This thesis investigates two systems that exploit multi-user interference in wireless communication in which multiple users transmit simultaneously. The thesis consists of two parts that aim for two different wireless network setups: the first part focuses on the decoding of convolutional-coded physical-layer network coding (PNC) in a two-way relay network (TWRN); the second part presents a scalable multi-user multiple-input multiple-output (MU-MIMO) architecture supporting real-time MIMO decoding. Both systems take advantage of multi-user interference to boost the system performance. The investigations of the two systems involve both theoretical analysis and system prototyping on software radio. / The first part of this thesis investigates the decoding process of asynchronous convolutional-coded PNC systems. Specifically, we put forth a layered decoding framework for convolutional-coded PNC consisting of three layers: symbol realignment layer, codeword realignment layer, and joint channel-decoding network coding (Jt-CNC) decoding layer. Our framework can deal with phase asynchrony (phase offset) and symbol arrival-time asynchrony (symbol misalignment) between the signals simultaneously transmitted by multiple sources. A salient feature of this framework is that it can handle both fractional and integral symbol misalignments. For the decoding layer, instead of Jt-CNC, previously proposed PNC decoding algorithms (e.g., XOR-CD and reduced-state Viterbi algorithms) can also be used with our framework to deal with general symbol misalignments. Our Jt-CNC algorithm based on belief propagation (BP) is BER-optimal for synchronous PNC and near optimal for asynchronous PNC. Extending beyond convolutional codes, we further generalize the Jt-CNC decoding algorithm for all cyclic codes. Our simulations show that Jt-CNC outperforms the previously proposed XOR-CD algorithm and reduced-state Viterbi algorithm by 2 dB for synchronous PNC. For both phase-asynchronous and symbolasynchronous PNC, Jt-CNC performs better than the other two algorithms. Importantly, for real wireless network experimentation, we implemented our decoding algorithm in a PNC prototype built on the USRP software radio platform. Our experiment shows that the proposed Jt-CNC decoder works well in practice. / The second part of the thesis presents BigStation, a scalable architecture that enables real-time signal processing in large-scale MU-MIMO systems with tens or hundreds of antennas. Our strategy to scale the system is to extensively parallelize the MU-MIMO processing on many simple and low-cost commodity computing devices. Our design can incrementally support more antennas by proportionally adding more computing devices. To reduce the overall processing latency, which is a critical constraint for wireless communication, we parallelize the MU-MIMO decoding process with a distributed pipeline based on its computation and communication patterns. At each stage of the pipeline, we further use data partitioning and computation partitioning to speed up the signal processing. As a proof of concept, we have built a BigStation prototype based on commodity PC servers and SORA software radio platform. Our prototype employs 15 PC servers and can support realtime signal processing of 12 software radio antennas. Our results show that the BigStation is able to scale to tens to hundreds of antennas. With 12 antennas, our BigStation prototype can increase wireless throughput by 6.8 x with a low mean latency of 860μs. While this latency is not yet low enough for the 802.11 MAC, it already satisfies the real-time requirements of many existing wireless standards such as LTE and WCDMA. / 本論文探討了兩種可以利用多個無線用戶同時傳輸時所造成的多用戶干擾的無線通信系統。本文由旨在針對不同的無線網絡結構的兩部分構成：第一部分論述了雙向中繼網絡（TWRN）中卷積編碼的非同步物理層網絡編碼（PNC）的解碼; 第二部分介紹了一種支持實時解碼的可擴展的多用戶多輸入多輸出（MU-MIMO）架構。這兩種系統都利用了無線通信中的多用戶干擾以提高系統性能。在對這兩個系統的探討中，我們都進行了理論分析，並搭建了基於軟件無線電的原型系統。 / 論文的第一部分研究卷積編碼的非同步物理層網絡編碼系統的解碼過程。具體來說，我們提出了一種非同步物理層網絡編碼的分層解碼架構。該架構由三層組成：符號對齊層，碼字對齊層，以及聯合信道解碼網絡編碼（Jt-CNC）解碼層。我們的這種架構能夠處理由多個用戶同時發送信號時造成的相位不同步（相位偏移）和符號到達時間不同步（符號偏移）。這種架構的一個顯著特點是，它可以同時處理分數和整數的符號偏移。對於解碼層，除了Jt-CNC算法，先前提出的各種PNC解碼算法（例如，XOR-CD 和減少狀態的Viterbi算法）也可以應用於我們的解碼架構用來解決符號偏移問題。我們基於置信度傳播理論（BP）的Jt-CNC算法，對於同步PNC是比特差錯率（BER）最優的，對於非同步PNC是接近最優的。除了卷積碼以外，我們進一步將Jt-CNC解碼算法推廣到了所有循環碼。我們的仿真表明，對於同步PNC，Jt-CNC的性能超過先前提出的XOR-CD算法和減少狀態Viterbi算法2 dB。對於相位非同步和符號非同步PNC，Jt-CNC均優於前述兩種算法。重要的是，為了進行真實環境下的無線網絡實驗，我們在基於USRP軟件無線電平台的PNC原型系統上實現了我們的解碼算法。我們的實驗顯示，該Jt-CNC解碼器在真實環境下中表現優越。 / 論文的第二部分介紹了BigStation，一個可擴展的MU-MIMO系統架構。該架構支持帶有幾十甚至上百根天線的大規模MU-MIMO系統的實時信號處理。我們的系統擴展策略是將MU-MIMO信號處理並行化到大量簡單廉價的商用計算單元上。我們的這種設計可以通過成比例地增加計算單元以支持更多的天線。為了減少系統的整體處理延遲，我們根據MU-MIMO信號處理的計算和通信模式，利用分布式流水線把解碼過程並行化。在流水線的各級，我們進一步使用數據分割和計算分割來加速信號處理。作為概念驗證，我們基於商用個人電腦和SORA軟件無線電平台開發了BigStation原型系統。我們的原型使用了15 台個人電腦並且可以支持12 根軟件無線電天線的實時解碼。我們的實驗結果表明，BigStation系統能夠擴展到幾十甚至幾百根天線。當有12 根天線時，我們的BigStation原型系統將無線網絡吞吐量提高了6.8 倍，而平均延遲則低至860 μs。雖然這個延遲對於802.11 的MAC 還不足夠低，但它已滿足了許多現有的無線標准諸如LTE和WCDMA的實時性要求。 / Yang, Qing. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2015. / Includes bibliographical references (leaves 101-108). / Abstracts also in Chinese. / Title from PDF title page (viewed on 21, December, 2016). / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only.