pith. sign in

arxiv: 2510.14939 · v2 · submitted 2025-10-16 · 📡 eess.SP

Decoding in the presence of ISI without interleaving -- ORBGRAND-AI

Ken R. Duffy , Moritz Grundei , Jane A. Millward , Muralidhar Rangaswamy , Muriel Medard This is my paper

Pith reviewed 2026-05-18 06:05 UTC · model grok-4.3

classification 📡 eess.SP
keywords inter-symbol interferencecolored noiseORBGRAND-AIdecodingblock error rateRFView channeldicode channelapproximate independence
0
0 comments X

The pith

ORBGRAND-AI achieves the same or better block error rates in inter-symbol interference channels without using interleaving.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces ORBGRAND-AI, a decoder that handles colored noise from equalization in ISI channels by leveraging approximate independence ideas from statistical physics. It demonstrates that this approach can match or surpass the performance of advanced soft decoders like CA-SCL that rely on interleavers. Readers should care because skipping interleaving reduces delay and system complexity in communication setups such as wireless links or storage devices. The work tests the idea on a simple two-tap dicode channel and on channels modeled from RFView data under imperfect channel state information.

Core claim

ORBGRAND-AI operates without the need for turbo equalization or interleaving by adapting guessing random additive noise decoding to colored noise via approximate independence. In ISI channels, it delivers the same or lower block error rate for the same energy per information bit compared to CA-SCL decoding with an interleaver.

What carries the argument

ORBGRAND-AI, Ordered Reliability Bits Guessing Random Additive Noise Decoding adapted for Approximate Independence to handle noise correlations after equalization.

If this is right

  • Systems can remove the latency and complexity of interleavers in ISI environments while preserving error performance.
  • The decoder works across delay-tap models and RFView-derived channels even with imperfect channel state information.
  • A second-order autoregressive model is adequate to capture the noise coloring effects from the RFView channel for decoding.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Lower overall system latency could result in real-time applications over dispersive channels.
  • The technique may apply to other forms of correlated noise beyond the autoregressive cases examined.

Load-bearing premise

The approximate independence concept applies sufficiently to the colored noise produced by equalization in the tested ISI channels.

What would settle it

Simulations on the dicode or RFView channels showing higher block error rate for ORBGRAND-AI than for CA-SCL with an interleaver at the same energy per information bit would disprove the central claim.

Figures

Figures reproduced from arXiv: 2510.14939 by Jane A. Millward, Ken R. Duffy, Moritz Grundei, Muralidhar Rangaswamy, Muriel Medard.

Figure 1
Figure 1. Figure 1: Signal processing chain of a bit interleaved communication system [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Illustration of RFView dataset for a single CPI. We [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Ranked normalized inverse block substitution likeli [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 5
Figure 5. Figure 5: The impact of channel correlation on ORBGRAND [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: The impact of block-size, b, on BLER performance of ORBGRAND-AI for ρ = 0.5 and an [128,116] RLC in the equalized dicode channel. checking circuits have been built [28], [32], [35]. Thus, com￾plexity for GRAND algorithms is generally compared by the average number of codebook queries until a decoding is found as this operation dominates the total energy consumption of the circuits [PITH_FULL_IMAGE:figures… view at source ↗
Figure 7
Figure 7. Figure 7: [256,240+11] 5G NR Uplink CA-Polar code with 11 [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Number of codebook queries it takes ORBGRAND-AI [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Comparison of BLER for different block sizes, [PITH_FULL_IMAGE:figures/full_fig_p010_9.png] view at source ↗
Figure 11
Figure 11. Figure 11: ORBRAND-AI’s sensitivity to a CSI quantization [PITH_FULL_IMAGE:figures/full_fig_p010_11.png] view at source ↗
Figure 13
Figure 13. Figure 13: Comparison of BLER for different block sizes, [PITH_FULL_IMAGE:figures/full_fig_p011_13.png] view at source ↗
Figure 12
Figure 12. Figure 12: Comparison of BLER for different block sizes, [PITH_FULL_IMAGE:figures/full_fig_p011_12.png] view at source ↗
read the original abstract

Inter symbol interference (ISI), which occurs in a wide variety of channels, is a result of time dispersion. It can be mitigated by equalization, which results in noise coloring. Inspired by the development of Approximate Independence in statistical physics, for such colored noise we propose a decoder called Ordered Reliability Bits Guessing Random Additive Noise Decoding (ORBGRAND-AI) that operates without the need for turbo equalization or interleaving. By foregoing interleaving, ORBGRAND-AI can deliver the same, or lower, block error rate (BLER) for the same amount of energy per information bit in an ISI channel as a state-of-the-art soft input decoder, such as Cyclic Redundancy Check Assisted-Successive Cancellation List (CA-SCL) decoding, with an interleaver. To assess the decoding performance of ORBGRAND-AI, we consider delay tap models and their associated colored noise. In particular, we examine a two-tap dicode ISI channel as well as an ISI channel derived from data from RFView, a physics-informed modeling and simulation tool. We investigate the dicode and RFView channel under a variety of imperfect channel state information assumptions and show that a second order autoregressive model adequately represents the RFView channel effect.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 3 minor

Summary. The manuscript proposes ORBGRAND-AI, an extension of Ordered Reliability Bits Guessing Random Additive Noise Decoding that incorporates approximate independence concepts from statistical physics to handle colored noise after MMSE equalization in ISI channels. It claims that this decoder, operating without interleaving or turbo equalization, achieves the same or lower block error rate (BLER) at equivalent Eb/N0 compared to CA-SCL decoding with an interleaver. The evaluation uses a two-tap dicode channel and an ISI channel derived from RFView data, with tests under imperfect CSI and a fitted second-order autoregressive noise model.

Significance. If the performance claims hold, the work could simplify receiver design for dispersive channels by removing interleaving, which often adds latency and complexity. The application of statistical-physics approximate independence to post-equalization noise and the use of physics-informed RFView simulations with imperfect CSI represent a practical extension of the GRAND decoder family. Credit is given for reproducible simulation-based comparisons on realistic channel models.

major comments (3)
  1. [Simulation results section] Simulation results section: The BLER performance curves comparing ORBGRAND-AI to interleaved CA-SCL lack error bars, confidence intervals, or the number of Monte Carlo trials, which is load-bearing for assessing whether the claimed parity or advantage at specific Eb/N0 points is statistically reliable.
  2. [RFView channel modeling subsection] RFView channel modeling subsection: The claim that a second-order autoregressive model adequately represents the RFView channel effects under imperfect CSI is asserted without quantitative fit metrics (e.g., residual autocorrelation or prediction error), which directly affects validation of the approximate independence assumption for colored noise.
  3. [Decoder algorithm description] Decoder algorithm description: No explicit equations, pseudocode, or step-by-step adaptation is provided showing how the ORBGRAND guessing procedure is modified to exploit approximate independence for the post-equalization noise correlation, which is central to the method's claimed operation without interleaving.
minor comments (3)
  1. [Abstract and introduction] The abstract and introduction refer to 'a variety of imperfect channel state information assumptions' but do not tabulate the specific error variances or models used in the experiments.
  2. [Figure captions] Figure captions for BLER plots should specify code length, rate, and interleaver parameters for the CA-SCL baseline to improve reproducibility.
  3. [System model section] Notation for the noise correlation matrix after equalization could be clarified with an explicit definition in the system model section.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their constructive and detailed review of our manuscript. The comments highlight opportunities to improve statistical rigor, quantitative validation, and algorithmic clarity. We address each major comment below and have incorporated revisions to strengthen the paper.

read point-by-point responses

  1. Referee: [Simulation results section] Simulation results section: The BLER performance curves comparing ORBGRAND-AI to interleaved CA-SCL lack error bars, confidence intervals, or the number of Monte Carlo trials, which is load-bearing for assessing whether the claimed parity or advantage at specific Eb/N0 points is statistically reliable.

    Authors: We agree that explicit reporting of Monte Carlo trial counts and error bars would better substantiate the statistical reliability of the BLER comparisons. In the revised manuscript, we have added the number of trials used for each Eb/N0 point and included error bars (representing standard error) on the performance curves in the simulation results section. revision: yes

  2. Referee: [RFView channel modeling subsection] RFView channel modeling subsection: The claim that a second-order autoregressive model adequately represents the RFView channel effects under imperfect CSI is asserted without quantitative fit metrics (e.g., residual autocorrelation or prediction error), which directly affects validation of the approximate independence assumption for colored noise.

    Authors: The referee is correct that quantitative fit metrics are needed to support the model adequacy claim and the approximate independence assumption. We have revised the RFView channel modeling subsection to include residual autocorrelation analysis and prediction error metrics for the second-order autoregressive fit under imperfect CSI conditions. revision: yes

  3. Referee: [Decoder algorithm description] Decoder algorithm description: No explicit equations, pseudocode, or step-by-step adaptation is provided showing how the ORBGRAND guessing procedure is modified to exploit approximate independence for the post-equalization noise correlation, which is central to the method's claimed operation without interleaving.

    Authors: We acknowledge that the adaptation of the ORBGRAND procedure for approximate independence could be described more explicitly. In the revised manuscript, we have added equations and pseudocode in the decoder algorithm description section that detail the modifications to the guessing procedure to account for the post-equalization noise correlation structure. revision: yes

Circularity Check

0 steps flagged

No circularity: performance claims rest on simulation comparisons using external channel models and statistical-physics inspiration

full rationale

The paper's central claim is an empirical performance comparison (ORBGRAND-AI BLER vs. interleaved CA-SCL at equal Eb/N0) obtained via Monte-Carlo simulation on two-tap dicode and RFView-derived ISI channels under imperfect CSI. The decoder itself is constructed by adapting the existing GRAND framework with an approximate-independence heuristic drawn from external statistical-physics literature; no equation or parameter is defined in terms of the target BLER result, no fitted quantity is relabeled as a prediction, and no load-bearing step reduces to a self-citation chain. The AR(2) model is presented as an adequate empirical representation of the RFView data rather than a derived necessity, keeping the derivation self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim rests on the applicability of approximate independence to equalized ISI noise and on the adequacy of a low-order autoregressive model for the RFView channel; these are stated as modeling choices rather than derived results.

axioms (1)
  • domain assumption Approximate independence from statistical physics applies to the colored noise after equalization in ISI channels
    Invoked to justify operating the decoder directly on colored noise without interleaving or turbo equalization.
invented entities (1)
  • ORBGRAND-AI decoder no independent evidence
    purpose: Decode in the presence of ISI without interleaving by guessing noise in reliability order
    New method introduced in the paper; performance is demonstrated only via the paper's own simulations.

pith-pipeline@v0.9.0 · 5771 in / 1448 out tokens · 55369 ms · 2026-05-18T06:05:20.395628+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

37 extracted references · 37 canonical work pages

  1. [1]

    Genera- tive Diffusion Models for Radio Wireless Channel Modelling and Sampling

    K. R. Duffy, M. Grundei, and M. M ´edard, “Using Channel Correlation to Improve Decoding – ORBGRAND-AI,” inIEEE Globecom, 2023, doi: 10.1109/GLOBECOM54140.2023.10437763

    work page doi:10.1109/globecom54140.2023.10437763 2023

  2. [2]

    Using Equalization-Induced Noise Coloring to Improve Error- Correcting Decoding,

    J. A. Millward, K. R. Duffy, M. Rangaswamy, and M. M ´edard, “Using Equalization-Induced Noise Coloring to Improve Error- Correcting Decoding,” in58th Asilomar Conference on Signals, Systems, and Computers. IEEE, 2024, pp. 1400–1403, doi: 10.1109/IEEECONF60004.2024.10943085

    work page doi:10.1109/ieeeconf60004.2024.10943085 2024

  3. [3]

    Michon, C

    ——, “Enhancing Guessing Random Additive Noise Decoding using Channel Estimation,” inIEEE International Workshop on Signal Pro- cessing Advances in Wireless Communications (SPA WC), 2024, pp. 676– 680, doi: 10.1109/SPAWC60668.2024.10694171

    work page doi:10.1109/spawc60668.2024.10694171 2024

  4. [4]

    R. G. Gallager,Information Theory and Reliable Communication. New York, NY , USA: John Wiley & Sons, Inc., 1968

    work page 1968

  5. [5]

    T. M. Cover and J. A. Thomas,Elements of Information Theory. John Wiley & Sons, 1991

    work page 1991

  6. [6]

    High Fidelity RF Clutter Modeling and Simulation,

    S. Gogineni, J. R. Guerci, H. K. Nguyen, J. S. Bergin, D. R. Kirk, B. C. Watson, and M. Rangaswamy, “High Fidelity RF Clutter Modeling and Simulation,”IEEE Aerospace and Electronic Systems Magazine, vol. 37, no. 11, pp. 24–43, 2022, doi: 10.1109/MAES.2022.3208862

    work page doi:10.1109/maes.2022.3208862 2022

  7. [7]

    Lin and D

    S. Lin and D. J. Costello,Error Control Coding: Fundamentals and Applications. Pearson/Prentice Hall, 2004

    work page 2004

  8. [8]

    Capacity-Achieving Guessing Random Additive Noise decoding,

    K. R. Duffy, J. Li, and M. M ´edard, “Capacity-Achieving Guessing Random Additive Noise decoding,”IEEE Trans. Inf. Theory, vol. 65, no. 7, pp. 4023–4040, 2019, doi: 10.1109/TIT.2019.2896110

    work page doi:10.1109/tit.2019.2896110 2019

  9. [10]

    Keep the Bursts and Ditch the Interleavers,

    W. An, M. M ´edard, and K. R. Duffy, “Keep the Bursts and Ditch the Interleavers,”IEEE Trans. Commun., vol. 70, pp. 3655–3667, 2022, doi: 10.1109/TCOMM.2022.3171798

    work page doi:10.1109/tcomm.2022.3171798 2022

  10. [11]

    High-Throughput VLSI Architecture for GRAND Markov Order,

    S. M. Abbas, M. Jalaleddine, and W. J. Gross, “High-Throughput VLSI Architecture for GRAND Markov Order,” inIEEE Workshop Sig. Proc. Sys., 2021

    work page 2021

  11. [12]

    Coding Techniques for Backscatter Communications—A Contemporary Sur- vey,

    F. Rezaei, D. Galappaththige, C. Tellambura, and S. Herath, “Coding Techniques for Backscatter Communications—A Contemporary Sur- vey,”IEEE Communications Surveys & Tutorials, vol. 25, no. 2, pp. 1020–1058, 2023

    work page 2023

  12. [13]

    Upper bounds on the natarajan dimensions of some function classes

    W. An, M. Medard, and K. R. Duffy, “Soft Decoding without Soft Demapping with ORBGRAND,” inIEEE ISIT, 2023, pp. 1080–1084, doi: 10.1109/ISIT54713.2023.10206762

    work page doi:10.1109/isit54713.2023.10206762 2023

  13. [14]

    Specific Information Gain for Interacting Markov Processes,

    W. G. Sullivan, “Specific Information Gain for Interacting Markov Processes,”Probab. Theory Relat. Fields, vol. 37, no. 1, pp. 77–90, 1976, doi: 10.1007/BF00536299

    work page doi:10.1007/bf00536299 1976

  14. [15]

    Entropy, Concentration of Probability and Conditional Limit Theorems,

    J. T. Lewis, C.-E. Pfister, and W. G. Sullivan, “Entropy, Concentration of Probability and Conditional Limit Theorems,”Markov Process. Relat. Fields, vol. 1, pp. 319–386, 1995

    work page 1995

  15. [16]

    Thermodynamical Aspects of Classical Lattice Systems,

    C.-E. Pfister, “Thermodynamical Aspects of Classical Lattice Systems,” Prog. Probab., pp. 393–472, 2002

    work page 2002

  16. [17]

    Large Deviations Estimates for Dy- namical Systems without the Specification Property. Application to the β-shifts,

    C.-E. Pfister and W. G. Sullivan, “Large Deviations Estimates for Dy- namical Systems without the Specification Property. Application to the β-shifts,”Nonlinearity, vol. 18, no. 1, p. 237, 2004, doi: 10.1088/0951- 7715/18/1/013

    work page doi:10.1088/0951- 2004

  17. [18]

    Burst-Correcting Codes for the Classic Bursty Channel,

    G. Forney, “Burst-Correcting Codes for the Classic Bursty Channel,” IEEE Trans. Commun. Technol., vol. 19, no. 5, pp. 772–781, 1971, doi: 10.1109/TCOM.1971.1090719

    work page doi:10.1109/tcom.1971.1090719 1971

  18. [19]

    Van Nee and R

    R. Van Nee and R. Prasad,OFDM for Wireless Multimedia Communi- cations. Artech House, Inc., 2000

    work page 2000

  19. [20]

    Frequency Domain Data Transmission using Reduced Computational Complexity Algorithms,

    A. Peled and A. Ruiz, “Frequency Domain Data Transmission using Reduced Computational Complexity Algorithms,” inIEEE ICASSP, vol. 5, 1980, pp. 964–967, doi: 10.1109/ICASSP.1980.1171076

    work page doi:10.1109/icassp.1980.1171076 1980

  20. [21]

    The Cyclic Prefix of OFDM/DMT-An Analysis,

    W. Henkel, G. Taubock, P. Odling, P. O. Borjesson, and N. Petersson, “The Cyclic Prefix of OFDM/DMT-An Analysis,” inInternational Zurich Seminar on Broadband Communications Access-Transmission- Networking, 2002, pp. 22–22, doi: 10.1109/IZSBC.2002.991762

    work page doi:10.1109/izsbc.2002.991762 2002

  21. [22]

    Tomlinson-Harashima Precoding in Space-Time Transmission for Low- Rate Backward Channel,

    R. F. Fischer, C. Windpassinger, A. Lampe, and J. B. Huber, “Tomlinson-Harashima Precoding in Space-Time Transmission for Low- Rate Backward Channel,” in2002 International Zurich Seminar on Broadband Communications Access-Transmission-Networking (Cat. No. 02TH8599). IEEE, 2002, pp. 7–7, 10.1109/IZSBC.2002.991747

    work page doi:10.1109/izsbc.2002.991747 2002

  22. [23]

    Decision-Feedback Equalization for Digital Communi- cation over Dispersive Channels,

    M. E. Austin, “Decision-Feedback Equalization for Digital Communi- cation over Dispersive Channels,” 1967

    work page 1967

  23. [24]

    B. D. Anderson and J. B. Moore,Optimal Filtering. Courier Corpora- tion, 2005

    work page 2005

  24. [25]

    An Information-Theoretic Proof of Burg’s Maximum Entropy Spectrum,

    B. Choi and T. M. Cover, “An Information-Theoretic Proof of Burg’s Maximum Entropy Spectrum,”Prof. IEEE, vol. 72, no. 8, pp. 1094– 1096, 1984, doi: 10.1109/PROC.1984.12981

    work page doi:10.1109/proc.1984.12981 1984

  25. [26]

    R. H. Shumway, D. S. Stoffer, and D. S. Stoffer,Time Series Analysis and its Applications, 4th ed. Springer, 2017

    work page 2017

  26. [27]

    A General Formula for Channel Capacity,

    S. Verd ´u and T. S. Han, “A General Formula for Channel Capacity,” IEEE Transactions on Information Theory, vol. 40, no. 4, pp. 1147– 1157, 1994, doi: 10.1109/18.335960

    work page doi:10.1109/18.335960 1994

  27. [28]

    Detection of post-traumatic stress disorder using learned time-frequency representations from pupillometry,

    K. R. Duffy, W. An, and M. M ´edard, “Ordered Reliability Bits Guessing Random Additive Noise Decoding,”IEEE Trans. Signal Process., vol. 70, pp. 4528–4542, 2022, doi: 10.1109/ICASSP39728.2021.9414615

    work page doi:10.1109/icassp39728.2021.9414615 2022

  28. [29]

    A sub-0.8-pj/bit Universal Soft- Detection Decoder Using ORBGRAND,

    A. Riaz, A. Yasar, F. Ercan, W. An, J. Ngo, K. Galligan, M. M ´edard, K. R. Duffy, and R. T. Yazicigil, “A sub-0.8-pj/bit Universal Soft- Detection Decoder Using ORBGRAND,”IEEE Journal of Solid-State Circuits, 2024, doi: 10.1109/JSSC.2024.3502240

    work page doi:10.1109/jssc.2024.3502240 2024

  29. [30]

    High-Throughput and Energy-Efficient VLSI Architecture for Ordered Reliability Bits GRAND,

    S. M. Abbas, T. Tonnellier, F. Ercan, M. Jalaleddine, and W. J. Gross, “High-Throughput and Energy-Efficient VLSI Architecture for Ordered Reliability Bits GRAND,”IEEE Trans. V ery Large Scale Integr . Syst., vol. 30, no. 6, pp. 681–693, 2022, doi: 10.1109/TVLSI.2022.3153605

    work page doi:10.1109/tvlsi.2022.3153605 2022

  30. [31]

    A Fixed Latency ORBGRAND Decoder Architecture with LUT-Aided Error-Pattern Scheduling,

    C. Condo, “A Fixed Latency ORBGRAND Decoder Architecture with LUT-Aided Error-Pattern Scheduling,”IEEE Trans. Circuits Syst. I Regul. Pap., 2022, doi: 10.1109/TCSI.2022.3150583

    work page doi:10.1109/tcsi.2022.3150583 2022

  31. [32]

    A Sub-0.8pJ/b 16.3Gbps/mm 2 Uni- versal Soft-Detection Decoder Using ORBGRAND in 40nm CMOS,

    A. Riaz, A. Yasar, F. Ercan, W. An, J. Ngo, K. Galligan, M. M ´edard, K. R. Duffy, and R. T. Yazicigil, “A Sub-0.8pJ/b 16.3Gbps/mm 2 Uni- versal Soft-Detection Decoder Using ORBGRAND in 40nm CMOS,” in IEEE ISSCC, 2023, doi: 10.1109/ISSCC42615.2023.10067519

    work page doi:10.1109/isscc42615.2023.10067519 2023

  32. [33]

    A Low-Latency and Area- Efficient ORBGRAND Decoder for Polar Codes,

    J. Xiao, Y . Zhou, S. Song, and Z. Wang, “A Low-Latency and Area- Efficient ORBGRAND Decoder for Polar Codes,” inIEEE ICTC, 2023, pp. 10–15, doi: 10.1109/ICTC57116.2023.10154861

    work page doi:10.1109/ictc57116.2023.10154861 2023

  33. [34]

    J. G. Proakis,Digital Communications. NY , USA: McGraw-Hill, 2001

    work page 2001

  34. [35]

    A sub-0.8-pJ/bit Universal Soft- Detection Decoder Using ORBGRAND,

    A. Riaz, A. Yasar, F. Ercan, W. An, J. Ngo, K. Galligan, M. M ´edard, K. R. Duffy, and R. T. Yazicigil, “A sub-0.8-pJ/bit Universal Soft- Detection Decoder Using ORBGRAND,”IEEE J. Solid-State Circuits, to appear

    work page

  35. [36]

    Best CRC Polynomials,

    P. Koopman, “Best CRC Polynomials,” Available at https://users.ece. cmu.edu/∼koopman/crc/ (2025/05/31)

    work page 2025

  36. [37]

    Tse and P

    D. Tse and P. Viswanath,Fundamentals of Wireless Communication. Cambridge university press, 2005

    work page 2005

  37. [38]

    Channel Estima- tion Techniques Based on Pilot Arrangement in OFDM systems,

    S. Coleri, M. Ergen, A. Puri, and A. Bahai, “Channel Estima- tion Techniques Based on Pilot Arrangement in OFDM systems,” IEEE Trans Brodcast, vol. 48, no. 3, pp. 223–229, 2002, doi: 10.1109/TBC.2002.804034

    work page doi:10.1109/tbc.2002.804034 2002