發(fā)布時(shí)間：2018-07-21 14:41

【摘要】：低壓高速電力線通信(PLC)系統(tǒng)利用已有的低壓電力配電網(wǎng)構(gòu)成的電力線作為通信信道進(jìn)行高速率數(shù)據(jù)傳遞和信息交換,已經(jīng)被廣泛應(yīng)用于用戶終端寬帶接入、智能電網(wǎng)、企業(yè)內(nèi)部網(wǎng)絡(luò)和智能家庭網(wǎng)絡(luò)等領(lǐng)域。然而,最初電力線是為傳輸50Hz或60Hz低損耗輸電要求而設(shè)計(jì),因此在電力線信道中傳輸高頻信號時(shí),信道會(huì)表現(xiàn)出時(shí)變、多徑、衰減、高噪聲干擾等特性,使得電力線信道通信環(huán)境極為惡劣。因此,如何在低壓電力線信道中實(shí)現(xiàn)信息的可靠傳輸,就成為一個(gè)有重要研究意義和應(yīng)用價(jià)值的課題。 應(yīng)用正交頻分復(fù)用(OFDM)調(diào)制技術(shù)的電力線通信系統(tǒng)能夠有效的對抗惡劣的電力線信道通信環(huán)境,為實(shí)現(xiàn)可靠的電力線高速數(shù)據(jù)傳輸提供了技術(shù)保障。本論文在對實(shí)際低壓電力線信道傳輸特性和信道噪聲進(jìn)行測量的基礎(chǔ)上,對它們的統(tǒng)計(jì)特性進(jìn)行了分析研究,進(jìn)而建立了信道模型和噪聲模型,并研究了基于OFDM電力線信道估計(jì)技術(shù)和多用戶自適應(yīng)動(dòng)態(tài)資源分配算法。論文的主要研究內(nèi)容和創(chuàng)新點(diǎn)如下： 1、針對復(fù)雜的低壓電力線信道特性,提出了一種基于周期PN序列相關(guān)法的測量方案并對室內(nèi)實(shí)際低壓電力線信道進(jìn)行測量分析,從而得到了信道統(tǒng)計(jì)特性,同時(shí)對信道建立了信號多徑傳輸模型。針對電力線信道多徑模型中復(fù)雜的多參數(shù)辨識問題,提出了一種基于匹配追蹤(MP)算法的電力線信道多徑傳輸模型參數(shù)辨識方法。在所提參數(shù)辨識系統(tǒng)中摒棄了以往MP算法中傳統(tǒng)的高斯原子,提出了種新的基于電力線信道多徑傳輸模型的原子結(jié)構(gòu),并將遺傳算法(GA)應(yīng)用到MP算法中有效地提高了系統(tǒng)搜尋最優(yōu)原子的效率,從而降低了MP算法運(yùn)算復(fù)雜度,提高了系統(tǒng)收斂速度和信道模型參數(shù)辨識精度。 2、對室內(nèi)外低壓電力線信道噪聲進(jìn)行了實(shí)際測量,并對實(shí)測噪聲數(shù)據(jù)分別進(jìn)行了時(shí)域和頻域統(tǒng)計(jì)分析從而得到噪聲特性,在此基礎(chǔ)上,研究了電力線信道中五類基本噪聲模型。重點(diǎn)研究了背景噪聲和脈沖噪聲,并對實(shí)測噪聲進(jìn)行了建模仿真。 3、研究了OFDM調(diào)制技術(shù)原理、核心算法及其技術(shù)特點(diǎn),并在實(shí)測信道數(shù)據(jù)基礎(chǔ)上建立了信道狀態(tài)空間模型,提出了分別基于判決反饋粒子濾波算法和基于梳狀導(dǎo)頻粒子濾波算法的兩種OFDM電力線時(shí)變信道估計(jì)方案,并分別與基于最小二乘(LS)、卡爾曼濾波(KF)和無先導(dǎo)卡爾曼濾波(UKF)算法的信道估計(jì)方法性能進(jìn)行了比較。因KF、UKF算法缺乏對普遍非線性、非高斯系統(tǒng)的處理能力,LS算法在對信道進(jìn)行估計(jì)時(shí)忽略了噪聲的影響,而PF算法具有較強(qiáng)的處理各種非線性、非高斯問題的能力,因此采用PF算法的信道估計(jì)方法獲得了更高的估計(jì)精度和更好的系統(tǒng)性能。 4、針對多用戶資源分配這種有約束多目標(biāo)優(yōu)化NP難問題,提出了一種基于粒子群優(yōu)化遺傳(PSO_GA)混合算法的多用戶自適應(yīng)OFDM系統(tǒng)資源分配方案,在此混合算法中將遺傳算法的全局搜索能力以及遺傳算子融合到粒子群優(yōu)化算法中可有效的避免粒子容易出現(xiàn)早熟收斂的現(xiàn)象并增強(qiáng)了粒子局部和全局搜索能力,從而提高了混合算法尋找最優(yōu)解的能力和系統(tǒng)收斂速度。
[Abstract]:Low voltage and high speed power line communication (PLC) system has been widely used in the fields of user terminal broadband access, smart grid, enterprise internal network and intelligent home network, using the power line formed by the existing low-voltage power distribution network as the communication channel. However, the initial power line is 5. 0Hz or 60Hz low loss transmission requirements are designed. Therefore, when high frequency signals are transmitted in the power line channel, the channel will show time-varying, multipath, attenuation, high noise interference and so on, making the communication environment of the power line channel extremely bad. Therefore, it is an important study to realize the reliable transmission of information in the low voltage power line channel. A subject of significance and applied value.
The power line communication system using orthogonal frequency division multiplexing (OFDM) modulation technology can effectively combat bad power line channel communication environment and provide technical support for reliable transmission of high speed data transmission of power lines. The statistical characteristics are analyzed, the channel model and the noise model are established, and the OFDM power line channel estimation technology and the multi-user adaptive dynamic resource allocation algorithm are studied. The main research contents and innovation points of this paper are as follows:
1, in view of the complex low voltage power line channel characteristics, a measurement scheme based on the periodic PN sequence correlation method is proposed and the indoor low voltage power line channel is measured and analyzed. The channel statistical characteristics are obtained. At the same time, the multipath transmission model of the channel is set up. A parameter identification method for multipath transmission model of power line channel based on matching tracking (MP) algorithm is proposed. In the proposed parameter identification system, the traditional Gauss atom in the previous MP algorithm is abandoned. A new atomic structure based on the multipath transmission model based on the power line channel is proposed, and the genetic algorithm (GA) is applied to the MP calculation. The method effectively improves the efficiency of the system searching for the optimal atom, thus reducing the computational complexity of the MP algorithm, improving the convergence speed of the system and the identification precision of the channel model parameters.
2, the noise of the indoor and outdoor low-voltage power line channel is measured, and the measured noise data are analyzed in time domain and frequency domain, and then the noise characteristics are obtained. On this basis, five basic noise models in the power line channel are studied. The background noise and impulse noise are studied, and the measured noise is modeled. Simulation.
3, the principle of OFDM modulation technology, the core algorithm and its technical characteristics are studied, and the channel state space model is established on the basis of the measured channel data. Two OFDM power line time-varying channel estimation schemes are proposed based on the decision feedback particle filter algorithm and the comb guided frequency particle filter algorithm, respectively, and are respectively based on the least squares based on the least squares. (LS) the performance of the channel estimation method of Calman filtering (KF) and the unguided Calman filter (UKF) algorithm is compared. Because KF, UKF algorithm lacks the processing capability of universal nonlinear, non Gauss system, LS algorithm neglects the influence of noise when estimating the channel, and the PF algorithm has a strong handling of various nonlinear, non Gauss problems. Therefore, the channel estimation method based on PF algorithm achieves higher estimation accuracy and better system performance.
4, a multi user adaptive OFDM system resource allocation scheme based on particle swarm optimization (PSO_GA) hybrid algorithm is proposed to solve the constrained multi-objective optimization problem of multi-user resource allocation. In this hybrid algorithm, the global searching ability of genetic algorithm and genetic operator can be fused into particle swarm optimization (PSO) algorithm in this hybrid algorithm. The effective avoidance of precocious convergence is easy to avoid, and the local and global searching ability of the particles is enhanced, thus the ability to find the optimal solution and the convergence rate of the system are improved.
【學(xué)位授予單位】：北京郵電大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2014
【分類號】：TN915.853

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 李豐;田海亭;王思彤;袁瑞銘;梁貴書;;低壓電力線窄帶載波通信信道阻抗與衰減特性的現(xiàn)場測量及分析[J];電測與儀表;2011年07期

2 呂海峰,王贊基,郭靜波;電力線正交頻分復(fù)用通信的實(shí)時(shí)信道估計(jì)[J];電力系統(tǒng)自動(dòng)化;2003年09期

3 馬建旭,于東海,張圣清,鄒采榮;用周期PN序列測量配電網(wǎng)信道傳輸特性[J];電力系統(tǒng)通信;2003年02期

4 王繼業(yè);中電飛華PLC寬帶運(yùn)營試驗(yàn)[J];電力系統(tǒng)通信;2005年01期

5 甄翔宇;楊慶慶;周曉方;陳建霖;;室內(nèi)低壓電力線信道噪聲分析與建模[J];電路與系統(tǒng)學(xué)報(bào);2011年04期

6 李黃強(qiáng);孫云蓮;;基于量子遺傳算法的寬帶電力線多用戶通信資源分配[J];電力自動(dòng)化設(shè)備;2009年10期

7 劉海濤,張保會(huì),譚倫農(nóng);低壓電網(wǎng)500kHz～10MHz頻段噪聲特性研究及信道容量估算[J];電網(wǎng)技術(shù);2004年03期

8 張旭輝;張禮勇;梁宵;;基于改進(jìn)粒子群優(yōu)化算法的電力線通信多徑傳輸模型參數(shù)辨識[J];電網(wǎng)技術(shù);2009年01期

9 徐志強(qiáng);翟明岳;趙宇明;唐明聰;;多用戶電力線通信自適應(yīng)OFDM系統(tǒng)中的資源分配與交換優(yōu)化[J];電網(wǎng)技術(shù);2009年06期

10 翟明岳;徐志強(qiáng);王九金;;寬帶電力線通信系統(tǒng)中的資源分配綜述[J];電網(wǎng)技術(shù);2010年05期

相關(guān)博士學(xué)位論文 前1條

1 張旭輝;基于自適應(yīng)OFDM的電力線高速數(shù)據(jù)通信關(guān)鍵技術(shù)研究[D];哈爾濱理工大學(xué);2009年

，

本文編號：2135853