與Wi-Fi系統共存的第五代行動通訊波束成型系統即時收發機之設計與整合

近年隨著通訊技術的發展，人們對網路的使用需求大幅提升，除了資料傳輸量急劇增加外，也追求更佳的傳輸品質，使原先授權頻帶內的頻譜資源逐漸不敷使用，因此許多研究致力於將行動通訊技術拓展至未授權頻帶上，以5G NR為例，其所使用的頻帶為美國聯邦通訊委員會(FCC)於2020年公布，提供給Wi-Fi與其他通訊技術使用。此舉雖然使5G技術獲取更多的頻帶資源，卻也同時面臨了隨之而來的挑戰，例如不同通訊技術之間的干擾，以及頻帶資源的競爭等，因此如何達到5G與Wi-Fi之間的網路共存成為了重要的研究課題，此亦為本論文的研究目標。本文引入空間分工多重接取(Spatial Division Multiple Access)技術，在5G發射端使用八天線搭配波束成型，集中訊號發射給5G的接收端，並且抑制對Wi-Fi傳輸的干擾，讓兩系統訊號即使發生碰撞也能成功解碼，達到5G與Wi-Fi網路在同頻代的共存。本文的第二章介紹了5G NR與Wi-Fi 6通訊標準，並對未授權頻譜進行簡介，以及現有使用未授權頻帶通訊技術的演變。後半部介紹本文使用的最小變異無失真響應(Minimum Variance Distortionless Response, MVDR)波束成型，說明其數學推導，以及如何將之應用於發射端，最後呈現了模擬八天線陣列所繪製的波束圖形。在第三章中，我們介紹了根據5G NR標準實作具備波束成型功能的八天線硬體發射機，以此作為5G基地台(Base Station, BS)的下行發射機，除了內部各模組介紹外，也說明了如何將多套發射機電路整合至賽靈思射頻系統單晶片(Xilinx RFSoC)上，搭配MVDR波束成型演算法，集中訊號發射給5G使用者(User Equipment, UE)同時抑制對Wi-Fi使用者(Station, STA)的干擾。5G UE、Wi-Fi STA與Wi-Fi存取點(Access Point, AP)皆使用軟體定義無線電設備NI USRP實作，建立起一套5G與Wi-Fi的共存網路系統。此外，我們使用另一片RFSoC開發出四天線接收機，用以監聽Wi-Fi STA的上行訊號，除了估計STA所在方位角，也解碼得知Wi-Fi當前傳輸情形；5G UE上行訊號則透過有線傳輸。此共存系統乃基於前一代系統[13]加以拓展與改良，從原先支援1個5G UE與1個Wi-Fi STA的共存，進步為支援2個5G UE與1個Wi-Fi STA的共存，主機端結合了裝置追蹤技術與拓展後的功率控制技術，實時計算並更新波束成型係數，即便在Wi-Fi STA移動或是Wi-Fi AP發射功率發生改變的情況下，此系統也能維持5G與Wi-Fi的共存，並且相較於前一代系統的總體有效傳輸率(goodput) 27.34 Mbps，經過本文改良後的goodput則進一步提升至46.69 Mbps。本章最後呈現OTA實驗的結果，驗證了如前述所開發的功能。第四章說明為了提升接收端的解碼效能，我們使用場效應可程式化邏輯閘陣列(Field Programmable Gate Array, FPGA)實作出一套接收機硬體電路，對各接收機模組逐一介紹其功能與電路架構，以及如何搭配主機端的C程式建立軟硬體之間的溝通機制。為了更進一步優化接收端的資料處理流程，我們結合了USRP硬體驅動程式(USRP Hardware Driver, UHD)與控制FPGA的C程式，將之改寫為平行化運作的程式，如此便支援了接收端實時接收訊號並同步進行解碼的功能，大幅提升接收機資料處理效能。第五章考慮了另一種通訊情境，也就是5G在非直視路徑下的訊號傳輸，在此情境下MVDR演算法不再適用，因此我們改為使用共軛波束成型，並且為了能夠抑制STA方向上的訊號能量，我們參考MVDR演算法的數學原理，將數學式中的引導向量取代為共軛波束成型係數，如此即可使5G BS集中訊號發射給未於非直視路徑處的UE，同時抑制對Wi-Fi STA方向上的干擾。針對本文所開發的發射機系統規格，我們建立一套估計各路徑通道效應的流程，並且設計兩種子情境的實驗，驗證了在非直視路徑的訊號傳輸下，本文提出的新方法相較於MVDR能夠達到更佳的網路共存效果。其中在第二種子情境下，相較於MVDR達到的共存網路總體goodput 19.7 Mbps，本文所提出的新波束成型演算法可以達到更高的總體goodput 23.69 Mbps。

關鍵字

未授權頻譜之新無線電技術；異質網路共存；空間分工多重接取；波束成型；第五代行動通訊新無線電；賽靈思射頻系統單晶片；空氣通道；功率控制；場效應可程式化邏輯閘陣列；全硬體正交分頻多工接收機；非直視路徑

並列摘要

In recent years, with the development of communication technology, there has been a significant increase in people's demand for network usage. Besides the sharp rise in data transmission volume, there is also a pursuit for better transmission quality. As a result, the spectrum resources within the licensed bands have gradually become insufficient. Therefore, many studies have been dedicated to extending mobile communication technology to unlicensed bands. Taking 5G NR as an example, the bands it uses were announced by the Federal Communications Commission (FCC) in 2020, made available for use by Wi-Fi and other communication technologies. While this move allows 5G technology to access more spectrum resources, it also brings challenges such as interference between different communication technologies and competition for spectrum resources. Therefore, achieving coexistence between 5G and Wi-Fi networks has become an important research topic, which is also the research objective of this paper. This paper introduces Spatial Division Multiple Access (SDMA) technology, using eight antennas combined with beamforming at the 5G transmitter to concentrate signal transmission to the 5G receiver while suppressing interference to Wi-Fi transmission. This allows the signals of both systems to be successfully decoded even if they collide, achieving coexistence of 5G and Wi-Fi networks on the same frequency band. Chapter 2 of this paper introduces the communication standards of 5G NR and Wi-Fi 6, provides an overview of unlicensed spectrum, and discusses the evolution of existing communication technologies using unlicensed bands. The latter part introduces the Minimum Variance Distortionless Response (MVDR) beamforming used in this paper, explains its mathematical derivation, and describes how it is applied at the transmitter. Finally, simulated beam patterns drawn for an eight-antenna array are presented. In Chapter 3, we introduce the implementation of an eight-antenna hardware transmitter with beamforming capabilities based on the 5G NR standard. This transmitter serves as the downlink transmitter for a 5G Base Station (BS). In addition to introducing the internal modules, we also explained how to integrate multiple transmitter circuits onto a Xilinx RFSoC (Radio Frequency System on Chip), along with the MVDR (Minimum Variance Distortionless Response) beamforming algorithm. This setup allows for the concentration of signals to be transmitted to 5G User Equipment (UE) while suppressing interference to Wi-Fi Station (STA). Both 5G UEs and Wi-Fi STA, as well as Wi-Fi Access Point (AP), are implemented using software-defined radio equipment, specifically NI USRP devices, establishing a coexisting network system of 5G and Wi-Fi. Additionally, we utilized another RFSoC to develop a four-antenna receiver for monitoring the uplink signal of Wi-Fi STA. Apart from estimating the azimuth of the STA, this receiver also decodes the signal to retrieve information about the current transmission status of Wi-Fi. The uplink signals of 5G UEs are transmitted via wired connections. This coexistence system is an extension and improvement of the previous-generation system [13]. From supporting the coexistence of 1 5G UE and 1 Wi-Fi STA, it has advanced to supporting 2 5G UEs and 1 Wi-Fi STA. At the host end, device tracking technology and enhanced power control technology are integrated, enabling real-time calculation and updating of beamforming coefficients. Even in scenarios where the Wi-Fi STA moves or the Wi-Fi AP's transmission power changes, this system can maintain the coexistence of 5G and Wi-Fi. Compared to the previous-generation system's overall effective throughput (goodput) of 27.34 Mbps, the improved goodput after the modifications presented in this paper further increases to 46.69 Mbps. This chapter concludes by presenting the results of over-the-air (OTA) experiments, validating the functionalities developed as described above. Chapter 4 describes the implementation of a receiver hardware circuit using Field Programmable Gate Array (FPGA) to improve the decoding performance at the receiver end. We introduce the functions and circuit architecture of each receiver module, as well as how to establish communication between software and hardware by using C programs on the host end. To further optimize the data processing flow at the receiver end, we combine the USRP Hardware Driver (UHD) with the C program controlling the FPGA, rewriting them for parallel operation. This enables real-time reception and decoding of signals at the receiver end, significantly enhancing the data processing efficiency of the receiver. Chapter 5 considers another communication scenario, which is the signal transmission of 5G in non-line-of-sight (NLOS) conditions. In this scenario, the MVDR algorithm is no longer applicable. Therefore, we switch to using conjugate beamforming. Additionally, to suppress the signal energy in the direction of the STA, we refer to the mathematical principles of the MVDR algorithm and replace the steering vector in the mathematical expression with conjugate beamforming coefficients. This allows the 5G BS to concentrate signal transmission to UE not located in the LOS paths while suppressing interference in the direction of Wi-Fi STA. For the transmitter system specifications developed in this paper, we establish a process to estimate the channel responses of various paths and design experiments for two sub-scenarios. These experiments verify that under NLOS signal transmission, the new method proposed in this paper achieves better network coexistence compared to MVDR. In the second sub-scenario, compared to the total network goodput achieved by MVDR at 19.7 Mbps, the new beamforming algorithm proposed in this paper achieves a higher total goodput of 23.69 Mbps.

並列關鍵字

New radio technologies for unlicensed spectrum (NR-U) ； heterogeneous network coexistence ； spatial division multiple access (SDMA) ； beamforming ； 5th generation mobile communication new radio (5G NR) ； Xilinx RFSoC (Radio Frequency System on Chip) ； over-the-air (OTA) ； power control ； field programmable gate array (FPGA) ； all-hardware orthogonal frequency-division multiplexing transceiver ； non-line-of-sight (NLOS) paths

參考文獻

Ericsson. “Mobile data traffic outlook.” ericsson.com. Accessed: Apr. 4, 2024. [Online]. Available: https://www.ericsson.com/en/reports-and-papers/mobility-report/dataforecasts/mobile-traffic-forecast

Google Scholar

RCR Wireless News. “How does support for unlicensed spectrum with NR-U transform what 5G can do for you?” rcrwireless.com. Accessed: Apr. 8, 2024. [Online]. Available: https://www.rcrwireless.com/20200616/5g/how-does-support-for-unlicensed-spectrum-with-nr-u-transform-what-5g-can-do-for-you

Google Scholar

Keysight. “Understanding the 5G NR Physical Layer.” keysight.com. Accessed: Apr. 20, 2024. [Online]. Available: https://www.keysight.com/upload/cmc_upload/All/Understanding_the_5G_NR_Physical_Layer.pdf

Google Scholar

Chia-Cheng Yang, “Design and FPGA-based Implementation of 5G NR Multi-User MIMO Beamforming Downlink Transmitter,” M.S. thesis, Graduate Institute of Electronics Engineering, National Taiwan University, Taipei, Taiwan, 2020.

Google Scholar

5G NR User Equipment (UE) radio transmission and reception Part 1: Range 1 Standalone, 3GPP TS 38.101-1 version 15.3.0 Release 15, Oct. 2018, Accessed: Apr. 10, 2024. [Online]. Available: https://www.etsi.org/deliver/etsi_ts/138100_138199/13810101/15.03.00_60/ts_13810101v150300p.pdf

Google Scholar

主題瀏覽