本研究計畫將開發應用於下一世代5G 無線通訊系統中之可見光通訊巨量多輸入多輸出收發機演 算法設計與硬體實現。利用分佈在室內天花板各處用作照明設備的一維或二維LED 陣列中多數的獨 立LED 單體做為資料傳輸的發射源，能夠同時兼具照明的功能。在傳送端部分，由於具有大量的LED 可以選擇，前編碼技術能夠在得知通道狀態資訊下幫助減緩通道間干擾，圖樣合成技術可以在面對多 重使用者時，減輕使用者間的互相干擾。在接收端部分，使用由數個感光元件組成的光感應器接收來 自上方帶有資料訊息的光照明，並針對個別使用者利用已知的通道狀態訊息來採用具有彈性的非線性 解碼演算法以獲得較好的位元錯誤率效能。在硬體實現方面，由於在室內很可能同時會有超過兩位以 上的使用者，因此在傳送端部分的前編碼電路必須是高速且可彈性支援多種組態的硬體架構；在接收 端部分的解碼電路則需要隨著不同使用者所處的接收環境作適當的彈性最佳化。在計畫的最終成果期 望達成室內的可見光通訊實際傳送與接收的展示。 ;This project focuses on developing transceiver algorithms and hardware designs of visible-light communication system with massive multiple-input multiple-output technology in next 5G wireless communication system. Thanks to 1D and 2D LED arrays distributed on a ceiling for illumination purpose, large number of LED units can be applied to transmit data. In the transmitter, a precoding scheme can improve inter-channel interference if channel state information is known, and a pattern synthesis method can decrease inter-user interference while multi-users service is required. In the receiver, channel state information of each user and non-linear decoding algorithm are applied to achieve better BER performance. The receiver which composes of several light sensors is adopted to collect light which carries data information. In aspects of hardware implementation, the precoding circuit at the transmitter requires high throughput ability and supports various hardware configurations flexibly; the decoding circuit at the receiver needs to be optimized adaptively according to different receiving environment of each user. In the final stage of this project, we expect to demonstrate real-time transmitting and receiving of visible light communication system.
