Quantum Machine Learning for 6G Communication Networks: State-of-the-Art and Vision for the Future

Probably not a substantial claim: may not pass S3 (load-bearing).

A model-free distributed reinforcement learning method for power allocation is proposed, in which CSI and QoS indicators are exploited to adapt transmit power.

“In this regard, a model-free distributed reinforcement learning method for power allocation is proposed [79], in which Channel State Information (CSI) and QoS indicators are exploited to adapt the transmit power.”

• cites [79] 79. M. I. Tiwana, S. J. Nawaz, A. A. Ikram, and M. I. Tiwana, "Self-organizing networks: A packet scheduling approach for cover- age/capacity optimization in 4G networks using reinforcement learning," Elektronika ir Elektrotechnika, vol. 20, no. 9, pp. 59-64, 2014. sourceDisputed

  The Abstract and Introduction describe a reinforcement-learning agent at each eNB, but explicitly identify the adapted variable as the α-fair packet-scheduling parameter. The claimed power-allocation method is contrary to the paper’s stated contribution; the Simulation Scenario lists transmit power only as a simulation parameter. No page numbers are available in our copy. (LLM, conf=0.88)

  Checked by a reasoning model — may be wrong.

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

A DL based framework for optimization of downlink beamforming is proposed.

“In this context, a DL based framework for optimization of downlink beamforming is proposed in [92].”

• cites [92] 92. F. Saleem and S. Wyne, "WLAN-based indoor localization using neural networks," Journal of Electrical Engineering, vol. 67, no. 4, pp. 299-306, 2016. sourceDisputed

  The Introduction (page unavailable) says this paper proposes methods for WLAN indoor localization, including handling missing access points and interpolating location estimates. That account of what the paper proposes conflicts with the attribution of a downlink-beamforming optimization framework. (LLM, conf=0.87)

  Checked by a reasoning model — may be wrong.

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Genetic algorithms at the physical layer have been used for optimal antenna selection in MIMO systems, power control, and symbol detection in MIMO systems.

“The Genetic Algorithm (GA) at physical layer of communication systems has been used for optimal antenna selection in MIMO systems, power control, and symbol detection in MIMO systems in [80,81], [82], [83,84], respectively.”

• cites [80] 80. N. Y. S and D. Guo, "Deep reinforcement learning for distributed dynamic power allocation in wireless networks," arXiv preprint arXiv:1808.00490, 2018. sourceDisputed

  The citation to this paper is attached to the power-control example, not the antenna-selection or symbol-detection examples. The Introduction (p. 1) identifies its approach as deep reinforcement learning for power control, and the Conclusion (p. 11) describes a deep Q-learning power-allocation algorithm—not a genetic algorithm. (LLM, conf=0.88)

  Checked by a reasoning model — may be wrong.

  Experimental — accuracy not yet measured
• cites [81] 81. J. Lain, "Joint transmit/receive antenna selection for MIMO systems: A real-valued genetic approach," IEEE Commun. Letters, vol. 15, no. 1, pp. 58-60, Jan. 2011. sourceNot checked — no full text of the source
• cites [82] 82. P. D. Karamalis, N. D. Skentos, and A. G. Kanatas, "Selecting array con- figurations for MIMO systems: an evolutionary computation approach," IEEE Trans. Wireless Commun., vol. 3, no. 6, pp. 1994-1998, Nov. 2004. sourceNot checked — no full text of the source
• cites [83] 83. B.-K. Lee, H.-W. Chen, and B.-S. Chen, "Power control of cellular radio systems via robust smith prediction filter," IEEE Trans. Wireless Commun., vol. 3, no. 5, pp. 1822-1831, Sep. 2004. sourceNot checked — no full text of the source
• cites [84] 84. H. Y.Lu and W. H. Fang, "Joint receive antenna selection and symbol detection for mimo systems: a heterogeneous genetic approach," IEEE Commun. Letters, vol. 13, no. 2, pp. 97-99, Feb. 2009. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

5G networks are aimed at providing a 1000-fold increase in aggregate throughput and a 10-fold increase in individual link throughput relative to 4G wireless networks.

“The 5G networks are aimed at providing a 1000-fold increase in the aggregate throughput and a 10-fold increase in the individual link throughput relative to the 4 th Generation (4G) wireless networks [46].”

• cites [46] 46. J. G. Andrews, S. Buzzi, W. Choi, S. V. Hanly, A. Lozano, A. C. K. Soong, and J. C. Zhang, "What will 5G be?" IEEE J. Sel. Areas Com- mun., vol. 32, no. 6, pp. 1065-1082, Jun. 2014. sourcePartly supported

  The “Data Rate” section supports a roughly 1000-fold increase in aggregate data rate from 4G to 5G. It does not support the claimed 10-fold increase in individual-link throughput: it describes the 5G edge-rate goal as about 100-fold above 4G, and the same section says per-link rates will increase by about 100-fold. (LLM, conf=0.93)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

QC-assisted communications is envisioned to hold promise for achieving extremely high data rates and link security in future 6G and beyond communications.

“The QC-assisted communications is another new research area, which is envisioned to hold promise for achieving extremely high data rates and link security in future 6G and beyond communications [30].”

• cites [30] 30. P. Botsinis, D. Alanis, Z. Babar, H. Nguyen, D. Chandra, S. X. Ng, and L. Hanzo, "Quantum algorithms for wireless communications," IEEE Commun. Surveys Tuts., vol. PP, Nov. 2018. sourcePartly supported

  The introduction (‘The next generation of wireless communications,’ page not provided) says quantum computing is being combined with classical communications to pursue throughput targets. It does not establish that quantum-assisted communications will improve link security or specifically situate that promise in 6G and beyond. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

Massive MIMO systems and millimeter-wave spectrum are believed to have an important role in addressing the capacity demands of future communication networks.

“Massive Multiple-Input Multiple-Output (M-MIMO) systems and millimeter-Wave (mmWave) spectrum exploiting high spatial resolution and multi-gigahertz bandwidth, respectively, are believed to have an important role in addressing the capacity demands of future communication networks [12,22].”

• cites [12] 12. M. G. Kibria, K. Nguyen, G. P. Villardi, O. Zhao, K. Ishizu, and F. Ko- jima, "Big data analytics, machine learning, and artificial intelligence in next-generation wireless networks," IEEE Access, vol. 6, pp. 32 328- 32 338, May 2018. sourceCould not verify

  The paper’s Abstract and “Moore’s law and superconducting electronics” section discuss quantum machine learning and hybrid computers, not wireless-network capacity. Searches for massive MIMO, millimeter-wave spectrum, spatial resolution, and bandwidth found no passage addressing the attributed claim. (LLM, conf=0.77)

  Experimental — accuracy not yet measured
• cites [22] 22. F. Jameel, S. Wyne, S. J. Nawaz, and Z. Chang, "Propagation channels for mmWave vehicular communications: State-of-the-art and future research directions," IEEE Wireless Communications, vol. 26, no. 1, pp. 144-150, Feb. 2019. sourcePartly supported

  The Introduction (page unavailable) says mmWave offers the large bandwidth needed for high-data-rate vehicular communications, supporting that part of the claim. The paper discusses MIMO channel modeling and large antenna arrays, but the passages found do not say that massive MIMO's spatial resolution helps meet future network capacity demands. (LLM, conf=0.85)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

ML can potentially assist big data analytics to realize self-sustaining and proactive wireless networks.

“ML can potentially assist big data analytics to realize self-sustaining and proactive wireless networks [12].”

• cites [12] 12. M. G. Kibria, K. Nguyen, G. P. Villardi, O. Zhao, K. Ishizu, and F. Ko- jima, "Big data analytics, machine learning, and artificial intelligence in next-generation wireless networks," IEEE Access, vol. 6, pp. 32 328- 32 338, May 2018. sourceCould not verify

  The Abstract discusses potential uses of quantum machine learning, and Societal Implications discusses computerized agents and web traffic. Neither addresses big-data analytics for self-sustaining, proactive wireless networks; searches for those terms and their paraphrases found no supporting passage in our copy. (LLM, conf=0.75)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

The Tactile Internet allows for real-time interaction of humans and machines that supports haptic input with audiovisual feedback to control machine operation in real-time.

“The TI allows for a real-time interaction of humans and machines that supports haptic input with audiovisual feedback to control the machine operation in real-time.”

• cites [56] 56. M. Simsek, A. Aijaz, M. Dohler, J. Sachs, and G. Fettweis, "5G-enabled tactile internet," IEEE J. Sel. Areas Commun., vol. 34, no. 3, pp. 460- 473, Mar. 2016. sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

A bank of ANNs is proposed for symbol decoding in MIMO-OFDM systems.

“A bank of ANNs is proposed for symbol decoding in MIMO-OFDM systems in [74].”

• cites [74] 74. Y. Li, P. Zhang, and R. Huang, "Lightweight quantum encryption for secure transmission of power data in smart grid," IEEE Access, vol. PP, 2019. sourceCould not verify

  The Introduction describes a lightweight quantum-cryptography scheme for transmitting smart-grid power data, and the cryptographic-algorithm section describes a stream cipher. Searches for a bank of ANNs, symbol decoding, and MIMO-OFDM found no passage addressing the attributed proposal in our copy. (LLM, conf=0.76)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

ANNs are considered as the most notable enablers of DL mechanism, in which understanding of data's distribution and important features are automatically learned to simulate an effective mapping function.

“ANNs are considered as the most notable enablers of DL mechanism [17], in which the understanding of the data's distribution and important features are automatically learned to simulate an effective mapping function.”

• cites [17] 17. M. D. A. Zappone, M. D. Renzo, "Wireless networks design in the era of deep learning: Model-based, AI-based, or both?" arXiv preprint arXiv:1902.02647, 2019. sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

The scope of GA for intelligent cognitive radio has been encouraged.

“The scope of GA for intelligent cognitive radio has been encouraged in [87].”

• cites [87] 87. S. J. Nawaz, M. I. Tiwana, M. N. Patwary, N. M. Khan, M. I. Tiwana, and A. Haseeb, "GA based sensing of sparse multipath channels with superimposed training sequence," Elektronika ir Elektrotechnika, vol. 22, no. 1, pp. 87-91, 2016. sourcePartly supported

  The Introduction (p. 1) describes GAs as useful for optimization and applies them to wireless-channel estimation; Section III (p. 4) reports improved performance for the proposed GA-based estimator. Our copy does not discuss intelligent cognitive radio or encourage GA use for that specific purpose (searched: cognitive radio, cognitive). (LLM, conf=0.78)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

The scope of GA for estimation of sparse UWAC channels has been investigated.

“The scope of GA for estimation of such channels has been investigated in the literature, see e.g., [84,86] for GA based estimation of sparse channels (UWAC), etc.”

• cites [84] 84. H. Y.Lu and W. H. Fang, "Joint receive antenna selection and symbol detection for mimo systems: a heterogeneous genetic approach," IEEE Commun. Letters, vol. 13, no. 2, pp. 97-99, Feb. 2009. sourceNot checked — no full text of the source
• cites [86] 86. S. K. Goudos, "Evolutionary algorithms for wireless communications -a review of the state-of-the-art," in Contemporary Issues in Wireless Communications. InTech, 2014. sourceCould not verify

  Section 3 (Related work; page unavailable) discusses genetic algorithms for wireless-communications problems, but does not address estimation of sparse underwater acoustic channels. Searches for “UWAC,” “underwater acoustic sparse channel,” and GA-based sparse-channel estimation found no passage addressing that specific application. (LLM, conf=0.72)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

Deployment of ANNs in communication networks is not new; ANNs have been deployed to perform and assist in various operations of communication systems.

“Deployment of ANNs in communication networks is not a new idea, instead, ANNs have been deployed to perform and assist in various operations of communication systems; e.g., ANN is proposed for symbol decoding for MIMO-OFDM systems in [74].”

• cites [74] 74. Y. Li, P. Zhang, and R. Huang, "Lightweight quantum encryption for secure transmission of power data in smart grid," IEEE Access, vol. PP, 2019. sourceCould not verify

  The Introduction and the section on quantum cryptography for smart grids discuss secure power-data communication, not ANN deployment. A search of our copy for ANN-based MIMO-OFDM symbol decoding found no passage addressing the cited example; the available passages therefore cannot verify the attribution. (LLM, conf=0.76)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Description of radio propagation channels characteristics for molecular communications is not well established in the literature; for such scenarios, an ANN based receiver design is proposed.

“Description of radio propagation channels characteristics for molecular communications is not well established in the literature [89]; for such communication scenarios, an ANN based receiver design is proposed in [90].”

• cites [89] 89. S. Navabi, C. Wang, O. Y. Bursalioglu, and H. Papadopoulos, "Pre- dicting wireless channel features using neural networks," arXiv preprint arXiv:1802.00107, 2018. sourceCould not verify

  The Introduction discusses millimeter-wave cellular channels, and the Problem Formulation describes neural networks that predict channel features at a base station. Neither passage addresses whether propagation-channel characteristics for molecular communications are established or proposes an ANN-based receiver for that setting. Searches for molecular communications and ANN receiver design found no supporting passage in our copy. (LLM, conf=0.77)

  Experimental — accuracy not yet measured
• cites [90] 90. N. Farsad, H. B. Yilmaz, A. Eckford, C. Chae, and W. Guo, "A com- prehensive survey of recent advancements in molecular communication," IEEE Commun. Surveys Tuts., vol. 18, no. 3, pp. 1887-1919, Feb. 2016. sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Applications of DL in communication networks include network planning, deployment, resources management, operations control, and maintenance.

“applications of DL in communication networks have been indicated in the literature, which includes network planning, deployment, resources management, operations control, and maintenance etc.”

• cites [17] 17. M. D. A. Zappone, M. D. Renzo, "Wireless networks design in the era of deep learning: Model-based, AI-based, or both?" arXiv preprint arXiv:1902.02647, 2019. sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

A DNN for real-time radio resource management at the physical layer of communication networks has been proposed.

“The DNN for real-time radio resource management at the physical layer of communication networks has been proposed in [101].”

• cites [101] 101. T. Wang, C.-K. Wen, H. Wang, F. Gao, T. Jiang, and S. Jin, "Deep learning for wireless physical layer: Opportunities and challenges," China Communications, vol. 14, no. 11, pp. 92-111, 2017. sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Quantum key distribution (QKD), quantum teleportation, and dense coding are promising quantum communication protocols for expanding classical data transmission in quantum-based systems.

“Some of the promising quantum communication protocols to expand the possibility of classical data transmission in quantum-based systems include quantum key distribution (QKD) [126,127], quantum teleportation [128] and dense coding [129].”

• cites [126] 126. M. Berta, J. M. Renes, and M. M. Wilde, "Identifying the information gain of a quantum measurement," IEEE Trans. Inf. Theory, vol. 60, no. 12, pp. 7987-8006, Dec. 2014. sourceCould not verify

  The Summary of Results (p. 5) discusses quantum-measurement simulation and channel coding, not a list of promising communication protocols. Searches for quantum key distribution, teleportation, and dense coding found only a QKD reference in the bibliography (p. 44), with no passage supporting the attributed claim. (LLM, conf=0.76)

  Experimental — accuracy not yet measured
• cites [127] 127. V. Sharma and S. Banerjee, "Analysis of quantum key distribution based satellite communication," in proc. of Int. Conf. on Computing, Commun. and Networking Technol., Jul. 2018, pp. 1-5. sourceNot checked — no full text of the source
• cites [128] 128. K. Inoue, "Quantum key distribution technologies," IEEE J. Sel. Topics Quantum Electron., vol. 12, no. 4, pp. 888-896, Jul. 2006. sourceNot checked — no full text of the source
• cites [129] 129. C. H. Bennett, G. Brassard, C. Crépeau, R. Jozsa, A. Peres, and W. K. Wootters, "Teleporting an unknown quantum state via dual classical and einstein-podolsky-rosen channel," Physical Review Letters, vol. 70, no. 13, pp. 1895-1899, Mar. 1993. sourceSupported

May combine two claims.

Quantum-assisted communications can enhance channel estimation, optimal MUD, design of optimal precoding matrix, and optimal routing by employing quantum algorithms.

“The quantumassisted communications can enhance various aspects of the existing classical communication networks including channel estimation, optimal Multi-User Detection (MUD), the design of optimal precoding matrix and the optimal routing by employing the quantum algorithms [30].”

• cites [30] 30. P. Botsinis, D. Alanis, Z. Babar, H. Nguyen, D. Chandra, S. X. Ng, and L. Hanzo, "Quantum algorithms for wireless communications," IEEE Commun. Surveys Tuts., vol. PP, Nov. 2018. sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Dynamic sharing of spectrum aims at enhancing spectrum utilization efficiency through access of under-utilized spectrum to secondary network users.

“Such dynamic sharing of spectrum aims at enhancing the spectrum utilization efficiency through the provision of access of under-utilized spectrum to the secondary network users [111].”

• cites [111] 111. X. Li, J. Fang, W. Cheng, H. Duan, Z. Chen, and H. Li, "Intelligent power control for spectrum sharing in cognitive radios: A deep reinforcement learning approach," IEEE Access, vol. 6, pp. 25 463-25 473, Apr. 2018. sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Cognitive radio techniques enable a radio system to sense, learn, and adapt based on the context of the surrounding environment.

“Cognitive radio techniques enables a radio system to sense, learn, and adapt based on the context of the surrounding environment [87].”

• cites [87] 87. S. J. Nawaz, M. I. Tiwana, M. N. Patwary, N. M. Khan, M. I. Tiwana, and A. Haseeb, "GA based sensing of sparse multipath channels with superimposed training sequence," Elektronika ir Elektrotechnika, vol. 22, no. 1, pp. 87-91, 2016. sourceCould not verify

  The Introduction (p. 1) discusses wireless-channel estimation and genetic algorithms, not cognitive radio techniques. Searches for cognitive radio and for sensing, learning, and adapting to environmental context found no passage addressing the attributed claim. (LLM, conf=0.76)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

Supervised and unsupervised ML-based cooperative spectrum sensing algorithms for cognitive radio networks have been proposed.

“In [109], supervised and unsupervised ML-based cooperative spectrum sensing algorithms for cognitive radio networks have been proposed.”

• cites [109] 109. S. K. Sharma, T. E. Bogale, S. Chatzinotas, B. Ottersten, L. B. Le, and X. Wang, "Cognitive radio techniques under practical imperfections: A survey," IEEE Commun. Surveys Tuts., vol. 17, no. 4, pp. 1858-1884, Jul. 2015. sourcePartly supported

  The survey describes cooperative spectrum sensing in “A. Spectrum Awareness Techniques” (p. 5) and a learning approach used for feature detection in “A. Cyclostationary Based Detector” (p. 11). Neither passage identifies supervised and unsupervised ML-based cooperative sensing algorithms, and the cited work is a survey rather than a paper proposing them. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

Quantum-proof randomness extractors can be considered an important building block for implementing classical and quantum cryptography in security applications.

“Quantum-proof randomness extractors can be considered as an important building block for implementing classical and quantum cryptography in security applications [160].”

• cites [160] 160. R. T. Konig and B. M. Terhal, "The bounded-storage model in the presence of a quantum adversary," IEEE Trans. Inf. Theory, vol. 54, no. 2, pp. 749-762, Feb. 2008. sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

The achievable rate for the transmission of classical information over a noisy quantum channel is given by the Holevo-Schumacher-Westmorel (HSW) coding theorem, which generalizes Shannon's theorem in quantum settings.

“For example, the achievable rate for the transmission of classical information over a noisy quantum channel is given by the Holevo-Schumacher-Westmorel (HSW) coding theorem [136], which generalizes the Shannon's theorem in quantum settings.”

• cites [136] 136. S. Boixo, T. F. Rønnow, S. V. Isakov, Z. Wang, D. Wecker, D. A. Lidar, J. M. Martinis, and M. Troyer, "Evidence for quantum annealing with more than one hundred qubits," Nature Physics, vol. 10, no. 3, p. 218â Ȃ Ş224, Feb. 2014. sourceCould not verify

  The Experimental results (p. 4) discuss quantum annealing and success probabilities, while the Conclusions (p. 5) discuss possible quantum speedup. Neither addresses the achievable rate for transmitting classical information over a noisy quantum channel. Searches for the HSW coding theorem, channel capacity, and Shannon's theorem found no relevant passage in our copy. (LLM, conf=0.77)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

QKD protocols are applicable in quantum-assisted satellite communication (SatCom) systems for secure communication between ground stations and the satellite.

“For example, authors in [126] discussed and analyzed the applicability of QKD protocols in quantumassisted SatCom systems in order to perform secure communication between ground stations and the satellite.”

• cites [126] 126. M. Berta, J. M. Renes, and M. M. Wilde, "Identifying the information gain of a quantum measurement," IEEE Trans. Inf. Theory, vol. 60, no. 12, pp. 7987-8006, Dec. 2014. sourceCould not verify

  The Summary of Results (p. 5) describes measurement simulation and state-splitting protocols, while Extensions and Applications (pp. 30–31) discusses information reconciliation and quantum side information. Searches for satellite, SatCom, QKD, and quantum key distribution found no passage addressing QKD between ground stations and a satellite. (LLM, conf=0.77)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

IBM has a plan of finalizing a gate-based quantum computer with 50 qubits by 2020.

“and IBM has a recent plan of finalizing a gate-based quantum computer with 50 qubits by 2020 [30].”

• cites [30] 30. P. Botsinis, D. Alanis, Z. Babar, H. Nguyen, D. Chandra, S. X. Ng, and L. Hanzo, "Quantum algorithms for wireless communications," IEEE Commun. Surveys Tuts., vol. PP, Nov. 2018. sourceSupported

May combine two claims.

QML is observed to exponentially speed-up the optimization problems involving quadratic objective functions subject to equality constraints and involving penalty functions subject to inequality constraints.

“QML is observed to exponentially speed-up the optimization problems involving quadratic objective functions subject to equality constraints and involving penalty functions subject to inequality constraints [41].”

• cites [41] 41. J. Biamonte, P. Wittek, N. Pancotti, P. Rebentrost, N. Wiebe, and S. Lloyd, "Quantum machine learning," Nature, vol. 549, no. 7671, pp. 195-202, Sep. 2017. sourcePartly supported

  The qBLAS-based optimization section (p. 8) describes an exponential algorithmic speedup for quadratic objectives subject to equality constraints, but only when the matrices are sparse or low rank. It also says inequality constraints are often handled with penalty functions and describes an exponentially faster method for polynomial optimization; it does not establish that all such constrained problems are sped up. The Perspectives section (p. 14) treats practical performance and benchmarking as open questions, so “observed” overstates the evidence. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

The concept of grid computing may further assist in realizing the ambitions by conceding the coordinated resource sharing and access to dynamic multi-institutional virtual organizations.

“To this end, the concept of grid computing may further assist in realizing the ambitions by conceding the coordinated resource sharing and access to dynamic multi-institutional virtual organizations [197].”

• cites [197] 197. M. P. Frank, L. Oniciuc, U. H. M.-Baese, and I. Chiorescu, "A space- efficient quantum computer simulator suitable for high-speed FPGA implementation," in Quantum Information and Computation VII, vol. 7342. International Society for Optics and Photonics, 2009, p. 734203. sourceCould not verify

  The Introduction (pp. 1–2) and Conclusion (p. 11) discuss a space-efficient quantum computer simulator and possible FPGA implementation. Searches for grid computing, coordinated resource sharing, and multi-institutional virtual organizations found no passage addressing the attributed claim. (LLM, conf=0.77)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

ML for optimal user clustering and power allocation in mmWave NOMA has been proposed.

“Also, ML for optimal user clustering and power allocation in mmWave NOMA has been proposed in [231].”

• cites [231] 231. D. A. Awan, R. L. G. Cavalcante, M. Yukawa, and S. Stanczak, "Detec- tion for 5G-NOMA: An online adaptive machine learning approach," in proc. of IEEE Int. Conf. on Commun., May 2018, pp. 1-6. sourcePartly supported

  The Introduction and Section III describe an online-learning method for detecting signals in a clustered NOMA uplink. They do not describe an ML method for optimizing user clustering or power allocation, or identify the setting as mmWave NOMA. The citation supports the ML-for-NOMA aspect, but not the proposed optimization attributed to it. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

An online adaptive ML approach for detection in NOMA in the context of 5G networks has been proposed.

“Online adaptive ML approach for detection in NOMA in the context of 5G networks has been proposed in [230].”

• cites [230] 230. R. Abozariba, M. K. Naeem, M. Patwary, M. Seyedebrahimi, and P. Bull, "NOMA based resource allocation and mobility enhancement framework for IoT in next generation cellular networks," IEEE Access, vol. PP, pp. 1-15, Jan. 2019. sourceCould not verify

  The paper’s discussion of NOMA in 5G describes dynamic spectrum allocation, and its resource-allocation formulation uses successive interference cancellation. No passage found in our copy describes a proposed online adaptive machine-learning approach for NOMA detection (searched for that phrase and for machine-learning detection). Page numbers are unavailable in the indexed copy. (LLM, conf=0.72)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

For FDD M-MIMO systems, a dictionary learning-based channel estimation method has been proposed.

“For such an FDD M-MIMO system, a dictionary learning-based channel estimation method has been proposed in [225].”

• cites [225] 225. F. A. Aoudia and J. Hoydis, "End-to-end learning of communications systems without a channel model," arXiv preprint arXiv:1804.02276, 2018. sourceCould not verify

  The Introduction describes end-to-end transmitter and receiver learning, and the Transmitter and Receiver Architectures section describes estimating a Rayleigh block-fading channel response. Searches for dictionary learning and FDD channel estimation found no passage proposing the claimed method for FDD massive-MIMO systems; the channel-response estimate alone does not establish that claim. (LLM, conf=0.76)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

NOMA-based MEC networks have received considerable attention in recent years.

“In addition, NOMA-based MEC network has also received considerable attention in the recent years, see e.g., [233,234].”

• cites [233] 233. G. Gui, H. Huang, Y. Song, and H. Sari, "Deep learning for an effective nonorthogonal multiple access scheme," IEEE Trans. Veh. Technol., vol. 67, no. 9, pp. 8440-8450, Sep. 2018. sourceNot checked — no full text of the source
• cites [234] 234. Y. Ye, G. Lu, R. Q. Hu, and L. Shi, "On the performance and op- timization for MEC networks using uplink NOMA," arXiv preprint arXiv:1901.04149, 2019. sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

6G communications will leverage robust learning capability at different network layers to perform diverse tasks such as network management, radio resource allocation, data security, and manipulation of smart surfaces.

“The foregoing discussions suggest that the 6G communications will leverage robust learning capability at different network layers to perform diverse tasks such as network management, radio resource allocation, data security, and manipulation of smart surfaces to name a few.”

Probably not a substantial claim: may not pass S3 (load-bearing).

5G URLLC targets end-to-end latencies of approximately 1 ms.

“Both the reliability aspect, with packet error rates ≤ 10 -5 , and end-to-end latencies ∼1ms aim at supporting new use cases such as factory automation, autonomous driving, e-health, building automation, and smart cities, to name a few [50].”

• cites [50] 50. International Telecommunication Union, "IMT Visionâ ȂŤ Framework and Overall Objectives of the Future Development of IMT for 2020 and Beyond," IMT, Tech. Rep. document M Series 2083, Sep. 2015.Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

RAN congestion in ultra-dense IoT networks may arise due to massive numbers of short-packet transmissions, huge signalling overhead per data packet, and very dynamic and sporadic device transmissions.

“For example, the RAN congestion in ultra-dense IoT networks may arise due to various reasons including the massive number of short-packet transmissions, huge signalling overhead per data packet, and very dynamic and sporadic nature of device transmissions [70].”

• cites [70] 70. X. Artiga, A. Perez-Neira, J. Baranda, E. Lagunas, S. Chatzinotas, R. Zetik, P. Gorski, K. Ntougias, D. Perez, and G. Ziaragkas, "Shared access satellite-terrestrial reconfigurable backhaul network enabled by smart antennas at mmwave band," IEEE Network, vol. 32, no. 5, pp. 46- 53, Sep. 2018. sourceNot checked — no full text of the source

May combine two claims.

ML-based proactive content caching at the edge node can avoid backhaul congestion and further reduce service latency.

“Moreover, ML-based proactive content caching at the edge node can also be exploited to avoid backhaul congestion and further reduce the service latency [66,67].”

• cites [66] 66. T. D. Ponnimbaduge Perera, D. N. K. Jayakody, S. K. Sharma, S. Chatzinotas, and J. Li, "Simultaneous wireless information and power transfer (SWIPT): Recent advances and future challenges," IEEE Com- mun. Surveys Tuts., vol. 20, no. 1, pp. 264-302, Feb. 2018. sourceNot checked — no full text of the source
• cites [67] 67. T. Hou, G. Feng, S. Qin, and W. Jiang, "Proactive content caching by exploiting transfer learning for mobile edge computing," in proc. of IEEE Global Commun. Conf., Dec. 2017, pp. 1-6. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

ML techniques can be significantly advantageous in addressing the access congestion problem in emerging ultra-dense IoT networks.

“For example, ML techniques can be significantly advantageous in addressing the access congestion problem in emerging ultra-dense IoT networks [14].”

• cites [14] 14. S. K. Sharma and X. Wang, "Collaborative distributed Q-learning for RACH congestion minimization in cellular IoT networks," IEEE Com- munications Letters, pp. 1-1, 2019. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S4 (based).

This work is the first to explicitly investigate the deployment of the QML framework for future communication networks.

“To the authors' best knowledge, this work is the first to explicitly investigate the deployment of the QML framework for future communication networks.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The mMTC service aims at providing internet connectivity to low-rate low-power devices.

“The mMTC service aims at providing internet connectivity to such devices.”

Probably not a substantial claim: may not pass S4 (based).

DL methods lack an efficient mechanism for prior evaluation of the best choice of training algorithm, size and structure of DNN, and parameters setting that suits the model/problem under consideration.

“However, DL methods lack an efficient mechanism for prior evaluation of the best choice of training algorithm, size and structure of DNN, and parameters setting that suits the model/problem under consideration.”

Probably not a substantial claim: may not pass S3 (load-bearing).

5G URLLC targets packet error rates of 10^-5 or lower.

“Both the reliability aspect, with packet error rates ≤ 10 -5 , and end-to-end latencies ∼1ms aim at supporting new use cases such as factory automation, autonomous driving, e-health, building automation, and smart cities, to name a few [50].”

• cites [50] 50. International Telecommunication Union, "IMT Visionâ ȂŤ Framework and Overall Objectives of the Future Development of IMT for 2020 and Beyond," IMT, Tech. Rep. document M Series 2083, Sep. 2015.Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

One potential application of DL in mmWave M-MIMO systems is estimation of radio channel quality.

“One of the potential applications of employing DL methods in such mmWave M-MIMO systems is the estimation of radio channel quality [23], which is essential for the design of transmission techniques such as beamforming.”

• cites [23] 23. H. He, C.-K. Wen, S. Jin, and G. Y. Li, "Deep learning-based channel estimation for beamspace mmWave massive MIMO systems," arXiv preprint arXiv:1802.01290, 2018. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

ML has a strong potential to replace conventional mathematical model-based algorithmic solutions, given the availability of adequate data and computational power.

“Furthermore, ML has a strong potential to replace the conventional mathematical mode-based algorithmic solutions, given the availability of adequate data and computational power.”

Probably not a substantial claim: may not pass S3 (load-bearing).

DL capabilities can be utilized for channel estimation and symbol detection in OFDM systems.

“As an example, DL capabilities can be utilized for channel estimation and symbol detection in Orthogonal Frequency-Division Multiplexing (OFDM) systems [20].”

• cites [20] 20. H. Ye, G. Y. Li, and B. Juang, "Power of deep learning for channel esti- mation and signal detection in OFDM systems," IEEE Wireless Commun. Lett., vol. 7, no. 1, pp. 114-117, Feb. 2018. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Following evolution trends, the bit/s throughput targets in 6G networks are expected to increase by an order of magnitude relative to those of 5G.

“Following the evolution trends of previous mobile network generations, the bit/s throughput targets in 6G networks are expected to increase by an order of magnitude relative to those of 5G.”

Probably not a substantial claim: may not pass S3 (load-bearing).

A deep reinforcement learning based decentralized resource allocation mechanism can be utilized to support highly dynamic communication applications.

“For example, a Deep Reinforcement Learning (DRL) based decentralized resource allocation mechanism can be utilized to support highly dynamic communication applications [21].”

• cites [21] 21. H. Ye and G. Y. Li, "Deep reinforcement learning for resource allocation in V2V communications," in IEEE Int. Conf. on Commun., May 2018, pp. 1-6. sourceNot checked — no full text of the source

May combine two claims.

UAV base stations can offload the data-traffic of static base stations and relocate dynamically to provide a more favorable propagation channel to edge users.

“These UAV BSs can not only offload the data-traffic of the static BSs but they can also relocate dynamically to provide a more favorable propagation channel to the edge users.”

• cites [59] 59. S. J. Nawaz et al.: Quantum ML for 6G Communication Networks: Vision for the Future full-duplex technology: Recent advances and research challenges," IEEE Commun. Surveys Tuts., vol. 20, no. 1, pp. 674-707, Feb. 2018.Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim. May combine two claims.

Higher mobility causes higher Doppler spread, which reduces coherence-time and leads to fast variability in channel statistics.

“higher mobility causes higher Doppler spread which further causes reduction in the coherence-time and this eventually leads to fast variability in the channel statistics.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Model-based learning usually optimizes performance indexes through available objective functions with high computational efficiency.

“Modelbased learning usually optimizes the performance indexes through available objectives functions with high computational efficiency.”

Probably not a substantial claim: may not pass S3 (load-bearing).

A promising application of reinforcement learning at the physical layer is power control and optimization.

“A promising application of reinforcement learning at the physical layer of communication networks is power control and optimization.”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

Integration of DL capability with smart city infrastructure can help in effective utilization of available big data for accomplishing the cognitive smart world of the future.

“The integration of DL capability with the smart city infrastructure can help in effective utilization of the available big data for accomplishing the dream of the cognitive smart world of the future [27].”

• cites [27] 27. M. Mohammadi and A. Al-Fuqaha, "Enabling cognitive smart cities using big data and machine learning: Approaches and challenges," IEEE Commun. Mag., vol. 56, no. 2, pp. 94-101, Feb. 2018. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

DL for radio resource management in 5G networks is suggested, and a users location-aware DL method for run-time optimal users association for maximization of sum-rate in M-MIMO based land mobile radio cellular networks is proposed.

“In addition, DL for radio resource management in 5G networks is also suggested in [94], and a users location-aware DL method for run-time optimal users association for the maximization of sum-rate in M-MIMO based land mobile radio cellular networks is proposed in [95].”

• cites [94] 94. Z. Qin, H. Ye, G. Y. Li, and B.-H. F. Juang, "Deep learning in physical layer communications," proc. of IEEE Veh. Technol. Conf. (Fall-19), vol. preprint arXiv:1807.11713, 2019. sourceNot checked — no full text of the source
• cites [95] 95. F. D. Calabrese, L. Wang, E. Ghadimi, G. Peters, L. Hanzo, and P. Soldati, "Learning radio resource management in rans: Framework, opportunities, and challenges," IEEE Commun. Mag., vol. 56, no. 9, pp. 138-145, Sep. 2018. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

A typical application of off-line (batch) and online learning in communication systems can be intelligent caching and channel tracking, respectively.

“A typical application of off-line (batch) and online learning in communication systems can be intelligent caching and channel tracking, respectively.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

The selection of encoding/precoding schemes for performance optimization is a potential application of unsupervised learning.

“The selection of encoding/precoding schemes for performance optimization is another potential application of unsupervised learning.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Generalization of an ANN's operations for efficiently dealing with new unseen input data samples needs training over a large amount of data (big data).

“The generalization of an ANN's operations, for efficiently dealing with every new unseen input data sample, needs its training over a large amount of data (i.e., big data).”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

At higher layers, unsupervised and semi-supervised learning have potential applications for grouping/pairing/clustering of nodes/points for optimal allocation of network/radio resources.

“Subsequently, at the higher layers, there are various potential applications of unsupervised and semi-supervised learning for grouping/pairing/clustering of nodes/points for optimal allocation of network/radio resources.”

Probably not a substantial claim: may not pass S3 (load-bearing).

An ANN-based method for predicting channel features for large-scale multi-antenna BSs is proposed.

“Recently, an ANN-based method for predicting channel features for large-scale multi-antenna BSs is proposed in [88].”

• cites [88] 88. T. W. Rondeau, B. Le, C. J. Rieser, and C. W. Bostian, "Cognitive radios with genetic algorithms: Intelligent control of software defined radios," in proc. of SDR Forum Technical Conf., vol. 100. Citeseer, 2004, pp. 3-8. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

Unsupervised learning can potentially be applied for points clustering, features extraction, features classification, distribution estimation, and distribution specific samples generation.

“Unsupervised learning can potentially be applied for performing the wide range of tasks related to points clustering, features extraction, features classification, distribution estimation, and distribution specific samples generation.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Revisited learning algorithms for exploiting numerous processing cores in advanced GPUs have demonstrated a remarkable performance gain.

“For example, the revisited learning algorithms for exploiting the numerous amount of available processing cores in advanced Graphics Processing Units (GPUs) has demonstrated a remarkable performance gain.”

Probably not a substantial claim: may not pass S3 (load-bearing).

In highly dynamic vehicular communications, semi-supervised and unsupervised learning can assist channel equalization and tracking operations.

“At the physical layer in highly dynamic scenarios of vehicular communications, where less coherence-time limits the available time and data for supervising the learning of channel equalizer; semi-supervised and unsupervised learning can make their way for assisting the channel equalization and tracking operations.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Another application of supervised learning at the physical layer for downlink communications can be optimal power allocation and interference cancellation.

“Another application of supervised learning at the physical layer for downlink communications can be optimal power allocation and interference cancellation.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

The size of an ANN and the amount of available data determine its performance and requisite computational power.

“The size of an ANN (number of neurons and hidden layers) and the amount of available data determine its performance and requisite computational power, as illustrated in Fig. 2.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

With 5G reaching its limits in the next decade or so, the design goals for its successor have already begun to be explored in the literature.

“With 5G reaching its limits in the next decade or so, the design goals for its successor have already begun to be explored in the literature.”

Probably not a substantial claim: may not pass S3 (load-bearing).

On-line training is a suitable solution for streaming data applications.

“On-line training is a suitable solution for such streaming data applications.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

A typical application of model-based and samples-based learning can be symbol decoding and contents demand prediction, respectively.

“A typical application of model-based and samples-based learning can be symbol decoding and contents demand prediction, respectively.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

5G wireless networks have recently started to be deployed in some parts of the world, but the goal of a fully intelligent network furnishing everything as a service and rendering a completely immersive user experience remains elusive.

“The 5G wireless networks have recently started to be deployed in some parts of the world but the goal of a fully intelligent network furnishing everything as a service and rendering a completely immersive user experience remains elusive.”

Probably not a substantial claim: may not pass S2 (checkable). The quote only partly states this claim.

ML and QC methods can play a significant enabling role in managing the network at different levels.

“To this end, ML and QC methods can play a significant enabling role.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Advanced TPUs have demonstrated tremendous parallel processing potential with manifold speed-ups and power efficiency in executing ML algorithms.

“Furthermore, the advanced Tensor Processing Units (TPUs) have demonstrated a tremendous parallel processing potential with manifold speed-ups and power efficiency in executing ML algorithms.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

At the physical layer, received noisy data symbols can be used to train a system by clustering sample points in the decision space for generating nonlinear decision boundaries for mapping symbols according to constellation maps.

“at the physical layer, the received noisy data symbols can be used to train a system by clustering the sample points in the decision space for generating effective nonlinear decision boundaries for mapping of the symbols according to the constellation maps.”

Probably not a substantial claim: may not pass S3 (load-bearing).

An agent can be associated with each serving station in a cellular network to assist in learning optimal scheduling parameters to enhance network QoS.

“An agent can be associated with each serving station in a cellular network to assist in learning the optimal scheduling parameters to enhance the network Quality-of-Service (QoS) [78].”

• cites [78] 78. M. Chen, W. Saad, C. Yin, and M. Debbah, "Echo state networks for proactive caching in cloud-based radio access networks with mobile users," IEEE Trans. Wireless Commun., vol. 16, no. 6, pp. 3520-3535, Jun. 2017. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Recent advances in parallel computing capabilities and distributed learning methods have enabled the deployment of data-driven DL approaches to complement conventional model-based approaches.

“The recent advances in parallel computing capabilities and distributed learning methods have enabled the deployment of data-driven DL approaches to complement the conventional model-based approaches.”

Probably not a substantial claim: may not pass S4 (based).

The available state-of-the-art in ML is isolated in terms of ML techniques and their operations across different layers of the protocol stack of communication networks.

“Also, the available state-ofart in ML is isolated in terms of ML techniques as well as their operations across different layers of the protocol stack of communication networks.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Big data-driven DL can enable deployment of ANNs in learning complex statistical structures and input-output operations of future communication networks.

“As anticipated in Fig. 2, big data-driven DL can enable the deployment of ANNs in learning complex (high dimensions, a high number of features, a high number of classes, etc) statistical structures and input-output operations of the future communication networks.”

Probably not a substantial claim: may not pass S3 (load-bearing).

These advances have demonstrated a profound impact of DL-based solutions across various multidimensional signal processing applications.

“These advances have demonstrated a profound impact of DL-based solutions across various multidimensional signal processing applications, e.g., medical image processing, natural languages processing, and wireless communications, to name a few.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Pure data samples-based learning exploits all available data samples to interpolate and/or extrapolate samples, with high memory and time requirements.

“On the other hand, the pure data samplesbased learning exploits all the available data samples to interpolate and/or extrapolate the samples, with high memory and time requirements.”

Probably not a substantial claim: may not pass S4 (based).

Training and processing of data through conventional ML algorithms on conventional CPUs with a limited number of cores has a limitation of large processing delays.

“The training and processing of data through conventional ML algorithms, executed on conventional Central Processing Units (CPUs) with a limited number of cores, has a limitation of large processing delays.”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

Given increasing complexity of future communication networks, ANNs have a wide scope of deployment in planning, optimization, estimation, tracking, controlling, and maintaining tasks.

“Given the increasing complexity of the future communication networks, ANNs have a wide scope of its deployment in performing various diverse tasks, e.g., planning, optimization, estimation, tracking, controlling, and maintaining tasks, etc.”

Probably not a substantial claim: may not pass S3 (load-bearing).

ANNs assisted indoor localization method exploiting two variants of fingerprinting approaches is proposed.

“Also, ANNs assisted indoor localization method exploiting two variants of the fingerprinting approaches is proposed in [91].”

• cites [91] 91. X. Qian and M. D. Renzo, "Receiver design in molecular communica- tions: An approach based on artificial neural networks," in Int. Symp. on Wireless Commun. Systems. IEEE, 2018, pp. 1-5. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Estimation and tracking of channels without well-defined models is challenging, and errors in channel estimate can significantly affect symbol detection performance.

“This makes the estimation and tracking of such channels a challenging task, where any error in channel estimate can significantly affect the symbol detection performance.”

• AssumptionRests on an assumption: For some communication scenarios, there is no well-defined model of propagation channel available, including UWAC, mmWave, high mobility, and molecular communication channels.

Probably not a substantial claim: may not pass S3 (load-bearing).

For fast time-varying channels, a hybrid learning method is proposed to assist in estimation and tracking of the channels.

“For such fast timevarying channels, a hybrid learning method is proposed in [75], to assist in estimation and tracking of the channels.”

• cites [75] 75. S. J. Nawaz, S. Mohsin, and A. A. Ikaram, "Neural network based MIMO-OFDM channel equalizer using comb-type pilot arrangement," in proc. of IEEE Int. Conf. on Future Computer and Commun., Apr. 2009. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Supervised ML can play an important role in determining users association with BSs on the basis of contents/media demand.

“Another potential scenario, where supervised ML can play an important role, is in determining the users association with BSs on the basis of contents/media demand.”

Probably not a substantial claim: may not pass S3 (load-bearing).

ML has a popular application in intelligent caching, which can help in reducing latency in satellite links.

“ML has a popular application in intelligent caching, which can help in reduc-ing the latency in satellite links.”

Classified as probably not a substantial claim.

The major limitation for ANNs is available computational power to deal with big data and deep structured ANNs.

“However, the major limitation is the available computational power to deal with big data and deep structured ANNs.”

Classified as probably not a substantial claim.

QC has recognized potential in manipulating large-dimensional data vectors in large tensor product spaces.

“The QC has a recognized potential in conveniently manipulating such large-dimensional data vectors in large tensor product spaces.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Another popular application of DNNs at the physical layer is power control and optimization.

“Another popular application of DNNs at the physical layer of communication networks is power control and optimization [102].”

• cites [102] 102. H. Sun, X. Chen, Q. Shi, M. Hong, X. Fu, and N. D. Sidiropoulos, "Learning to optimize: Training deep neural networks for wireless re- source management," in proc. of IEEE Int. Workshop on Signal Process- ing Advances in Wireless Commun., Jul. 2017, pp. 1-6.Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Transfer learning facilitates learning by reducing required data and relaxes the i.i.d. condition between training and validation data.

“This learning technique offers the advantage of facilitating the learning process by reducing the amount of required data, and relaxes the condition on the training data to be independently and identically distributed (i.i.d.) with the validation data.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The capacity of quantum channels is not completely understood and various measures are available in the literature.

“Although the capacity of classical channels has been well understood within the framework of classical information theory, the capacity of quantum channels is not completely understood and various measures are available in the literature.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

For optimal beamforming at the physical layer, latency induced by conventional iterative methods makes it outdated for future networks.

“An example of such delay critical applications at the physical layer of communication networks is optimal beamforming; where the latency induced by conventional iterative methods makes it outdated for the future networks.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For enabling quantum communications, information signal can be encoded by modulating photons' polarization, detected by single-photon detectors, and photons' phase, measured by homodyne detection.

“For enabling the quantum communications, the information signal can be encoded in different ways such as by modulating the photons' polarization, which is usually detected by utilizing single-photon detectors, and the photons' phase, usually measured by the homodyne detection [120].”

• cites [120] 120. J. D. Franson, "Quantum communications," in proc. of Technical Digest. Summaries of papers presented at the Quantum Electronics and Laser Science Conf. (Post conf. Technical Digest), May 2001, pp. 171-. sourceNot checked — no full text of the source

May combine two claims.

Quantum principles provide significant benefits to communication networks including enhanced channel capacity, quantum teleportation, and quantum cryptography.

“In summary, quantum principles provide significant benefits to the communication networks including enhanced channel capacity, the ability to transmit an unknown quantum state, i.e, quantum teleportation, and to deliver secure information, i.e., quantum cryptography by utilizing a number of advanced communication protocols which will not be possible with the classical techniques [119].”

• cites [119] 119. S. Imre, "Quantum communications: Explained for communication engi- neers," IEEE Commun. Mag., vol. 51, no. 8, pp. 28-35, Aug. 2013. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Quantum measurements can be asymptotically represented by the amount of classical communication equivalent to the quantum notion of mutual information of the measurement, if shared randomness is available.

“In this regard, authors in [125] have shown that shared randomness is available if necessary, quantum measurements can be asymptotically represented by the amount of classical communication equivalent to the quantum notion of the mutual information of the measurement.”

• cites [125] 125. I. Devetak, "The private classical capacity and quantum capacity of a quantum channel," IEEE Trans. Inf. Theory, vol. 51, no. 1, pp. 44-55, Jan. 2005.Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S2 (checkable).

QC and ML can create close synergies towards providing joint benefits in communication systems.

“QC and ML can create close synergies with each other towards providing their joint benefits in communication systems.”

Probably not a substantial claim: may not pass S3 (load-bearing).

A DNN based auto-encoder for jointly optimizing BER and PAPR in OFDM systems is proposed.

“A DNN based auto-encoder for jointly optimizing Bit-Error-Rate (BER) and Peak-to-Average Power Ratio (PAPR) in OFDM systems is proposed in [97].”

• cites [97] 97. N. C. Luong, D. T. Hoang, S. Gong, D. Niyato, P. Wang, Y. C. Liang, and D. I. Kim, "Applications of deep reinforcement learning in commu- nications and networking: A survey," arXiv preprint arXiv:1810.07862, 2018. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Deep transfer learning is a new direction of research to reduce dependence of learning on large amounts of data.

“Deep transfer learning is another new direction of research to reduce the dependence of learning from the required large amount of data.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

A single photon source can be considered as an ideal source for generating quantum information and can generate pulses with mean number equal to one and zero variance.

“Regarding the quantum information sources, a single photon source can be considered as an ideal source for generating quantum information and this can generate pulses having the mean number of pulses equal to one and zero variance [120].”

• cites [120] 120. J. D. Franson, "Quantum communications," in proc. of Technical Digest. Summaries of papers presented at the Quantum Electronics and Laser Science Conf. (Post conf. Technical Digest), May 2001, pp. 171-. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S4 (based).

The limitation of required computational power for training an ANN creates challenges for practical applications.

“Despite ANNs being promising, the limitation of required computational power for the training of an ANN creates challenges for their practical applications.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

For polarization encoding, polarization is not preserved while transmitting quantum signals via optical fibers, and non-intrusive polarization control is necessary to preserve information.

“For the first approach, the polarization is not preserved while transmitting quantum signals via optical fibers and it is necessary to have a non-intrusive polarization control to preserve the information being transmitted in the quantum domain.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Realization of an ideal photon source in practical quantum communications is very challenging because it occupies very large space and requires trained technicians.

“However, the realization of such an ideal photon source in practical quantum communications is very challenging since it occupies very large space and requires trained technicians.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Spread spectrum based multiple access techniques, which can send photons of multiple users via the medium by sharing frequency band, time window and route, are promising in the context of quantum communications.

“Furthermore, spread spectrum based multiple access techniques, which can send the photons of multiple users via the medium (optical or free-space) by sharing the frequency band, time window and the route, seem promising in the context of quantum communications [130].”

• cites [130] 130. C. H. Bennett and S. J. Wiesner, "Communication via one-and twoparti- cle operators on einstein-podolsky-rosen states," Phys. Rev. Lett., vol. 69, no. 20, p. 2881, Nov. 1992. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

Fainted lasers are mostly used in quantum cryptographic key distribution systems, while four-wave mixing is useful for optical processing devices such as parametric amplifiers and wavelength converters.

“Out of these, fainted lasers is mostly used in quantum cryptographic key distribution systems while the four-wave mixing process is useful for optical processing devices such as parametric amplifiers and wavelength converters.”

May combine two claims.

Construction of network entities with quantum Internet requires quantum switches/routers and repeaters, which is difficult due to no-cloning theorem.

“The first challenge is regarding the construction of network entities with quantum Internet which requires quantum switches/routers and repeaters, which becomes difficult due to no-cloning theorem 122].”

• cites [122] 122. L. Gyongyosi, S. Imre, and H. V. Nguyen, "A survey on quantum channel capacities," IEEE Commun. Surveys Tuts., vol. 20, no. 2, pp. 1149-1205, Jan. 2018.Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

One important advantage of utilizing quantum domain in communications is high degrees of freedom.

“One important advantage of utilizing quantum domain in communications is high degrees of freedom.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Emerging applications of DL in communication networks include dynamic spectrum access, joint user association, data caching and offloading, security, connection preservation, traffic routing, resource sharing, and power control.

“where the highlighted applications are: dynamic spectrum access, joint user association, data caching and offloading, security, connection preservation, traffic routing, resource sharing, and power control.”

• cites [96] 96. A. Zappone, L. Sanguinetti, and M. Debbah, "User association and load balancing for massive MIMO through deep learning," arXiv preprint arXiv:1812.06905, 2018. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Recent advances in GPUs have provided a breakthrough by speeding up training through simultaneous multiple weight calculations.

“the recent advances in Graphical processing units (GPUs) have provided the initial breakthrough by speeding up the training process through their ability to simultaneously perform multiple weight calculations operations.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

DNN based architectures for symbol detection in MIMO systems have been proposed, and a DNN-based localization method exploiting fingerprint-based and channel measurements for M-MIMO systems has been proposed.

“Furthermore, DNN based architectures for symbol detection in MIMO systems have been proposed in [98], and a DNN-based localization method exploiting fingerprintbased and channel measurements for M-MIMO systems has been proposed in [99].”

• cites [98] 98. M. Kim, W. Lee, and D. Cho, "A novel PAPR reduction scheme for OFDM system based on deep learning," IEEE Commun. Letters, vol. 22, no. 3, pp. 510-513, Mar. 2018. sourceNot checked — no full text of the source
• cites [99] 99. N. Samuel, T. Diskin, and A. Wiesel, "Learning to detect," arXiv preprint arXiv:1805.07631, 2018. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

Quantum communications has great potential to provide absolute randomness and security, carry much more information, and significantly enhance transmission quality compared to classical binary communications.

“In comparison to the classical binary based communications systems, quantum communications has the great potential to provide absolute randomness and security, to carry much more information and to significantly enhance the transmission quality.”

Probably not a substantial claim: may not pass S3 (load-bearing).

A deep reinforcement learning based power control method for spectrum sharing in cognitive radios has been proposed.

“A deep reinforcement learning based power control method for spectrum sharing in cognitive radios has been proposed in [110].”

• cites [110] 110. K. M. Thilina, K. W. Choi, N. Saquib, and E. Hossain, "Machine learning techniques for cooperative spectrum sensing in cognitive radio networks," IEEE J. Sel. Areas Commun., vol. 31, no. 11, pp. 2209-2221, Nov. 2013. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

A detector design by unfolding iterative calculations into ANN layers for MIMO decode and forward relay channels has been proposed.

“A detector design by unfolding iterative calculations into ANN layers for MIMO decode and forward relay channels has been proposed in [106].”

• cites [106] 106. J. R. Hershey, J. L. Roux, and F. Weninger, "Deep unfolding: Model-based inspiration of novel deep architectures," arXiv preprint arXiv:1409.2574, 2014. sourceNot checked — no full text of the source

May combine two claims.

Quantum-based techniques are able to execute tasks much faster and beyond the capability of classical systems.

“Furthermore, quantum-based techniques are able to execute the tasks much faster and beyond the capability of the classical systems [121].”

• cites [121] 121. A. N. Pinto, N. A. Silva, N. J. Muga, Ã. J. Almeida, and D. F. Pereira, "Quantum communications: An engineering approach," in proc. of Int. Conf. on Transparent Optical Networks, Jul. 2017, pp. 1-4. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Recent advancements in DNNs have enabled deployment in delay critical applications through offline training and online operations.

“The recent advancements in DNNs has enabled its deployment in even delay critical applications; which is achieved through offline training the DNNs and then performing the online operations (tracking, optimization, etc).”

Probably not a substantial claim: may not pass S3 (load-bearing).

Quantum channel/error correction coding is of significant importance for practical design of quantum-assisted communication protocols to approach theoretical achievable capacity.

“Furthermore, quantum channel/error correction coding is of significant importance for the practical design of quantumassisted communication protocols to approach closer to the theoretical achievable capacity.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

DL can be seen as an end-to-end solution for replacing sequential blocks based processing methods for estimation and decoding at receivers.

“DL can be seen as an end-to-end solution for replacing the sequential blocks based processing methods for estimation and decoding of information at the receivers.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

For phase encoding, polarization control is not required, but an optical carrier is needed to retrieve phase information.

“While the second approach based on the photons' phase does not require the polarization control but an optical carrier (either propagated along with the quantum signal or generated locally at the receiving side) is needed in order to retrieve the phase information.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

The combination of QC and ML in QML can produce and recognize statistical data patterns which classical computers and classical ML are unable to perform effectively.

“Also, it is envisioned that the combination of QC and ML features together in the framework of QML can produce and recognize the statistical data patterns which classical computers and classical ML are unable to perform effectively.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The Quantum Decision Theory (QDT) model proposed in [150] is a generic theory applicable to any type of decision making, and its mathematical structure is common to both decision theory and quantum measurements.

“In contrast to the previous quantumlike models which mainly utilize several fitting parameters for the construction of some models to describe particular effects of a use-case, the QDT model proposed in [150] is considered as a generic theory applicable to any type of decision making, and its mathematical structure is common to both the decision theory and quantum measurements.”

• cites [150] 150. M. Ashtiani and M. A. Azgomi, "A survey of quantum-like approaches to decision making and cognition," IEEE Trans. Inf. Theory, vol. 75, pp. 49-80, May 2015. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

The recent success of quantumassisted and data-driven learning methods in communication networks has a clear motivation to consider these as enablers of future heterogeneous networks.

“The recent success of quantumassisted and data-driven learning methods in communication networks (discussed in previous sections) has a clear motivation to consider these as enablers of future heterogeneous networks.”

May combine two claims.

Harmful quantum perturbations are a crucial issue for the application of quantum technology in communications, and their harmful effects can be mitigated by employing Quantum Error Correction Codes (QECCs).

“One of the crucial issues for the application of quantum technology in communications related applications is the presence of harmful quantum perturbations, whose harmful effects can be mitigated by employing Quantum Error Correction Codes (QECCs) [141].”

• cites [141] 141. M. Hsieh and M. M. Wilde, "Trading classical communication, quantum communication, and entanglement in quantum shannon theory," IEEE Trans. Inf. Theory, vol. 56, no. 9, pp. 4705-4730, Sep. 2010.Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

For the realization of consciousness and responsiveness in massively connected heterogeneous networks of future, very robust processing and learning capabilities will be required.

“For the realization of such consciousness and responsiveness in massively connected heterogeneous networks of future, very robust processing and learning capabilities will be required.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Long-distance quantum communication can be suitably realised through the physical platform of photons.

“Long-distance quantum communication can be suitably realised through the physical platform of photons, where an open research problem is the loss of photons in quantum channels.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Development of transducers to photonic states can help in traveling long distances with minimal decoherence.

“Development of transducers to photonic states can help in traveling long distances with minimal decoherence.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Recently, Google has announced a 72 qubit superconducting quantum computer.

“Recently, Google has announced a 72 qubit superconducting quantum computer [195].”

• cites [195] 195. K. Klemic and J. Zeigler, Eds., Future Directions of Quantum Informa- tion Processing: A Workshop on the Emerging Science and Technology of Quantum Computation, Communication, and Measurement. Basic Research Innovation and Collaboration Center in Arlington, VA: Virginia Tech Applied Research Corporation (VT-ARC), Aug. 2016.Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

Intelligent proactive caching helps in reducing the delay and power consumption in data routing and delivery, and it offers a significant performance improvement for all type of users.

“This concept helps in reducing the delay and power consumption in data routing and delivery, and it offers a significant performance improvement for all type of users; e.g., the smartphones designed for previous generations also equally benefit from this.”

May combine two claims.

Quantum Search Algorithm (QSA) can reduce the computational complexity of mmWave and VLC-based localization algorithms while achieving the same performance as that of a full search.

“In this regard, authors in [35] showed the possibilities of utilizing QSA for reducing the computational complexity of mm-Wave based and VLC-based localization algorithms while achieving the same performance as that of a full search.”

• cites [35] 35. P. Botsinis, D. Alanis, S. Feng, Z. Babar, H. V. Nguyen, D. Chandra, S. X. Ng, R. Zhang, and L. Hanzo, "Quantum-assisted indoor localization for uplink mm-Wave and downlink visible light communication systems," IEEE Access, vol. 5, pp. 23 327-23 351, Jul. 2017. sourceNot checked — no full text of the source

May combine two claims.

A novel quantum routing mechanism executing quantum circuits in parallel at intermediate nodes is independent of the number of routing hops and is closer to the optimum in terms of time taken to teleport a quantum state.

“To address this, a novel approach of quantum routing mechanisms by executing quantum circuits in parallel at the intermediate nodes has been recently proposed in [134] and it has been shown that the proposed quantum routing approach is independent of the number of routing hops and is closer to the optimum in terms of time taken to teleport a quantum state.”

• cites [134] 134. J. Du, "Terahertz essential devices technique research based on quantum communication," in proc. of IEEE Advanced Information Technology, Electronic and Automation Control Conf., Oct. 2018, pp. 1927-1930.Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

The quantum Internet enables the transmission of qubits from one quantum computer to another.

“Also, another promising application area is the quantum Internet, which enables the transmission of Qubits from one quantum computer to another [119].”

• cites [119] 119. S. Imre, "Quantum communications: Explained for communication engi- neers," IEEE Commun. Mag., vol. 51, no. 8, pp. 28-35, Aug. 2013. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Quantum teleportation utilizes the quantum entanglement principle to transfer a particular quantum state to another place using classical bits rather than quantum bits.

“Another important application of quantum communications is quantum teleportation, which utilizes the quantum entanglement principle to transfer a particular quantum state to another place with the quantum devices by using the classical bits rather than the quantum bits [128].”

• cites [128] 128. K. Inoue, "Quantum key distribution technologies," IEEE J. Sel. Topics Quantum Electron., vol. 12, no. 4, pp. 888-896, Jul. 2006. sourceNot checked — no full text of the source

May combine two claims.

Quantum Support Vector Machines (QSVMs) have been implemented, and an exponential speed-up of QSVMs over classical SVMs has been reported.

“The implementation of Quantum SVMs (QSVMs) has been presented in [185], where an exponential speed-up of QSVMs over classical SVMs has been reported.”

• cites [185] 185. A. Kapoor, N. Wiebe, and K. Svore, "Quantum perceptron models," in proc. of Advances in Neural Information Processing Systems, 2016, pp. 3999-4007.Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S4 (based).

The provision of AI in 5G networks is only limited to perform different isolated tasks of optimization, control, and management nature.

“However, this provision is only limited to perform different isolated tasks of optimization, control, and management nature.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Superconducting qubits and ion traps are the popular forms of the current architecture for quantum computation, which are arrays of interacting qubits that are continuously controlled via external pulses to implement the desired operations.

“Superconducting qubits and ion traps are the popular forms of the current architecture for quantum computation, which are arrays of interacting qubits that are continuously controlled via external pulses to implement the desired operations [192].”

• cites [192] 192. M. K. Saggi and S. Jain, "A survey towards an integration of big data analytics to big insights for value-creation," Information Processing & Management, vol. 54, no. 5, pp. 758-790, 2018. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

An open research problem is the loss of photons in quantum channels.

“Long-distance quantum communication can be suitably realised through the physical platform of photons, where an open research problem is the loss of photons in quantum channels.”

Probably not a substantial claim: may not pass S4 (based).

The voluminous amount of data produced from massive IoT devices can only be effectively utilized to provide instantaneous services if a very high computational capability is leveraged to the system.

“In the concept of fully integrated smart cities in 6G and beyond communications, the voluminous amount of data instantaneously produced from the massive amount of IoT devices (almost everything connected to the network) can only be effectively utilized to provide instantaneous (runtime) services if a very high computational capability is leveraged to the system.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Intelligently caching the data at serving stations in the proximity of mobile users can enable offloading heavy traffic from the network backhaul and reducing the network latency for popular contents.

“Intelligently caching the data at serving stations (e.g., BSs) in the proximity of mobile users can enable offloading heavy traffic from the network backhaul and reducing the network latency for popular contents, through an instantaneous service from the network edge.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The framework of DL has a recognized potential in multi-objective optimization.

“Moreover, the framework of DL has a recognized potential in multi-objective optimization, see e.g., [203].”

• cites [203] 203. H. Yetgin, K. T. K. Cheung, and L. Hanzo, "Multi-objective routing optimization using evolutionary algorithms," in IEEE Wireless Commun. and Networking Conf. IEEE, 2012, pp. 3030-3034. sourceNot checked — no full text of the source

May combine two claims.

Quantum-assisted precoding provides better BER performance compared to the conventional PSO algorithm, while keeping the same computational complexity.

“Via numerical results, it was shown that quantumassisted precoding provides better BER performance as compared to the conventional PSO algorithm, while keeping the same computational complexity.”

• cites [38] 38. P. Botsinis, D. Alanis, Z. Babar, H. V. Nguyen, D. Chandra, S. X. Ng, and L. Hanzo, "Quantum-aided multi-user transmission in non-orthogonal multiple access systems," IEEE Access, vol. 4, pp. 7402-7424, Jul. 2016.Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Integrating ML in almost everything has recognized exposure of new types of privacy and security vulnerabilities, while the current understanding of these aspects is very limited.

“For example, in the paradigm of integrating ML in almost everything has recognized exposure of new types of privacy and security vulnerabilities, while the current understanding of these aspects is very limited [208].”

• cites [208] 208. S. Verma, Y. Kawamoto, Z. M. Fadlullah, H. Nishiyama, and N. Kato, "A survey on network methodologies for real-time analytics of massive IoT data and open research issues," IEEE Commun. Surveys Tuts., vol. 19, no. 3, pp. 1457-1477, Apr. 2017. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S4 (based).

The simulation of a quantum computer on a classical computer is a computationally hard problem.

“This is because the simulation of a quantum computer on a classical computer is a computationally hard problem.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The achievable rate for the transmission of quantum data over a quantum channel is given by the Lloyd-Shor-Devetak (LSD) coding theorem.

“Also, regarding the transmission of quantum data over a quantum channel, the achievable rate is given by the Lloyd-Shor-Devetak (LSD) coding theorem [124,137].”

• cites [124] 124. C. H. Bennett, P. W. Shor, J. A. Smolin, and A. V. Thapliyal, "Entanglement-assisted capacity of a quantum channel and the reverse shannon theorem," IEEE Trans. Inf. Theory, vol. 48, no. 10, pp. 2637- 2655, Oct. 2002. sourceNot checked — no full text of the source
• cites [137] 137. A. S. Holevo, "The capacity of the quantum channel with general signal states," IEEE Trans. Inf. Theory, vol. 44, no. 1, pp. 269-273, Jan. 1998. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

The role of quantum cryptography is highly recognized as a strong potential for the future of security/privacy.

“The role of quantum cryptography, through its features of generating secret keys to multiple legitimate parties in a Vernam one-time pad cryptosystems, is highly recognized as a strong potential for the future of security/privacy [127].”

• cites [127] 127. V. Sharma and S. Banerjee, "Analysis of quantum key distribution based satellite communication," in proc. of Int. Conf. on Computing, Commun. and Networking Technol., Jul. 2018, pp. 1-5. sourceNot checked — no full text of the source

May combine two claims.

For various learning tasks, QML algorithms can provide exponential speed-ups over classical ML algorithms involving large dimensional data, in both supervised and unsupervised learning approaches.

“For various learning tasks, QML algorithms can provide exponential speed-ups over classical ML algorithms [42], involving large dimensional data, in both supervised and unsupervised learning approaches.”

• cites [42] 42. S. Lloyd, M. Mohseni, and P. Rebentrost, "Quantum algorithms for supervised and unsupervised machine learning," arXiv preprint arXiv:1307.0411, 2013. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

This approach will face scalability issues even if they are highly capable of maintaining quantum coherence for a longer duration.

“However, this approach will face scalability issues even if they are highly capable of maintaining quantum coherence for a longer duration.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

To implement such quantum repeaters, the decoherence effects imposed by the quantum channels need to be dealt with.

“Moreover, to implement such quantum repeaters, the decoherence effects imposed by the quantum channels need to be dealt with.”

Probably not a substantial claim: may not pass S3 (load-bearing).

QKD provides a secret key to two legitimate parties in a Vernam one-time pad cryptosystem.

“This issue can be addressed by the QKD, also called the quantum cryptography, which provides a secret key to two legitimate parties in a Vernam one-time pad cryptosystem [127].”

• cites [127] 127. V. Sharma and S. Banerjee, "Analysis of quantum key distribution based satellite communication," in proc. of Int. Conf. on Computing, Commun. and Networking Technol., Jul. 2018, pp. 1-5. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

The development of highly consistent and controllable qubits and quantum logical operations is a fundamental need in the realization of reliable large-scale quantum computers.

“The development of highly consistent and controllable qubits and quantum logical operations is a fundamental need in the realization of reliable large-scale quantum computers, where quantum error correction methods can be used in bringing improvements in system reliability.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Fault-tolerant quantum computers are required to solve harder problems.

“More specifically, fault-tolerant quantum computers are required to solve harder problems.”

Probably not a substantial claim: may not pass S3 (load-bearing).

A quantum annealing chipset is commercially available from the company D-Wave.

“In terms of practical implementation, a quantum annealing chipset is already commercially available from the company D-Wave1 [135].”

• cites [135] 135. S.-T. Cheng, C.-Y. Wang, and M.-H. Tao, "Quantum communication for wireless wide-area networks," IEEE J. Sel. Areas Commun., vol. 23, no. 7, pp. 1424-1432, Jul. 2005. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Parallelization can partially facilitate the resolution by allowing the simulation of more qubits in less time.

“Parallelization can partially facilitate the resolution by allowing the simulation of more qubits in less time.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The scope of DL and evolutionary game theory for dynamic spectrum access in a cognitive radio network has been discussed.

“Alos, the scope of DL and evolutionary game theory for dynamic spectrum access in a cognitive radio network has been discussed in [237].”

• cites [237] 237. S. Singhal, S. Gupta, and A. Sinha, "Role of artificial intelligence in cognitive radio networks," in proc. of Sensing Techniques for Next Generation Cognitive Radio Networks. IGI Global, 2019, pp. 258-279. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

ML methods have been thoroughly reviewed for cognitive radios.

“Therefore, ML methods have thoroughly been reviewed for cognitive radios, see e.g., [235].”

• cites [235] 235. Z. Ding, D. W. K. Ng, R. Schober, and H. V. Poor, "Delay minimization for NOMA-MEC offloading," IEEE Signal Process. Lett., vol. 25, no. 12, pp. 1875-1879, Dec. 2018. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The TDD scheme is usually preferred over FDD in large-scale multi-antenna systems to avoid estimating the massive amount of downlink channels at the user end, but this causes the pilot contamination problem that limits capacity, which can become severe in high mobility scenarios.

“The TDD scheme is usually preferred over FDD in such large-scale multi-antenna systems for the reasons to avoid the tedious task of estimation of the massive amount of downlink channels at the userend, however, this causes the pilot contamination problem imposing limitations over the capacity, which can get severe in high mobility scenarios.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

This popularity of auto-encoding of a single end-to-end communication link in a DNN is a strong motivation to extend this idea of auto-encoding an entire heterogeneous wireless access network in a QC-assisted very deep ANN.

“This popularity of auto-encoding of a single end-to-end communication link in a DNN, is a strong motivation to extend this idea of auto-encoding an entire heterogeneous wireless access network in a QC-assisted very deep ANN to enable an instantaneous unique encoding response for service provisions through hybrid 1∼6G network infrastructures.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The NOMA principle is believed to be a convenient method for realizing massive connectivity in the context of IoT in emerging 5G networks.

“The NOMA principle is also believed to be a convenient method in realizing massive connectivity in the context of IoT in emerging 5G networks [227][228][229].”

• cites [227][228][229]Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

The user-centric transmission not only overcomes the inter-cell interference encountered in the conventional cellular networks but also provides macro-scale diversity.

“The user-centric transmission not only overcomes the inter-cell interference encountered in the conventional cellular networks but also provides macro-scale diversity [219].”

• cites [219] 219. X.-Z. Cui, X.-L. Yin, H. Chang, Y.-L. Guo, Z.-J. Zheng, Z.-W. Sun, G.-Y. Liu, and Y.-J. Wang, "Analysis of an adaptive orbital angular momentum shift keying decoder based on machine learning under oceanic turbulence channels," Optics Communications, vol. 429, pp. 138-143, 2018. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

Truly intelligent self-sustainability in communication networks will be materialized in 6G and beyond communication, with QC-assisted ML and QML as the potential enablers.

“The aforementioned indications assert that the truly intelligent self-sustainability in communication networks will be materialized in 6G and beyond communication with QCassisted ML and QML as the potential enablers.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

The advanced ML concepts of DL, online-learning, transfer learning, and auto-encoder can together help overcome the performance-limiting factors of NOMA.

“Moreover, the advanced ML concepts of DL, online-learning, transfer learning, and auto-encoder can together help to overcome these limitations.”

• AssumptionRests on an assumption: NOMA has recognized fundamental performance-limiting factors: high computation complexity for SIC with a massive number of users, inefficient transmission-time consumption, and estimation/feeding of CSI for a large number of users, especially for very fast time-varying channels such as mmWave V2V channels.

Probably not a substantial claim: may not pass S2 (checkable).

Because both NOMA and MEC are intelligence- and computing-power-demanding technologies, QML can be seen as a strong potential enabler for them.

“Both NOMA and MEC being intelligence and computing power demanding technologies, QML can be seen as a strong potential enabler.”

• AssumptionRests on an assumption: NOMA has recognized fundamental performance-limiting factors: high computation complexity for SIC with a massive number of users, inefficient transmission-time consumption, and estimation/feeding of CSI for a large number of users, especially for very fast time-varying channels such as mmWave V2V channels.

Probably not a substantial claim: may not pass S3 (load-bearing).

Hardware support for successive interference cancellation (SIC) at user nodes has been released as a NOMA-chipset.

“The hardware support for such SIC at user nodes is also released as NOMA-chipset [226].”

• cites [226] 226. Y. Ding and B. D. Rao, "Dictionary learning-based sparse channel representation and estimation for FDD massive MIMO systems," IEEE Trans. Wireless Commun., vol. 17, no. 8, pp. 5437-5451, Aug. 2018. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

The antenna tweaking parameter for configuring antennas at the user side opens the possibility of intelligently manipulating them through ML for overall performance optimization.

“The antenna tweaking parameter for configuring antennas at the userside opens the possibility of intelligently manipulating them through ML for overall performance optimization.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The role of AI in enabling cognitive radio networks has been strongly endorsed.

“The role of AI in enabling cognitive radio networks has been strongly endorsed in [236].”

• cites [236] 236. C. Clancy, J. Hecker, E. Stuntebeck, and T. O'Shea, "Applications of machine learning to cognitive radio networks," IEEE Wireless Commu- nications, vol. 14, no. 4, pp. 47-52, Aug. 2007. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

All the tasks for operating cognitive radios (sensing spectrum gaps, network interference analysis, CSI estimation, power control, and dynamic resources management) have natural connections with the capabilities offered by ML methods.

“All the tasks for operating cognitive radios, e.g., sensing spectrum gaps, network (spatial) interference analysis, CSI estimation, power control, and dynamic resources management, etc, have natural connections with the deals offered by ML methods.”

Probably not a substantial claim: may not pass S4 (based). May combine two claims.

The training of DNNs is usually performed through backpropagating the gradient of the loss function, however, the unavailability of prior knowledge of channel statistics, in end-to-end DNN systems, prevents the back-propagation of gradients.

“The training of DNNs is usually performed through backpropagating the gradient of the loss function, however, the unavailability of prior knowledge of channel statistics, in end-to-end DNN systems, prevents the back-propagation of gradients.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The 5G era is expected to lay the foundation of intelligent communication networks by introducing basic AI-based operations such as intelligent resource management, intelligent management of services provision, and intelligent control.

“The 5G era is expected to lay the foundation of intelligent communication networks [54] by introducing some basic AI based operations, such as intelligent resource management, intelligent management of services provision, intelligent control, etc.”

• cites [54] 54. R. Li, Z. Zhao, X. Zhou, G. Ding, Y. Chen, Z. Wang, and H. Zhang, "Intelligent 5G: when cellular networks meet artificial intelligence," IEEE Wireless Communications, vol. 24, no. 5, pp. 175-183, Oct. 2017. sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

With the growing size and complexity of communication networks, the future of cognitive radios seems to be in sensing and optimization through advanced ML methods assisted with quantum speed-ups.

“With the growing size and complexity of communication networks, the future of cognitive radios seems to be in the sensing and optimization through advanced ML methods assisted with quantum speed-ups.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Software-defined (fluid) antennas are expected to provide rich diversity at the user nodes at a level similar to what M-MIMO systems provide at the BS side.

“The software-defined (fluid) antennas are expected to provide rich diversity at the user nodes, of similar level which M-MIMO systems can provide at the BS side.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

These studies involve the assumption of prior available channel statistics (i.e., available differentiable channel model).

“These studies involve the assumption of prior available channel statistics (i.e., available differentiable channel model).”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

The processing and learning capability required for manipulating multi-dimensional, large-sized, and highly-dynamic (V2V) streaming data in such large-scale software-defined multi-antenna systems is far beyond the reach of classical computational and learning methods.

“The processing and learning capability required for manipulating multi-dimensional, large-sized, and highly-dynamic (V2V) streaming data in such large-scale software-defined multi-antenna systems; is far beyond the reach of classical computational and learning methods.”

May combine two claims.

The number of available configurable parameters at a communication node is rapidly increasing and is projected to be 2000+ parameters in a typical 5G node.

“The amount of available configurable parameters at a communication node is rapidly increasing, which is projected to be 2000+ parameters in a typical 5G node [54].”

• cites [54] 54. R. Li, Z. Zhao, X. Zhou, G. Ding, Y. Chen, Z. Wang, and H. Zhang, "Intelligent 5G: when cellular networks meet artificial intelligence," IEEE Wireless Communications, vol. 24, no. 5, pp. 175-183, Oct. 2017. sourceNot checked — no full text of the source