第五代行動通訊新無線電多使用者多輸入單輸出波束成形系統即時內接收機之設計與實現

時代一直在進步，無論是一般大眾或者是公司企業，對於無線連網的需求變得越來越高，應用範疇更是包羅萬象，舉凡居家、交通、娛樂、醫療、工廠生產，很多以前科幻電影的畫面，甚至是人們從沒想過的場景都被逐步地實現，而能夠實現的原因很大一部份要歸功於新世代通訊系統能夠提供更高的資料傳輸速率，還有容許更多裝置在同一時間內連網。當同時間連網的裝置變多的時候，傳統的無線通訊系統是透過時間或是頻率的分隔來達到多重接取的效果，如時間分工多重接取 (TDMA)、頻率分工多重接取 (FDMA)等等，但由於頻譜資源有限，若連網裝置的數量持續增加勢必會遇到資源不夠的情況，這時利用空間分工多重接取 (Spatial Division Multiple Access, SDMA)將不同使用者所傳送的資料分開來，就能再提升整體系統的效能。本論文基於現有的波束成形理論，選擇Xilinx Alveo U250 FPGA加速平台，並參考5G NR的規範來實作多使用者多輸入單輸出波束成形 (Multi-User MISO Beamforming)硬體內接收機 (Inner Receiver)。此內接收機負責將使用者的時域IQ資料經過載波頻率偏移補償以及傅立葉轉換轉成頻域後，進行通道估測，並且將訊號等化後輸出星座點給軟體進一步做外接收機 (Outer Receiver)解碼。本內接收機系統包含了軟體端和硬體端，分別負責輸入輸出資料的控制和資料的運算，而這樣的分配正是基於軟體的高度彈性和硬體的強大運算能力。要正確地解出資料就需要軟體和硬體之間的合作，而這兩者順利溝通的關鍵就是我們透過狀態暫存器 (Status Register)來讓雙方知道目前系統運作的狀況，實際上做法就是軟體端會去存取FPGA上特定記憶體位置的資料，而硬體內部的RTL程式也會存取同樣記憶體位置的資料，所以我們可以事先定義不同的數字代表什麼狀態，讓軟體端和硬體端根據當下的狀態去存取這個狀態暫存器來達到溝通效果。使用硬體來實現內接收機就是為了加快接收端解碼的速度，進一步實現即時(Real-time)解碼的效果。為了驗證本論文所設計的系統之正確性及可行性，我們在空氣通道 (Over-The-Air)的環境裡傳送影片檔給本接收端系統，並將內接收機解碼完的星座點送給外接收機進行錯誤更正碼解碼。最後解碼後的結果可以達到Block Error Rate = 0，並順利在接收端將影片重新播放出來。而解碼時間大約等同於影片傳送的時間。

關鍵字

即時 (Real-time) ；空氣通道 (OTA) ；現場可編程邏輯閘陣列 (FPGA) ；第五代行動通訊新無線電 (5G NR) ；內接收機 (Inner Receiver) ；多輸入單輸出 (MISO)

並列摘要

Over the past few years, people have been asking for more from wireless Internet. There are growing demands from education, traffic, entertainment, medical purpose, factory production, etc. Many of the applications that only existed in sci-fi movies are now realized in our world. One of the reasons for these real-world sci-fi applications is that the new generation communication systems provide higher data transmission rates and allow more devices to connect to the Internet at the same time. Traditionally, wireless communication systems separate user data in the time domain or the frequency domain to achieve multiple access, such as Time Division Multiple Access (TDMA) or Frequency Division Multiple Access (FDMA). However, the spectrum is precious and limited, and it will inevitably become insufficient when more and more devices are trying to get connected. Therefore, separate user data in the space domain, namely Spatial Division Multiple Access (SDMA), can help ease the insufficiency and improve the overall system performance. We’ve built a Multi-User MISO (MU-MISO) Beamforming Inner Receiver system based on the existing beamforming theories and 5G NR specifications. This inner Rx system is built on the Xilinx Alveo U250 FPGA platform. It can compensate the CFO of the input time-domain IQ data, take FFT to convert it to frequency-domain data, and then perform channel estimation and equalization. The equalized constellations will be sent to a software-based outer receiver for further decoding. This inner receiver system contains both software and hardware, which are in charge of input/output data control and baseband processing, respectively. This is because the software has high flexibility, while the hardware owns outstanding performance. In order to decode the data correctly, the software must cooperate with the hardware, and the key to make this happen is to use a “status register” to let them communicate with each other. The software program will access the data stored in specific SRAM on the FPGA, and the hardware module will access the same data as well. By using pre-defined values that stand for different status, both software program and hardware module can write a status register to notify each other of their current status or read the status register to know each other’s status, and when one side executes, the other side waits. Adopting hardware to implement inner receiver aims for speeding up the decoding progress and achieving the effect of decoding data in real-time. In order to verify the correctness and feasibility of the system in this thesis, we transmitted a video through OTA channel, and decoded the OTA signals with this inner Rx system and sent the output constellations to software outer Rx for LDPC decoding and so on so forth. The decoded results showed that the BLER was 0, and the video could be played at the receiver side without any error. The elapsed time for decoding data about the same as the OTA transmission time.

並列關鍵字

Real-time ； Over-The-Air (OTA) ； FPGA ； 5G NR ； inner receiver ； MISO

參考文獻

[1] NCC. 行動寬頻業務 5G釋照電子式競價系統-最新競價結果 Accessed: Dec. 8, 2020. [Online]. Available: https://mbb5g.ncc.gov.tw/

Google Scholar

[2] ITU. Setting the Scene for 5G: Opportunities Challenges. Accessed: Dec. 8, 2020. [Online]. Available: https://www.itu.int/dms_pub/itu-d/opb/pref/D-PREF-BB.5G_01-2018-PDF-E.pdf

Google Scholar

[3] S. Gautam and T. R. Rao, “Beamforming Investigation for Wireless Communication at 60 GHz,” in Proc. Int. Conf. WiSPNET, Chennai, India, March 2017, pp. 1765-1768.

Google Scholar

[4] E. Nurvitadhi, D. Sheffield, J. Sim, A. Mishra, G. Venkatesh, and D. Marr, “Accelerating binarized neural networks: Comparison of FPGA, CPU, GPU, and ASIC,” in Proc. Int. Conf. Field-Program. Technol. (FPT), Xi'an, China, December 2016, pp. 77-84.

Google Scholar

[5] Keysight. Understanding the 5G NR Physical Layer. Accessed: Dec. 8, 2020. [Online]. Available: https://www.keysight.com/upload/cmc_upload/All/Understanding_the_5G_NR_Physical_Layer.pdf

Google Scholar

國際替代計量

第五代行動通訊新無線電多使用者多輸入單輸出波束成形系統即時內接收機之設計與實現

全文下載

主題瀏覽