References & Further Reading
References
- H. El Gamal, G. Caire, and M. O. Damen, The MIMO ARQ channel: Diversity-multiplexing-delay tradeoff, 2006
THE foundational paper of this chapter. El Gamal, Caire, and Damen generalise the Zheng-Tse DMT to MIMO channels with ARQ feedback, introducing the delay dimension $L$ and proving the closed-form tradeoff $d_\mathrm{ARQ}(r, L) = L \cdot d^{*}(r/L)$. The paper also constructs the first explicit ARQ-DMT-optimal code family — Incremental-Redundancy Lattice Space-Time (IR-LAST) codes — and proves their optimality via MMSE-GDFE lattice decoding with common random dithering. Cited in every 3GPP contribution on HARQ design.
- L. Zheng and D. N. C. Tse, Diversity and multiplexing: A fundamental tradeoff in multiple-antenna channels, 2003
The static DMT theorem of Ch. 12 that this chapter generalises to the ARQ setting. The Wishart large-deviations machinery and Gaussian random-coding achievability argument used here are direct lifts of Zheng-Tse's proof pattern. Backward reference throughout.
- H. El Gamal, G. Caire, and M. O. Damen, Lattice coding and decoding achieve the optimal diversity-multiplexing tradeoff of MIMO channels, 2004
The static-DMT LAST code paper, providing the lattice-decoder (MMSE-GDFE) machinery invoked in Thm. <a href="#thm-ir-last-achieves-arq-dmt" class="ferkans-ref" title="Theorem: IR-LAST Codes Achieve the ARQ-DMT" data-ref-type="theorem"><span class="ferkans-ref-badge">T</span>IR-LAST Codes Achieve the ARQ-DMT</a>. Forward reference to Ch. 17 (LAST codes). The 2006 IR-LAST paper is a direct extension of this 2004 result to the ARQ setting.
- G. Caire and D. Tuninetti, The throughput of hybrid-ARQ protocols for the Gaussian collision channel, 2001
Pre-DMT analysis of HARQ throughput over Gaussian / collision channels. Established the formula $\eta_\mathrm{eff} = R \cdot (1 - P_\mathrm{out}(LR, L)) / \mathbb{E}[\mathrm{rounds}]$ used in §4 and Ex. 12. Sets up the IR-HARQ analysis that El Gamal-Caire-Damen later lift to the MIMO block-fading setting.
- J. M. Wozencraft and I. M. Jacobs, Principles of Communication Engineering, John Wiley & Sons, 1965
The original formalisation of ARQ protocols (stop-and-wait, go-back-N, selective-repeat) in a communications textbook. Pre-dates the concept of diversity gain on fading channels; the 1961 lecture notes on which this textbook is based treated ARQ purely as a throughput-reliability tradeoff on AWGN. Historical context for §1.
- D. Chase, Code combining—A maximum-likelihood decoding approach for combining an arbitrary number of noisy packets, 1985
The Chase-combining paper. Proves that summing LLRs across repetitions is asymptotically the maximum-likelihood decoder for the repeated-packet ensemble on a memoryless channel. The name "Chase combining" honours this 1985 paper. Theoretical baseline for the CC-HARQ diversity analysis in Thm. <a href="#thm-cc-limited-diversity" class="ferkans-ref" title="Theorem: Chase Combining Achieves at Most $L \cdot d^{*}(r)$ Diversity" data-ref-type="theorem"><span class="ferkans-ref-badge">T</span>Chase Combining Achieves at Most $L \cdot d^{*}(r)$ Diversity</a>.
- D. M. Mandelbaum, An adaptive-feedback coding scheme using incremental redundancy, 1974
The earliest formulation of incremental redundancy as an ARQ mechanism. Mandelbaum proposed transmitting parity in chunks (rather than repeating the whole codeword) and decoding when enough parity accumulated. The engineering implementation in LTE/NR traces back to this 1974 correspondence.
- 3GPP, NR; Multiplexing and channel coding, 2022. [Link]
5G NR specification for channel coding, including the LDPC mother code, the circular-buffer rate matcher, and the redundancy-version definitions RV_0 through RV_3. §5.4 is the primary reference for the rate-matching / HARQ interface.
- 3GPP, NR; Physical layer procedures for data, 2022. [Link]
5G NR specification for the HARQ protocol and scheduling: 16 parallel HARQ processes, per-numerology RTT tables, and the URLLC mini-slot HARQ design. §5.1, §5.3, and §11 are the primary references for the engineering notes in §5.
- 3GPP, NR; Physical layer procedures for control, 2022. [Link]
5G NR specification for the uplink control channel (PUCCH) that carries HARQ ACK/NACK feedback. §9.2 describes the PUCCH formats and the silent-failure (NACK-to-ACK) rate budget referenced in <a href="#pitfall-nr-ack-nack-error" class="ferkans-ref" title="Common Mistake: ACK/NACK Is Not Error-Free" data-ref-type="pitfall"><span class="ferkans-ref-badge">⚠</span>ACK/NACK Is Not Error-Free</a>.
- D. Tuninetti, On the benefits of partial channel state information for repetition protocols in block-fading channels, 2011
Extends the ARQ-DMT to settings with partial CSI at the transmitter (e.g., imperfect CSIT via feedback). Quantifies when the ARQ-DMT prediction of $L$-fold gain is achievable under realistic feedback quality. Relevant to the correlated- round pitfall of §1.
- B. Liu and V. K. N. Lau, Design principles of HARQ protocols, 2016
Accessible magazine-style overview of HARQ protocol design, covering CC vs IR tradeoffs, redundancy-version selection, and finite-length corrections. Useful pedagogical companion to §4 of this chapter.
- İ. E. Telatar, Capacity of multi-antenna Gaussian channels, 1999
The original MIMO ergodic-capacity paper. At $L \to \infty$, the ARQ-DMT reduces to the Telatar ergodic-capacity regime (Remark <a href="#rmk-s02-ergodic-limit" class="ferkans-ref" title="Remark: The $L \to \infty$ Limit: Ergodic Capacity Regime" data-ref-type="remark"><span class="ferkans-ref-badge">R</span>The $L \to \infty$ Limit: Ergodic Capacity Regime</a>). Backward reference to Ch. 10.
- G. Caire, G. Taricco, and E. Biglieri, Optimum power control over fading channels, 1999
Earlier CommIT-group work on adaptive-rate / adaptive-power policies over fading channels. The water-filling power allocation analysed there is a simpler form of the rate-adaptation that appears implicitly in ARQ-DMT (each round adapts the effective rate to the realised channel).
- P. Elia, K. R. Kumar, S. A. Pawar, P. V. Kumar, H.-f. Lu, and G. Caire, Explicit space-time codes achieving the diversity-multiplexing gain tradeoff, 2006
The CDA-based explicit static-DMT construction (Ch. 13 of this book) that provides the nonvanishing-determinant building block for IR-LAST codes. The IR-LAST construction of §3 is a temporal extension of the CDA codes of this 2006 companion paper.
- D. Tse and P. Viswanath, Fundamentals of Wireless Communication, Cambridge University Press, 2005
The definitive wireless-communications textbook. Chapter 9 covers the DMT and its extensions including a brief mention of the ARQ-DMT. The treatment complements the information- theoretic emphasis of this chapter with a more practical perspective.
- D. N. C. Tse, P. Viswanath, and L. Zheng, Diversity-multiplexing tradeoff in multiple-access channels, 2004
Extension of the Zheng-Tse DMT to the MIMO multiple-access channel. Parallel development to the ARQ-DMT: both add a structural dimension (ARQ round budget / user index) to the static DMT framework.
Further Reading
For readers interested in deeper or related aspects of ARQ coding for MIMO and wireless standards.
Finite-blocklength ARQ — non-asymptotic corrections
F. Gomez-Cuba, J. Du, M. Medard, and E. Erkip, "Unified capacity limit of non-coherent wideband fading channels," IEEE Trans. Inform. Theory, vol. 63, no. 1, pp. 43–68, Jan. 2017.
Extends the finite-blocklength information-theory machinery (Polyanskiy-Poor-Verdú) to ARQ settings. Quantifies the $O(1/\sqrt{N})$ penalty of finite per-round blocklength on the effective ARQ-DMT. Essential for URLLC design where per-round blocks are 2–7 OFDM symbols.
ARQ with delayed CSI feedback and rate-adaptive schedulers
P. Wu and N. Jindal, "Performance of hybrid-ARQ in block-fading channels: A fixed outage probability analysis," IEEE Trans. Commun., vol. 58, no. 4, pp. 1129–1141, Apr. 2010.
Analyses practical HARQ schedulers that adapt the per-round MCS based on observed outages, rather than the fixed-rate model assumed in the ARQ-DMT. Shows that adaptive-MCS gains can partially compensate for finite-SNR gaps, essentially realising "link adaptation" on top of HARQ.
Multi-user ARQ: broadcast and multiple-access with feedback
S. Pfletschinger and M. Navarro, "Throughput analysis of HARQ protocols with adaptive modulation and coding," IEEE Trans. Wireless Commun., vol. 16, no. 3, pp. 1528–1540, Mar. 2017.
Extends the throughput-latency tradeoff of §4 to multi-user downlink with MCS adaptation. Relevant for cellular scheduling where many UEs share the same HARQ pipeline.
HARQ for mmWave and NR FR2
R. Kovalchukov, D. Moltchanov, A. Samuylov, A. Ometov, S. Andreev, Y. Koucheryavy, and K. Samouylov, "Evaluating SIR in 3D millimeter-wave deployments: Direct modeling and feasible approximations," IEEE Trans. Wireless Commun., vol. 18, no. 2, pp. 879–896, Feb. 2019.
Discusses the mmWave-specific challenges of HARQ: short coherence time, high blockage rates, and the resulting need for aggressive frequency hopping across retransmissions. Ties the ARQ-DMT analysis to NR FR2 deployments.
URLLC design philosophy: grant-free, repetition, and duplication
C. Bockelmann, N. Pratas, H. Nikopour, K. Au, T. Svensson, C. Stefanović, P. Popovski, and A. Dekorsy, "Massive machine-type communications in 5G: Physical and MAC-layer solutions," IEEE Commun. Mag., vol. 54, no. 9, pp. 59–65, Sept. 2016.
Overview of URLLC and mMTC design. Explains why URLLC often avoids HARQ entirely in favour of one-shot low-rate codes with $K$-repetition — a choice directly motivated by the latency-reliability tradeoff of §5.