《電子技術(shù)應(yīng)用》
您所在的位置:首頁(yè) > 其他 > 設(shè)計(jì)應(yīng)用 > 基于SDNSR-Net深度網(wǎng)絡(luò)的大規(guī)模MIMO信號(hào)檢測(cè)算法
基于SDNSR-Net深度網(wǎng)絡(luò)的大規(guī)模MIMO信號(hào)檢測(cè)算法
2022年電子技術(shù)應(yīng)用第11期
曾相誌,申 濱,,陽(yáng) 建
重慶郵電大學(xué) 通信與信息工程學(xué)院,,重慶400065
摘要: 大規(guī)模多輸入多輸出(MIMO)系統(tǒng)能有效地提高頻譜效率,當(dāng)天線規(guī)模漸進(jìn)趨向于無(wú)窮時(shí),,最小均方誤差(MMSE)檢測(cè)算法能達(dá)到接近最優(yōu)的檢測(cè)性能,。然而由于算法中存在矩陣求逆的步驟,帶來(lái)極高的計(jì)算復(fù)雜度,,在大規(guī)模MIMO系統(tǒng)中難以實(shí)現(xiàn),。理查森(Richardson)算法能夠在不對(duì)矩陣求逆的情況下,以迭代的形式達(dá)到MMSE算法的檢測(cè)性能,,但該算法受其松弛參數(shù)影響較大,。在結(jié)合最陡梯度下降算法的Richardson算法(SDNSR)中,松弛參數(shù)的誤差可由梯度下降算法彌補(bǔ),,卻提高了計(jì)算復(fù)雜度,。首先通過深度展開的思想,將SDNSR的迭代過程映射為深度檢測(cè)網(wǎng)絡(luò)(SDNSR-Net);然后,,通過修改網(wǎng)絡(luò)結(jié)構(gòu)及添加可訓(xùn)練參數(shù)來(lái)降低計(jì)算復(fù)雜度并提高檢測(cè)精度,。實(shí)驗(yàn)結(jié)果表明,在上行鏈路大規(guī)模MIMO系統(tǒng)中不同信噪比和天線配置的情況下,,SDNSR-Net都優(yōu)于其他典型的檢測(cè)算法,可作為實(shí)際中有效的待選檢測(cè)方案,。
中圖分類號(hào): TN925
文獻(xiàn)標(biāo)識(shí)碼: A
DOI:10.16157/j.issn.0258-7998.222520
中文引用格式: 曾相誌,,申濱,陽(yáng)建. 基于SDNSR-Net深度網(wǎng)絡(luò)的大規(guī)模MIMO信號(hào)檢測(cè)算法[J].電子技術(shù)應(yīng)用,,2022,,48(11):84-88.
英文引用格式: Zeng Xiangzhi,Shen Bin,,Yang Jian. Signal detection based on SDNSR-Net deep network for massive MIMO systems[J]. Application of Electronic Technique,,2022,48(11):84-88.
Signal detection based on SDNSR-Net deep network for massive MIMO systems
Zeng Xiangzhi,,Shen Bin,,Yang Jian
School of Communication and Information Engineering,Chongqing University of Posts and Telecommunications,, Chongqing 400065,,China
Abstract: Massive multiple-input multiple-output(MIMO) systems can effectively improve the spectrum efficiency. When the antenna scale gradually tends to infinity, the minimum mean square error(MMSE) detection algorithm can achieve near-optimal detection performance. However, due to the matrix inversion required in the algorithm, which brings extremely high computational complexity, it is difficult to implement in a massive MIMO system. The Richardson algorithm can achieve the detection performance of the MMSE algorithm in an iterative form without matrix inversion, but the algorithm is greatly affected by its relaxation parameters. In the Richardson algorithm combined with the steepest gradient descent algorithm (SDNSR), the error of the relaxation parameter can be compensated by the gradient descent algorithm, but the computational complexity is increased. This paper firstly uses the idea of deep expansion to map the iterative process of SDNSR to a deep detection network (SDNSR-Net); then, by modifying the network structure and adding trainable parameters,the computational complexity is reduced and the detection accuracy is improved. The experimental results show that SDNSR-Net is superior to other typical detection algorithms in the case of different signal-to-noise ratios and antenna configurations in the uplink massive MIMO system and can be used as an effective detection scheme in practice.
Key words : massive MIMO system,;signal detection,;modern driven;deep learning

0 引言

    大規(guī)模MIMO系統(tǒng)中存在信道硬化現(xiàn)象,,即由信道矩陣生成的Gram矩陣的對(duì)角項(xiàng)遠(yuǎn)大于非對(duì)角項(xiàng),。在該情況下最小均方誤差(Minimum Mean Square Error,MMSE)檢測(cè)算法已證明可以達(dá)到次優(yōu)的檢測(cè)性能[1],。然而該算法中存在矩陣求逆運(yùn)算,,因此難以適用于大規(guī)模MIMO系統(tǒng)。

    為降低線性檢測(cè)算法的計(jì)算復(fù)雜度,,出現(xiàn)了Richardson迭代[2],、Jacobi迭代[3]和逐次超松弛(Successive Over Relaxation,SOR)迭代[4]等迭代檢測(cè)算法,。然而,,在大規(guī)模MIMO系統(tǒng)中,隨著用戶增加,,該類算法的檢測(cè)性能退化嚴(yán)重,。

    深度學(xué)習(xí)技術(shù)作為一種流行的人工智能技術(shù),目前已開始應(yīng)用于解決信號(hào)檢測(cè)的問題,。例如:Ye[5]等人提出利用深度神經(jīng)網(wǎng)絡(luò)進(jìn)行OFDM系統(tǒng)的信道估計(jì)和信號(hào)檢測(cè),;Samuel[6]等人提出的DetNet通過將投影梯度下降算法的迭代過程展開為網(wǎng)絡(luò),,從而獲得了良好的檢測(cè)性能;He[7]等人提出了OAMPNet,,在傳統(tǒng)的OAMP檢測(cè)算法的基礎(chǔ)上增加了一些可優(yōu)化參數(shù),,在不增加額外復(fù)雜度的同時(shí)獲得了更好的檢測(cè)性能。




本文詳細(xì)內(nèi)容請(qǐng)下載:http://forexkbc.com/resource/share/2000005010,。




作者信息:

曾相誌,,申  濱,陽(yáng)  建

(重慶郵電大學(xué) 通信與信息工程學(xué)院,,重慶400065)




wd.jpg

此內(nèi)容為AET網(wǎng)站原創(chuàng),,未經(jīng)授權(quán)禁止轉(zhuǎn)載。