《電子技術(shù)應(yīng)用》
您所在的位置:首頁(yè) > 通信與網(wǎng)絡(luò) > 設(shè)計(jì)應(yīng)用 > GFDM中基于高階長(zhǎng)短時(shí)記憶神經(jīng)網(wǎng)絡(luò)的自適應(yīng)均衡器
GFDM中基于高階長(zhǎng)短時(shí)記憶神經(jīng)網(wǎng)絡(luò)的自適應(yīng)均衡器
2022年電子技術(shù)應(yīng)用第8期
牛安東1,2,苗 碩1,2,劉佳寧1,2,李英善1,2
1.南開大學(xué) 電子信息與光學(xué)工程學(xué)院,天津300350;2.南開大學(xué) 光電傳感器與傳感網(wǎng)絡(luò)技術(shù)重點(diǎn)實(shí)驗(yàn)室,天津300350
摘要: 在廣義頻分復(fù)用系統(tǒng)(GFDM)中,為解決5G網(wǎng)絡(luò)下車載移動(dòng)通信在Sub-6 GHz頻段信道中信號(hào)嚴(yán)重失真的問(wèn)題,提出一種基于高階長(zhǎng)短時(shí)記憶神經(jīng)網(wǎng)絡(luò)(HO-LSTM)結(jié)構(gòu)的自適應(yīng)均衡器。HO-LSTM自適應(yīng)均衡器在傳統(tǒng)高階前饋神經(jīng)網(wǎng)絡(luò)(HO-FNN)的基礎(chǔ)上,采用復(fù)雜度更低的廣義記憶多項(xiàng)式模型(GMP)代替Volterra模型,并引入LSTM神經(jīng)網(wǎng)絡(luò)使其更適用于復(fù)雜非線性模型的預(yù)測(cè)。結(jié)果表明,相比于傳統(tǒng)HO-FNN均衡器和LSTM均衡器,所提出的HO-LSTM均衡器的均衡效果顯著提升,系統(tǒng)性能也得到進(jìn)一步改善。
中圖分類號(hào): TN911.7
文獻(xiàn)標(biāo)識(shí)碼: A
DOI:10.16157/j.issn.0258-7998.212382
中文引用格式: 牛安東,苗碩,劉佳寧,等. GFDM中基于高階長(zhǎng)短時(shí)記憶神經(jīng)網(wǎng)絡(luò)的自適應(yīng)均衡器[J].電子技術(shù)應(yīng)用,2022,48(8):95-100.
英文引用格式: Niu Andong,Miao Shuo,Liu Jianing,et al. An adaptive equalizer based on high order LSTM in GFDM[J]. Application of Electronic Technique,2022,48(8):95-100.
An adaptive equalizer based on high order LSTM in GFDM
Niu Andong1,2,Miao Shuo1,2,Liu Jianing1,2,Li Yingshan1,2
1.College of Electronic Information and Optical Engineering,Nankai University,Tianjin 300350,China; 2.Tianjin Key Laboratory of Optoelectronic Sensor and Sensing Network Technology,Nankai University,Tianjin 300350,China
Abstract: In the generalized frequency division multiplexing system(GFDM), in order to solve the problem of severe signal distortion in the sub-6 GHz frequency band channel of the vehicle-mounted mobile communication under the 5G network, an adaptive equalizer based on high order long short-term memory(HO-LSTM) neural network structure is proposed. Based on the traditional high-order feedforward neural network(HO-FNN), HO-LSTM adaptive equalizer uses the generalized memory polynomial model(GMP) with lower complexity instead of Volterra model, and introduces LSTM neural network to make it more suitable for the prediction of complex nonlinear models. The results show that, compared with the traditional HO-FNN equalizer and LSTM equalizer, the equalization effect of the proposed HO-LSTM equalizer is significantly improved, and the system performance is further improved.
Key words : generalized frequency division multiplexing(GFDM);long short-term memory(LSTM);high order neural network;generalized memory polynomial(GMP)

0 引言

    近年來(lái),第五代移動(dòng)通信技術(shù)(5th Generation Mobile Communication Technology,5G)受到了極大的關(guān)注。廣義頻分復(fù)用技術(shù)(Generalized Frequency Division Multiplexing,GFDM)作為5G候選波形,由于其能夠有效地克服碼間干擾,讓依賴于超可靠低時(shí)延通信的車聯(lián)網(wǎng)等業(yè)務(wù)從中受益[1]。在中國(guó)工信部出臺(tái)的針對(duì)5G通信規(guī)劃中,將Sub-6 GHz頻段作為商用頻段。相比于第四代移動(dòng)通信(4th Generation Mobile Communication Technology,4G)中1.8 GHz~2.7 GHz的低頻段信道,Sub-6 GHz的高頻段信道導(dǎo)致的信號(hào)失真會(huì)更加嚴(yán)重[2]。

    目前接收端均衡技術(shù)是提高通信質(zhì)量的有效方法之一,傳統(tǒng)的均衡器分為線性均衡器和非線性均衡器兩種類型。其中非線性均衡器通常有兩種常用的設(shè)計(jì)方式:基于Volterra濾波器的方法[3-4]和基于神經(jīng)網(wǎng)絡(luò)的方法[5-7]




本文詳細(xì)內(nèi)容請(qǐng)下載:http://theprogrammingfactory.com/resource/share/2000004659。




作者信息:

牛安東1,2,苗  碩1,2,劉佳寧1,2,李英善1,2

(1.南開大學(xué) 電子信息與光學(xué)工程學(xué)院,天津300350;2.南開大學(xué) 光電傳感器與傳感網(wǎng)絡(luò)技術(shù)重點(diǎn)實(shí)驗(yàn)室,天津300350)




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。