{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T12:10:55Z","timestamp":1785413455808,"version":"3.56.0"},"reference-count":32,"publisher":"MDPI AG","issue":"14","license":[{"start":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T00:00:00Z","timestamp":1783468800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["42527803"],"award-info":[{"award-number":["42527803"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"award":["42527803"],"award-info":[{"award-number":["42527803"]}],"id":[{"id":"https:\/\/ror.org\/01h0zpd94","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>IEEE 1588 Precision Time Protocol (PTP) synchronization accuracy is degraded by delay asymmetry in physical links. Accurate prediction and compensation of this asymmetry are essential for high-precision timekeeping. This paper presents a multi-horizon delay asymmetry prediction framework validated on a cesium-referenced testbed. The experimental platform integrates a cesium atomic clock (frequency stability &lt;5\u00d710\u221212), a Syncedge C10 PTP analyzer, and an RK3568 embedded TimeReceiver, providing hardware-timestamped ground-truth measurements over 24 h periods on both copper and fiber links under real operating conditions. A unified multi-output encoder architecture is compared against per-horizon recurrent models and classical baselines to predict delay asymmetry at five future horizons (h\u2208{1,2,4,8,16}, corresponding to 0.125\u20132.0 s ahead). On copper links at an 8 s observation window, the unified model achieves a mean absolute error of 2.15 ns and an R2 of 0.714 at the nearest horizon, statistically tied with the best per-horizon model (2.16 ns, R2=0.713) while using 1.9\u20136.1\u00d7 fewer total parameters (655 K vs. 1.2\u20134.0M). On fiber links, all models achieve comparable accuracy with an MAE of 2.16\u20132.18 ns. The method requires ground-truth labels from a calibrated 1PPS or cesium reference; both training and inference require the same physical link and environmental conditions. Generalization to other settings remains untested.<\/jats:p>","DOI":"10.3390\/s26144335","type":"journal-article","created":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T16:16:18Z","timestamp":1783527378000},"page":"4335","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Multi-Horizon Delay Asymmetry Prediction and Compensation for IEEE 1588 Precision Time Protocol on Copper and Fiber Links by Means of Ground-Truth Measurements"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-0367-5103","authenticated-orcid":false,"given":"Junchao","family":"Wang","sequence":"first","affiliation":[{"name":"School of Electronics, Peking University, Beijing 100871, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shengping","family":"Xu","sequence":"additional","affiliation":[{"name":"School of Electronics, Peking University, Beijing 100871, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chuwen","family":"Tang","sequence":"additional","affiliation":[{"name":"School of Electronics, Peking University, Beijing 100871, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chen","family":"Yang","sequence":"additional","affiliation":[{"name":"School of Electronics, Peking University, Beijing 100871, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiayue","family":"Shen","sequence":"additional","affiliation":[{"name":"School of Electronics, Peking University, Beijing 100871, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianye","family":"Zhao","sequence":"additional","affiliation":[{"name":"School of Electronics, Peking University, Beijing 100871, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,7,8]]},"reference":[{"key":"ref_1","unstructured":"(2008). IEEE Standard for a Precision Clock Synchronization Protocol for Networked Measurement and Control Systems (Standard No. IEEE Std 1588-2008)."},{"key":"ref_2","unstructured":"(2019). IEEE Standard for a Precision Clock Synchronization Protocol for Networked Measurement and Control Systems (Revision) (Standard No. IEEE Std 1588-2019)."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"155660","DOI":"10.1109\/ACCESS.2020.3013669","article-title":"IEEE 1588 for Clock Synchronization in Industrial IoT and Related Applications: A Review on Contributing Technologies, Protocols and Enhancement Methodologies","volume":"8","author":"Idrees","year":"2020","journal-title":"IEEE Access"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"1551","DOI":"10.1109\/LCOMM.2014.2341604","article-title":"Clock Synchronization Over Communication Paths With Queue-Induced Delay Asymmetries","volume":"18","author":"Chaloupka","year":"2014","journal-title":"IEEE Commun. Lett."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"20575","DOI":"10.1109\/ACCESS.2021.3054164","article-title":"Clock Synchronization Algorithms Over PTP-Unaware Networks: Reproducible Comparison Using an FPGA Testbed","volume":"9","author":"Freire","year":"2021","journal-title":"IEEE Access"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"61370","DOI":"10.1109\/ACCESS.2020.2983421","article-title":"An Adaptive Fuzzy-PI Clock Servo Based on IEEE 1588 for Improving Time Synchronization Over Ethernet","volume":"8","author":"Nguyen","year":"2020","journal-title":"IEEE Access"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Zhang, Y., Li, H., Wang, S., and Chen, F. (2024). A Fuzzy-PI Clock Servo with Window Filter for Compensating Queue-Induced Delay Asymmetry in IEEE 1588 Networks. Sensors, 24.","DOI":"10.3390\/s24072369"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Wei, X., Xiong, X., Luo, Z., Wang, J., and Cheng, K. (2021, January 19\u201321). Fuzzy Adaptive PI Clock Synchronization Algorithm Based on Kalman Filter. Proceedings of the 2021 IEEE International Conference on Information Communication and Software Engineering (ICICSE), Chengdu, China.","DOI":"10.1109\/ICICSE52190.2021.9404084"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Novaes, C., Freire, I., Klautau, A., Almeida, I., and Medeiros, E. (2021, January 17\u201319). Analysis of Kalman Filtering for Clock Synchronization in PTP-Unaware Networks. Proceedings of the 2021 IEEE Latin-American Conference on Communications (LATINCOM), Santo Domingo, Dominican Republic.","DOI":"10.1109\/LATINCOM53176.2021.9647759"},{"key":"ref_10","unstructured":"Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A., Kaiser, L., and Polosukhin, I. (2017, January 4\u20139). Attention Is All You Need. Proceedings of the 31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"1181","DOI":"10.1016\/j.ijforecast.2019.07.001","article-title":"DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networks","volume":"36","author":"Salinas","year":"2020","journal-title":"Int. J. Forecast."},{"key":"ref_12","unstructured":"Oreshkin, B., Carpov, D., Chapados, N., and Bengio, Y. (2020, January 26\u201330). N-BEATS: Neural Basis Expansion Analysis for Interpretable Time Series Forecasting. Proceedings of the 8th International Conference on Learning Representations (ICLR 2020), Addis Ababa, Ethiopia."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"1748","DOI":"10.1016\/j.ijforecast.2021.03.012","article-title":"Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecasting","volume":"37","author":"Lim","year":"2021","journal-title":"Int. J. Forecast."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"69683","DOI":"10.1109\/ACCESS.2020.2984785","article-title":"Clock Synchronization in Wireless Sensor Networks Based on Bayesian Estimation","volume":"8","author":"Yang","year":"2020","journal-title":"IEEE Access"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Dutra, R., Freire, I., Bemerguy, P., Klautau, A., Almeida, I., and Medeiros, E. (2021, January 7\u201311). An LSTM-based Approach for Holdover Clock Disciplining in IEEE 1588 PTP Applications. Proceedings of the 2021 IEEE Global Communications Conference (GLOBECOM), Madrid, Spain.","DOI":"10.1109\/GLOBECOM46510.2021.9685776"},{"key":"ref_16","first-page":"11106","article-title":"Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting","volume":"35","author":"Zhou","year":"2021","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"ref_17","unstructured":"Wu, H., Xu, J., Wang, J., and Long, M. (2021, January 6\u201314). Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting. Proceedings of the 35th Conference on Neural Information Processing Systems (NeurIPS 2021), Online."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"20200209","DOI":"10.1098\/rsta.2020.0209","article-title":"Time-Series Forecasting with Deep Learning: A Survey","volume":"379","author":"Lim","year":"2021","journal-title":"Philos. Trans. R. Soc. A"},{"key":"ref_19","unstructured":"Nie, Y., Nguyen, N., Sinthong, P., and Kalagnanam, J. (2023, January 1\u20135). A Time Series is Worth 64 Words: Long-Term Forecasting with Transformers. Proceedings of the 11th International Conference on Learning Representations (ICLR 2023), Kigali, Rwanda."},{"key":"ref_20","unstructured":"Zhang, Y., and Yan, J. (2023, January 1\u20135). Crossformer: Transformer Utilizing Cross-Dimension Dependency for Multivariate Time Series Forecasting. Proceedings of the 11th International Conference on Learning Representations (ICLR 2023), Kigali, Rwanda."},{"key":"ref_21","unstructured":"Wu, H., Hu, T., Liu, Y., Zhou, H., Wang, J., and Long, M. (2023, January 1\u20135). TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis. Proceedings of the 11th International Conference on Learning Representations (ICLR 2023), Kigali, Rwanda."},{"key":"ref_22","unstructured":"Liu, Y., Hu, T., Zhang, H., Wu, H., Wang, S., Ma, L., and Long, M. (2024, January 7\u201311). iTransformer: Inverted Transformers Are Effective for Time Series Forecasting. Proceedings of the 12th International Conference on Learning Representations (ICLR 2024), Vienna, Austria."},{"key":"ref_23","unstructured":"Bai, S., Kolter, J.Z., and Koltun, V. (2018). An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling. arXiv."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"31729","DOI":"10.1038\/s41598-025-14908-x","article-title":"Comparative Analysis of Deep Learning Architectures in Solar Power Prediction","volume":"15","author":"AbdelMoety","year":"2025","journal-title":"Sci. Rep."},{"key":"ref_25","first-page":"11121","article-title":"Are Transformers Effective for Time Series Forecasting?","volume":"37","author":"Zeng","year":"2023","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"ref_26","unstructured":"Hyndman, R., and Athanasopoulos, G. (2018). Forecasting: Principles and Practice, OTexts. [3rd ed.]."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Chitty-Venkata, K., Mittal, S., Emani, M., Vishwanath, V., and Somani, A. (2023). A Survey of Techniques for Optimizing Transformer Inference. arXiv.","DOI":"10.1016\/j.sysarc.2023.102990"},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Ahn, H., Chen, T., Alnaasan, N., Shafi, A., Abduljabbar, M., Subramoni, H., and Panda, D. (2023, January 1\u20134). Performance Characterization of Using Quantization for DNN Inference on Edge Devices. Proceedings of the 7th IEEE International Conference on Fog and Edge Computing (ICFEC 2023), Bangalore, India.","DOI":"10.1109\/ICFEC57925.2023.00009"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"142400","DOI":"10.1109\/ACCESS.2021.3120769","article-title":"Timely Survey of Time-Sensitive Networking: Past and Future Directions","volume":"9","author":"Seol","year":"2021","journal-title":"IEEE Access"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Zaccaro, E., Ronchetti, L., Lometti, A., Troia, S., and Bregni, S. (2020, January 18\u201320). Performance Evaluation of PTP Synchronization Networks for 5G Front\/Mid-Haul Transport. Proceedings of the 2020 IEEE Latin-American Conference on Communications (LATINCOM), Santo Domingo, Dominican Republic.","DOI":"10.1109\/LATINCOM50620.2020.9282341"},{"key":"ref_31","unstructured":"(2017). IEEE Standard Profile for Use of IEEE 1588 Precision Time Protocol in Power System Applications (Standard No. IEEE Std C37.238-2017)."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"157398","DOI":"10.1109\/ACCESS.2021.3127852","article-title":"Towards an Increased Detection Sensitivity of Time-Delay Attacks on Precision Time Protocol","volume":"9","author":"Hamad","year":"2021","journal-title":"IEEE Access"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/14\/4335\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,10]],"date-time":"2026-07-10T11:48:53Z","timestamp":1783684133000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/14\/4335"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,8]]},"references-count":32,"journal-issue":{"issue":"14","published-online":{"date-parts":[[2026,7]]}},"alternative-id":["s26144335"],"URL":"https:\/\/doi.org\/10.3390\/s26144335","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,8]]}}}