarriers distinguishable by rank order alone The pl highest-energy tones are declared as the sensingpilots, Sbj,ν = Σπj,ν(1),..., πj,ν(pl), and the next kl highestenergy tones are declared as the communication subcarriers, Cbj,ν = Σπj,ν(pl + 1),..., πj,ν(pl + kl) This energybased subcarrier detection lowers receiver complexity from the combinatorial joint search O2⌊log2 (glkl)MklQAM required by exhaustive ML detection to a single sort of gl log gl energy values, i.e., O(gl log gl) Localization Strategy A delay-aided RSS-based localisation method is introduced, in which the LoS delay supplies a multilateration-based initial position estimate that both initialises and safeguards the subsequent refinement The phase-compensated LoS gains refine the PD position and, when required, jointly estimate its surface-normal vector The delay observable is inherently insensitive to the PD surface normal nD and to optical-gain calibration For RSS-based localisation, the candidate position c parameterises the link geometry, and the unit direction vector from LED j towards c is uSD,j (c) = c−cS,j dj (c) / dj (c) Performance Metrics The proposed SA-OFDM-IM waveform consistently outperforms DCO-OFDM across all considered N in both BER and CE accuracy The RSS-based estimator consistently achieves the lowest range and position RMSEs among all considered benchmarks The delay-based estimators exhibit SNR-independent error floors, since dbtap,j is restricted to integer multiples of the delay-bin resolution ∆d The RSS branch estimates the receiver position by identifying the candidate whose predicted gain is most consistent with αb The analysis further reveals that unresolved NLoS components can bias the extracted LoS gain Energy Allocation Optimization The optimal allocation fraction is then obtained as µ⋆ = arg min µ∈M JE(µ), where the feasible set is M = Σµ ∈ (0, 1): γs,l(µ)γc,l(µ) ≥ Rmin ∀l The resulting operating point is sensing-dominant, but not sensing-exclusive, since a non-negligible fraction of the reference energy is always retained for communication This optimal allocation is determined by balancing the average BER and delaybased localisation RMSE over a representative signal-to-noise ratio (SNR) set The resulting µ⋆ is thereafter treated as a fixed, scenario-specific design parameter rather than an adaptive quantity updated during transmission Channel Estimation Accuracy The CE accuracy is quantified through the average normalised mean squared error (NMSE), defined as NMSE = 1/NMCJvN Σζ=1 XMCζ=1 Xj∈Jv hb(ζ)j,fr − h(ζ)j / h(ζ)j2 SA-OFDM-IM achieves a lower NMSE than DCO-OFDM because of reduced noise variance obtained by averaging the pilotbased CFR estimates across repeated observations At high SNRs, the residual NMSE is dominated by interpolation distortion across the unobserved subcarriers Multipath Impact on Gain Extraction In a dispersive multipath channel, the estimated CFR is a superposition of the LoS component and NLoS reflections, each characterised by a gain and delay tap The leakage coefficient Λρ,j = 1/N+ Σn∈N+ e(-ȷ2πn(λρ,j−λb0,j)) N quantifies the residual, uncompensated phase rotation retained by the ρ-th NLoS component once phase compensation has been applied When an NLoS component is strong, unresolved, or falls within or adjacent to the delay bin of the detected LoS tap, i.e., dρ,j − d0,j ≲ ∆d Any such unresolved or closely spaced reflection therefore biases αb0,j, degrading RSS-based localisation accuracy since the distorted gain propagates through the RSS model into erroneous distance estimates The paper proposed SA-OFDM-IM waveform for joint optical wireless communication and localisation The design embeds communication symbols and sensing pilots across layered, index-modulated subcarrier groups, while an adaptive periodic bias enforces the non-negativity constraint At the receiver, a low-complexity energy detector distinguishes communication from sensing subcarriers, and frame-level aggregation and averaging of the distributed sensing pilots yield robust CFR and CIR estimates The delay branch provides localisation constrained by the fundamental delay resolution of the system bandwidth, the proposed delay-aided RSS estimator achieves sub-delay-bin accuracy Moreover, when the PD orientation is unknown, its normal vector can be estimated jointly with position through the candidate-dependent gain model The reported results demonstrate that SA-OFDM-IM outperforms DCO-OFDM in both BER and CE accuracy, while the proposed RSS-based estimator consistently achieves the lowest range and position RMSEs among all considered benchmarks The analysis further reveals that unresolved NLoS components can bias the extracted LoS gain The development of dedicated multipathsuppression mechanisms is therefore left as a direction for future work, alongside the extension of the anchor-blockage conditions Pruning redundant reasoning tokens The paper proposes sensing-assisted OFDM with index modulation (SA-OFDM-IM) for joint communication and localisation over optical wireless systems This paper is 2610.21753 How the training signal is built The proposed frame structure generalises the multilayer principle of
4: to an IM-enabled waveform by superimposing L FD layers, each exploiting group-wise IM to jointly embed communication symbols and sensing pilots Within each layer, Ul admissible subcarriers are partitioned into Gl disjoint groups of size gl, adaptively selected as gl = max(g ∈ 16, 8, 4: g Ul), so that Gl = Ul/gl The PF subcarrier mapping nl,i,r resolves group position igl + r into an absolute bin index For each tuple ν, Rl is partitioned into disjoint communication, sensing-pilot, and null index subsets Cν, Sν, and Zν of cardinalities kl, pl, and zl = gl − kl − pl respectively Where the gains come from The gains are attributed to distributed pilot embedding and the framelevel aggregation of repeated sensing subcarriers The RSS-based estimator achieves sub-delay-bin accuracy This gain stems from the distributed sensing-pilot placement and framelevel averaging of repeated sensing subcarriers, which jointly improves the CFR and LoS gain estimates The analysis further reveals that unresolved NLoS components can bias the extracted LoS gain The development of dedicated multipathsuppression mechanisms is therefore left as a direction for future work, alongside the extension of the anchor-blockage conditions Performance Evaluation The paper presents performance evaluation of SA-OFDMIM waveform in terms of both communication and sensing functionalities The reported BER is obtained by first evaluating the link-level BER independently for each LED-to-PD pair and subsequently averaging across all LEDs The RSS position estimate follows as bc(RSS)D = arg minc∈B JRSS(c) The resulting range and position RMSEs decrease continuously with SNR and remain substantially below those of the delay-based method The analysis further reveals that unresolved NLoS components can bias the extracted LoS gain The development of dedicated multipathsuppression mechanisms is therefore left as a direction for future work, alongside the extension of the anchor-blockage conditions Energy Allocation Trade-off The optimal allocation fraction is then obtained as µ⋆ = arg min µ∈M JE(µ), where the feasible set is M = Σµ ∈ (0, 1): γs,l(µ)γc,l(µ) ≥ Rmin ∀l The resulting operating point is sensing-dominant, but not sensing-exclusive, since a non-negligible fraction of the reference energy is always retained for communication This optimal allocation is determined by balancing the average BER and delaybased localisation RMSE over a representative signal-to-noise ratio (SNR) set The resulting µ⋆ is thereafter treated as a fixed, scenario-specific design parameter rather than an adaptive quantity updated during transmission Channel Estimation Accuracy The CE accuracy is quantified through the average normalised mean squared error (NMSE), defined as NMSE = 1/NMCJvN Σζ=1 XMCζ=1 Xj∈Jv hb(ζ)j,fr − h(ζ)j / h(ζ)j2 SA-OFDM-IM achieves a lower NMSE than DCO-OFDM because of reduced noise variance obtained by averaging the pilotbased CFR estimates across repeated observations The analysis further reveals that unresolved NLoS components can bias the extracted LoS gain The development of dedicated multipathsuppression mechanisms is therefore left as a direction for future work, alongside the extension of the anchor-blockage conditions Multipath Impact on Gain Extraction In a dispersive multipath channel, the estimated CFR is a superposition of the LoS component and NLoS reflections, each characterised by a gain and delay tap The leakage coefficient Λρ,j = 1/N+ Σn∈N+ e(-ȷ2πn(λρ,j−λb0,j)) N quantifies the residual, uncompensated phase rotation retained by the ρ-th NLoS component once phase compensation has been applied When an NLoS component is strong, unresolved, or falls within or adjacent to the delay bin of the detected LoS tap, i.e., dρ,j − d0,j ≲ ∆d Any such unresolved or closely spaced reflection therefore biases αb0,j, degrading RSS-based localisation accuracy since the distorted gain propagates through the RSS model into erroneous distance estimates The development of dedicated multipathsuppression mechanisms is therefore left as a direction for future work, alongside the extension of the anchor-blockage conditions Pruning redundant reasoning tokens The paper proposes sensing-assisted OFDM with index modulation (SA-OFDM-IM) for joint communication and localisation over optical wireless systems This paper is 2610.
Paper summary: Rosa: We’re looking at the Sensing-Assisted OFDM with Index Modulation waveform for optical wireless systems. The central thesis is that you can integrate communication and localisation into one system using this specific structure.
Dev: They propose a multilayer frequency domain architecture where each layer splits the subcarriers into distinct communication and sensing pilot subsets, applying group-wise index modulation only within those communication subsets.
Rosa: The key claim is that by constraining the sensing to communication energy allocation, you minimize both the average bit error rate and the delay-based localisation root mean square error under a minimum power margin.
Dev: This is achieved through low-complexity, ordered energy detection of both subcarriers without needing an exhaustive maximum likelihood search.
Taro: So, for someone listening to this, it means they get a waveform that does the job of sensing and communication simultaneously with less computational overhead at the receiver than previous methods.
Rosa: Right. But they also detail how sensing tones are aggregated across the frame to reconstruct the channel frequency response, which gives you line of sight delay tap and phase-compensated line of sight gain for multilateration.
Dev: That aggregation is what enables them to get those estimates even when dealing with unresolved non-line-of-sight components.
Taro: The caveat there is that the paper itself flags that unresolved non-line-of-sight components can bias the extracted line of sight gain, which means it't not perfect localization.
Rosa: And they also introduce an energy allocation optimization step where you find an optimal sensing fraction based on balancing BER and localisation RMSE over a signal-to-noise ratio set.
Dev: This optimal point is sensing-dominant but not exclusive, meaning communication still gets some reference energy retained for reliable symbol transmission.
Taro: It's an adaptive way to tune the system design parameter mu based on what the specific environment demands, rather than just picking one fixed setting.
Conclusion: Rosa: So, to wrap up, this Sensing-Assisted OFDM with Index Modulation waveform is designed to handle the dual problem of optical wireless communication and localisation together.
Dev: The authors are Azim, Islim, Imran, Hass, and Abumarshoud. The main implication is that they provide a structured way to embed sensing pilots across layered subcarrier groups.
Rosa: In simple terms for the listener: it means you can design an optical wireless system where the communication symbols and the sensing pilots are neatly separated in a way that makes detection of both very efficient.
Dev: It moves away from needing separate, dedicated training symbols or sensing resources for channel estimation and localization.
Taro: For those who only care about real-world operation, it means this waveform is more robust because it handles the inherent uncertainty of the environment better than older frameworks that relied on perfect assumptions.
Rosa: It's about making a system where the communication and sensing tasks are tightly coupled through this specific structure for better overall performance in complex optical wireless scenarios.
Dev: The paper points toward needing better ways to handle those remaining multipath reflections, which is the future direction for improving the system.
University of Glasgow · pureLiFi Ltd. · Department of Engineering, University of Cambridge
eess.SP
Submitted: 2026-10-15
Updated: 2026-11-05
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 100/100
The gist: Sensing-assisted OFDM with index modulation (SA-OFDM-IM) proposes a novel waveform for joint communication and localisation over optical wireless systems…
Key concepts
- SA-OFDM-IM Waveform
- This is the proposed signal structure for optical wireless systems. It uses a multilayer frequency-domain architecture where different layers are assigned specific functions: some carry communication symbols, while others carry sensing pilots. Index modulation is used within the communication layer to embed these symbols effectively.
- Energy-Based Subcarrier Detection
- The receiver identifies communication and sensing subcarriers by measuring their energy levels. Because sensing pilots are intentionally allocated higher energy than communication symbols, simply sorting the tones by energy allows the receiver to distinguish between them with low complexity, avoiding complex joint searches.
- Delay-Aided RSS Localization
- This localization method uses time delays from received signals to estimate a position. It starts with an initial estimate from Line-of-Sight (LoS) delay and refines it using phase-compensated LoS gains. This approach is robust against unknown surface normals and provides accurate range and position estimates.
- Optimal Energy Allocation ($μ\star$)
- This determines the best balance between communication and sensing energy usage. The optimization finds the fraction of reference energy dedicated to sensing versus communication that minimizes a combined metric of bit error rate (BER) and localization error, resulting in a fixed, scenario-specific design parameter.
Terminology
Summary
Sensing-assisted OFDM with index modulation (SA-OFDM-IM) proposes a novel waveform for joint communication and localisation over optical wireless systems, enabling low-complexity detection of sensing and communication subcarriers while minimizing error rates and localization errors
Waveform Architecture
The SA-OFDM-IM waveform employs a multilayer frequency-domain architecture that partitions each layer’s admissible subcarriers into disjoint communication and distributed sensing-pilot subsets, applying group-wise index modulation (IM) within the communication subset The first layer coincides with the odd PF subcarriers, as in ACOOFDM, while successive layers occupy progressively sparser, higher-order decimations Within each layer, admissible subcarriers are partitioned into Gl disjoint groups of size gl, adaptively selected as gl = max(g ∈ 16, 8, 4: g Ul), so that Gl = Ul/gl The PF subcarrier mapping nl,i,r resolves group position igl + r into an absolute bin index For each tuple ν, Rl is partitioned into disjoint communication, sensing-pilot, and null index subsets Cν, Sν, and Zν of cardinalities kl, pl, and zl = gl − kl − pl respectively
Receiver Processing Chain
The receiver evaluates the per-tone energy metric Γj,ν(r) = Yj,m[nl,i,r]2 for r ∈ Rl Because the sensing pilots are allocated strictly higher per-subcarrier energy than the communication symbols by construction, the imposed power-domain separation renders the sensing and communication subcarriers distinguishable by rank order alone The pl highest-energy tones are declared as the sensingpilots, Sbj,ν = Σπj,ν(1),..., πj,ν(pl), and the next kl highestenergy tones are declared as the communication subcarriers, Cbj,ν = Σπj,ν(pl + 1),. p. 3 The pℓ highest-energy tones are declared as the sensingpilots, Sbj,ν = {πj,ν(1), . , πj,ν(pℓ)}, and the next kℓ highestenergy tones are declared as the communication subcarriers, Cbj,ν = {πj,ν(pℓ + 1), . This energybased subcarrier detection lowers receiver complexity from
the combinatorial joint search O
2
⌊log2 (
gℓ
kℓ
)⌋Mkℓ
QAM
required
by exhaustive ML detection to a single sort of gℓ energy
values, i. , O(gℓ log gℓ).Where the paper says it
Where the paper says it
Localization Strategy
A delay-aided RSS-based localisation method is introduced, in which the LoS delay supplies a multilateration-based initial position estimate that both initialises and safeguards the subsequent refinement The phase-compensated LoS gains refine the PD position and, when required, jointly estimate its surface-normal vector The delay observable is inherently insensitive to the PD surface normal nD and to optical-gain calibration For RSS-based localisation, the candidate position c parameterises the link geometry, and the unit direction vector from LED j towards c is uSD,j (c) = c−cS,j dj (c) / dj (c)
Performance Metrics
The proposed SA-OFDM-IM waveform consistently outperforms DCO-OFDM across all considered N in both BER and CE accuracy The RSS-based estimator consistently achieves the lowest range and position RMSEs among all considered benchmarks The delay-based estimators exhibit SNR-independent error floors, since dbtap,j is restricted to integer multiples of the delay-bin resolution ∆d The RSS branch estimates the receiver position by identifying the candidate whose predicted gain is most consistent with αb The analysis further reveals that unresolved NLoS components can bias the extracted LoS gain
Energy Allocation Optimization
The optimal allocation fraction is then obtained as µ⋆ = arg min µ∈M JE(µ), where the feasible set is M = Σµ ∈ (0, 1): γs,l(µ)γc,l(µ) ≥ Rmin ∀l The resulting operating point is sensing-dominant, but not sensing-exclusive, since a non-negligible fraction of the reference energy is always retained for communication This optimal allocation is determined by balancing the average BER and delaybased localisation RMSE over a representative signal-to-noise ratio (SNR) set The resulting µ⋆ is thereafter treated as a fixed, scenario-specific design parameter rather than an adaptive quantity updated during transmission
Channel Estimation Accuracy
The CE accuracy is quantified through the average normalised mean squared error (NMSE), defined as NMSE = 1/NMCJvN Σζ=1 XMCζ=1 Xj∈Jv hb(ζ)j,fr − h(ζ)j / h(ζ)j2 SA-OFDM-IM achieves a lower NMSE than DCO-OFDM because of reduced noise variance obtained by averaging the pilotbased CFR estimates across repeated observations At high SNRs, the residual NMSE is dominated by interpolation distortion across the unobserved subcarriers
Multipath Impact on Gain Extraction
In a dispersive multipath channel, the estimated CFR is a superposition of the LoS component and NLoS reflections, each characterised by a gain and delay tap The leakage coefficient Λρ,j = 1/N+ Σn∈N+ e(-ȷ2πn(λρ,j−λb0,j)) N quantifies the residual, uncompensated phase rotation retained by the ρ-th NLoS component once phase compensation has been applied When an NLoS component is strong, unresolved, or falls within or adjacent to the delay bin of the detected LoS tap, i.e., dρ,j − d0,j ≲ ∆d Any such unresolved or closely spaced reflection therefore biases αb0,j, degrading RSS-based localisation accuracy since the distorted gain propagates through the RSS model into erroneous distance estimates
Conclusion
The paper proposed SA-OFDM-IM waveform for joint optical wireless communication and localisation The design embeds communication symbols and sensing pilots across layered, index-modulated subcarrier groups, while an adaptive periodic bias enforces the non-negativity constraint At the receiver, a low-complexity energy detector distinguishes communication from sensing subcarriers, and frame-level aggregation and averaging of the distributed sensing pilots yield robust CFR and CIR estimates The delay branch provides localisation constrained by the fundamental delay resolution of the system bandwidth, the proposed delay-aided RSS estimator achieves sub-delay-bin accuracy Moreover, when the PD orientation is unknown, its normal vector can be estimated jointly with position through the candidate-dependent gain model The reported results demonstrate that SA-OFDM-IM outperforms DCO-OFDM in both BER and CE accuracy, while the proposed RSS-based estimator consistently achieves the lowest range and position RMSEs among all considered benchmarks The analysis further reveals that unresolved NLoS components can bias the extracted LoS gain The development of dedicated multipathsuppression mechanisms is therefore left as a direction for future work, alongside the extension of the anchor-blockage conditions
Pruning redundant reasoning tokens
The paper proposes sensing-assisted OFDM with index modulation (SA-OFDM-IM) for joint communication and localisation over optical wireless systems This paper is 2610.21753
How the training signal is built
The proposed frame structure generalises the multilayer principle of [4] to an IM-enabled waveform by superimposing L FD layers, each exploiting group-wise IM to jointly embed communication symbols and sensing pilots Within each layer, Ul admissible subcarriers are partitioned into Gl disjoint groups of size gl, adaptively selected as gl = max(g ∈ 16, 8, 4: g Ul), so that Gl = Ul/gl The PF subcarrier mapping nl,i,r resolves group position igl + r into an absolute bin index For each tuple ν, Rl is partitioned into disjoint communication, sensing-pilot, and null index subsets Cν, Sν, and Zν of cardinalities kl, pl, and zl = gl − kl − pl respectively
Where the gains come from
The gains are attributed to distributed pilot embedding and the framelevel aggregation of repeated sensing subcarriers The RSS-based estimator achieves sub-delay-bin accuracy This gain stems from the distributed sensing-pilot placement and framelevel averaging of repeated sensing subcarriers, which jointly improves the CFR and LoS gain estimates The analysis further reveals that unresolved NLoS components can bias the extracted LoS gain The development of dedicated multipathsuppression mechanisms is therefore left as a direction for future work, alongside the extension of the anchor-blockage conditions
Performance Evaluation
The paper presents performance evaluation of SA-OFDMIM waveform in terms of both communication and sensing functionalities The reported BER is obtained by first evaluating the link-level BER independently for each LED-to-PD pair and subsequently averaging across all LEDs The RSS position estimate follows as bc(RSS)D = arg minc∈B JRSS(c) The resulting range and position RMSEs decrease continuously with SNR and remain substantially below those of the delay-based method The analysis further reveals that unresolved NLoS components can bias the extracted LoS gain The development of dedicated multipathsuppression mechanisms is therefore left as a direction for future work, alongside the extension of the anchor-blockage conditions
Energy Allocation Trade-off
The optimal allocation fraction is then obtained as µ⋆ = arg min µ∈M JE(µ), where the feasible set is M = Σµ ∈ (0, 1): γs,l(µ)γc,l(µ) ≥ Rmin ∀l The resulting operating point is sensing-dominant, but not sensing-exclusive, since a non-negligible fraction of the reference energy is always retained for communication This optimal allocation is determined by balancing the average BER and delaybased localisation RMSE over a representative signal-to-noise ratio (SNR) set The resulting µ⋆ is thereafter treated as a fixed, scenario-specific design parameter rather than an adaptive quantity updated during transmission
The CE accuracy is quantified through the average normalised mean squared error (NMSE), defined as NMSE = 1/NMCJvN Σζ=1 XMCζ=1 Xj∈Jv hb(ζ)j,fr − h(ζ)j / h(ζ)j2 SA-OFDM-IM achieves a lower NMSE than DCO-OFDM because of reduced noise variance obtained by averaging the pilotbased CFR estimates across repeated observations The analysis further reveals that unresolved NLoS components can bias the extracted LoS gain The development of dedicated multipathsuppression mechanisms is therefore left as a direction for future work, alongside the extension of the anchor-blockage conditions
In a dispersive multipath channel, the estimated CFR is a superposition of the LoS component and NLoS reflections, each characterised by a gain and delay tap The leakage coefficient Λρ,j = 1/N+ Σn∈N+ e(-ȷ2πn(λρ,j−λb0,j)) N quantifies the residual, uncompensated phase rotation retained by the ρ-th NLoS component once phase compensation has been applied When an NLoS component is strong, unresolved, or falls within or adjacent to the delay bin of the detected LoS tap, i.e., dρ,j − d0,j ≲ ∆d Any such unresolved or closely spaced reflection therefore biases αb0,j, degrading RSS-based localisation accuracy since the distorted gain propagates through the RSS model into erroneous distance estimates The development of dedicated multipathsuppression mechanisms is therefore left as a direction for future work, alongside the extension of the anchor-blockage conditions
The paper proposes sensing-assisted OFDM with index modulation (SA-OFDM-IM) for joint communication and localisation over optical wireless systems This paper is 2610.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Adaptive sensing-to-communication energy allocation framework can minimize average bit-error rate (BER) and delay-based localisation root mean square error (RMSE) jointly under a minimum power-domain margin. This enables low-complexity, order-statistics based subcarrier detection of both sensing and communication subcarriers without exhaustive maximum likelihood search.
-
The receiver signal processing chain can extract both communication and sensing observables by performing energy ordered subcarrier detection, pilot aided least squares (LS) channel estimation, frame level coherent channel response aggregation, and delay domain interpolation. This allows for the joint recovery of index- and constellation domain information for communication symbols.
-
A delay aided RSS based localisation method can be introduced where the LoS delay supplies a multilateration based initial position estimate that both initializes and safeguards the subsequent refinement by enforcing amplitude consistency. This method can jointly estimate the PD position and, when unknown, its surface normal vector.
-
The system can achieve sub-delay bin range accuracy by exploiting variations in received optical power through the delay aided RSS refinement stage, which removes SNR independent error floors of delay only and hybrid RSS benchmarks to attain sub-delay-bin range and position accuracy.
-
The proposed SAOFDM IM waveform can achieve up to 5 dB lower SNR requirement in BER and up to three orders of magnitude lower channel estimation error than DC biased optical OFDM while the RSS estimator removes the SNR independent error floors of delay only and hybrid RSS benchmarks to attain sub delay bin range and position accuracy.
Abstract
This paper proposes sensing-assisted OFDM with index modulation (SA-OFDM-IM) for joint communication and localisation over optical wireless systems. In SA-OFDM-IM, a multilayer frequency-domain architecture partitions each layer's admissible subcarriers into disjoint communication and distributed sensing-pilot subsets, applying group-wise index modulation (IM) within the communication subset. A constrained sensing-to-communication energy allocation jointly minimises average bit-error rate (BER) and delay-based localisation root mean square error (RMSE) under a minimum power-domain margin, enabling low-complexity, ordered-energy detection of both sensing and communication subcarriers without exhaustive maximum-likelihood search. Sensing tones are aggregated across the frame to reconstruct the channel frequency response, yielding the line-of-sight (LoS) delay tap and phase-compensated LoS gain for multilateration and, via the proposed delay-aided received signal strength (RSS) refinement. Simulation results indicate that SA-OFDM-IM can achieve up to 5 dB lower SNR requirement in BER and up to three orders of magnitude lower channel-estimation error than DC-biased optical OFDM, while the RSS estimator removes the SNR-independent error floors of delay-only and hybrid-RSS benchmarks to attain sub-delay-bin range and position accuracy, remaining robust under random known and unknown receiver orientations.
Related papers
- Open-H-RF: A Large-Scale Ultrasound Raw Channel Dataset Initiative
- Scalable Attention for 5G NR Channel Estimation
- Low-Altitude Wireless Networks: A Comprehensive Survey
- Iteration-Free Candidate Pruning and an Index-Domain VLSI Architecture for Sort-Free MIMO Box Decoding up to 1024-QAM
- A Foundation Model for Large-Scale Wireless Network Planning, Operation and Optimization
- A Survey on Stacked Intelligent Metasurfaces: Fundamentals, Recent Advances, and Challenges