Full metadata record
DC FieldValueLanguage
dc.contributorDepartment of Electrical and Electronic Engineeringen_US
dc.contributor.advisorYu, Changyuan (EEE)en_US
dc.creatorLi, Cheng-
dc.identifier.urihttps://theses.lib.polyu.edu.hk/handle/200/14662-
dc.languageEnglishen_US
dc.publisherHong Kong Polytechnic Universityen_US
dc.rightsAll rights reserveden_US
dc.titleResearch on performance optimization of low-latency high-speed passive optical networksen_US
dcterms.abstractPassive Optical Network (PON) technology, as the core solution for optical fiber access, has become the key to supporting the construction of “dual gigabit” networks due to its advantages, like low cost, high bandwidth, and easy scalability. With the explosive growth of the digital economy and intelligent scenarios such as industrial Internet, 8K live streaming, and virtual reality/ augmented reality, traditional gigabit passive optical networks are gradually facing bandwidth bottlenecks. In 2025, the Ministry of Industry and Information Technology of China officially launched the pilot program of 50G optical networks. Based on 50G PON technology, it promoted the upgrade of network infrastructure towards high speed, high reliability, low latency, and facilitated the realization of the “Digital China” strategy. Huawei has also launched the world's first commercial 50G PON solution. The “three-mode integration” optical module and hard slicing technology realize on-demand bandwidth allocation and deterministic experience, which applies to high-value scenarios such as smart parks and telemedicine.en_US
dcterms.abstractIntensity modulation direct detection (IMDD) technology is used in traditional PON. IMDD technology has a simple structure and low cost, and can achieve a 50G transmission rate. However, time division multiplexing (TDM) is difficult to meet the application requirements of many low-latency scenarios. Therefore, this paper proposes a frequency division multiplexing (FDM) -PON architecture to meet the low latency requirements. After the rate exceeds 50G, the detection sensitivity of IMDD is not enough, and coherent detection techniques are considered to improve the detection sensitivity. Founded on digital signal processing (DSP) technology, coherent PON can support high-speed and long-distance optical communication transmission. However, there are problems such as slow convergence in coherent detection. We also proposed a 200G PON coherent optical burst mode digital signal processing (Co-BM-DSP) which can achieve faster convergence and is suitable for coherent burst mode digital signal processing.en_US
dcterms.abstractIn this thesis, we proposed some structures of high-speed PON uplink systems to achieve low latency and fast convergence. A multi-point to-point FDM-PON structure is proposed to meet the low delay requirements of PON. The experimental results show that by combining the proposed FDM-PON architecture with entropy loading and probabilistic constellation shaping techniques, multiple users can obtain the optimal bit rate while meeting the 20% forward error correction threshold, thereby maintaining a total throughput of 50 Gbps.en_US
dcterms.abstractWe also explored the next generation using a 200G PON system. To achieve the fast convergence processing of burst signals in the 200G Coherent Time Division Multiple Access (CO-TDMA) system, 10ns preambles were designed. The designed preambles can be decomposed into a frequency-tone Preamble A containing 128 symbols and Preamble B containing 192 equal-amplitude zero autocorrelation (CAZAC) symbols. Preamble A can be used to estimate the states of polarization, frequency offset, and sampling phase offset (SPO). The Preamble B CAZAC symbol can be used to estimate the frame synchronization and tap coefficients of a multiple-input multiple-output (MIMO) equalizer. Built upon the designed preamble, the feedforward algorithm can quickly estimate the SPO values of the timing recovery and tap coefficient of the MIMO equalizer, and initialize them to ensure that the feedback algorithm is in the correct state. The experimental results also show that the preamble we designed successfully helped the 200G PON achieve fast convergence. The Co-BM-DSP with a compact preamble is more likely to be applied to the future CO-TDMA.en_US
dcterms.abstractIn short, we expect that the next generation of PON technology will move towards a 200G rate, and the research on low delay and fast convergence we conducted in this paper may provide some help for the development of the next generation of PON technology.en_US
dcterms.extentxvii, 113 pages : color illustrationsen_US
dcterms.isPartOfPolyU Electronic Thesesen_US
dcterms.issued2026en_US
dcterms.educationalLevelEng.D.en_US
dcterms.educationalLevelAll Doctorateen_US
dcterms.accessRightsrestricted accessen_US

Files in This Item:
File Description SizeFormat 
9107.pdfFor All Users (off-campus access for PolyU Staff & Students only)4.13 MBAdobe PDFView/Open


Copyright Undertaking

As a bona fide Library user, I declare that:

  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.

By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

Show simple item record

Please use this identifier to cite or link to this item: https://theses.lib.polyu.edu.hk/handle/200/14662