非正交多工(Non-orthogonal multiple access，NOMA)是在無線通訊系統中實現高頻譜效率的一種有前途的技術。與傳統的正交多工(Orthogonal multiple access，OMA)不同，NOMA允許多個使用者通過將其信號疊加來共享相同的頻率資源，根據每個用戶的通道狀況去分配用戶的輸送功率並與多天線合併可以增加系統的吞吐量和降低傳輸的錯誤率。然而，NOMA 也會在使用者之間引入干擾，這給信號檢測和解碼帶來了重大挑戰。在本文中，我們提出了一種新的方法，將梯度搜尋與干擾消除結合起來，以減輕 NOMA 系統中的干擾。
在本論文中，我們在下行鏈路傳輸中使用多輸入多輸出系統非正交多工存取傳輸方式，在接收端先使用迫零偵測以及最小均方誤差偵測當作初始序列，並藉由此序列與梯度搜尋結合為更有效的偵測方式，在此系統中，我們會分別在初始偵測以及連續干擾消除後使用其偵測法，由線性偵測所取得的初始序列結合梯度搜尋取得更好的強用戶訊號序列，再將訊號重建並進行干擾消除，降低用戶間的錯誤傳播造成的影響，並且在使用同樣的偵測法偵測剩餘用戶，使得能夠有效降低其系統的錯誤率

Non-orthogonal multiple access (NOMA) is a promising technique for achieving high spectral efficiency in wireless communication systems. Unlike traditional orthogonal multiple access (OMA), NOMA allows multiple users to share the same frequency resources by superimposing their signals. By allocating user's transmit power based on each user's channel conditions and combining it with multiple antennas, NOMA can increase system throughput and reduce transmission error rates. However, NOMA also introduces interference between users, which poses significant challenges for signal detection and decoding.

In this thesis, we propose a novel approach that combines gradient search with interference cancellation to mitigate interference in NOMA systems. We use a multiple-input multiple-output (MIMO) system with non-orthogonal multiple access in the downlink transmission. At the receiver, we employ zero-forcing detection and minimum mean square error detection as initial sequences and combine them with gradient search for more efficient detection. In this system, we use the detection method obtained from linear detection as the initial sequence, combine it with gradient search to obtain a better sequence for strong user signals, reconstruct the signals, and perform interference cancellation to reduce the impact of error propagation among users. We also use the same detection method to detect the remaining users, effectively reducing the system's error rate.

Therefore, we present an algorithmic approach that integrates gradient search with interference cancellation techniques to mitigate interference in NOMA systems. By utilizing this approach, we aim to enhance the performance of NOMA systems in terms of throughput and error rate, ultimately improving the overall efficiency of wireless communication systems.

[1]. Z. Ding, F. Adachi and H. V. Poor, "Performance of MIMO-NOMA Downlink Transmissions," 2015 IEEE Global Communications Conference (GLOBECOM), San Diego, CA, USA, 2015, pp. 1-6, d10.1109/GLOCOM.2015. Pancaldi, G. M. Vitetta, R. Kalbasi, N. Al-Dhahir, M. Uysal, and H. Mheidat,
[2]. X. Huang, “Diversity performance of precoded OFDM with MMSE equalization,” ISCIT 2007, pp. 802-807, Oct. 2007.
[3]. G. Song and X. Wang, “Comparison of interference cancellation schemes for non¬orthogonal multiple access system,” in Proc.2016 IEEE 83rd Veh. Technol. Conf., May 2016, pp. 1–5.Y. Li and X. Huang, :
[4]. Lin, C.; Chang, Q.; Li, X. A Deep Learning Approach for MIMO-NOMA Downlink Signal Detection. Sensors 2019, 19, 2526.
[5]. The Simulation of Independent Rayleigh Faders,” IEEE Trans. Commun., vol. 50, no. 9, pp. 1503-1514, Sept. 2002.
[6]. X. Wang, D. Zhang, B. Chen, T. Liu, Y. Xin and Y. Xu, "Deep Transfer Learning for Model-Driven Signal Detection in MIMO-NOMA Systems," in IEEE Transactions on Vehicular Technology
[7]. M.-X. Chang, “Characterization of Single-Carrier Block Transmission under the Precoded OFDM Architecture,” in ISWPC 2010, pp. 381-385, May. 2010.
[8]. J. Benesty, Y. Huang, and J. Chen, “A fast recursive algorithm for optimum sequential signal detection in a BLAST system,” IEEE Trans. Signal Process., vol. 51, pp. 1722-1731, Jul. 2003.
[9]. M.-X. Chang and W.-Y. Chang, “Efficient maximum likelihood detection for the MIMO system based on differential metrics,” in Proc. IEEE WCNC 2015, pp. 603-608, Mar. 2015.
[10]. M.-X. Chang and W.-Y. Chang, “Efficient detection for MIMO systems based on gradient search,” IEEE Trans. Veh. Technol., vol. 65, no, 12, pp. 10057-10063, Dec. 2016.
[11]. M.-X. Chang and W.-Y, Chang, “Maximum likelihood detection for MIMO systems based on differential metrics,” IEEE Trans. Signal Process., vol. 65, no. 14, pp. 3718-3732, Jul. 2017.
[12]. Lin, C.; Chang, Q.; Li, X. A Deep Learning Approach for MIMO-NOMA Downlink Signal Detection. Sensors 2019, 19, 2526.