Therefore this paper proposes the delay equalization algorithm which can both guarantee the qos of high priority traffic and effectively reduce the delay of ubr traffic in broadband satellite network . it is based on the prediction and equalization the delay of vbr or abr traffic so that the saved slots can be allocated to ubr traffic . the approach of dynamic alternation slots between vbr and ubr traffic can improve the mean tdma frame utilization while reducing the delay of ubr traffic 因此本文在基于流量估计的资源调度算法基础上,提出了一种在宽带卫星网络下能够保证高级别类业务qos ,同时又能够有效地降低ubr业务时延的时延均衡( delayequalization )调度算法,它是基于对vbr或abr业务时延的预测,均衡时延的方法,保证vbr或者abr业务qos的同时,将节省的时隙分配给ubr业务。
In the technique of pre - processing on night vision images , such algorithms as image registration , non - uniformity correction , image enhancement are mainly studied and realized . an innovatory algorithm of full - scale non - uniformity correction on ir images is presented and an arithmetical progression histogram equalization algorithm in the enhancement of image is validated and developed in this dissertation 在夜视图像预处理技术方面,主要研究实现了图像配准、非均匀校正、图像增强等算法,并创新性地提出了红外图像基于全局的非均匀校正算法,验证并发展了等差数列直方图均衡图像增强方法。
The proposed algorithm lows the complexity by choosing decision delay in advance and then making solution of the equalizer tap coefficient , a new decision delay choice method is proposed , solution by using weight method , diagonal matrix transform , z extension and a new energy restrictive condition which restrains noise enhance , a new decision delay choice method is proposed , the simulation result shows that the proposed algorithm has better equalization effect and enhanced performance comparing to the general mmse algorithm , simulation research on the precision and dynamic scope of parameters in digital realization time domain equalization algorithm , design software and hardware of time domain equalizer . fourthly , there exist a lot of interfere in dsl line , especially , dsl works in multi - user mode , the near end interference is serious 通过先选择判决时延,再进行时域均衡器抽头系数求解的方法降低了时域均衡计算复杂度;对于均衡器抽头系数的求解使用了加权技术,通过对角矩阵变换, z扩展,使用不同的能量约束条件对算法求解,结果表明这种约束有效的抑制了噪声增强,与常用删se比较,该算法有更佳的均衡效果,算法性能得到了提高:论文还对数字化实现时域均衡算法中每一部分参数的精度、变量的动态范围进行了模拟研究,对时域均衡的软硬件实现进行了设计。
The main works of this paper can be summarized as follows : 1 . in this paper , the inherent limitation of cma is studied which greatly affect the convergence properties , variable step size thoeries is put into cma , and several variable step size cm blind equalization algorithm is proposed which can overcome the tradeoff between convergence rate and misadjustment 本文的主要贡献在于: 1 、本文对制约恒模算法收敛性能提高的因素做了深入分析,将变步长理论应用于恒模算法,得到了一系列恒模改进算法,克服了恒模算法收敛速度和收敛精度之间存在的内在矛盾。
In term of the orthogonal property of alamouti scheme and the character of the ofdm system equivalent model expression of receiver signal for stbc - ofdm system was obtained by flexible transform method so that the inverse of matrix was avoided using direct - decision mode and the complexity of computation and receiver was decreased . the simulation results show the validity and efficiency of the proposed receiver . ( 6 ) an adaptive equalization algorithm used in space - time block coded ofdm ( stbc - ofdm ) system with alamou 该算法充分利用了alamouti编码的正交特性和多发射天线ofdm系统特点,通过灵活的变换对接收信号进行等价的表示;通过对11矩阵求逆的研究,将块最小二乘递推算法( brl )中的zkxzk的矩阵求逆分解为k个2x2矩阵的求逆,从而降低了自适应算法的运算量和接收机的复杂度。
Secondly , introduce discrete multi - tone modulation principle in detail , have a systemic and comprehensive analysis and explanation on minimum mean square error ( mmse ) channel shorten time domain equalizer design methods which are based on all kinds of cost functions , analyze their advantage and disadvantage . research on time domain equalizer structure , compare all the time domain equalization algorithm with simulation which afford a valuable reference for the choice of equalization algorithm and equalizer structure when design time domain equalizer 详细介绍离散多音频调制原理,对mmse信道缩短法和基于其他代价函数的时域均衡器设计方法的进行了系统全面的分析和阐述,分析了各自的优缺点;对时域均衡器结构进行研究;仿真比较了各种时域均衡方法,为进行时域均衡器设计时均衡算法、均衡器结构的选择提供了有价值的参考。
The improved algorithm not only inherits the fast convergence trait from rls and realizable systolic array from qr _ rls , but also eliminates square - root operation and gets directly equalizer output signals from systolic array . so this algorithm has more simple operation , faster executing speed , less hardware resource and lower hardware cost . some different equalization algorithm are imitated on the qpsk communication system with multipath channel 其次对一些成熟的自适应均衡算法(如lms算法、 rls算法、 qr _ rls算法、逆qr _ rls算法)进行了分析;介绍了一种改进后的无平方根的qr _ rls算法,该算法不但继承了rls的快速收敛特性和qr _ rls算法的systolic阵列可实现性,还取消了qr _ rls算法的平方根运算,使算法在硬件实现时运行速度更快,占用资源也更少,同时该算法还可由systolic阵列直接得到均衡后的输出信号,运算量更小。
Based on the characteristics of the underwater acoustic channel and channel equalization theory mentioned above , this thesis focuses on the sato and the cma blind equalization algorithms , both of which belong to the bussgang class algorithm . with the work of a lot of computer simulations , the performances of these two algorithms are surveyed from the aspects of equalizer - tap initialization scheme , step - size of the blind algorithms , the order of equalizer , and the signal - to - noise ratio 在水声多径信道特性和信道均衡原理的基础上,本文研究了bussgang盲均衡算法中的sato算法和cma算法,通过大量的计算机仿真工作,采用sato盲算法和cma盲算法对信道进行均衡,分别从均衡器的抽头初始化方案、盲均衡算法的迭代步长和均衡器的阶数以及信噪比几个方面入手,对sato 、 cma盲均衡算法的性能进行了研究。
This thesis investigates a few of blind equalization algorithms that are broadly used . these equalizers " structures and principles are introduced in detail . the convergence properties of these equalizers applying in certain underwater acoustic channels are analyzed mathematically by computer simulations 本论文对几种应用比较广泛的盲均衡算法进行了研究,对它们的原理和结构进行了描述,并且分别对其在特定的水声信道中的性能进行了理论仿真,得到了满意的结果,并且对这几种算法进行了比较。
Tricepstrum equalization algorithm ( btea ) and super - exponential ( se ) algorithm based on block data estimation is studied , and these algorithms use hos explicitly . their performance , such as estimation variance and bias , is analyzed . a kind of sparse cross cumulant and sparse equalizer is proposed to simplify the se algorithm , and the simulating results show efficient reduction in complication ?研究了几种直接使用高阶统计量的算法,包括基于数据段估计的倒三谱算法和超指数算法,分析了算法的估计方差和偏差等性能;由于超指数算法计算量较大,不利于实时均衡,利用水声信道的稀疏性,提出了一种基于稀疏互四阶累积量和稀疏权的算法,有效降低了超指数算法的计算量。