CN1284345C - System and method for peak power reduction in spread spectrum communications systems - Google Patents

System and method for peak power reduction in spread spectrum communications systems Download PDF

Info

Publication number
CN1284345C
CN1284345C CN 01808283 CN01808283A CN1284345C CN 1284345 C CN1284345 C CN 1284345C CN 01808283 CN01808283 CN 01808283 CN 01808283 A CN01808283 A CN 01808283A CN 1284345 C CN1284345 C CN 1284345C
Authority
CN
Grant status
Grant
Patent type
Prior art keywords
peak
filter
symbol
spread spectrum
output
Prior art date
Application number
CN 01808283
Other languages
Chinese (zh)
Other versions
CN1425242A (en )
Inventor
M·J·亨顿
Original Assignee
电力波技术公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Grant date

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B1/00Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/69Spread spectrum techniques
    • H04B1/707Spread spectrum techniques using direct sequence modulation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00Baseband systems
    • H04L25/02Details ; Arrangements for supplying electrical power along data transmission lines
    • H04L25/03Shaping networks in transmitter or receiver, e.g. adaptive shaping networks ; Receiver end arrangements for processing baseband signals
    • H04L25/03006Arrangements for removing intersymbol interference
    • H04L25/03343Arrangements at the transmitter end
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00Modulated-carrier systems
    • H04L27/26Systems using multi-frequency codes
    • H04L27/2601Multicarrier modulation systems
    • H04L27/2614Peak power aspects
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00Modulated-carrier systems
    • H04L27/32Carrier systems characterised by combinations of two or more of the types covered by groups H04L27/02, H04L27/10, H04L27/18 or H04L27/26
    • H04L27/34Amplitude- and phase-modulated carrier systems, e.g. quadrature-amplitude modulated carrier systems
    • H04L27/36Modulator circuits; Transmitter circuits
    • H04L27/366Arrangements for compensating undesirable properties of the transmission path between the modulator and the demodulator
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B2201/00Indexing scheme relating to details of transmission systems not covered by a single group of H04B3/00 - H04B13/00
    • H04B2201/69Orthogonal indexing scheme relating to spread spectrum techniques in general
    • H04B2201/707Orthogonal indexing scheme relating to spread spectrum techniques in general relating to direct sequence modulation
    • H04B2201/70706Orthogonal indexing scheme relating to spread spectrum techniques in general relating to direct sequence modulation with means for reducing the peak-to-average power ratio
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00Modulated-carrier systems
    • H04L27/26Systems using multi-frequency codes
    • H04L27/2601Multicarrier modulation systems
    • H04L27/2614Peak power aspects
    • H04L27/2621Reduction thereof using phase offsets between subcarriers

Abstract

一种在扩频通信系统中用于信号峰值降低的系统和方法,通信系统是包括用于限制从系统发送的符号的信号带宽的滤波器的类型的。 A system and method for reducing signal peaks in a spread spectrum communication system, the communication system is a type of signal for limiting the bandwidth of the transmitted symbols from the filter system. 一个信号峰值降低单元(122)被提供在滤波器(126)之前。 A peak signal reduction unit (122) is provided before the filter (126). 包括一个滤波器预测器(146),通过使用与滤波器脉冲响应函数对应的滤波器系数值预测滤波器对输入符号的影响。 Comprising a predictor filter (146), by using the filter impulse filter coefficient values ​​affect the prediction filter function corresponding to the input symbols in response. 被预测使输出信号超过预先确定的峰值限值的输入符号被调节。 It predicted the output signal exceeds a predetermined limit peak input symbol is adjusted. 公开了几个适当的算法的例子,用于计算要被施加到输入符号上的必要的峰值降低。 Discloses several examples of appropriate algorithms for calculating the necessary to be applied to the peak reduced input symbol. 峰值降低单元提供已调节的符号到滤波器用于处理和通信系统的输出。 Peak reducing units have been adjusted for outputting symbols to the filter processing and communications systems.

Description

在扩频通信系统中降低峰值功率的系统和方法 System and method for reducing peak power in a spread spectrum communication system

技术领域 FIELD

本发明涉及码分多址通信系统和相关的操作方法。 The present invention relates to code division multiple access communications systems and related methods of operation. 更具体而言,本发明涉及蜂窝通信系统和信号处理设备以及在蜂窝通信系统中采用的方法。 More particularly, the present invention relates to cellular communication systems and signal processing apparatus and a method employed in a cellular communication system.

背景技术 Background technique

采用在基站与多个移动用户之间的传输的无线通信系统是现代通信基础设施的关键部件。 Using the plurality of mobile users between base stations and a wireless communication transmission system is a key component of modern communications infrastructure. (为了简短和不将术语“蜂窝”限制到特定类型的通信系统或有时与该术语相关联的特定频段,将这种无线通信系统在此称为“蜂窝”通信系统)。 (For brevity and not to the term "cellular" limited to a particular type of communication system or a particular frequency band that is sometimes associated with the term, such a wireless communication system, referred to herein as "cellular" communication system). 这些蜂窝系统被置身于不断增加的性能要求之下,对可用的设备,尤其是蜂窝基站设备的能力加重了负担。 The cellular system is under exposure to increasing performance requirements, the available equipment, especially the ability to base station apparatus burdens. 这些不断增加的性能要求是由于在给定的蜂窝区域内不断增加的用户数以及对于给定信道的带宽要求引起的。 These increasing performance requirements is due in a given cellular area, and increasing the number of users for a given channel bandwidth caused. 不断增加的蜂窝电话用户数当然是容易明白的,并且由于蜂窝电话的方便这种趋势是不可能变慢的。 Increasing the number of cellular phone users is of course easy to understand, and because of the convenience of a cell phone this trend is unlikely to slow down. 第二种考虑很大程度上是由于由蜂窝电话系统所提供的功能类型的不断增加所引起的,如因特网接入和通过蜂窝电话系统的数据传递的其他形式。 The second consideration is largely due to the increasing function of the type provided by the cellular telephone system caused, such as Internet access and other forms of data transmitted through the cellular telephone system. 这些考虑已经导致对于在提供给蜂窝电话载波的可用频谱内更多的信道以及对每个信道更大带宽的需要。 These considerations have led to the inner provided to the cellular telephone carrier frequency spectrum available for more channels and each channel requires greater bandwidth.

使尽可能多的信道适应可用的频谱的传统方法是将每个信道放在窄频带中。 So that as many channels available to adapt the conventional method is the spectrum of each channel in a narrow band. 然而,为了避免各个蜂窝系统用户之间显著的干扰,各个信道在频率上必须充分地远离。 However, to avoid significant interference between the various cellular system users, each channel must be sufficiently remote in frequency. 对于给定信道的频带越窄,对于特定信道的可用带宽也越小。 The narrower the band for a given channel, the available bandwidth for a particular channel is also smaller.

在给定频谱内提供最大信道数的另一种方法是码分多址扩频通信,这已在越来越多的数字蜂窝系统中被采用。 Another method for providing the maximum number of channels in a given spectrum code division multiple access spread spectrum communications, which has been used in an increasing number of digital cellular systems. 当数字信息被从一个地方发送到另一个地方时,在发送前将数据位变换成数据符号。 When digital information is transmitted from one place to another, before the transmission data bits into data symbols. 所发送的信号的带宽是对每个被发送的数据位所发送的符号数的函数。 The bandwidth of the transmitted signal is a function of the number of symbols transmitted for each data bit to be transmitted. 在码分多址扩频通信中,所发送的符号比要发送的数据位多。 In a CDMA spread spectrum communication, the transmitted symbols than the data bits to be transmitted. 具体是,对于每个要发送的数据位发送多个符号代码。 Specifically, for transmitting a plurality of code symbols for each data bit to be transmitted. 知道该代码的接收机将所发送的信号解码,恢复被发送的数据位。 The receiver knows the code will decode the transmitted signal, recover the data bits to be transmitted. 通过适当选择的独特代码,许多用户可以没有干扰地在相同的带宽中通信,因为通过编码使每个信道是正交的。 By appropriately selecting the unique codes, many users can not interfere with the communication in the same bandwidth as by encoding each channel are orthogonal. 在码分多址扩频蜂窝系统中,典型情况下,选择扩展代码,将来自各个信道的数据扩展在比较宽的频谱上,当然是在对于给定的蜂窝提供者可用的频谱范围内。 Spread spectrum code division multiple access cellular system, typically, selected spreading code, the data from each channel is spread over a relatively wide spectrum, of course, within a given cellular provider in range of available spectrum. 这样做使信道之间的干扰为最小,使可用频谱中信道数为最大。 Doing so allows interference between channels is minimized, so that the number of channels available spectrum is maximum. 目前,存在两种涉及码分多址蜂窝通信系统的标准。 Currently, there are two standard relates to code division multiple access cellular communication system. 对于码分多址和宽带码分多址通常将这些标准称为CDMA和WCDMA。 For the CDMA and WCDMA standards typically referred to as CDMA and WCDMA. 由于高度有效地利用可用的频谱,CDMA和WCDMA越来越多地被用为适应不断增长的蜂窝用途的选择方案。 Due to the highly efficient use of available spectrum, CDMA and WCDMA are increasingly being used to accommodate the growing cell selection scheme uses.

然而,在实际实现扩频蜂窝系统时,由于将多用户信道组合产生扩频信号而存在一个问题。 However, in actual implementation Spread Spectrum Cellular System, since the multi-user channel spread signal generating composition exists a problem. 这可通过参考图1来理解,图1示出在一种典型的现有技术蜂窝基站实施方案中扩频信号的产生。 This may be understood by referring to FIG. 1, FIG. 1 shows a spread spectrum signal is generated in a typical prior art cellular base station embodiments. 如图1中所示,在一种扩频系统中,代码多路复用信号发生器10接收多个数据信道D,例如,与可被容纳的用户数对应的数目n。 As shown in FIG. 1, in the spread spectrum system, the code multiplexed signal generator 10 receives a plurality of data channels D, e.g., the number of users can be accommodated with a corresponding number n. 通过对每个信道的输入符号乘以一个分离的正交代码,为每个通信信道产生符号序列。 By the input symbol for each channel multiplied by a separate orthogonal code to generate a symbol sequence for each communication channel. 每个信道的幅度可根据各个信道功率的需要而有差别。 The amplitude of each channel may differ according to the needs of each channel power. 然后将每个符号序列相加以产生单一的代码多路复用的符号序列(在图1中具有同相和正交分量,V1和V2)。 Each symbol sequence is then added to produce a single code symbol sequence multiplexed (having inphase and quadrature components in FIG. 1, V1 and V2). 然后该代码多路复用符号序列通过一个滤波器20以产生所希望的输出信号。 The code symbol sequence is then multiplexed through a filter 20 to produce a desired output signal. 这个滤波器起着关键的作用,因为它将一个“频谱掩膜”(spectral mask)强加在符号序列上,保证广播信号保持在分配给蜂窝载波的频谱内。 This filter plays a key role, since it a "spectrum mask" (spectral mask) imposed on the sequence of symbols, to ensure that the broadcast signal remains within the frequency spectrum allocated to the cellular carrier. 不遵守在频谱分配上的这些限制可能违背联邦政府的规定以及在给定载波的邻近频带中产生噪声。 Failure to observe these limits on spectrum allocation might violate federal regulations and the noise is generated in neighboring a given carrier frequency band. 然后将该输出信号提供给数模转换器30,得到在调制器40中与载波信号混合的模拟信号。 The signal is then output to the digital to analog converter 30, to obtain an analog signal in the modulator 40 is mixed with the carrier signal. 将所得的RF信号提供给RF功率放大器50并播放给蜂窝用户。 The resultant RF signal to the RF power amplifier 50 and played to the cellular user.

问题是从在图1的代码多路复用器10中将多个符号序列组合开始的。 The problem is that from 10 will be combined in a plurality of symbol sequences code multiplexer of FIG. 1 starts. 因为将许多单独的符号序列组合,从滤波器输出的总信号的峰值功率将取决于组合的各个符号的幅度。 Because the combination of many individual sequence of symbols, the peak power of the total signal output from the filter will depend on the amplitude of each symbol combinations. 各个信道的符号将相加产生非常大的组合符号峰值在统计上是可能的。 Symbols of each channel will be summed to produce very large peak in the combined symbol is statistically possible. 尽管在统计上不是通常的情况,在整个系统设计中必须适应这些非常大的符号峰值。 Although statistically not usually the case, the entire system design must adapt to these very large symbol peak. 在整个系统中适应这些大的符号峰值产生实际实施方面的问题。 Accommodate these large peaks symbols in the overall system aspects of practical implementation problems. 例如,从滤波器输出的信号中存在潜在的非常大的峰值对于数模变换器来说需要使用一种非常高分辨率的数模变换器。 For example, the presence of the signal from the output of the filter in potentially very large peak for the digital to analog converters, it requires the use of a very high resolution digital to analog converter. 这就增加了整个系统的成本和复杂性。 This increases the cost and complexity of the overall system.

与码分多址扩频系统中潜在的非常大信号峰值有关的另一个问题涉及由RF功率放大器提供对信号线性放大的困难。 Another potential problem with very large peak signal related to CDMA spread spectrum system involves difficulty in providing a signal amplified by a linear RF power amplifier. 在蜂窝系统中,对广播信号提供线性放大是非常重要的。 In a cellular system, a broadcast signal to provide a linear amplification is very important. 这是因为对信号的非线性放大可导致信号畸变以及产生可以干扰其他蜂窝频带的频谱边带。 This is because the non-linear amplification of the signal may cause signal distortion and interference can produce a spectrum of a sideband of other cellular band. 因为蜂窝频段被严格地控制,必须小心地设计蜂窝系统,使得避免这种在所分配的频段外的噪声产生。 Because the cellular band is strictly controlled, a cellular system must be designed carefully, so as to avoid such noise outside the frequency band allocated generation. 因此,在蜂窝基站中线性的RF放大是必要的。 Thus, cellular base stations in the linear RF amplification is necessary. 然而,为了使放大器运行在它的线性范围内,需要放大器运行在相当低的功率模式中。 However, in order to make operation of the amplifier in its linear range, it is necessary to run the amplifier at a relatively low power mode. 如果信号中大的随机峰值要由这样一种放大器适应并仍然保持它运行在线性状态中,就需要较高功率的RF放大器。 If the random peak signal to adapt to the large and still maintain such an amplifier which operate in the linear state, we need a high-power RF amplifiers. 高功率,高质量的RF放大器是非常费钱的,这样就对整个基站系统增加很大的成本。 High-power, high-quality RF amplifier is very expensive, thus adding significant cost to the entire base station system.

因此信号中大随机峰值的问题在实际实施扩频蜂窝通信系统中是一个重大的问题。 Thus random peaks in the signal large problem is a significant problem in the practical embodiment of a spread spectrum cellular communication system.

大随机信号峰值问题的重要性已在现有技术中有所认识,对这个问题的解决方案已被尝试过。 The importance of large random signal peaks problem has been recognized in the prior art, solutions to this problem have been tried. 例如,解决这个问题的一种方法被描述在授于Oishi等的美国专利No.6,009,090中,在图2中示出了美国专利No.6009090的方法。 For example, one solution to this problem is described in U.S. Patent No.6,009,090 granted Oishi and the like, in FIG. 2 illustrates a method of U.S. Patent No.6009090. 将信号峰值抑制单元60放置在将各个符号序列相加在一起的代码多路复用器10后面的信号产生路径中。 The signal peak suppression unit 60 is placed behind the 10 sequences are added together in each symbol code multiplexer signal generation path. 这个信号峰值抑制单元将多路复用的符号与最大允许值作比较,然后直接将那些超过最大允许值的符号截去。 This signal peak suppression unit multiplexed symbols compares the maximum allowable value, then those direct symbols exceeding the maximum permissible truncated. 虽然这个峰值抑制单元解决了大符号的问题,然而它没有除去必须由D/A变换器和功率放大器处理的所有的大信号峰值。 Although this solves the problem of peak suppression unit of a large sign, but it does not remove all of the large signal peaks must be processed by the D / A converter and a power amplifier. 另外,当一个符号被截止时,发送不太理想的符号,这将增加通信的差错。 Further, when a symbol is turned off, the transmitter is not ideal symbol, which will increase the error communication. 通过细致地考虑当这些符号连续通过信号产生路径时信号峰值抑制单元对这些符号的影响,可以理解这一点。 By carefully considering these symbols is continuously generated when the signal path through the peak suppression means influence on these symbols, it will be appreciated.

如图2中所示,峰值被抑制的符号离开峰值抑制单元以后,它们通过一个滤波器20。 As shown, the peak value is suppressed after leaving the symbol 2 in FIG peak suppression unit, which via a filter 20. 滤波器20可以用一个脉冲响应函数来表示。 Filter 20 may be represented by a pulse response function. 一种典型的扩频脉冲响应函数被示于图3中(WCDMA,根升余弦,α=0.22)。 A typical spread spectrum pulse response function is shown in Figure 3 (WCDMA, root raised cosine, α = 0.22). 当符号通过滤波器时,滤波器的脉冲响应对每个代码多路复用的符号施加影响。 When the symbol through a filter, the filter impulse response influence each code symbol multiplexing. 滤波器脉冲响应对符号的影响可以增加或减少在符号间隔上的峰值,和可以在符号时间之间产生新的峰值。 Effect filter impulse response of the symbol may increase or decrease in the peak symbol interval, and a new peak can be generated between symbol time. 更具体而言,图4示出滤波器输出峰值可以如何不同于输入符号峰值。 More specifically, FIG 4 illustrates how the filter output peak can be different from the input symbols peak. 图4显示出由两个相继的幅度1的输入符号引起的滤波器输出。 Figure 4 shows the filter output due to the amplitude of the two successive input symbols 1. 两个输入符号在滤波器输出上产生在图4中用实线和虚线表示的滤波器脉冲响应函数。 Two input symbols at the filter output of the filter impulse generated in FIG. 4 by the solid line and dotted line response function. 真正的滤波器输出因而将是这两种响应的组合(但为了说明容易,这种相加未在图4中实施)。 Thus the real filter output will be a combination of these two responses (but for ease of illustration, this embodiment is not adding in FIG. 4). 在符号时间0,一个脉冲响应在其最大值,而另一个则稍微有点负。 In symbol time 0, the impulse response of a maximum value thereof, while the other is slightly negative. 因此对于这种情况,信号输出将低于符号时间0上输入符号幅度。 Thus this case, the signal output will be lower than the symbol time 0 input symbol amplitude. (如果第二个符号已经是负而不是正,信号将大于符号时间0上的输入符号)。 (If the second symbol has a negative, rather than positive, the input signal is greater than the symbol on the symbol time 0). 当两个滤波器响应相加产生关于1.2的组合输出时,输出信号将在符号时间0.5(符号间)达到最大值。 When the two filter responses are summed to produce combined output about 1.2 when the output signal will reach a maximum at 0.5 times the symbol (inter-symbol). 在一种实际的输出信号中,在滤波器中同时出现的其他符号的影响将增强这些效果。 In an actual output signal, the influence appears in the filter while the other symbols will increase these effects.

图5A和5B用作说明一个给定的输入符号和在符号序列中该符号前后的符号,当这些符号通过滤波器时,在统计上可以产生的输出信号值的范围。 5A and 5B illustrate one given as input symbols before and after the symbol and the symbol in the symbol sequence, when the value range of the output signal of the symbol by the filter, can be statistically generated. 图5A和5B是用作说明一个输入符号作为来自复数平面的一个向量(同相和正交信号分量)的复数向量图。 5A and 5B are illustrative of an input vector from the symbol as a complex plane (in-phase and quadrature signal components) of the complex vector of FIG. 图5A示出稍微超过所希望的峰值限值(用虚线表示)的输入符号。 5A shows a peak slightly exceeded the desired limit (represented by a dotted line) input symbols. 在图5B中,输入符号是精确地落在限值线上。 In FIG 5B, the input symbol value is precisely fall line. 被滤波的输出信号是输入符号和滤波器脉冲响应函数的一个函数。 The filtered output signal is the input symbol and a function of filter impulse response function. 正如从图4的讨论很明显的那样,输出信号峰值将随机地区别于输入符号峰值,因为差别是由于对在时间上领先和随后于该符号的随机符号的滤波器响应引起的。 As apparent from the above discussion of Figure 4, the output signal of the peak will be different from the input symbols random peak, because the difference is due to the filter in response to the leading and subsequent random sign of the symbol in time induced. 这种随机效应统计上在图中用标有“预测的滤波器输出”的实线圆表示。 This has the statistical random effects in the figure labeled "predicted filter output" is represented by a solid line circle.

当滤波器对通过滤波器的符号序列的影响被考虑时,以上标有`090专利的信号峰值抑制单元的结果被显著地改变。 When the effects of the filter on the symbol sequence through a filter is considered, the above signal peaks marked `090 patent result of suppressing unit is changed significantly. 例如,假定图5A中所示的输入符号,当它超过限值时`090专利将始终对这个符号进行峰值抑制。 For example, assuming that the input symbols shown in FIG 5A, when it exceeds the limits of the '090 patent that the symbol will always peak suppression. 因而始终被这个过程引入某些畸变。 Thus always introduce some distortion of the process. 然而,被D/A变换和RF放大的实际值是被滤波的输出,在统计上用一个圆来表示。 However, the amplified D / A conversion and the actual RF output value is filtered, statistically represented by a circle. 正如可以看到的那样,某些时候,被滤波的值将在限值里面,并不需要限值。 As can be seen, some time will be filtered value limits which do not require limit. 另一方面,某些时候,被滤波的值将超过限值,大于输入符号。 On the other hand, sometimes, the filtered value exceeds the limit, greater than the input symbol. 即使输入符号被截取为限值,也将被不适当地峰值调节。 Even if the input symbols are taken to limit the peak will be adjusted improperly. 依次在图5B的例子中,输入符号不超过限值,在`090专利的方法中所有这些符号将不受影响地通过。 In the example of FIG. 5B sequentially, the input symbol does not exceed the limit, in the method of the '090 patent all of these symbols will not be affected by. 然而,正如从图5B中被滤波的输出圆可以理解的那样,滤波器的影响在于输出信号将实际上大大地超过限值。 However, as the filtered output can be appreciated from a circle as in FIG. 5B, the influence that the output signal of the filter will actually greatly exceeds the limits. 因此,对于这种情况,采用`090专利单元根本不解决信号峰值问题。 Thus, in this case, using the `090 patent does not solve the unit peak signal problem. 因而以上标明`090专利的方法不仅将不必要的畸变引入峰值抑制不必要的信号中,而且完全没有消除许多在输出信号中多余的峰值,这正是要设计解决的问题。 Thus more marked `090 patent's method is not only unnecessary distortion introduced peak suppress unwanted signals, and did not eliminate many unwanted peaks in the output signal, which is designed to solve the problem.

虽然在以上指出的专利中未讨论,一种替代方法可以直接将峰值抑制单元放置在图2中所示的滤波器20的下游侧。 Although not discussed in the above noted patent, an alternative method may be directly downstream side of the peak suppression unit shown in FIG 2 is placed in the filter 20. 然而这也引入一个问题,因为峰值抑制单元的存在将不可避免地使滤波器输出信号畸变。 However, this also introduces a problem because of the presence of peak suppression unit will inevitably filter output signal distortion. 这将产生延伸到该滤波器被设计保持的频谱掩模外的谱噪声。 This produces noise spectrum extends to the outside of the filter is designed to maintain the spectrum mask. 如以上指出的那样,通过对信号的滤波产生的频谱掩模在蜂窝系统中是关键性的,因为超出频谱的分配可能潜在地违背联邦政府的规章。 As noted above, a spectrum mask generated filtered signal is critical in a cellular system, because the spectrum beyond the allocated may potentially violate federal regulations.

因此,无论峰值抑制单元被放置在滤波器前或滤波器后,很清楚,这样一种解决方案对于解决输出信号中大峰值问题是完全不适合的,这样的解决方案或者没有消除峰值,或者引入同等重要的问题。 Thus, regardless of peak suppression unit is placed before the filter or after the filter, it is clear that such a solution to solve the problem of large peaks in the output signal is completely unsuitable, or such a solution does not eliminate the peaks, or the introduction of equally important issues. 因而,这样一种方法在现实世界的应用中是不能工作的。 Thus, such an approach in real-world applications are not working.

因此,将会理解,目前对于提供码分多址扩频蜂窝传输的系统和方法存在一种避免以上指出的大信号峰值问题和伴随而来的与对这样的大峰值RF放大和数模变换有关的约束和成本地需要。 Thus, it will be appreciated that currently exists for providing a cellular code division multiple access spread spectrum transmission system and method to avoid large signal peaks problems noted above and the accompanying digital analog conversion and amplification of such a large peak RF relevant and cost constraints need. 而且,将会理解,存在一种这样的不对系统引入重大的附加的新问题和可被没有过分的成本或其他实施复杂性的方式实施的系统和方法的需要。 Moreover, it will be appreciated that there is a misalignment such systems introduce significant additional new problems and can not be excessive cost or other embodiments of systems and methods are implemented complex manner.

发明内容 SUMMARY

本发明提供一种用于克服以上指出的问题,降低码分多址扩频通信系统中信号峰值功率的系统和方法。 The present invention provides a method for overcoming the problems noted above, the system and method for reducing code division multiple access spread spectrum communication system signal peak power. 而且,本发明以并不重大改变信号的谱特性,并不对信号引入重大的不希望的畸变,和并不对整个系统增加重大复杂性的方式提供这样一种系统和方法。 Further, the present invention is not a significant change in the spectral characteristics of the signal, the signal is not introduced significant undesirable distortion, and does not add significant complexity to the overall system of the embodiment to provide a system and method.

在第一方面中,本发明提供一种接收由该系统输出的,与多个组合的分离数据信道对应的扩频符号的扩频通信系统。 In a first aspect, the present invention provides a receiver, in combination with a plurality of separate data channels corresponding to the spread symbol output by a spread-spectrum communication system of the system. 通信系统包括一个用于在从系统输出前对符号滤波的滤波器。 Communication system comprising a system for the filter before the output of the symbol filter. 在滤波器前提供一个峰值降低单元,连接在输入扩频符号源和滤波器之间。 Before filter provides a peak reducing unit is connected between the input source and a filter spreading symbols. 峰值降低单元从扩频符号源接收输入扩频符号并利用与滤波器脉冲响应函数对应的已知的滤波器系数值作为输入,预测滤波器对符号的影响。 The peak reduction unit input source symbols from received spread spectrum symbols spread using known filter coefficient values ​​corresponding to the function as input to the filter impulse response, predict the impact of the filter on the symbol. 峰值降低单元只对被预测使滤波器输出超过预先确定的峰值限值的那些扩频符号实行峰值降低处理。 Implement only the peak reduction unit is predicted that the peak of the filter output exceeds a predetermined limit that spread symbols peak reduction process. 然后,峰值降低单元提供已处理的符号到滤波器供滤波和由通信系统输出。 Then, the peak reduction unit to provide symbol processed and output by the filter for filtering the communication system.

峰值降低单元最好对在滤波之前被上采样(up sample)的扩频符号流操作。 Peak reduction unit is preferably spread symbol streams prior to the operation to be filtered samples (up sample) a. 上采样通过在扩频符号之间插入零幅度的符号增加符号速率而没有改变扩频符号之间的时间。 Without changing the sampling time between the spread symbols by inserting zero amplitude symbol of the symbol rate increases between spread symbols. 通过插入这些零幅度符号使得到的符号带宽增加。 Symbols by inserting zero amplitude so that these symbols bandwidth is increased. 这种符号带宽的增加对于滤波器的通带和阻带产生了空间。 This increased bandwidth for the symbol filter passband and stopband generating space. 峰值降低单元最好在上采样被执行以前操作,但也可以在上采样以后操作。 Peak reducing means preferably on the sampling is performed before operation, but can also be operated after the sampling.

当在上采样以前操作时,峰值降低单元最好采用从滤波器脉冲响应函数周期性采样的系数。 When the previous sampling operation, the unit is preferably used to reduce the peak factor from the filter impulse response function sampled periodically. 例如,如果上采样过程在扩频符号之间增加三个零幅度的符号,将有四个不同的周期性的采样可用。 For example, if the symbol sampling process increases between three zero amplitude spread symbols, there will be four different periodic sampling available. 所取的样本之间的时间是与扩频符号之间的时间相同的。 The time between samples is taken and the time between the spreading of the same sign. 这种滤波器脉冲响应的周期性采样被用于产生一个滤波器输出预测器。 Periodically sampled impulse response of such a filter is used to produce a predictor filter output. 将根据所取的周期性采样的定时进行预测。 The prediction based on the timing of samples taken periodically. 至少这些周期性的采样应该在符号上(on-symbol)定时和符号间(inter-symbol)定时上取得。 These periodic samples between at least be on the symbol timing and the symbol (on-symbol) to obtain the (inter-symbol) Timing. 然后峰值降低单元将包括根据符号上的定时提供峰值降低的第一级和根据符号间的定时提供峰值降低的第二级。 Then peak reduction unit comprises providing a second stage of the peak reduction stage and providing a first peak reduction based on the timing between symbols in the symbol based on the timing. 另一种做法是,可以提供一种多级的实施方案。 Another approach may be to provide a multi-stage embodiment. 这样一种多级的实施方案可以采用多级的峰值降低处理,每级对应于一种不同的周期性的滤波器采样定时,以串行配置或并行配置的方式实施。 Such a multi-stage embodiment may employ a multi-stage peak reducing process, each level corresponding to a different sampling timing of a periodic filter, a serial configuration or a parallel configuration embodiments. 每级峰值降低单元可进一步包括一个反馈环,提供返回到滤波器预测器的峰值降低值。 Each stage of the peak reduction unit may further comprise a feedback loop is provided to return to the peak value decreases predictor filter. 因为峰值降低也被以临时方式应用,复制的峰值功率降低级可以提供进一步的峰值因数改进。 Because the peak is also reduced application in a temporary manner, to reduce the peak power level replication can provide further improved crest factor. 当按串行或并行方式处理时,这些复制级应该跟随一整套所有其他的周期性采样。 When the process in a serial or parallel fashion, these replication level should follow a set of all the other periodically sampled.

当在上采样后操作时,直接按上采样速率从滤波器脉冲响应函数取得系数。 When the upsampling operation, according to the sampling rate directly from the filter coefficients of the impulse response function to obtain. 然后这些系数被用于产生一个滤波器输出预测器。 These coefficients are then used to produce a predictor filter output. 将在上采样速率上完成滤波器输出预测。 The complete upper prediction filter output sampling rate. 将从这些预测进行峰值降低。 These forecasts were reduced from the peak. 这种方法对于某些应用和/或某些脉冲响应函数并不是优选的。 For certain applications this approach / or certain impulse response function and are not preferred. 当在上采样率上操作时,可以对在上采样期间添加的零值符号提供峰值降低。 When operating on a sampling rate, the peak reduction can be provided the zero-value symbols during the sampling added. 数字通信系统,如CDMA和WCDMA使用为使符号间干扰最小设计的发送和接收滤波器。 Digital communications systems, such as the use of CDMA and WCDMA transmit and receive filters such that the minimum intersymbol interference designed. 为了在通信期间保持最小的符号间干扰,在上采样期间添加的零值符号必须仍然是零值。 In order to maintain the minimum inter-symbol interference during communication, a zero value symbols during the sampling value added must still be zero. 在上采样速率上的操作也需要在较高速率上执行处理。 Operation on the sampling rate is also necessary to perform processing at a higher speed. 一般,处理速率越高,处理部件的成本越贵。 In general, the higher the processing speed, the more expensive the cost of the processing means. 然而可能有这样一些应用,这些成本是无关紧要的和不需要被考虑,则上采样后的处理是优选的。 There may be some applications, however, these costs are irrelevant and need not be considered, the sample treatment is preferable.

更具体而言,在一种详细的实施方案中,扩频通信系统包括一个峰值降低单元,其中包括一个滤波器输出预测器和一个峰值降低计算电路,峰值降低是基于滤波器输出预测和预先确定的滤波器输出限值。 More specifically, in one detailed embodiment, the spread spectrum communication system comprising a peak reducing unit, which includes a predictor filter output and a peak reducing calculation circuit, the peak reduction based on the filter output and a predetermined prediction the filter output value. 峰值降低单元最好也包括一个组合器,用于将算得的峰值降低值组合到以滤波器输出预测器为中心的扩频符号上。 Peak reduction unit also preferably includes a combiner for reducing the peak value is calculated to be combined into a predictor filter output centered spread symbols. 这样,峰值降低单元将扩频符号延时滤波器预测器长度的一半,并以时间同步的方式,在逐个符号的基础上提供峰值降低校正。 Thus, the peak reduction filter prediction delay spread symbols half the length of the unit, and time synchronization manner, to provide the peak reduction based on the correction-by-symbol basis. 组合器可以包括一个乘法器电路和峰值降低值增益,当与时间同步的扩频符号相乘时提供峰值被调节的符号。 Composition may comprise a multiplier circuit, and the peak value of the gain reduction, provides a peak when the symbol is adjusted by multiplying spread symbols with time synchronization. 另一种做法是,组合器可以包括一个加法电路和峰值降低值向量,当被加到扩频符号时提供峰值被调节的符号。 Another approach, the combiner comprises an adder circuit, and may reduce the peak value vector, it is adjusted to provide a peak of the symbol when the symbol is added to the spread.

在另一个方面中,本发明提供一种用于降低峰值信号值的系统,该系统被适配成供在包括一个滤波器的通信系统中使用,该滤波器在从所述的系统输出信号以前提供对符号的滤波。 In another aspect, the present invention provides a system for reducing the peak signal value, the system is adapted for use in a communication system comprising a filter, the filter output signal from the system before provide filtered symbols. 用于降低峰值信号值的系统包括一个滤波器预测器装置,用于在被所述的滤波器滤波以前接收扩频符号,并预测所述的滤波对所述的符号的影响,和连到滤波器预测器装置的装置,用于降低在经受所述的滤波后被预测超过峰值限值的扩频符号峰值。 The system for reducing the peak signal value predictor comprises a filter means for receiving a spread symbols are filtered before the filter, and to predict the impact of the filtering of the symbol, and is connected to the filter prediction means is means for reducing the spread symbols peak filter is subjected after the prediction exceed the peak limit. 在通信系统中的滤波器可以由预先确定的脉冲响应函数表示,滤波器预测器装置最好包括用于接收与在多个周期性的采样点上的脉冲响应函数对应的滤波器系数的装置。 Filter in a communication system may be pre-determined by the response function of a pulse, said means preferably comprises a prediction filter for receiving the periodic pulses at a plurality of sampling points means a filter coefficient corresponding to the response function.

在一种优选实施方案中,用于接收滤波器系数的装置至少在符号上的定时和符号间的定时上从滤波器脉冲响应函数接收滤波器系数。 In a preferred embodiment, the means for receiving the at least receiving filter coefficients for the filter coefficients at the timing of the timing between the symbols on the symbol and the filter impulse response function. 滤波器系数也可以在附加的脉冲函数时间上被提供,以符号速率被从脉冲响应函数周期性地采样。 The filter coefficients may also be provided on an additional pulse function of time, the symbol rate is a function of the impulse response from the sampled periodically. 另外,任何或全部符号系数时间可被重复,以便估计由于峰值降低处理的因果关系性质引起的峰值降低误差。 Additionally, any or all of the coefficients of the symbol time can be repeated, since the peak to reduce the peak to estimate the nature of the causal relationship to reduce the error caused by the processing.

在另一个方面,本发明提供一种用于在包括滤波器类型的扩频通信系统中降低峰值信号值的方法。 In another aspect, the present invention provides a method for including a filter type of spread spectrum communication system for reducing a peak signal value. 该滤波器可以用一个脉冲响应函数表示,在从所述的系统输出信号以前提供对符号的滤波。 The filter response function can be represented by a pulse, to provide filtered symbols from the output signal of the previous system. 该方法包括在被所述的滤波器滤波以前接收扩频符号,并预测滤波对符号的影响。 The method includes receiving a spread spectrum symbols are filtered before the filter, and to predict the impact of filtering on the symbols. 该方法进一步包括调节那些被预测使滤波器输出超过峰值限值的符号值。 The method further comprises adjusting the symbol values ​​are predicted so that the filter output exceeds the peak limit.

在本发明的一种优选应用中,本发明可在一种扩频蜂窝通信系统,如CDMA或WCDMA系统中被实施。 In a preferred application of the invention, the present invention can be implemented in a spread spectrum cellular communication system, such as CDMA or WCDMA system. 例如,本发明可在这样一种扩频蜂窝应用的基站中被实施。 For example, the present invention can be implemented in a spread spectrum cellular base station applications. 在这样一种应用中,避免了大峰值的线性RF放大器的问题,并可采用已降低成本的RF放大器。 In such an application, it avoids the problems of large linear RF amplifier peak, and may employ the RF amplifier has been to reduce costs. 对费钱的D/A变换器的需要也被避免。 The need for expensive D / A converter can also be avoided. 而且,在滤波以前完成峰值降低消除边带的产生和对频谱分配规则可能的违背。 Further, in the filtering completed before the peak reduction eliminate sidebands and possible violation of rules spectrum allocation.

通过查阅以下的本发明的详述将会理解本发明的其它特征和优点。 By review of the following detailed description of the present invention will be appreciated that other features and advantages of the present invention.

附图说明 BRIEF DESCRIPTION

图1是一种现有技术扩频通信系统的方框简图。 FIG. 1 is a block diagram of a prior art spread spectrum communication system.

图2是一种采用信号峰值抑制单元的现有技术扩频通信系统的方框简图。 FIG 2 is a block diagram of peak suppression signal using the prior art spread spectrum communication system units.

图3是一种用于现有技术扩频通信系统滤波器的脉冲响应函数简图。 FIG 3 is a prior art spread spectrum communication system impulse filter response function for a schematic view.

图4是示出两个相继的符号和在现有技术扩频通信系统中它们的滤波器响应函数。 FIG 4 is a diagram illustrating two consecutive symbols and their response filter prior art spread spectrum communication system functions.

图5A和5B是用作说明在一种现有技术扩频通信系统中滤波对任意的相继符号序列的影响的复数向量图。 5A and 5B are any influence on successive symbol sequence in a prior art filtering spread spectrum communication system in a complex vector as described in FIG.

图6是用作说明依据本发明提供峰值降低的一种扩频通信系统的方框简图。 FIG 6 is provided for illustrative block diagram of a spread spectrum communication system according to the present invention, the peak reduction.

图7是用作说明图6的峰值降低单元一种优选实施方案的方框简图。 FIG. 7 is described as reducing the peak in FIG. 6 a schematic block diagram of a preferred embodiment cell.

图8是示出在符号上间隔和符号间间隔上滤波器系数的一种滤波器脉冲响应函数图。 FIG 8 is a diagram illustrating inter-symbol interval in the symbol interval and the A filter impulse response function of the filter coefficients of FIG.

图9是用作说明峰值降低过程一种优选实施方案的方框简图。 9 is a block diagram of the peak reduction described as a preferred embodiment of the process.

图10是用作说明利用反馈的峰值降低过程一种替代的实施方案的方框简图。 FIG 10 is as described with reference to a block diagram of the peak reduction process feedback to an alternative embodiment.

图11是用作说明依据本发明的一种替代的实施方案的一种多级峰值降低单元方框简图。 FIG 11 is a schematic block diagram for illustrative reduction unit according to a multi-stage peak of an alternative embodiment of the present invention.

图12-21是示出在多个不同的周期性定时上滤波器系数的滤波器脉冲响应函数图,依据本发明,图11的多级峰值降低单元可以采用这些滤波器系数。 12-21 is shown in FIG plurality of different periodic timing of the filter coefficient of the filter impulse response function in FIG, according to the present invention, a multi-stage peak reduction unit 11 filter coefficients may be used.

图22是示出依据本发明采用并行实施峰值降低级的一种多级峰值降低单元方框简图。 FIG 22 is a diagram showing a parallel embodiment of the present invention according to a multi-stage peak stage of the peak reduction unit reducing block schematic.

图23是示出依据本发明采用并行实施峰值降低级的一种替代的多级峰值降低单元方框简图。 23 is a diagram illustrating embodiments according to the present invention employs a parallel multi-stage peak stage of the peak reduction unit to reduce an alternative schematic block.

图24是示出依据本发明的图22和23的多级峰值降低单元中一个峰值降低级的方框简图。 24 is a diagram illustrating a peak-reducing unit to reduce the level block diagram of a multi-stage peak according to the present invention of FIG. 22 and 23.

图25是示出依据本发明的一种峰值降低算法中采用的向量的复数向量图。 FIG 25 is a complex vector diagram vector algorithm employed in accordance with one peak is shown to reduce the present invention.

图26是示出依据本发明的一种替代的峰值降低算法中采用的向量的复数向量图。 FIG 26 is a complex vector diagram vector reduction algorithm employed an alternative based on the peak of the present invention.

图27是示出依据本发明对于一个特定的输入符号值的例子的预测滤波器输出值和峰值降低操作的复数图。 27 is a diagram illustrating the operation of the complex according to FIG reduced specific example of a symbol value input prediction filter output value and the peak value of the present invention.

图28是示出依据本发明在不同的输入符号值上的预测滤波器输出值和峰值降低操作的复数图。 FIG 28 is a graph showing predicted filter output values ​​in the present invention is different input symbol values ​​and the peak reduction based on the complex operation of FIG.

图29是示出依据本发明的峰值降低单元中一级的详细实施方案方框简图。 29 is a diagram illustrating a detailed embodiment of the reduction unit a block diagram of a peak according to the present invention.

图30是示出依据本发明实施一种近似的峰值降低算法的峰值降低单元中一级的详细实施方案方框简图。 FIG 30 is a diagram illustrating an approximate peak embodiment according to the peak reduction algorithm reducing block diagram of a detailed embodiment of a unit according to the present invention.

图31是示出依据本发明采用反馈的峰值降低单元中一级的详细实施方案方框简图。 FIG 31 is a diagram illustrating a detailed embodiment of the reduction unit a block diagram of a peak according to the present invention employs feedback.

图32是示出依据本发明的峰值降低单元中一级的一种替代的实施方案方框简图。 32 is a diagram illustrating an alternative means to reduce the block diagram of an embodiment of the present invention according to the peak.

图33是示出依据本发明利用反馈时图32中所示的替代实施方案方框简图。 FIG 33 is a block diagram of an alternative embodiment shown in accordance with the present invention, when using the feedback 32 in FIG.

图34是示出依据本发明在符号间间隔上操作的峰值降低单元中一级的一种详细实施方案方框简图。 FIG 34 is a block diagram of a detailed embodiment shown a reduction unit according to the peak interval between symbols operation of the invention.

图35是示出利用反馈的图34中所示的详细实施方案方框图。 FIG 35 is a block diagram showing in detail using the feedback of the embodiment shown in FIG. 34.

图36A和36B是示出依据本发明采用图23中所示的并行实施峰值降低级的一种多级峰值降低单元详细实施方案方框简图。 FIGS 36A and 36B are diagrams illustrating a parallel according to the invention shown in FIG. 23 embodiment of a multi-stage peak stage of the peak reduction unit reducing block diagram of a detailed embodiment.

具体实施方式 detailed description

参考图6,示出一种依据本发明采用峰值功率降低的扩频通信系统优选实施方案。 Referring to Figure 6, according to the present invention shows a spread spectrum communication system using a preferred embodiment of a peak power reduction. 虽然所示的扩频通信系统可以在一种无线蜂窝网,如WCDMA或CDMA网中实施,并且这些系统提供本发明的一种优选的应用,应该认识到对于本发明的其他应用和环境也是可能的。 Although the spread spectrum communication system shown may be implemented in a radio cellular network, such as CDMA or WCDMA network, and that these systems provide a preferred application of the invention, it should be appreciated that other applications and environments of the present invention are possible of.

如图所示,数量为N的多个信道被提供为进入系统的数据输入。 As shown, the number N of the plurality of channels are provided as an input system for entering data. 数据信道可以包括例如在数字蜂窝应用中的声频数据,或者可以包括希望通过通信系统发送的任何其他形式的数据。 The data channel audio data may include, for example, in digital cellular applications, or may comprise any other desired form of data transmitted through a communication system. 然后在每个信道中的数据通过一个数据-符号变换器100,从输入的数据位流提供符号流。 Then in each channel data by a data --100 provides a stream of symbols from the input data bitstream sign changer. 各种各样不同的符号编码方案可被用于从输入的数据位流提供符号流(QPSK或“正交相移键控”被用在WCDMA中)。 A variety of different symbols encoding scheme may be used to provide a data symbol stream from the input bit stream (QPSK or "quadrature phase-shift keying" is used in WCDMA). (在图6中意味着,在数据-符号变换器之后,所有的处理路径是复数,包括同相和正交分量)。 (In FIG. 6 means, the data - after the symbol converter, all of the processing path is plural, comprising inphase and quadrature components). 接着,在每个信道中的符号流被提供给一个混合器110,在其中将每个信道中的输入符号流与由扩展代码电路112所提供的扩展代码混合。 Subsequently, the mixer 110 is supplied to a symbol stream for each channel, in which the input symbol stream for each channel by the spreading code is mixed with a spreading code provided by the circuit 112. 例如,在扩频蜂窝通信系统中可以采用一种Walsh码。 For example, in a spread spectrum cellular communication system may use a Walsh code. 每个信道接收一个唯一的正交扩展代码,允许在接收机端利用一种匹配的去扩展代码恢复各个信道。 Each channel receives a unique orthogonal spreading code that allows the receiver side to restore the code matching using one of the respective despread channel. 在与唯一的扩展代码组合以后每个信道可再次被提供给另一个混合器114,在其中将每个信道中的信号与来自扰频代码电路116的扰频代码组合。 In combination with a unique spreading code for each channel may later be re-supplied to another mixer 114, in which each channel signal is scrambled with the scrambling codes from the scrambling code combination circuit 116. 扰频代码被用在蜂窝应用中供单元位置的识别。 Scrambling code is used in cellular applications for position identification unit. 典型情况下扰频代码被应用在蜂窝通信系统中,但也可被分配在其他的应用中。 Typically scrambling code is used in a cellular communication system, but may be assigned in other applications. 然后每个信道的输出被提供给加法电路120,将来自各个信道中每一个的符号流组合并将它们组合到单一的输出符号流中(对于两种复数正交相位中每一个)。 Then the output of each channel 120 is provided to the channels from each of the respective symbol stream compositions and combinations thereof to a single output symbol stream (for both quadrature phase of each complex). Adder circuit

因为在加法电路120中N个信道被组合在一起,如在以上背景部分中所讨论过的那样,存在产生非常大的符号峰值的可能性。 Because are combined in an adder 120 N channel circuits, as above discussed in the background section above, the possibility of a very large peak exists symbol generation. 本发明提供一种峰值降低单元122,降低或消除从滤波器126输出的信号峰值,这些峰值将超过为特定的通信系统应用选取的给定的最大信号峰值功率电平。 The present invention provides a peak reduction unit 122 reduces or eliminates peak signal output from the filter 126, these peaks will exceed a specific communication system to select a given maximum signal peak power level. 如在图6中进一步示出的那样,峰值降低单元122的输出被提供给一个上采样电路124,将符号速率变换为满足或超过对于感兴趣的频率范围的带宽要求的频率。 As further shown in FIG. 6, the peak reduced output unit 122 is supplied to a sampling circuit 124, converted into the symbol rate to meet or exceed the frequency bandwidth of the frequency range of interest. 典型情况下上采样过程将直接将零插入符号流,使信号流能够工作在所希望的较高钟速率以满足或超过带宽的需要。 Upsampling process directly inserted symbol stream zero Typically, a signal flow can work at the desired high clock rates to meet or exceed the bandwidth needed. 因此,对于每个输入符号将插入M个零到符号流中,其中M是为提供所希望的上变换选取的一个整数。 Thus, for each input symbol M zeros inserted into symbol streams, where M is the desired transform to provide a selected integer. 典型情况下整数M将至少是1或更多。 Typically an integer of at least 1 M or more. 然后上采样的信号流被提供给一个滤波器126。 Then the up-sampled signal stream is provided to a filter 126. 在某些情况下上采样电路124的操作可被并入滤波器126的操作中而不是在分离的电路中进行。 In some cases the operation of the sampling circuit 124 may be incorporated in the operation of the filter 126 rather than in a separate circuit. 滤波器输出被提供给一个数模变换器128以提供模拟信号。 Filter output is supplied to a digital to analog converter to provide an analog signal 128. 这个模拟信号将由对此细节未被示出的同相和正交分量组成。 The inphase and quadrature components of the analog signal by the details are not shown on this composition. 这个模拟信号在混合器134中与来自RF源136的RF载波混合,然后RF调制的信号被提供给一个RF放大器130,然后,例如在无线蜂窝通信基站的应用中,被提供给一个RF发射机132。 This analog signal is mixed in mixer 134 with the RF carrier from the RF source 136, the modulated RF signal is then supplied to an RF amplifier 130, and then, in applications such as wireless cellular communications base station, is supplied to an RF transmitter 132. 混合器134在这种应用中实际上是一种对此细节未示出的正交上变换混频器。 In such applications, the mixer 134 is actually this one kind of orthogonal transform details not shown mixer. 正如对本领域的技术人员是众所周知的那样,从滤波器输出的信号可有选择地从同相和正交信号变换成偏离零Hz的基带中心频率的实信号。 As those skilled in the art is well known, the signal output from filter may be selectively converted into a real signal deviates from zero Hz of the center frequency of the baseband in-phase and quadrature signals. 如果这被完成,D/A变换器可被用于产生实中频输出,然后可利用一种简单的标准混频器混频成RF频率。 If this is done, D / A converter may be used to generate a real IF output, may then be mixed with a simple standard mixer into an RF frequency.

如在以上背景部分中所讨论过的那样,由滤波器126所提供的滤波操作在许多采用扩频通信的应用中是极其重要的。 As in the above discussed in the background section above, the filtering operation provided by the filter 126 employed in many applications of spread spectrum communication are extremely important. 特别是,在蜂窝通信系统中,由于必须保持所发送的信号在预先规定的频带内这样的滤波是关键性的。 In particular, in a cellular communication system, since the transmitted signal must be maintained within a predetermined frequency band such filtering is critical. 滤波器126也将对由符号流产生的信号峰值具有重大的影响,因此可以显著地影响在这样的滤波以前所实施的任何峰值降低。 Filter 126 will also have a peak signal produced by the symbol stream having a significant impact, and therefore can significantly affect any of the peak filter in such embodiments previously reduced.

峰值降低单元122通过根据符号流输入预测滤波器输出126,然后根据滤波后所预测的信号峰值对符号实行峰值降低处理。 Peak reduction processing unit 122 to reduce input symbol streams output prediction filter 126, and then the implementation of the peak of the peak signal symbols according to the predicted filtered through. 峰值降低单元122的这种滤波器预测操作可被非常精确地实现,因为滤波器126的脉冲响应函数是事先已知的。 Such a filter unit to reduce a peak prediction operations may be implemented very precisely 122, 126 because the filter impulse response function is known in advance. 因此,滤波器对任何给定的符号流的影响可被一种电路精确地预测到任何所希望的精确度,该电路具有作为对该电路的输入的所希望数目的滤波器系数,以便适当地模拟滤波器脉冲响应函数。 Thus, the effect of the filter for any given symbol streams can be accurately predicted. A circuit to any desired accuracy, the circuit has the filter coefficients as an input to the desired number of circuits, in order to properly analog filter impulse response function. 这些滤波器系数被加到输入符号流上,模拟滤波器对输入符号流的影响,产生预测的滤波输出流。 These filter coefficients are applied to the input symbol stream, the filter simulated input symbol stream, generating a prediction of the filtered output stream. 然后预测的滤波输出流被经受峰值降低计算,确定是否预测的滤波器输出流将超过信号峰值限值,如果是的,就需要校正。 Then the predicted filter output stream is subjected to reduced peak calculation, determining whether the prediction filter output signal stream exceeds the peak limit, if yes, it needs to be corrected. 如果限值被超过,则实际的符号流被经受峰值降低处理,在逐个符号的基础上将所得的信号峰值降低到所希望的限值。 If the limit value is exceeded, then the actual symbol stream is subjected to peak reducing process decreases to the desired value in the peak signal obtained on the basis of symbol by symbol.

用这种方法,只有需要的峰值降低被执行,最少的畸变被引入信号。 In this way, only the peak needs to be performed to reduce the minimum signal distortion is introduced. 峰值降低处理的精度以及滤波器预测过程的精度可根据峰值降低电路的速度和复杂性和相关联的成本的权衡来选取。 Accuracy and precision of the peak reduction process filter prediction process may be selected according to the trade-off peak velocity and reducing circuit complexity and cost associated. 然而,在大多数应用中,滤波器预测处理以及峰值降低处理可被适当地实施而不对整个系统增加很大的复杂性或成本。 However, in most applications, the filter and the peak reducing process prediction processing may be suitably implemented without greatly increasing the complexity or cost of the entire system.

参考图7和8,示出一种峰值降低单元122的优选实施方案。 7 and 8, shows a preferred embodiment of the peak reduction unit 122. 更具体而言,在图7中示出一种峰值降低单元122的方框简图,在图8中用作说明在峰值降低电路122中采用的典型滤波器系数,示出一种典型滤波器的脉冲响应函数。 More specifically, in FIG. 7 shows a block diagram of the peak reduction unit 122, be described as a typical filter coefficient reducing circuit 122 employed in the peaks in Figure 8, illustrates a typical filter the impulse response function. 如图所示,峰值降低单元122最好包括两级140,142,分别对应于在符号上的间隔,t=0,和符号间的间隔,t=0.5上的峰值降低。 As shown, the peak reduction unit 122 preferably includes two stages 140, 142, respectively corresponding to the symbol interval, t = the interval between 0 and symbol, t = 0.5 to reduce the peak. 这两级的次序是随意的。 Both levels of the order is arbitrary. 正如以上关于图6所讨论过的那样,符号流典型情况下在通过滤波器126以前被上采样。 As discussed above with respect to FIG off, typically before the symbol stream is sampled by the filter 126. 因此,附加的采样点将被添加在符号序列中实际符号之间,滤波器脉冲响应函数将被加在添加的采样点以及符号采样点上。 Thus, additional sampling points are added between the actual symbol in the symbol sequence, to the filter impulse response function is applied to the sampling point and the symbol sampling points added. 假定对符号流中每个符号的一个添加的采样点上采样,滤波器的脉冲响应函数将被加在符号上的间隔和符号之间一半的位置,也就是符号间的间隔上的上采样符号流上。 Assumed that the sampling points of a symbol stream added to each symbol sample, the filter impulse response function will be added half way between the spacer and the symbols on the symbol, i.e. the sampling interval between the symbols on the symbol stream. 这被示于图8中,在其中在符号上间隔上的滤波器系数被用星号(*)表示,在符号间的间隔上滤波器系数被用十字号(+)表示。 This is shown in Figure 8, wherein the filter coefficients in the intervals (*) indicated by an asterisk in the symbol filter coefficients are (+) is represented by a cross in the space between the number of symbols. 因为符号和符号间滤波器系数的影响当它从滤波器126出现时将被加在符号流上,为了准确地预测在峰值降低单元中滤波器的响应,必须既考虑在符号上的间隔上的滤波器系数,也考虑在符号间间隔上的滤波器系数。 Because of the influence of filter coefficients between symbols and when it emerges from filter 126 will be added to the symbol stream, in order to accurately predict the response of the filter in the peak reduction unit, we must consider both the spacing in the symbol the filter coefficients, the filter coefficients are also contemplated in the inter-symbol interval. 图7中所示的两级过程使这种符号上和符号间的处理能够被串行地实施。 Two process shown in FIG. 7 and process make inter-symbol can be implemented on such symbols serially. 这种串行实施方式比起同时地,也就是并行地进行符号上和符号间处理可以使用较少的硬件,或较不复杂的DSP程序。 This serial embodiment than simultaneously, i.e. in parallel between the symbols and the process may use less hardware or less complex DSP programs. 不过,应该认识到,在一种替代的实施方案中,可以完成这样一种同时处理方式,以下描述这样一种实施方案。 However, it should be appreciated that in an alternative embodiment, simultaneous processing can be completed in such a way, such an embodiment is described below. 在对于符号序列中每个符号具有多于1个添加的符号的上采样符号流的情况下,为了滤波器预测处理可以增加附加的滤波器定时点。 In the case of the samples having more than one symbol stream added to the symbol for each symbol in the symbol sequence, to filter prediction process may add additional filter timing point. 也应该认识到,虽然至少一个符号上和符号间的处理目前是优选的,提供只有符号上(或符号间)的峰值降低处理并且仍然获得某些有益的结果也是可能的。 It should also be recognized that while processing between at least one of the symbols and are presently preferred to provide only a peak on the symbol (or inter-symbol) reduction process and still achieve some beneficial results are also possible. 另外,为了估计由于峰值降低处理的因果关系的性质引起的峰值降低误差,任何或全部的级可被重复。 Further, since the peak to reduce the peak to estimate the properties of causality process to reduce the error due to any or all of the stages may be repeated.

虽然图6示出了在上采样电路124以前的峰值降低单元,也可以配置在上采样电路(但在滤波器126以前)以后。 While FIG. 6 illustrates a reduction in the peak before the sampling circuit unit 124 may be disposed on a sampling circuit (filter 126 but before) later. 当在上采样以后操作时,直接按上采样速率从滤波器脉冲响应函数取得系数。 When operating in the later sampling, the sampling rate directly from the press filter impulse response function to obtain coefficients. 然后这些系数被用于产生滤波器输出预测器。 These coefficients are then used to generate the predictor filter output. 将在上采样速率上完成滤波器输出预测。 The complete upper prediction filter output sampling rate. 将从这些预测结果完成峰值降低。 These forecast results from the completion of peak reduction. 这种方法对于某些应用和/或某些脉冲响应函数并不是优选的。 For certain applications this approach / or certain impulse response function and are not preferred. 当在上采样速率上工作时,可以对在上采样期间添加的零值符号提供峰值降低。 When working on the sampling rate, the peak reduction can be provided the zero-value symbols during the sampling added. 数字通信系统,如CDMA和WCDMA,使用为使符号间干扰最小设计的发送和接收滤波器。 Digital communication systems such as CDMA and WCDMA, the use of transmit and receive filters such that the minimum intersymbol interference designed. 为了在通信期间保持最小的符号间干扰,在上采样期间添加的零值符号必须仍然是零值。 In order to maintain the minimum inter-symbol interference during communication, a zero value symbols during the sampling value added must still be zero. 在上采样速率上工作也需要在较高速率上进行处理。 Work needs to be processed at a higher speed on the sampling rate. 一般,处理速率越高,处理部件的成本越贵。 In general, the higher the processing speed, the more expensive the cost of the processing means. 然而,可能有这样一些应用,在其中这些成本是无关紧要的,不需要被考虑,则在上采样以后处理被优先采用。 However, there may be some applications in which the cost does not matter, need not be considered, after the upsampling process is given priority. 对于以下所描述的实现上采样以后的峰值降低单元的特定的实施方案必要的任何修改对于本领域的技术人员将是明显的,在此是不言而喻的。 Any modification particular embodiment the reduction means for sampling the peak after the realization described below is necessary to those skilled in the art will be apparent, this is self-evident.

图7中的两级可以使用图9中所示的降低过程。 Two in FIG. 7 may be used to reduce the process shown in FIG. 参考图9,降低过程包括一个滤波器系数的源144。 Referring to Figure 9, the process comprising reducing a source 144 of filter coefficients. 这些滤波器系数被取自符号上或符号间的间隔上的滤波器脉冲响应函数,这取决于图7中的处理级140或142。 The filter impulse filter coefficients are taken from the space between the upper symbol or symbol response function, depending on the processing stage 140 or 142 in FIG. 7. 这些系数源可以采取存储滤波器系数的存储器144的形式,例如,对于适当的符号间隔在图8中所示的系数。 These sources may take the form factor of the memory 144 stores filter coefficients, e.g., for a suitable symbol intervals in 8 coefficients shown in FIG. 当然,其他的滤波器实施方案可以具有不同的滤波器响应函数,因此不同的滤波器系数将被存储在滤波器系数存储器144中。 Of course, other embodiments of a filter may have different filter response functions, thus different filter coefficients will be stored in the filter coefficient memory 144. 这些滤波器系数被提供给一个滤波器预测器146,接收沿线路148提供的输入符号流并模拟在所选的符号间隔上滤波器126对符号流的影响,从滤波器预测器提供两个输出,一个输出147是使用所有输入滤波器系数和相等数目的时间微分的输入符号的滤波器系数加权和。 These filter coefficients are supplied to a prediction filter 146, the input symbol stream received along path 148 and provided at the symbol interval analog effects filter 126 selected symbol stream and provides two outputs from the filter predictor , a filter coefficient output 147 is all of the input filter coefficients and an equal number of time differential input symbols and weighted. 另一个输出145是只使用中心滤波器系数和相匹配的中心符号的滤波器系数加权和。 Another is to use only the output 145 of filter coefficients and weighting coefficients that match the filter center and the center of the symbol. 当奇数数目的滤波器系数被使用时,第二输出是用于计算第一输出147的中心系数和匹配的中心符号。 When an odd number of filter coefficients are used for calculating a first output a second output 147 symbols of the center and the center of the matching coefficients. 当偶数数目的系数被使用时,两个中心系数将具有相同的值,可被随同两个匹配的中心符号使用计算第一输出147。 When an even number of coefficients is used, two centers will have the same coefficient value, may be calculated along the center of the symbols used to match the two first output 147. 每次一个新的符号进入滤波器预测器时由146产生两个输出。 Each time a new symbol entering the filter from the predictor 146 produces two outputs. 每个新的输出对被在逐个符号的基础上提供给峰值降低算法处理电路152。 Each new output reduction peak arithmetic processing circuit 152 is supplied to a symbol by symbol basis. 峰值降低算法处理电路152将第一预测滤波器输出147的幅度与预先确定的最大允许的峰值限值L作比较。 Peak reducing arithmetic processing circuit 152 to output a first prediction filter 147 to the peak amplitude of the maximum permissible limits of a predetermined L for comparison. 如果第一预测滤波器输出147超过限值,则峰值降低算法电路152计算对第二滤波器预测器输出145的调节量,这将导致被图6的滤波器126处理以后滤波器输出将仍然在峰值限值内。 If the first prediction filter 147 output exceeds the limit, then the algorithm 152 calculates the peak reduction circuit 145 outputs the adjustment amount for the second predictor filter, which will cause the filter is after processing 126 of FIG. 6 will remain in the filter output peak within limits. 然后这个调节量被组合器168在逐个符号的基础上施加到被延时电路166延时的相应的符号上。 This amount is then adjusted on the basis of the combiner 168 is applied symbol by symbol to the corresponding symbol delay circuit 166 is delayed. 然后峰值已调节的符号流被沿着线路154输出。 Then the adjusted peak symbol stream is output along line 154. 各种各样不同的算法可被应用在峰值降低算法电路152中。 A variety of different algorithms may be applied to reduce the peak in the arithmetic circuit 152. 可以根据对于特定的应用所希望的精确度和可达到的处理速度和/或所希望的硬件的复杂程度选取特定的算法。 Can be selected according to the complexity of the algorithm is specific for the particular application desired accuracy and processing speed can be achieved and / or desired hardware. 例如,在许多应用中一种近似算法可能是完全可接受的,在符号序列中给出所希望的峰值降低。 For example, in many applications an approximate algorithm may be perfectly acceptable, given the desired reduction in peak symbol sequence.

图9的一种替代实施方案被示于图10中,在其中图9的滤波器预测器146,延时器166和组合器168被合并为滤波器预测器的部件。 FIG 9 is an alternative embodiment is shown in FIG. 10, FIG. 9 in which the predictor filter 146, delay 166 and combiner 168 are combined into a predictor filter member. 由峰值降低算法152提供的调节量被回送到滤波器预测器中以便将目前的调节量引入未来的预测中。 Adjusting the amount of reduction algorithm provided by the peak 152 is returned to the filter for the current predictor adjustment amount introduced in the future prediction. 当滤波器预测器实施方案被讨论时,将较详细地讨论这种引入。 When the predictor filter embodiments are discussed, such incorporation will be discussed in greater detail.

参考图11,(图6的)峰值降低单元122的一种替代的实施方案被示出。 Referring to Figure 11, (FIG. 6) the peak reduction unit 122 An alternative embodiment is shown. 图11的实施方案提供一种利用多级串行排列的多级峰值降低单元。 Fig 11 embodiment provides a multi-stage peak using a multi-stage reduction unit arranged in series.

更具体地,参考图11,所示出的峰值降低单元包括多个单独的级320。 More specifically, referring to FIG. 11, the peak reduction unit shown comprises a plurality of individual stages 320. 在峰值降低单元中每一级320利用一套与特定的滤波器输出定时对应的周期性采样的滤波器系数施加滤波器预测操作。 Reducing peak in each unit 320 is applied to a prediction filter operation using the filter coefficients periodically sampled with a particular set of filter output corresponding to the timing. 例如,如果上采样在符号之间插入9个零,则产生的实际的滤波器操作对于每个符号间隔将包括10个滤波器定时点。 For example, if the sample is inserted between the nine zero symbol, the actual operation of the filter is generated for each symbol interval filter 10 comprises timing point. 为了提供滤波器操作完全精确的模型,这些滤波器定时点中每一点将需要被包括在峰值降低处理中。 In order to provide a filter operation is completely accurate models, these filters each timing point reduction processing point needs to be included in the peak. 这样,滤波器脉冲响应函数对于每个符号间隔将需要在10个不同的位置上被采样。 Thus, the filter impulse response function would need to be sampled for each symbol interval at 10 different positions. 对于一个特定的脉冲响应函数在每个符号间隔的10个滤波器采样位置的这种例子被示于图12-21中。 For a specific example in which the impulse response function 10 filters the sampling positions of each symbol interval is shown in FIGS. 12-21. 这10个图中每一个示出符号间隔内不同的滤波器系数采样定时。 FIG 10 which each different filter coefficients illustrated symbol interval sampling timing. 特别是,图12表示在从符号上的间隔的-0.5的定时,也就是在负的时间方向(符号间)两个符号之间半程点的偏离上开始的符号间隔上对脉冲响应函数周期性的采样。 In particular, FIG 12 shows timing of a symbol interval from -0.5 spaced on the symbol that is offset from the halfway point between (inter-symbol) in the time direction the negative sign two starting period of the impulse response function of sampling. 图13表示从-0.4的定时开始的符号间隔上周期性地采样的脉冲响应函数。 13 shows the pulse symbol interval starting from the timing -0.4 sampled periodically in response function. 图14-21依次表示从-0.3至+0.4偏离的顺序的采样定时。 14-21 showing sequentially from -0.3 to +0.4 sampling timing offset order. 因此图12-21合起来表示对于符号上间隔对称的10个滤波器系数样本位置。 FIG. 12-21 thus combined to represent the symbol 10 symmetrical filter coefficients spaced sample position. 图11中峰值降低单元的每一级320在一个单独的采样定时点上执行滤波器预测操作。 FIG peak 11 of each reduction unit 320 performs a prediction filter operation on a separate sampling timing point. 因此,对于图12-21中所示的特定的脉冲响应函数和采样,10个分离的滤波器级320将被提供,每级在与图12-21中的一个对应的一个定时点上提供滤波器预测操作。 Thus, for a particular pulse shown in FIG response function and samples 12-21, 10 separate filter stages 320 are provided, each stage providing a filter on a timing point corresponding to 12-21 in FIG. prediction operations. 具体地说,级320-1可以对应于图12中所示的采样定时,级320-2对应于在图13中所示的采样定时,等。 Specifically, stage 320-1 may correspond to the sampling timing shown in FIG. 12, a timing corresponding to the sample stage 320-2 shown in FIG. 13, and the like.

当然,将认识到在图11和图12-21中的10个采样点和10级的例子纯粹是用作说明,可以提供较多或较少数目的采样点和级。 Of course, it is appreciated that the sampling points and 10 in FIGS. 12-21 and 10 in FIG. 11 example purely as described, can provide greater or fewer number of sampling points and level. 级320的数目也不需要与发生的上采样的特定数量对应,可以采用比实际的上采样点的数量少的级和系数采样点。 The number of stages 320 need not correspond to a specific number of samples on the occurrence, it can be employed less than the actual number of sample points on the stage, and the coefficient of the sample points. 图11也示出按时间次序从t=-0.5到t=0.4排列的每个峰值降低过程。 FIG 11 also shows a chronological order from -0.5 t = 0.4 to t = arranged in each peak reduction processes. 在图11中峰值降低级的时间关系可以是按任何随意的次序排列。 Reducing the time of the peak level in FIG. 11 may be arranged in any arbitrary order. 另外,任何或所有的级可被重复以便估计由峰值降低处理的因果关系性质引起的峰值降低误差。 Additionally, any or all of the stages may be repeated in order to estimate the peak reducing process by the causal nature of the peak reduction caused by the error.

参考图22,峰值降低单元的一种替代的实施方案被示出,包括一种以并行方式实施的多级峰值降低处理。 Referring to Figure 22, an alternative embodiment of the peak reduction unit is shown in a multi-stage peak comprise one embodiment of the reduction process in a parallel manner. 以上关于图11描述的峰值降低单元实施方案是按串行方式实施的一种多级峰值降低过程。 FIG 11 described above with respect to the peak of the reduction means is a serial embodiment of a multi-stage embodiment of the peak reduction process. 在图22中所示的实施方案中,一种类似的多级峰值降低过程是以并行方式实施的。 In the embodiment shown in FIG. 22, a similar multi-stage embodiment of the peak reducing process is in parallel. 也就是,在图22中所示的每级360根据滤波器脉冲响应函数中滤波器系数的不同定时实施峰值降低处理,与对输入符号的上采样后发生的较高速率的滤波相对应。 That is, the function corresponding to the different timings of the filter coefficients of each peak reducing process stage embodiment 360 in accordance with the filter impulse response, filter and sampled to a higher rate of occurrence of the input symbols in FIG. 22. 例如,在图22中所示的每级360可采用在图12-21中所示的不同定时上采样的滤波器系数执行峰值降低处理。 For example, the peak reducing process performed at each stage of the filter coefficients 360 can be sampled at different timing shown in FIG. 22 12-21 shown in FIG. 然而,因为在图11的实施方案的情况下,可以采用不同的定时和不同的特定的滤波器脉冲响应函数,图12-21的特定的滤波器系数和系数采样定时纯粹是用作说明。 However, since in the case of the embodiment of FIG. 11 may be employed various different specific timing and the filter impulse response function, and the particular filter coefficients coefficients 12-21 of FIG sampling timing is purely for illustrative. 另外,任何的或所有的符号系数定时可被重复,以便估计由峰值降低处理的因果关系性质引起的峰值降低误差。 Further, any or all of the coefficients of the timing symbols may be repeated, so that the estimated peak reduction processing by the causal nature of the peak reduction caused by the error.

在以前所描述的图11的实施方案中,在多级峰值降低单元中每个相继的级接收来自前级的峰值已调节的符号作为其输入。 In the embodiment of Figure 11 described previously, the received symbols from the preceding stage of the peak regulated as an input unit in the multi-stage peak decrease in each successive stage. 因此,采用这种串行实施方案避免了对以前已调节过的符号的峰值调节作不必要的重复。 Accordingly, with this embodiment, a serial-to-peak has been avoided previously adjusted symbols for conditioning unnecessary duplication. 在图22的并行实施方案中,最好采用一种多级反馈方法,将峰值降低值从并行峰值降低级回送到其他的级以达到类似的结果。 In the parallel embodiment of Figure 22, it is preferable to adopt a multi-stage feedback method, the peak reduction peak values ​​from the parallel back to the other reduction stage to achieve a similar level results. 更具体而言,图22的分支0,顶部的分支的峰值降低过程接收来自它自已内部的峰值降低算法的反馈,也将这种反馈提供给所有的较低的分支。 More specifically, FIG branch 022, the peak top of the branches receive feedback from the process of reducing the peak reduced its own internal algorithms, which will also provide feedback to all of the lower branch. 较低的分支,如分支i,接收来自它自己内部的峰值降低算法的反馈和来自在它之上所有支路的反馈。 Feedback lower branches, such as branches i, the reception peak decreased from its own internal algorithms and feedback from all of the branches in the top of it. 最后的分支,分支N-1,接收包括来自它自己的内部峰值降低算法的所有分支的反馈。 The last branch, which is N-1, comprising receiving from its own internal feedback branch all peak reduction algorithm. 如图22中所示,底部的级输出全部峰值已调节的符号流。 As shown in FIG. 22, the entire stage output symbol streams peak regulated bottom.

参考图23,示出一种对并行处理实施方案的替代方案。 Referring to Figure 23, there is shown an alternative embodiment of the parallel processing. 这种实施方案除了来自所有分支的反馈被提供给每个分支的峰值降低单元外,与图22的方案是相同的。 Apart from such an embodiment, feedback is provided to all branches of each branch of the peak reduction unit, the Scheme 22 is the same as FIG. 这种替代的实施方案改进了来自每个分支的峰值降低计算,因为每个计算将基于最近已调节的符号。 This alternative embodiment improves the reduction peak from each branch is calculated, based on the latest calculation because each adjusted symbol.

因为在图22和23的实施方案中峰值降低处理是以并行方式完成的,可以比以前所描述的串行实施方案完成得快。 Because the peak reducing process in FIG. 22 are parallel and embodiment 23 is completed can be completed faster than the serial embodiments previously described. 因此,在某些情况下,图22和23的实施方案可能被优先选择。 Thus, in some cases, the embodiment of Figure 22 and 23 may be preferentially selected. 因为处理也可被更快地完成,在更多的系数定时和/或利用更多的系数实施滤波器预测,从而利用图22的并行实施方案增加峰值降低处理的精确度也是可能的。 Because the process can be completed more quickly, more coefficients in the timing and / or to use more filter prediction coefficients embodiment, to take advantage of the embodiment of Figure 22 in parallel to increase the accuracy of the peak reducing process are possible.

参考图24,示出图22和23的并行峰值降低单元的一级360。 Referring to Figure 24, a diagram showing the parallel peaks 22 and 23 decrease in a 360 unit. 如图所示,对于级360的输入符号首先被提供给滤波器预测器372,利用在特定的偏置-符号定时上的被示出从滤波器系数源144提供的滤波器系数实施滤波器预测处理。 As shown, for input symbols stage 360 ​​is first supplied to the prediction filter 372, utilizing specific bias - prediction filter embodiment is shown on the symbol timing filter coefficients from the filter coefficients 144 provided by the source deal with. 滤波器系数源144可被硬线连入电路,或者可采取适当的存储器的形式,如在峰值降低单元的一种被适当编程的DSP实施方案中的一个寄存器。 Filter coefficient source 144 may be hard-wired into the circuit, or may take the form of suitable memory, such as a reduction in the peak register one kind of embodiment is suitably programmed DSP unit in. 如以上所指出的那样,对于供应给滤波器预测器372的滤波器系数的特定的采样偏置定时对应于在系统中所采用的实际滤波器的上采样速率上的所选的采样偏置定时。 The sampling timing of the bias on the selected sampling rate As noted above, the offset for a particular sample is supplied to the prediction filter 372, the filter coefficients corresponding to the actual timing of the filter employed in the system of . 因此,例如,如图12-21中所示的采样定时可被应用于存储在滤波器系数源144中用于每个特定的级360的滤波器系数。 Thus, for example, the sampling timings shown in FIG. 12-21 can be applied to the storage for each particular stage 360 ​​of filter coefficients in the filter coefficient source 144.

滤波器预测器372输出两种在图9中描述过的预测的已滤波输出到峰值降低算法电路152,在其中实施一种适当的峰值降低算法以确定峰值降低值,如果必要的话,将峰值降低到所希望的限值。 Predictor filter 372 outputs two kinds described in FIG. 9 has been predicted peak reduction algorithm to the filtered output circuit 152, in which embodiment a suitable algorithm to determine a peak reducing peak reduction value, if necessary, peak reduction to the desired limit. 算得的峰值降低值被作为反馈值F1从级360输出,被提供给它自己的内部滤波器预测器372和图22和23中其他并行分支360-1的滤波器预测器。 Value is calculated as the peak reducing feedback value F1 from the output stage 360 ​​is supplied to its own internal filter prediction filter 372 and the predictor 22 and 23 in the other parallel branch 360-1.

如以上指出的,各种各样不同的算法可在图9,10和24的峰值降低算法电路152中实施。 As noted above, a variety of different algorithms may be implemented in the arithmetic circuit 152 to reduce the peak of FIGS. 9, 10 and 24. 一种这样的算法被示于图25中。 One such algorithm is shown in FIG. 25. 图25示出一种复数向量图,用作说明基于利用所有的输入系数的输入符号的滤波器系数加权和,和只有中心滤波器系数和匹配的中心符号被使用的输入符号的滤波器系数加权和的滤波器输出。 FIG 25 shows a complex vector diagram as described and weighting filter coefficient, only the center and the input symbol and the symbol center matched filter coefficients to be used based on the use of all input symbols of the input weighting coefficients and the filter output. 对于适合将滤波器输出放置在预置的限值L内的滤波器为中心的符号从这两个输入计算校正值。 The filter output is suitable for placement within a preset limit value L of the filter centered symbol correction value is calculated from these two inputs. 在图25中所示的算法被特别设计用以只将幅度误差引入输入符号。 The algorithm shown in FIG. 25 is specifically designed for only amplitude errors into the input symbol. 某些通信系统对于幅度误差比相位误差更能容忍。 For some communication systems are more tolerant of errors than the amplitude of the phase error.

更具体地参考图25,基于中心输入符号的滤波器输出被用向量A表示。 Referring more particularly to FIG. 25, a filter output based on the center of the input symbols is represented by a vector A. 基于多输入符号,包括中心输入符号,的预测滤波器输出被用向量B表示。 Multi input symbols, comprising a central input symbols, the prediction filter output is represented by a vector B. 通过取这两个向量的差计算向量D。 This is calculated by taking the difference between two vectors vector D. 因此向量D表示从滤波器输出的缺少中心符号向量A的多个符号,输出gA表示向量A的一种增益已调节的形式,当被添加到D上时,将滤波器输出拉回到限值电平L。 Accordingly vector D represents a plurality of symbols missing from the center of the symbol vector filter output A, output gain gA vector A shows a regulated form, when added to D, back to the filter output value level L. 在图25中所示的其余的向量被用于计算增益g。 It is used to calculate the gain vector g in the remaining 25 shown in FIG. 通过以下一系列向量计算计算增益g。 Calculating the gain g calculated by the following series of vectors.

L2=z2+y2z=L2-y2]]>g|A|=L2-y2-x]]>x=D·A|A|]]>y=|D-(D·A)A|A|2|]]>g=(L2-|D-|(D·A)A|A|2|2-(D·A)|A|)1|A|---(1)]]>这种算法可通过使用一种被适当地编程的DSP或其他的处理器在图9,10和24中所示的峰值降低算法电路中实施。 L2 = z2 + y2z = L2-y2]]> g | A | = L2-y2-x]]> x = D & CenterDot; A | A |]]> y = | D- (D & CenterDot; A) A | A | 2 |]]> g = (L2- | D- | (D & CenterDot; A) A | A | 2 | 2- (D & CenterDot; A) | A |) 1 | A | --- (1)]]> this algorithms implemented in the arithmetic circuit may be reduced in the peak shown in Figures 9, 10 and 24 by using a DSP or other a processor is suitably programmed. 因为该算法包含取若干向量量的平方根,这可能多少是高强度的计算,因此可能希望采用一种近似算法,在大多数情况下也将提供适当的符号校正。 Because the method comprising taking the square root of the quantity of a plurality of vectors, which may be calculated how much of the high strength, it may be desirable to employ an approximation algorithm, in most cases, will also provide proper correction of symbols. 特别是,如果假定,在预测的输出B和单一的符号输出A之间的差相当小,也就是在以上等式中D是很小。 In particular, if it is assumed, the difference between the predicted output and the B output of a single symbol A relatively small, i.e. in the above equation D is small. 可以使用以下的近似公式计算将滤波器输出降低到限值L所需的增益g。 It can be calculated using the following approximate equation will be reduced to the limit value L required for filter output gain g.

g≈L+|A|-|B||A|,]]>对于小的'D'(2) g & ap; L + | A || B || A |,]]> For small 'D' (2)

然后利用以上的准确的或近似的算法算得的增益值被施加到图9中所示的组合器168。 Then using more accurate approximation algorithm or calculated gain value is applied to the combiner 168 shown in FIG. 9. 组合器可以就是一个乘法器电路,将增益g和产生输出向量A的输入符号相乘。 Combiner is a multiplier circuit can, the gain g and generating output vector A multiplied input symbol. 另一种方案是,如果在以上的(1)和(2)得到的增益计算被转换成通过以下的等式(3)所提供的向量调节量,组合器可被改变成一个加法电路。 Alternatively, if the calculation is converted by the following equation in the above (1) and a gain (2) obtained in (3) adjusting the amount of vector, the mix may be changed to an adder circuit. 值gc是由滤波器施加到图25中产生向量A的中心符号上的增益。 Gc value is applied by the filter 25 to produce a gain on the vector A center mark.

V=(A-gA)1gc---(3)]]>如以上所指出的,可将各种各样不同的算法用于计算将预测滤波器输出放置在限值L内的符号调节量。 V = (A-gA) 1gc --- (3)]]> As noted above, a variety of different algorithms may be used to calculate the prediction filter output value is placed in the sign of the adjustment amount L . 一种这样的附加的算法被示于图26中,示出通过相加与中心输入符号组合的一个校正向量的计算。 One such additional algorithm is shown in FIG. 26, shown is calculated by adding a correction vector to the center of the input symbol combination. 该算法与图25的算法不同,在其中调节量允许有相位误差,以便使加上的总畸变能量为最小。 The same algorithm is different from FIG. 25, in an amount which allows adjusting the phase error so as to minimize the total harmonic distortion plus energy. 更具体而言,如图26中所示,向量A,B和D与图25中具有相同的意义。 More specifically, in FIG vectors A, B and D in FIG. 26 and FIG. 25 have the same meaning. 在图26中值C是为了使预测滤波器输出放置在限值L上对预测滤波器输出所做的附加的调节。 Value C in FIG. 26 is provided to enable the output of the predictive filter is placed in the upper limit value L filter output additional adjustment made to the prediction. 虽然各种各样调节向量可被附加上使所得的向量放置在限值L上,使向量C的大小从而校正量为最小是所希望的,因为对输入符号所做的任何改变都有可能导致信号中某些畸变。 Although various vectors that the adjustment may be attached on the resulting vector is placed in the upper limit value L, so that the magnitude of the vector C so that the correction amount is minimized is desirable, because any changes made to the input symbol may lead to some distortion of the signal. 将基本的向量代数应用到图26中所示的向量得到以下的等式,确定为了得到所希望的在限值L内的峰值降低输出,要被施加到中心符号滤波器输出A的校正量C:C=(LB|B|-B)---(4)]]>在与符号调节量相加以前,以上的向量必须是被施加到用于计算向量C的中心符号的反相滤波器增益所调节的增益。 The basic vector algebra applied to the vector shown in Figure 26 to give the following equation, is determined in order to obtain the desired limiting the peak output decreased in the L, a correction amount to be applied to the center symbol of the filter output A C : C = (LB | B | -B) --- (4)]]> symbol before adding the adjustment amount, the above vector must be applied to the inverse filter calculated for the center mark of the vector C gain adjustment gain. 在(5)中给出为了调节符号使峰值降低所得到的算法。 Given in (5) in order to adjust the peak reduction algorithm symbol obtained. 这种算法可在被适当编程的DSP或其他的硬件或软件实现的电路中实施。 This algorithm may be implemented in a suitably programmed DSP circuit or other hardware or software implementation.

V=(LB|B|-B)1gc---(5)]]>然后在以上的等式(5)中算得的校正向量值被施加到图9中所示的组合器168上。 V = (LB | B | -B) 1gc --- (5)]]> is then calculated in the above equation (5) is applied to the correction value to the combiner shown in FIG. 9 168. 组合器可以就是一个加法电路,将向量V加到产生输出向量A的中心符号上。 Is a combination may adder circuit, the vector V is added to produce the output vector A center mark. 另一种方案是,如果在(4)中得到的向量计算被转换成通过以下等式(6)所提供的增益调节量,组合器可被改变成一个乘法器电路。 Another embodiment is obtained if the vector calculated in (4) is converted into the amount of gain adjustment is provided, it may be changed to a combination of a multiplier circuit by the following equation (6).

g=(A+C)1A---(6)]]>本领域的技术人员应该认识到,分别在图25和26中所示的两种算法实际上纯粹是用作说明,各种各样不同的算法可被适当地采用并可在DSP或包括图9,10和24中所示的峰值降低算法电路152的其他电路中实施。 g = (A + C) 1A --- (6)]]> skilled in the art will recognize that, and two algorithms are shown in FIGS. 25 26 in fact purely as described, each of the various different algorithms like may be suitably employed and may reduce other circuits in the arithmetic circuit 152 in the embodiment shown in FIGS. 9, 10 and peaks 24 or the DSP included.

参考图27和28,依据本发明的峰值降低处理的例子被示于两个复数向量图中。 Referring to Figure 27 and 28, the peak reduction processing according to the present invention, an example is shown in FIG two complex vector. 图27和28适用于只有一个中心符号被调节的场合。 Figures 27 and 28 only applies to a case where the center symbol is adjusted. 然而,所示的中心符号可以表示两个中心符号的组合。 However, as shown in the center of the symbol may represent a combination of symbols of the two centers. 如图27和28中所示,输入符号被提供滤波器预测器处理。 27 and FIG, 28 are provided input symbol filter prediction processing. 因为滤波器输出在时间上与许多符号有关,不仅仅是被调节的中心符号,输出可被表示为中心符号和表示时间上相邻的符号的影响的圆。 Because the filter output in time with the number of symbols related to not only the center of the symbol is adjusted, the output can be represented as symbols and shows the influence of the center of the adjacent circle symbol time.

虽然起初在两个例子中中心符号超过限值,在图27的例子中,表示已滤波输出的圆的一部分实际上位于限值线内。 Although initially the center symbol exceeds the limit In both cases, in the example of FIG. 27 shows a portion of a circle limit lines actually in the filtered output. 因此这些输出值并不经受峰值降低处理。 Thus these output values ​​are not subjected to peak reducing process. 在图27和28的复数图的粗线部分所示的输出的其余部分按已滤波的符号超过限值线的程度而变化的量经受峰值降低处理。 Filtered symbols by an amount greater degree than limit lines varies in the rest of the output of the portion indicated by thick lines in FIG. 27 and FIG. 28 the complex is subjected to peak reducing process. 然后符号被调节使预测的滤波器输出被拉回到限值线,如图27和28中限值线粗线部分所示。 Symbol is then adjusted so that the predicted value of the filter output line is pulled back, the limit lines 27 and 28, the bold line shown in FIG. 因此,将认识到,不需要峰值限处理的符号仍然未被触动过,从而减少通过这样的降低引入的任何畸变,而且需要峰值降低处理的符号被提供为将已滤波输出置于限值内必要的最低限度的峰值降低量。 Thus it will be appreciated that the symbol does not need to limit the peak remains untouched through the process, thereby reducing any distortion introduced by such a reduction, but also symbols peak reducing process is provided as the filtered output disposed within limits necessary peak minimum to reduce the amount. 否则,起初看来不需要峰值降低处理的符号将被峰值调节,如果滤波器预测表明该滤波器将超过限值。 Otherwise, not seem necessary to first reduce peak symbol processing will be adjusted peak, it indicates that the prediction filter if the filter exceeds the limit. 因此,将认识到,本发明提供高度有效的信号峰值降低而同时使引入符号序列的畸变为最小。 Thus it will be appreciated that the present invention provides a highly effective peak reducing signal distortions introduced while leaving a minimum symbol sequence.

图29到35示出利用等式(1),(2),(3),(5)和(6)中给出的算法的不同的峰值降低单元实施方案。 FIGS 29-35 shown by the equation (1), different peak algorithm (2), (3), (5) and (6) reduction unit given embodiment. 图29,30,32和34表示图9中所示的峰值降低处理。 Peaks 29,30,32 and 34 shown in FIG. 9 represents a reduction process. 图31,33和35表示图10中所示的降低处理。 And reduction processing shown in FIG. 31, 33, 35 represented in FIG. 10. 在图23和24中所示的并行处理的一个例子将在图36A和36B中给出。 It will be given in FIGS. 36A and 36B in FIG. 23 and a parallel processing example shown in FIG 24.

参考图29,利用等式(1)的一种峰值降低单元实施方案被以方框简图形式示出。 Referring to FIG 29 A peak using equation (1) reduction unit embodiments are shown in block schematic form. 如图29中所示,采用一种多抽头的滤波器200可方便地将延时电路166和图9中所示的滤波器预测器电路146组合。 As shown in Figure 29, using the prediction filter 200 easily combination circuit 146 delay circuit 166 shown in FIG. 9, and a multi-tap filter may. 滤波器200包括多个单独的存储器寄存器202,其中5个被示于图29的特定的实施方案中。 Filter 200 comprises a plurality of separate memory register 202, which is five in the particular embodiment shown in FIG. 29. 然而应该认识到,附加的或较少的延时存储器寄存器可被提供,一般N个这样的存储器寄存器202将被提供组成N个元件的移位寄存器,通过将一个存储器寄存器输出抽头,例如,在中心存储器寄存器处,可以提供一个延时的符号序列,使得峰值校正可以在组合器168上,在被正确定时的逐个符号的基础上完成。 However, it should be appreciated that additional or fewer delay memory register may be provided, typically N such memory register 202 is composed of N elements provide a shift register, a memory register by the output taps, e.g., in at the center of the memory register, a delay may be provided symbol sequence such that the peak of the correction may be accomplished on the basis of the correct timing on a symbol by symbol combiner 168. 这样一种从N元件存储器寄存器的延时输出用线205示出,它相应于图9中所示的延时电路166的输出。 Such a cable 205 is shown with N elements output from the delay memory registers, which corresponds to the delay circuit 166 shown in FIG. 9 output. 沿着线204所提供的输出是从与中心滤波器系数相乘以后的中心延时线的一个抽头得到的。 Along the output line 204 is supplied from the delay after multiplication with the center line of a center of the filter tap coefficients obtained. 这条线表示中心的已滤波符号输出(图9中的线145和图25中的向量A),被提供给如图29中所示的峰值降低算法处理电路152。 This line represents the center of the filtered output symbols (line 145 in FIG. 9 and FIG. 25 vector A), is provided to a peak as shown in FIG. 29 reduction algorithm processing circuit 152. 来自存储器寄存器202中每一个的延时输出被提供给相应的乘法器206,该乘法器也接收相应的滤波器系数作为它的一个输入。 From memory register 202 of each of the delayed output is provided to a respective multiplier 206, the multiplier also receives a respective filter coefficient as one of its inputs. 因此每个滤波器系数起着一个增益gN的作用,N=1到5,与来自相应的延时级202的符号输出相乘。 Thus each filter coefficient plays a role gN gain, N = 1 to 5 multiplied by the sign of the output from the respective delay stage 202. 滤波器系数gN可以对应于图8或图12-21中所示的符号间隔系数中任何一个,取决于在图9或11的处理中哪一级被示出。 GN filter coefficients may correspond to a symbol interval coefficients shown in FIG. 8 or any one of 12-21, depending on which process is shown in FIG. 9 or 11 a. 当然,可以使用各种各样不同的滤波器响应函数,这取决于被预测的特定的滤波器,其系数将相应地改变。 Of course, different filter response functions of a wide variety, depending on the particular filter is predicted coefficients will change accordingly. 也将认识到,利用实际上纯粹是用作说明的五个系数的例子,可以从脉冲响应函数的任何符号间隔上取得附加的系数,对于特定的实施方案可以采用多于或少于五个系数,这取决于被模型化的特定的脉冲响应函数以及采用的处理系统的速度和所希望的精确度。 It will also be appreciated that, in fact, is purely an example of using five coefficients as described, to obtain an additional coefficient may be any symbol interval from the pulse response function, for the specific embodiments more or less than five coefficients may be employed , rate response function and a processing system used and the desired degree of accuracy depending on the particular pulses being modeled.

再参考图29,来自乘法器电路206的输出被提供给加法电路208,将多个输出相加并将它们沿着线210提供。 Referring again to FIG. 29, the output from the multiplier circuit 206 is supplied to an adder circuit 208, a plurality of summed output along line 210 and provides them. 沿着线210的输出对应于在特定的符号间隔上取得的符号的滤波器系数加权和,因而对应于在此间隔上滤波器对符号的影响的一个模型。 Output along line 210 corresponds to a symbol weighting filter coefficients acquired in a specific symbol interval and, thus corresponds to a model of the influence on the filter symbol in this interval. 这种预测的滤波器符号输出被沿着线210提供,作为对峰值降低算法电路152的一个输入。 This symbol prediction filter output is provided along line 210 as an input arithmetic circuit 152 to reduce the peak. 如以上指出的那样,峰值降低算法电路152也沿着线204接收延时的中心符号滤波器输出。 As noted above, the peak reduction algorithm circuit 152 also receives the delayed filter output symbol center along line 204. 这种延时的中心符号滤波器输出流被沿着线212提供给减法电路216,和沿着线214提供给算法处理器218。 This delay central filter output symbol stream is provided along line 212 to a subtracting circuit 216, and supplied to the arithmetic processor 218 along line 214. 这样,减法电路216接收沿着线210提供的已滤波输出作为对它的一个输入,接收沿着线212的延时的中心符号滤波器输出作为对它的第二个输入。 Thus, subtraction circuit 216 receives the filtered output line 210 provided as one input to it, the filter output symbol reception center 212 along the delay line as its second input along. 减法电路216取这两个输出流的差,提供逐个符号的差值D(利用图25的术语,D=BA),沿着线220给算法处理器218。 Two output streams subtraction circuit 216 takes, symbol by symbol to provide a difference D (using the terminology of FIG. 25, D = BA), along a line 220 to processor 218 algorithm. 算法处理器218接收沿着线220和214的两个输入符号流,也接收限值L作为输入。 Algorithm processor 218 receives two input symbol streams along lines 220 and 214, which also receives as an input value L. 算法处理器218利用等式(1)计算增益g,以便将已滤波输出降低到位于限值L内的值。 Algorithm processor 218 using Equation (1) calculates the gain g, to the filtered output is reduced to a value located within a limit L.

在更一般的情况下,减法电路216将与算法处理器218组合,建立一种更通用的算法处理器。 In the more general case, subtracting circuit 216 and the arithmetic processor 218 in combination establish a more general algorithm processor. 通过对图29的这种少量修改,基于来自线204,210的输入和限值L,可以使用各种各样不同的算法。 Such minor modifications by FIG. 29, based on input from the line 204, 210 and limit L, a variety of different algorithms may be used. 在这种更通用的情况下,可以使用在等式(2)中给出的近似算法或在等式(6)中给出的,基于图26的算法。 The algorithm in this more general case, may be used in Equation (2) given or approximation algorithm given in Equation (6) 26 based on FIG.

在图29中所示的特定情况下,或所描述的更通用的情况下,从算法处理器218算得的增益值g被沿着线232输出到选择开关230。 In the particular case shown in FIG. 29, or the more general case described, the algorithm processor 218 from the calculated gain value g is output along line 232 to the selector switch 230.

再参考图29,沿着线210提供的已滤波输出流也被提供给量值检测电路222。 Referring again to FIG. 29, the filtered output stream along line 210 is also supplied to the detection circuit 222 magnitude. 最值检测电路222确定已滤波输出的量值,也就是,包含该输出的复数向量量的绝对值,该量值被沿着线224作为输出提供。 The most value detecting circuit 222 determines the magnitude of the filtered output, i.e., the absolute value of complex vector comprising an amount of the output, the magnitude of which is provided as an output along line 224. 这个量值被提供给比较器226,将已滤波符号的量值与限值L作比较。 This value is supplied to the comparator 226, the magnitude of the filtered symbols with the limit value L for comparison. 如果已滤波符号的量值超过限值L,从比较器226的输出取第一值(例如“1”)。 If the magnitude of the filtered symbols exceeds a limit value L, the output of the comparator 226 takes a first value (e.g. "1"). 如果已滤波符号的量值小于限值L,则从比较器226的输出是一个第二值(例如,“0”)。 If the magnitude of the filtered value symbol is less than L, from the output of the comparator 226 is a second value (e.g., "0"). 然后,这个值,也就是“0”或“1”被作为输出沿着线228提供给选择开关230。 Then, this value is "0" or "1" is used as an output along line 228 to the selector switch 230. 如果沿着线228到选择开关230的输入是0,则从选择开关230的输出是一个单位信号,它对沿着线205提供给组合器168(在图29的特定的实施方案中作为一个乘法器示出)的符号流没有影响。 If the select input along line 228 to switch 230 is 0, selects the output from the switch 230 is a unit signal, it is supplied to line 205 along a combiner 168 (as a multiplication in the particular embodiment of FIG. 29 symbol streams is shown) is not affected. 如果沿着线228提供给选择开关230的信号是1,对应于已滤波符号值超过限值L的情况,则从算法处理器218提供的算得的增益值g被输出到乘法器168。 If provided along line 228 to the selection switch signal 230 is 1, the symbol corresponding to the filtered value exceeds the limit value L, the gain calculated from the arithmetic processor 218 provides the value of g is outputted to the multiplier 168. 只有必要在逐个符号的基础上用这种方法,沿着线205提供的符号流将是被通过算法算得的适当值降低的增益,适当的峰值被调节的符号将在线154上输出。 Only in this way the necessary symbol by symbol based on the symbol stream provided along line 205 will be reduced by an appropriate value calculated gain algorithm, the appropriate symbol peak regulated output on line 154.

将认识到,在图29中所示的各种电路部件可以只用硬件,只用软件,也就是作为一个被适当编程的DSP或其他的处理器来实现,或者可作为硬件和软件的一种组合来实现。 A appreciated in the various circuit components shown in FIG 29 may be implemented in hardware only, software only, which is to be implemented as a suitably programmed DSP or other processor, or may be implemented as hardware and software combination to achieve. 例如,滤波器200被作为硬件实现,而算法处理器218被作为适当编码的DSP处理器来实现可能是有利的。 For example, filter 200 is implemented as hardware, but the algorithm processor 218 is encoded as an appropriate DSP processors may be quite advantageous. 另一种方案是,算法处理器218的电路可被作为一个可编程的门阵列电路来实现。 Alternatively, the arithmetic processor circuit 218 may be used as a programmable gate array circuits. 滤波器200和/或差电路216及量值检测器222也可被作为一个门阵列电路来实现,并与基于处理器的电路218组合。 Filter 200 and / or the magnitude of the difference circuit 216 and the detector 222 may also be used as a gate array circuit is achieved, and 218 in combination with a processor-based circuit. 因此,将认识到,在图29中所示的电路的各种各样不同的实施方案的组合是可能的。 Thus it will be appreciated that various combinations in various embodiments of the circuit shown in FIG 29 is possible.

参考图30,示出一种在图29中所示的峰值降低单元的替代的实施方案。 Referring to FIG 30, there is shown an alternative embodiment for reducing peak unit 29 shown in FIG. 在图30的实施方案中,在一种简化的峰值降低算法电路152中实施峰值降低算法,利用一个对于峰值降低的近似等式以便应用到输入符号上。 In the embodiment of Figure 30, the peak reduction algorithm a simplified embodiment circuit 152 peak reduction algorithm, to use a reduced peak approximation equation for application to the input symbol. 特别是,在图30中所示的特定的实施方案可以实施以上所描述的等式(2),提供对于施加到符号向量以便将它带到限值L的增益g的一种近似计算。 In particular, in the particular embodiment shown in FIG. 30 may be implemented in equation (2) described above, to provide an approximate calculation is applied to the symbol vector in order to limit it to one kind of L gain g.

正如可以看到的那样,该等式包含相当简单的计算。 As can be seen, this equation contains fairly simple calculations. 涉及限值L,中心符号滤波器输出A的量值和预测的滤波输出B的量值。 It relates to the magnitude value L, the filter output A of the center symbol and the predicted magnitude of the filtered output B. 将这个等式与图30中所示的峰值降低算法电路152作比较表明电路222,250,252,254和256以直截了当的方式实现等式(2)。 The peak with the equation shown in FIG. 30 for reducing algorithm circuit 152 comparator circuits 222,250,252,254, and 256 show a straightforward manner in equation (2). 更具体而言,线204连同量值检测电路250提供中心符号滤波器输出A,从滤波器200到量值检测电路222提供预测的滤波输出,确定预测的滤波输出B的量值。 More specifically, the line 204 together with the magnitude of the center of the symbol detection circuit 250 provides the filter output A, the detection value from the filter circuit 200 to provide the output prediction filter 222, determines the magnitude of the filtered output B predicted. 这两个量值被提供给减法电路252,如行符号量值的减法以提供值|A|-|B|。 These two values ​​are supplied to the subtraction circuit 252, such as the line to provide a signed magnitude subtraction value | A | - | B |. 然后加器电路254(如果电路252交换它的输入,它可以是一种减法电路)将从电路252提供的这个值加到限值L上。 Then adder circuit 254 (if its input switching circuit 252, it may be a subtracting circuit) supplied from the circuit 252 is added to the upper limit value L. 从电路250提供的中心滤波输出被提供给除法电路256,它也接收电路254的输出,以提供由以上等式(2)给出的近似峰值降低增益g。 Center filtered output is supplied from the circuit 250 provided to the divider circuit 256, which also receives the output of circuit 254, to be provided by the above equation (2) is approximately given by the peak reduction of the gain g.

因此,将认识到在图30中示出的用于峰值降低算法电路152的电路实施方案提供一种相当简单的可容易用硬件提供的实施方案。 Thus, a peak will be appreciated that illustrated in FIG. 30 reduction circuit embodiments arithmetic circuit 152 to provide a relatively simple embodiment of a hardware can be readily supplied. 这种硬件可采用可编程门阵列的形式或其他的硬件实施方案,或者用在DSP或其他处理器中实施的相当简单的程序来实现。 This may take the form of hardware, or other programmable gate array, hardware implementation, or in a DSP or other processor in the embodiment of relatively simple program. 在图30中所示的实施方案的这种相当的简单的特点可以具有成本和/或速度的优点,在特定的应用中可被优先选用。 May have the advantage of cost and / or speed in such a relatively simple features of an embodiment shown in FIG. 30, it may be preferred in certain applications. 在图30中所示的实施方案中电路的其余部分可与图29中所示的精确地相同,并可用与以上所描述的准确地相同的方式操作。 The rest of the circuit may be exactly the same in the embodiment shown in FIG. 30 and shown in FIG. 29, and may be exactly the same manner as the above described operation. 因此,为了描述图30的实施方案,将不重复这种公共电路的操作。 Thus, the purpose of describing the embodiment of Figure 30, will not be repeated operation of such a common circuit.

参考图31,示出一种峰值降低单元的替代的实施方案。 Referring to FIG 31, shows an alternative embodiment of the peak reduction unit. 图31的实施方案采用从峰值降低算法电路152的输出到滤波器预测器的反馈,以便增加滤波器预测操作的准确度。 Embodiment of Figure 31 uses the peak reducing arithmetic circuit 152 is output to a feedback filter prediction, in order to increase the accuracy of the prediction filter operation. 因此图31表示图10的一种实施方案。 Thus FIG. 31 shows an embodiment 10 of FIG. 更具体而言,如在以前已描述的实施方案中那样,滤波器预测器和延时电路被优先地组合在一个有限元滤波器200中,将接收输入符号的多个存储器寄存器合并,作为一个N元的移位寄存器操作。 More specifically, as above, the prediction filters and delay circuits are preferentially combined in a finite element filter 200, the received input symbols of the plurality of memory registers incorporated in embodiments described previously, as a membered N shift register operations. 如在图29的实施方案的情况中那样,存储器寄存器的输出被提供给乘法器电路206,它也接收滤波器系数作为它的输入。 As above, the output of the memory register is supplied to the multiplier circuit 206 in the case of the embodiment of Figure 29, which also receives the filter coefficients as its input. 也如图29的实施方案的情况中那样,乘法器输出被提供给加法器电路208,以提供已滤波的输出符号。 FIG situation embodiment as in embodiment 29, the multiplier output is supplied to adder circuit 208 to provide the filtered output symbols. 然而,图31中所示的峰值降低算法电路152也与图29中的对应,如在关于以前的实施方案以上已讨论过的那样,它可被修改以实施各种各样不同的算法。 However, the peak 31 shown in FIG reduction algorithm circuit 152 in FIG. 29 corresponds to, as described in relation to the previous embodiments have been discussed above, it may be modified to implement a variety of different algorithms.

与图29的实施方案相对照,在图31中峰值降低算法电路152的输出被回送到滤波器200。 In contrast to the embodiment of FIG. 29, the output of the peak reduction algorithm circuit 152 in FIG. 31 is returned to the filter 200. 特别是,峰值降低算法电路152的输出被沿着线162回送到乘法器,提供由电路152算得的峰值降低增益到滤波器200的N级存储器寄存器中中心延时级的输出。 In particular, the output of the peak reduction algorithm circuit 152 is supplied to the multiplier 162 along a return line, provided by the circuit 152 calculated peak reducing an N-stage gain to a filter memory register 200 in the center of the delay stage output. 结果,提供给存储器寄存器下游级的乘法器168的输出包括已经增益降低的符号值。 As a result, to the memory output register 168 downstream stage includes a multiplier symbol values ​​has a gain reduction. 这将更准确地反映滤波器126进行的实际处理(参考图6),因为增益降低的符号将被滤波器200包括在已滤波符号的计算中。 This more accurately reflect the actual processing filter 126 (see FIG. 6), since the gain will be reduced symbol filter 200 included in the calculation of the filtered symbols. 因此,在图31中所示的实施方案在许多情况下提供更精确的滤波器预测,在许多应用中可被优先选用。 Thus, a more accurate prediction filter in many cases, in the embodiment shown in FIG. 31, in many applications may be preferred.

对图30中所示的实施方案可以进行一种类似的反馈延伸,其中等式(2)被专门用一个独特的方框图限定。 Of the embodiment shown in FIG. 30 can be extended for feeding back a similar, wherein the equation (2) is specifically defined by a unique block diagram. 这种延伸应该容易被本领域的技术人员所理解。 This extension should be readily understood by those skilled in the art.

如上所述,反馈修改对于所有随后的峰值调节计算提供已调节的符号。 As described above, the calculations are modified to provide feedback regulated symbols for all subsequent peak shaving. 然而在修改前,预先调节的符号被用于在调节时间前计算峰值调节量。 However, prior to modification, preconditioned symbols are used to calculate the amount of adjustment before adjusting the peak time. 这意味着当已调节的符号流通过滤波器(图6中的20)时,已调节的符号将参与建立已调节的符号前后的峰值。 This means that when the conditioned symbol streams through (20 in FIG. 6) a filter, adjusted to participate in the establishment of the symbol before and after the peak regulated symbols. 这样,新的峰值可在已调节的符号前建立。 Thus, a new peak can be established prior to the adjusted symbol. 这些新的峰值是峰值调节过程的因果关系或非期望性质的结果。 The new peak is a peak adjustment process causation or undesirable properties result. 简单地重复图7,11,22和23中所示的峰值降低过程可容易地消除这些新的峰值。 Simply repeat the peak shown in FIG. 23 7,11,22 and reducing process may be readily eliminate these new peaks.

以前的实施方案描述了基于增益校正的符号调节,这些增益校正可从等式(1),(2)和(6)算出。 The previous embodiments describe adjustment based on the gain correction code, which may be a gain correction from the equation (1), (2) is calculated, and (6). 符号调节也可基于可从等式(3)和(5)算出的附加向量。 Adjusting the symbol may be based on the equations (3) and (5) calculating an additional vector. 参考图32,示出一种基于附加向量调节符号的实施方案。 Referring to FIG 32, illustrates a vector adjustment based on the additional symbols embodiment. 在图32的实施方案中,图9的延时电路被作为一系列存储器寄存器的部件来实现,以与以前描述过的实施方案类似的方式组成滤波器预测器电路146的部件。 In the embodiment of FIG. 32, FIG delay circuit 9 is implemented as a series of memory register means, similar to the previously described embodiments of embodiment the component parts of the filter prediction circuit 146. 更具体而言,滤波器200包括多个存储器寄存器202,可以作为N元的移位寄存器工作,专门的说明图形是一个7元移位寄存器。 More specifically, the filter 200 comprises a plurality of memory registers 202, as N-membered shift register operation, special instructions pattern shift register is a 7-membered. 存储器寄存器中心级的输出被抽头作一个输出,沿着线205提供延时的符号到组合器168。 A memory output register tapped central stage as an output along line 205 provides a delay to the symbol combiner 168. 组合器168作为一个加法器电路示出。 A combiner adder circuit 168 as shown. 存储器寄存器级的输出被提供给乘法器电路206,它也接收在与该级对应的特定的符号定时上的滤波器系数值作为它的输入。 Output stage memory register is supplied to a multiplier circuit 206, which also receives the filter coefficient values ​​on a particular symbol timing corresponding to the stage as its input. 乘法器206的输出被提供给加法电路,沿着线210输出预测的滤波器输出值,与以前描述过的实施方案类似。 The output of multiplier 206 is supplied to an adder circuit, the output line 210 along the predicted filter output values, similar to the previously described embodiments.

预测的滤波器输出沿着线210提供给峰值降低算法电路218,在预测滤波器输出上实施特定的峰值降低算法,并提供降低值,如果有的话,到组合器168。 Predicted filter output along line 210 to provide the peak reduction algorithm circuit 218, a specific embodiment of the peak reduction algorithms, and provides a predicted reduction value at the filter output, if any, to the combiner 168. 在图32的实施方案中,可以实施一种特别简单的算法,它不需要沿着线204来自中心滤波符号输出的输入,而且简单地在沿着线210提供的预测的滤波器输出上操作。 In the embodiment of Figure 32, may be implemented in a particularly simple algorithm, it does not require an input along line 204 from the center of the filter output symbols, and simple operation on the predictive filter output along a line 210 is provided. 对于所示的特定的实施方案该算法也采用所希望的限值L和滤波器预测器的中心抽头的增益g4作为输入。 For the particular embodiment shown in this embodiment also uses algorithm desired limit value L and the center tap of the filter prediction gain g4 as input. 这样一种算法可对应于以上关于图26所描述的等式(5)。 Such an algorithm corresponding to the above equation may be described with respect to FIG. 26 (5). 然而其他的算法也可被应用在电路338中。 However, other algorithms may also be applied in the circuit 338. 这些其他的算法可能需要沿着线204提供的来自符号流的输入,对于图32中电路152的这样一种输入的可能性在这些替代的实施方案的情况下得到理解。 Such other algorithms may be needed to provide an input along line 204 from the symbol stream, for the possibility of circuit 152 in FIG. 32 the input of such a case are understood in these alternative embodiments. 恰好在等式(3)中给出这样一种算法,其中等式(3)中的值g是在等式(1)中算得的。 Such an algorithm is given exactly in equation (3), wherein the value of G in equation (3) is calculated in Equation (1).

在所示的电路152的实施方案中,沿着线210提供的预测的滤波器输出的量值是由量值检测电路222确定的。 In the embodiment shown in circuit 152, the line 210 provided along the predicted magnitude of the filter output is determined by the magnitude of the detection circuit 222. 这个量值被提供给一个比较器226,将预测的滤波器输出值的量值与阈值L作比较。 This value is provided to a comparator 226, a threshold magnitude of the predicted value L filter output values ​​for comparison. 预测的滤波器输出也被提供给算法处理器电路218,可以是一个被适当编程的DSP或其他的处理器,实施用于所示的特定实施方案的等式或对预测的符号操作的其他适当的算法。 Other suitable prediction filter output is also supplied to the arithmetic processor circuit 218 may be a DSP or other processor appropriately programmed to implement the equation for the specific embodiments shown or predicted operation symbol algorithm. 另一种方案是,算法处理器218可以用门阵列结构或其他的硬件实施方案实现。 Alternatively, the algorithm processor 218 may be implemented using gate arrays or other hardware structure embodiments. 从算法处理器218的输出被提供给选择器开关230,该开关也接收比较器226的输出。 Output from the arithmetic processor 218 is supplied to the selector switch 230, the switch 226 also receives the output of the comparator. 如果预测的符号值大于阈值L,则来自比较器的输出使开关230能够输出峰值校正值到组合器168。 If the symbol value is greater than the predicted threshold value L, then the output from the comparator 230 can output the switch peak correction value to the combiner 168. 否则,如果预测的符号值小于或等于限值L,则比较器对选择器开关230的输出选择零输出到组合器168,对应于对符号流无峰值调节。 Otherwise, if the predicted value is less than or equal symbol value L, the comparator compares the output of the selector switch 230 selects the output of the combiner 168 to zero, corresponding to no peak symbol stream regulation.

参考图33,示出图32的一种替代的实施方案,如图10中所给出的那样,采用峰值调节对滤波器预测器的反馈。 Referring to FIG 33, there is shown an alternative embodiment of FIG. 32, as given in FIG. 10, using feedback to adjust the peak filter predictor. 更具体而言,在图33的实施方案中,滤波器200沿着线262从峰值降低算法电路218接收反馈的峰值调节值。 More specifically, in the embodiment of Figure 33, the filter 200 to reduce the peak value of the adjustment circuit 218 receives the feedback algorithm from the peak along the line 262. 在图33中示出的滤波器200,如在图32的情况中那样,可以对应于一个N级的实施方案,因此不需要详细描述。 In filter 33 shown in FIG. 200, as above, it may correspond to an N-stage embodiment in the case of FIG. 32, and therefore need not be described in detail. 如图所示,沿着线262提供的峰值调节量可被提供给组合器168,在图33中作为一个加法器示出,被配置在存储器寄存器的中心级之后,组成滤波器200的部件。 As shown, the peak amount of adjustment is provided along line 262 may be provided to a combiner 168, an adder as illustrated in FIG. 33, the center stage is arranged after the memory register of the component parts of the filter 200. 因此,对符号的峰值调节被包括在滤波器的相继的级中,对滤波器200的预测能力提供附加的改进。 Therefore, adjustment of the peak symbol in successive stages are included in the filter, the prediction filter 200 is the ability to provide additional improvements. 将认识到,利用存储器寄存器中不同的反馈位置可以方便地实现不同的滤波器实施方案。 It will be appreciated, can easily achieve different filter embodiments utilize different memory register feedback position. 因此,在图33中所示的特定的实施方案纯粹是用作说明,不应该在实际上当作限制。 Thus, in the particular embodiment shown in FIG. 33 is purely for illustrative and should not be taken as limiting in practice.

参考图34,示出一种图29的峰值降低电路单元的替代的实施方案方框简图。 Referring to FIG 34, shows a reduction peak 29 of FIG alternative embodiment of a block circuit diagram of the unit. 为了理解这种替代的实施方案的合理性,想起由两个相邻的类似幅度的符号占优势的符号间峰值。 To understand the reasonableness of this alternative embodiment, remembered between two adjacent symbols by the amplitude of a similar predominant peak symbol. 这是在以上参考图4中描述过的。 This is described above in reference to FIG. 如果只完成符号上和符号间的调节,在符号上处理以后将有大量的类似幅度的符号。 If the adjustment is completed and only the symbols between symbols, the symbols after the processing of a large number of symbols will have similar amplitudes. 除了延时更长-单元和滤波器预测器存储寄存器中居中的两个相邻符号被调节外,图34基本上与图29对应。 In addition to the delay time longer - a filter unit and a storage register predictor centered two adjacent symbols is adjusted, the FIG. 34 substantially corresponds to FIG. 29. 两个中心系数具有相同的值。 Two centers coefficients have the same value.

在图34中所示电路的实施方案基本上与图29的对应,因此同样的数字被应用于同样的部件,因而每个部件的详细描述将不再重复。 In the embodiment shown in FIG circuit 34 substantially corresponds to FIG. 29, so the same numerals are applied to like parts and therefore detailed description of each part will not be repeated. 如以上关于图29所讨论的那样,该电路利用一个多抽头滤波器200,在符号间的间隔上进行预测滤波器脉冲响应对输入符号的影响,并以此为基础提供峰值降低处理。 As discussed above with respect to FIG. 29, the circuit using a multi-tap filter 200, the filter impulse in the prediction interval between symbols in response effect on the input symbol, and thus provide a basis for the peak reducing process. 滤波器200除了以下的若干修改外,一般来说与图29中的滤波器200相对应。 In addition to several modifications filter 200 or less, 200 generally correspond to the filter 29 in FIG. 对乘法器206的输入gN,N=1到6,被选自符号间间隔的滤波器系数(t=0.5),如在此示出的特定的脉冲响应函数在图8中用十字符号表示。 GN 206 to the multiplier input, N = 1 through 6 filter coefficients (t = 0.5) is the spacing between selected symbols, as in this particular impulse response function shown by cross symbols shown in FIG. 8. 如关于图29所讨论的那样,在图8中特定的滤波器系数实际上纯粹是用作说明,所以输入gN并不限于在此所示的特定的符号间的值。 As discussed with respect to FIG. 29, as, in fact, is purely for illustrative specific filter coefficient in FIG. 8, the input gN is not limited to a specific value between the symbol shown herein. 为了对主要影响符号间峰值的两个符号提供符号校正,来自两个中心滤波器抽头202-3和202-4的滤波器输出被提供给一个加法电路240以便建立线204。 For two main influence between symbols to provide a peak symbol sign correction, the filter tap 202-3 and 202-4 output from the two centers of filters 240 is supplied to line 204 in order to establish an adder circuit. 线204等效于在图25和26中所示的单一的符号滤波器输出A。 Line 204 is equivalent to the single filter output symbols shown in FIGS. 25 and 26 A. 然后通过使用单个单元的存储器寄存器244和乘法器242,利用施加到两个中心抽头的增益校正如以前那样处理峰值调节。 242 Then, by using the memory by applying the multiplier register 244 and to a single unit of two center tap gain correction processing such as peak before adjustment. 因此,将认识到在线154上的输出符号流提供在符号间的间隔上,逐个符号进行调节的一个适当的峰值已调节的符号流,而不是与关于图29所讨论的方法相同。 Therefore, the appreciated output symbol stream on line 154 is provided on the space between the symbols, the symbol stream for a suitable adjustment of the adjusted peak symbol by symbol, rather than with respect to the method discussed in FIG. 29 is the same. 虽然处理最好在图34中与图29中相同,在某些情况下,可能希望在图34中实施与图29不同的算法或者另一种方法是将在符号上的间隔上处理修改为在符号间的间隔上处理。 Although the process is preferably the same as FIG. 29 in FIG. 34, in some cases, it may be desirable in the embodiment of FIG. 34 and FIG. 29 different algorithm or another method is to modify the process on the sign of the interval processing interval between symbols.

参考图35,示出一种图34的替代的实施方案,也按关于图31所描述的方式应用反馈。 Referring to FIG 35, shows an alternative embodiment of a FIG. 34, but also the manner described with respect to FIG. 31 feedback applications. 更具体而言,如图35中所示,滤波器200包括来自峰值降低算法电路152的反馈。 More specifically, as shown in FIG. 35, from a feedback filter 200 includes a peak reducing algorithm circuit 152. 这个反馈回路提供由峰值降低算法电路152算得的用于峰值降低的增益g,沿着线262到组成滤波器200的部件的N级存储器寄存器。 This feedback loop provides for reducing the peak by the peak gain g calculated reduction algorithm circuit 152, along line 262 to the N-stage memory register composed of a filter member 200. 在所示的特定的实施方案中,这个反馈的增益被提供给第四个存储器寄存器的相对侧上的乘法器168-1和乘法器168-2。 In the particular embodiment illustrated, the gain of the feedback is provided to the multiplier on the opposite side of the fourth memory register 168-1 and the multiplier 168-2. 这样把符号增益引入最影响符号间峰值的两个符号。 So the introduction of two symbols symbol gain influence among the most symbolic peaks. 这种实施方案是基于图8的特定的符号间脉冲响应函数和滤波器延时级的特定选择。 This embodiment is based on a filter impulse response function and a particular delay stage selected among specific symbols as in Fig. 8. 因此,将认识到利用不同的滤波器实施方案或对于不同的脉冲响应函数可以提供不同的方式引入反馈增益到存储器寄存器级。 Therefore, the appreciated that using different filters for different embodiments or impulse response function can be introduced in different ways to provide a feedback gain to the memory register stages. 如图所示从存储器寄存器的最后的级,沿着线154提供级的降低过程的输出。 As shown, provides an output along line 154 to reduce the process stages from the last stage memory register. 正如在图31的实施方案的情况中那样,符号降低的反馈引入存储器寄存器可以改进滤波器200的预测能力和在某些情况下可被优先选用。 As, in the case of the symbol decrease the embodiment of Figure 31 incorporated in the feedback memory register can improve the predictive power of the filter 200 and in some cases may be preferred.

图34和35描述对图29和31的替代的实施方案。 34 and FIG. 35 depicts an alternative embodiment of the embodiment of FIGS. 29 and 31. 可以对图30和32完成类似的替代的实施方案。 Can perform similar alternative embodiment of FIGS. 30 and 32. 这些替代的实施方案从以上的解释对于本领域的技术人员应该是明白的。 These alternative embodiment from the above explanation to those skilled in the art should understand.

参考图36A和36B,示出图23的并行多级峰值降低单元的一种详细的实施方案。 With reference to FIGS. 36A and 36B, a diagram illustrating one kind of parallel multi-stage peak reduction unit 23 of the detailed embodiments. 在图36A中所示的特定的实施方案包括10个并行的峰值降低处理级。 Particular embodiment shown in FIG. 36A comprises 10 parallel peak reducing process stages. 然而,将认识到,这纯粹是用作说明,取决于特定的应用情况可以采用较多或较少数目的级。 However, it will be appreciated that this is purely for illustrative, depending on the particular application of more or less number of stages can be employed. 如以前所提到的那样,由峰值降低的因果关系性质引起的峰值可被重复的级除去。 As previously mentioned, a peak by the peak reduction caused by the nature of causality can be removed by repeated stages. 在并行处理中,通过对在图12到21中实行的脉冲响应函数的周期性采样的连续进行实现这种重复,在这些图中表示周期性采样在-0.5到0.4上取得。 In parallel processing, by periodically sampling the continuous implementation of FIG. 12 to 21 in response to the pulse repetition achieve this function, the periodic represents samples taken at the 0.4 to -0.5 in these figures. 在t=0.5上取得的样本,将与在t=-0.4上取得的样本相同。 Samples taken on t = 0.5, will be the same on at t = -0.4 acquired sample. 这种采样图形的复制将在间隔t=-0.5到0.5上连续进行。 This sampling pattern replication in the continuous interval t = -0.5 to 0.5. 然后这些重复的采样可被加到图36A的底部作为并行线10,11等。 Duplicate samples are then added to the bottom of Fig. 36A may be used as parallel lines 10, 11 and the like.

每级包括一个延时电路370,如图所示,可作为一系列存储器寄存器来实施,其中的每一个将符号流延时与符号之间的时间对应的时间。 Each stage includes a delay circuit 370, as shown, may be implemented as a series of storage registers, the time between each of the delayed symbol stream where the symbol corresponding to the time. 如以前那样,抽头被取自这些存储器寄存器,以便计算存储在存储器寄存器中的符号的滤波器系数加权和。 As above, the tap is taken from the previous memory register, in order to calculate the filter coefficients stored in the weighting of the symbol and the memory register. 如图36A中所示,一个附加的延时的存储器寄存器必须被加到相继的并行级。 As shown in FIG. 36A, an additional delay must be added to the successive memory register stages in parallel. 这些延时寄存器供来自并行级的反馈符号调节的正确定时用。 These delays from the parallel registers for adjusting the level of feedback symbols with correct timing.

这些来自每个并行延时级Ti,n,i=0到10,n=0到7的单独的抽头被提供给一个滤波器预测器200,预测器提供输出到乘法器206(参考图36B,示出对于第i级的滤波器预测器),乘法器接收各个滤波器系数gi,n作为它的第二输入。 These parallel from each of the delay stages Ti, n, i = 0 to 10, n = 0 to 7 individual tap is supplied to a predictor filter 200, provides an output to multiplier predictor 206 (see FIG. 36B, shows that for i-th stage predictor filter), a multiplier receiving the respective filter coefficients gi, n as its second input. 乘法器206的输出被提供给加法电路208,加法器提供存储在存储器寄存器中的符号的滤波器系数加权和。 The output of multiplier 206 is supplied to an adder circuit 208, an adder providing the filter weighting coefficients stored in the symbol memory and registers. 因此沿着线210的输出表示在与目前的滤波器系数gi,n有关的定时上滤波器输出(图6的126)的预测。 Thus represents the prediction on the current filter coefficients gi, n related to the timing of filter output (126. 6) along the output line 210.

沿着线210提供的滤波器预测器200的输出被提供给峰值降低算法计算电路218。 Along line 210 to provide the output prediction filter 200 is supplied to the peak reducing algorithm circuit 218. 在所示的特定的实施方案中,峰值降低算法计算器电路包括一个量值检测电路222,该检测电路接收沿着线210的已预测的滤波器输出,并检测它的幅度。 In the particular embodiment illustrated, the peak reduction algorithm value calculator circuit comprises a detection circuit 222, the detection circuit receives the predicted filter output along a line 210, and its magnitude is detected. 已预测的滤波输出被检测到的量值被提供给比较器226,比较器也接收预先确定的限值L和中心滤波器抽头增益g4。 The predicted filtered output the detected values ​​are supplied to the comparator 226, the comparator also receives the filter center value L and the predetermined tap gain g4. 如在前面的实施方案的情况下,如果预测的滤波器符号值超过限值,则一个开关使能信号被提供给选择器开关230。 As in the case of the foregoing embodiment, if the prediction filter symbol value exceeds the limit, then a switch enable signal is supplied to the selector switch 230. 否则,如果预测的滤波器符号值小于或等于限值,开关230被使能,以便不提供调节峰值,例如,在所选的实施方案中的零值。 Otherwise, if the prediction filter symbol value less than or equal value, switch 230 is enabled, so as not to provide peak shaving, e.g., a zero value in the selected embodiment. 预测的滤波器输出也被提供给算法处理器216,可以实施许多的适当的峰值降低算法中任何一个。 Predicted filter output is also supplied to the arithmetic processor 216, it may be implemented in a number of suitable algorithms to reduce any peak. 在所示的特定的实施方案中。 In the particular embodiment shown in FIG. 只接收预测的滤波器符号值,限值L,和施加到滤波器预测器g4的中心抽头的增益作为输入,由电路实施的适当的算法可以是等式(5)。 Filter receives only symbol values ​​predicted value L, and is applied to the center of the filter tap predictor gain g4 as an input, a suitable algorithm may be implemented by the circuit equation (5). 然后,算法处理器218的输出被作为一个反馈峰值降低值Fi给其他级,如果对于这样一个输出,选择器开关230被比较器226使能的话。 Then, the algorithm processor 218 is output as a peak-reducing feedback value Fi to the other stages, if for such an output, the selector switch 230 is enabled by comparator 226 words.

每个并行分支产生反馈的符号调节量,这些反馈调节量被提供给并行分支中每一个,使得最近的符号值可被包括在未来的滤波器预测值中。 Each parallel branch adjustment amount generated symbol feedback, the feedback is provided to adjust the amount of each of the parallel branches, so that the most recent symbol value may be included in a future predicted value filter. 对分支的反馈可用两种方式实施。 Feedback branch of two embodiments may be used. 这两种方式被示于图22和23中。 Both methods are illustrated in FIGS. 22 and 23. 图36表示在图23中所示的实施方案的实现方式。 36 shows an implementation in the embodiment shown in FIG. 23. 每个并行分支的符号调节被提供给所有的并行分支。 Each parallel branch symbol adjustment is provided for all the parallel branches. 来自下面的分支的反馈并未在上面的分支中示出,因为反馈将发生在最后抽头的存储器寄存器之后。 Feedback branch from the following are not shown in the above branch, it will occur because the feedback taps after the last memory register. 如果每个单独的分支的反馈被回送到它自身和所有下面的分支,图36可被修改来表示图22。 If each individual feedback branch is returned to itself and all the lower branch, in FIG. 36 may be modified to 22 represented in FIG. 图22的实施方案比图23的精确性稍差,因为所有分支的未来的预测将并不基于最当前的符号值。 FIG embodiment 22 of somewhat less accuracy than that of FIG. 23, because the future will not predict all branches based on the most current symbol value. 然而图23将提供有效的峰值降低。 However, FIG. 23 provides a valid peak reduction.

本领域的技术人员应该认识到图36A和36B示出图33的一种并行实施方案,包括来自其他并行级的附加反馈。 Those skilled in the art will recognize that one kind of FIGS. 36A and 36B in FIG. 33 of the embodiment shown in parallel, comprising the additional feedback from other parallel stages. 本领域的技术人员也应该认识到所有包括象图31和35那样的反馈的实施方案可用同样的方式修改以便在以上的并行实施方案中使用。 Those skilled in the art should also recognize that all FIGS. 31 and 35 as comprising a feedback embodiments as available for use to modify the same manner as in the above embodiment parallel.

本领域的技术人员也应该认识到,通过在一个长的多级移位寄存器中对每个存储器寄存器提供多级抽头也可以产生在图22,23和36A中所示的并行实施方案。 Those skilled in the art should also recognize that, by providing a multistage tap for each of the memory registers of a long multi-stage shift register may also be produced in parallel in the embodiment shown in FIGS. 22, 23 and 36A. 通过将反馈抽头对于对应的滤波器系数适当地分组,用图24和36B中所示的方法以并行方式计算反馈校正值。 By suitably feedback taps for the filter coefficient corresponding to the packet, by the method shown in FIG. 24 and 36B, in parallel, calculate the feedback correction value. 然后并行计算得到的反馈值将被回送到图36A线9所示的反馈点。 Feedback value is then calculated in parallel to the feedback point is returned to the line 9 shown in FIG. 36A.

对本发明的许多不同的实施方案已参照各种图形作了描述。 Many different embodiments of the present invention has been described with reference to various figures. 然而,本领域的技术人员将认识到在本发明的教导的范围内各种各样附加的实施方案都是可能的。 However, those skilled in the art will recognize that within the scope of the teachings of the present invention, various additional embodiments are possible. 例如,采用本发明的讲授内容可以提供实施特定算法的各种各样特定的电路,空间的限制不能列出所有可能的电路实施方案的详细目录或列举所有可能的算法。 For example, using the teachings of the present invention may provide a wide variety of specific circuit implementations particular algorithm, space limitations not inventory lists all possible embodiments of a circuit or algorithm to enumerate all possible. 各种各样其他可能的修改和附加的实施方案显然也是可能的,并且落在本发明的范围内。 Variety of other possible modifications and additional embodiments are also clearly possible and fall within the scope of the present invention. 因此,所描述的特定的实施方案和过程在任何意义上都不应该被看作是实际上的限制,而只是用作说明本发明。 Thus, specific embodiments and processes described herein should not be viewed as in any sense limiting practical, but rather the present invention is used as described.

虽然所示的峰值降低系统和本发明的方法已经作为在扩频通信系统,如CDMA和WCDMA蜂窝网中的实践被示出,这些系统提供一种本发明的优选的应用,应该认识到,对于本发明的峰值降低系统和方法的其他应用和环境也是可能的。 While the method and system of reducing the peak value as shown in the present invention have as a spread spectrum communication system, such as practical CDMA and WCDMA cellular network is shown, these systems provide a preferred application of the present invention is one kind, it should be appreciated that, for peak reduction system and method of the present invention to other environments and applications are possible. 例如,本发明的峰值降低系统和方法也可被方便地应用在不一定是扩频通信系统的多载波蜂窝基站中。 For example, the present invention reduces the peak system and method may also be advantageously applied in a spread spectrum communication system is not necessarily a multi-carrier cellular base stations. 因此,对于本发明的峰值降低系统和方法所描述的特定的应用和环境在任何意义上都不应该被看作是实际上的限制,而只是用作说明本发明。 Thus, the particular application and environment of the present invention, the peak reduction system and method described herein should not be viewed as in any sense limiting practical, but rather the present invention is used as described.

Claims (79)

  1. 1.一种扩频通信系统,包括:扩频符号发生器,用于接收多个分离的数据信道并将它们组合在一起,以提供扩频数据符号;滤波器,用于根据滤波器脉冲响应函数提供滤波操作;和峰值降低单元,耦合在所述扩频符号发生器和所述滤波器之间,并从所述扩频符号发生器接收所述数据符号,该峰值降低单元包括:滤波器预测器,该滤波器预测器使用与所述滤波器的滤波器脉冲响应函数相对应的滤波器系数值来提供预测的滤波输出;峰值降低计算电路,用于接收预测的滤波输出和预先确定的滤波输出限值,并根据预测的滤波输出超过限值的数量来确定峰值降低值;和组合器,用于将峰值降低值和数据符号组合,该峰值降低单元提供峰值已调节的符号给该滤波器,以便滤波并由该通信系统输出这些符号。 A spread spectrum communication system, comprising: spreading a symbol generator for receiving a plurality of separate data channels and combines them to provide a spread data symbols; filter according to the filter impulse response function provides filtering operation; and a peak reduction means coupled between said generator and said filter spread symbols, and receiving the data symbols from said spreading symbol generator, the peak reduction unit comprises: a filter predictor, the prediction filter uses a filter coefficient value of the filter impulse response function of the filter to provide corresponding filtered output predicted; peak reduction calculating circuit for receiving the output of the prediction filter and the predetermined filter output value, and is determined according to the predicted filter output exceeds the limit number of peak reduction value; and a combiner for reducing the peak value and the data symbol combination, the peak reducing unit provides a regulated peak filter to the symbol , a communication system so that the filtered output by these symbols.
  2. 2.如权利要求1所述的扩频通信系统,还包括上采样电路,耦合在峰值降低单元和滤波器之间,用于在滤波之前增加数据符号的采样率和提供上采样的数据符号。 2. The spread spectrum communication system according to claim 1, further comprising a sampling circuit, coupled between the peak reduction unit and a filter, for increasing the data symbols prior to filtering the sampled sampling rate and provides data symbols.
  3. 3.如权利要求2所述的扩频通信系统,其中该滤波器以上采样滤波速率对上采样的数据符号提供所述滤波操作。 3. A spread spectrum communication system according to claim 2, wherein the filter is provided above the sampling rate of the filter filtering operation on the data symbol samples.
  4. 4.如权利要求1所述的扩频通信系统,其中该峰值降低单元还包括一个延时电路,用于延时数据符号,以使组合器在逐个符号的基础上以时间同步的方式接收峰值降低值和数据符号。 4. The spread spectrum communication system according to claim 1, wherein the peak reduction unit further comprises a delay circuit for delaying the data symbols, so that the combiner receives a peak on a symbol by symbol basis in a time-synchronized manner and reducing the value of the data symbols.
  5. 5.如权利要求2所述的扩频通信系统,其中该滤波器预测器接收与在符号上的间隔上的滤波器脉冲响应函数的样本对应的第一组滤波器系数和与在符号间的间隔上的滤波器脉冲响应函数的样本对应的第二组滤波器系数。 The spread spectrum communication system according to claim 2, wherein the predictor filter and the receiving filter pulse at the symbol intervals in response to a sampling function corresponding to a first set of filter coefficients and in the inter-symbol filter impulse response of the second interval of the set of filter coefficients corresponding to the sampling function.
  6. 6.如权利要求5所述的扩频通信系统,其中该峰值降低单元包括基于符号上定时提供峰值降低的第一级和基于符号间定时提供峰值降低的第二级。 Spread spectrum communication system as claimed in claim 5, wherein the peak reduction based on the symbol timing unit comprises a second stage of the peak reduction based on the first stage and the inter-symbol timing of the peak reduction provided.
  7. 7.如权利要求1所述的扩频通信系统,其中该组合器包括一个乘法器电路,和其中该峰值降低值包括一个增益,该增益当它与数据符号相乘时提供已调节的符号,这将产生峰值降低的输出信号。 7. The spread spectrum communication system according to claim 1, wherein the composition comprises a multiplier circuit, and wherein the value comprises a peak reduction gain, which provide a regulated symbol when it is multiplied with a data symbol, this will produce an output signal of the peak reduction.
  8. 8.如权利要求1所述的扩频通信系统,其中该组合器包括一个加法电路,和其中该峰值降低值包括一个值,该值当它被加到符号上时提供已调节的符号,这将产生峰值降低的输出信号。 8. The spread spectrum communication system according to claim 1, wherein the combiner comprises a summing circuit, and wherein the value comprises a peak value decreases, the value is adjusted to provide a symbol when it is applied to the symbols, which It will produce an output signal of the peak reduction.
  9. 9.如权利要求4所述的扩频通信系统,其中该滤波器预测器包括含有多个延时级的存储寄存器,和其中一个或多个延时级包括所述延时电路。 9. A spread spectrum communication system according to claim 4, wherein the predictor filter comprises a delay stage comprising a plurality of storage registers, and wherein one or more of said delay circuit comprising delay stages.
  10. 10.如权利要求1所述的扩频通信系统,其中该滤波器预测器包括含有多个延时级的存储寄存器,其中每个所述延时级对应于一个独立的滤波器系数输入。 10. The spread spectrum communication system according to claim 1, wherein the predictor filter comprises a delay stage comprising a plurality of storage registers, wherein each of said delay stages corresponding to a separate input filter coefficients.
  11. 11.如权利要求10所述的扩频通信系统,其中该滤波器预测器包括多个乘法器,这些乘法器的数量与含有所述存储寄存器的多个延时级相等,和其中每个所述乘法器接收所述滤波器系数之一。 11. The spread spectrum communication system of claim 10 wherein each of the claims, wherein the predictor filter comprises a plurality of multipliers, number of multipliers is equal to the delay stage comprising a plurality of said storage register, and the multiplier receiving said one of said filter coefficients.
  12. 12.如权利要求11所述的扩频通信系统,其中该滤波器预测器还包括一个加法电路,接收所述多个乘法器的输出。 12. The spread spectrum communication system according to claim 11, wherein the prediction filter further comprises an adder circuit receiving the output of said plurality of multipliers.
  13. 13.如权利要求12所述的扩频通信系统,其中具有N级所述存储寄存器、N个乘法器和N个滤波器系数输入,其中N小于或等于所述滤波器的滤波器系数的数目。 13. The number of filter coefficients of the spread spectrum communication system as claimed in claim 12, wherein said storage register having N stages, N multipliers inputs and N filter coefficients, wherein N is less than or equal to the filter .
  14. 14.如权利要求1所述的扩频通信系统,其中所述峰值降低计算电路包括:量值检测电路,用于检测输入的预测滤波输出的量值;和比较器,用于将输入的预测滤波输出的量值与信号输出峰值的限值进行比较。 14. The spread spectrum communication system according to claim 1, wherein said peak reducing calculation circuit comprises: a magnitude detection circuit for detecting the magnitude of the output of prediction filtering input; and a comparator, the input for predicting the peak magnitude of the output signal of the filtered output value is compared.
  15. 15.如权利要求14所述的扩频通信系统,其中该峰值降低单元还包括一个选择器开关,耦合到比较器,并且当所述开关被比较器启动时使峰值降低值能够被输出到组合器。 15. The spread spectrum communication system according to claim 14, wherein the reducing means further includes a peak selector switch coupled to the comparator, and the peak value is lowered when the switch is activated can be output to the comparator combination device.
  16. 16.如权利要求14所述的扩频通信系统,其中该峰值降低计算电路还包括一个算法处理器,用于根据限值和预测的滤波输出实施一种峰值降低算法。 16. The spread spectrum communication system according to claim 14, wherein the peak reduction algorithm calculation circuit further comprises a processor, according to the algorithm for reducing the filter output value and the predicted implement a peak.
  17. 17.如权利要求16所述的扩频通信系统,其中该峰值降低计算电路还包括一个减法电路。 17. The spread spectrum communication system according to claim 16, wherein the calculation circuit further comprises a peak reducing a subtraction circuit.
  18. 18.如权利要求16所述的扩频通信系统,其中该峰值降低计算电路还包括一个除法电路。 18. The spread spectrum communication system according to claim 16, wherein the calculation circuit further comprises a peak reducing a division circuit.
  19. 19.如权利要求1所述的扩频通信系统,其中该峰值降低单元还包括一个反馈回路,该反馈回路将峰值降低值提供给滤波器预测器。 19. The spread spectrum communication system according to claim 1, wherein the peak reduction unit further comprises a feedback loop, the feedback loop will decrease the peak value to the filter predictor.
  20. 20.如权利要求19所述的扩频通信系统,其中该滤波预测器包括含有多个延时级的存储寄存器,和其中该反馈回路提供所述峰值降低值给这些延时级之间的所述存储寄存器。 20. The spread spectrum communication system according to claim 19, wherein the predictor filter comprises a delay stage comprising a plurality of storage registers, and wherein the feedback loop provides the peak value to reduce the delay between these stages said storage register.
  21. 21.如权利要求1所述的扩频通信系统,还包括数模变换器,被耦合以接收滤波器的输出和该数模变换器的RF放大器输出。 21. The spread spectrum communication system according to claim 1, further comprising a digital to analog converter, coupled to receive the output of the RF amplifier and filter output of the digital to analog converter.
  22. 22.如权利要求21所述的扩频通信系统,还包括耦合到RF放大器输出的发送天线。 22. The spread spectrum communication system according to claim 21, further comprising an antenna coupled to the RF transmission amplifier output.
  23. 23.如权利要求1所述的扩频通信系统,其中该扩频符号发生器包括:多个用于接收多个分离的数字数据信道的输入信道;多个扩展代码电路,这些扩展代码电路的数量等于分离的数字数据信道的数量,每个扩展代码电路提供不同的扩展代码;和多个乘法器电路,用于将扩展代码之一与分离的数字数据信道之一相乘,并提供多个扩频信道。 A plurality of spreading codes circuits, these spreading codes circuits; a plurality of input channels for receiving a plurality of separate digital data channel: 23. The spread spectrum communication system as claimed in claim 1, wherein the symbol generator comprises spreading number of digits equal to the number of data channels separate each circuit to provide different spreading codes of spreading code; and a plurality of multiplier circuits for multiplying the spreading codes with a separate one of one of the digital data channel, and providing a plurality of spreading channel.
  24. 24.如权利要求23所述的扩频通信系统,还包括扩频信道组合电路,用于将各个扩频信道组合在一起,以形成组合的数据符号。 24. The spread spectrum communication system according to claim 23, further comprising a spreading channel combining circuit for combining each spread spectrum channel, to form a data symbol combinations.
  25. 25.一种扩频通信系统,包括:扩频符号发生器,用于接收多个分离的数据信道并将它们组合在一起,以提供扩频数据符号;上采样电路,用于增加输入给它的符号的采样速率并提供上采样符号;滤波器,用于根据滤波器脉冲响应函数来提供滤波操作,具有与基于所述增加的采样率的定时对应的多个滤波器系数;和峰值降低单元,耦合在所述扩频符号发生器和所述滤波器之间,并从所述扩频符号发生器接收所述数据符号,该峰值降低单元包括多个峰值降低级,每级使用与所述滤波器的滤波器脉冲响应函数的一部分对应的滤波器系数值来预测该滤波器对数据符号的影响,以提供预测的滤波输出,并对超过峰值限值的预测滤波输出提供峰值降低处理,该峰值降低单元在所述多级峰值降低处理之后提供峰值已调节的符号给所述滤波器,其中所述滤波器对这 25. A spread spectrum communication system, comprising: spreading a symbol generator for receiving a plurality of separate data channels and combine them together to provide the spread data symbols; sampling circuit for increasing an input to it the sampling rate and provides symbol-sampled symbol; filter, for providing a filtering operation in accordance with the filter impulse response function, with a timing corresponding to the plurality of filter coefficients based on said increased sampling rate; and a peak reducing unit coupled between said generator and said filter spread symbols, and receiving the data symbols from said spreading symbol generator, the peak reducing peak reduction unit comprises a plurality of stages, each stage using the filter impulse response of the filter portion of filter coefficient values ​​corresponding to the function to project the impact of the filter on the data symbols to provide a predicted filtered output, and to provide peak reducing process prediction filter output exceeds the peak limit, the peak reduction unit after the multi-stage peak reducing process provides a peak regulated symbols to said filter, wherein said filter this 峰值已调节的符号进行滤波,随后由该通信系统输出这些符号。 Peak filtered conditioned symbols, these symbols are then output by the communication system.
  26. 26.如权利要求25所述的扩频通信系统,其中以串行方式提供所述多级的所述峰值降低单元。 Spread spectrum communication system according to claim 26. 25, wherein said peak in a serial manner to provide the multi-stage reduction unit.
  27. 27.如权利要求25所述的扩频通信系统,其中以并行方式提供所述多级的所述峰值降低单元。 Spread spectrum communication system according to claim 27. 25, wherein said peak in a parallel manner to provide the multi-stage reduction unit.
  28. 28.如权利要求25所述的扩频通信系统,其中每级的峰值降低单元采用与符号速率偏离整数量的上采样定时的定时偏移相对应的滤波器系数。 28. The spread spectrum communication system according to claim 25, wherein each stage of the peak reduction unit with a symbol rate offset from an integer number of sampling timing offset filter coefficients corresponding to the timing.
  29. 29.如权利要求25所述的扩频通信系统,其中每级的峰值降低单元包括:滤波器预测器,用于接收与滤波器脉冲响应函数的所述部分对应的滤波器系数,并提供所述预测的滤波输出;和峰值降低计算电路,用于根据由该滤波器预测器提供的预测的滤波输出来计算峰值降低值。 29. The spread spectrum communication system according to claim 25, wherein each stage of the peak reduction unit comprises: a prediction filter for receiving the filter impulse response filter coefficient of the portion corresponding to the function, and provide the predictive filter output; and a peak reduction calculation circuit for calculating according to the filter provided by the predictor filter output peak predicted reduction value.
  30. 30.如权利要求29所述的扩频通信系统,其中每级的峰值降低单元还包括组合器,用于接收计算的峰值降低值并将它们与输入符号进行组合。 30. The spread spectrum communication system according to claim 29, wherein each stage of the peak reduction unit further comprises a combiner for receiving the calculated peak value and decreases with the input symbol combination thereof.
  31. 31.如权利要求30所述的扩频通信系统,其中每级的峰值降低单元还包括延时电路,用于延时输入符号和提供延时的符号给该组合器,以便在正确的定时上逐个符号地组合峰值降低值和延时的符号。 31. The spread spectrum communication system according to claim 30, wherein each stage of the peak reduction on the timing unit further comprises a delay circuit for delaying the input symbols and provides the delayed symbols to a combiner so that the correct symbol by symbol combining symbol values ​​and peak reducing delay.
  32. 32.如权利要求30所述的扩频通信系统,其中所述组合器包括乘法器。 30, the spread spectrum communication system as claimed in claim 32, wherein said composition comprises a multiplier.
  33. 33.如权利要求30所述的扩频通信系统,其中所述组合器包括加法器。 30, the spread spectrum communication system as claimed in claim 33, wherein said combiner comprises an adder.
  34. 34.一种用于降低峰值信号输出值的系统,该系统适于在包括滤波器的通信系统中使用,该滤波器在从所述通信系统输出信号之前提供符号滤波,该系统包括:滤波器预测器装置,用于在利用所述滤波器进行滤波之前接收符号和预测所述滤波对所述符号的影响;和符号调节器装置,耦合到该滤波器预测器装置,用于调节被预测在经受所述滤波之后将超过峰值限值的符号。 34. A system for reducing the peak value of the output signal, which system comprises a communication system adapted to use a filter, the filter provides filtered symbols from the output signal prior to the communication system, the system comprising: a filter prediction means for predicting the received symbols, and filtering effect on the symbols prior to filtering with the filter; and the symbol adjustment means, coupled to the prediction filter means for adjusting is predicted after being subjected to the filtered symbol will exceed the peak limit.
  35. 35.如权利要求34所述的用于降低峰值信号输出值的系统,其中所述滤波器具有预先确定的脉冲响应函数,其中所述滤波器预测器装置包括用于在多个采样点上接收与所述脉冲响应函数对应的滤波器系数的装置。 35. A system as claimed in claim for reducing the peak value of the output signal 34, wherein said filter having a predetermined impulse response function, wherein said filter means comprises a predictor for receiving a plurality of sampling points on It means a function of filter coefficients corresponding to the impulse response.
  36. 36.如权利要求34所述的用于降低峰值信号输出值的系统,其中所述通信系统是一种扩频蜂窝通信系统。 36. A system as claimed in claim for reducing the peak value of the output signal 34, wherein said communication system is a spread spectrum cellular communication system.
  37. 37.如权利要求34所述的用于降低峰值信号输出值的系统,其中所述的用于调节符号的装置包括用于计算峰值降低值的装置和用于将峰值降低值与所述符号组合的装置。 37. A system as claimed in claim for reducing the peak value of the output signal 34, wherein means for adjusting said symbol comprises means for calculating a value for the peak reducing and reducing the peak value and the symbol combination s installation.
  38. 38.如权利要求37所述的用于降低峰值信号输出值的系统,其中所述的用于计算的装置包括DSP。 38. A system as claimed in claim for reducing the peak value of the output signal 37, wherein said means for calculating comprises a DSP.
  39. 39.如权利要求37所述的用于降低峰值信号输出值的系统,其中所述的峰值降低值包括增益。 39. A system as claimed in claim for reducing the peak value of the output signal 37, which includes the gain peak reducing the value.
  40. 40.如权利要求37所述的用于降低峰值信号输出值的系统,其中所述的用于组合的装置包括乘法器电路。 40. A system as claimed in claim for reducing the peak value of the output signal 37, wherein said means for combining comprises a multiplier circuit.
  41. 41.如权利要求37所述的用于降低峰值信号输出值的系统,其中所述的用于组合的装置包括加法或减法电路。 41. A system as claimed in claim for reducing the peak value of the output signal 37, wherein said means for combining comprises an adder or a subtracting circuit.
  42. 42.如权利要求35所述的用于降低峰值信号输出值的系统,其中所述的用于接收滤波器系数的装置在采样定时和符号间采样定时上接收滤波器系数。 42. A system as claimed in claim for reducing the peak value of the output signal 35, the reception timing of the filter coefficient means for receiving the filter coefficients wherein said sampling timing between samples and symbols.
  43. 43.如权利要求35所述的用于降低峰值信号输出值的系统,其中所述的用于接收滤波器系数的装置在多个偏移符号采样时间上接收滤波器系数。 43. A system as claimed in claim for reducing the peak value of the output signal 35, means for receiving the filter coefficients wherein said receiving filter coefficients over a plurality of symbol sampling time offset.
  44. 44.如权利要求35所述的用于降低峰值信号输出值的系统,其中所述的滤波器预测器装置包括多个级,每级接收表示不同的滤波输出定时的不同的滤波器系数。 44. A system as claimed in claim for reducing the peak value of the output signal 35, wherein said prediction filter means comprises a plurality of stages, each stage receives a timing different from a filter coefficient different from the filter output.
  45. 45.如权利要求44所述的用于降低峰值信号输出值的系统,其中以串行方式配置所述级。 45. A system as claimed in claim for reducing the peak value of the output signal 44, wherein the stage arranged in a serial manner.
  46. 46.如权利要求44所述的用于降低峰值信号输出值的系统,其中以并行方式配置所述级。 46. ​​A system as claimed in claim for reducing the peak value of the output signal 44, which is disposed parallel to the stage.
  47. 47.一种在包括滤波器的通信系统中用于调节符号值的方法,该滤波器在从所述的系统输出信号之前提供符号滤波,该方法包括:在利用所述滤波器进行滤波之前,接收符号;采用所述滤波器的已知特性来预测所述滤波对所述符号的影响;和在经受所述滤波之后,调节被预测将超过峰值限值的符号的值,以降低从该系统输出的信号的峰值。 47. A communication system comprising a filter in a method for adjusting the symbol value, the filter provides filtered symbols before the system output signal from, the method comprising: prior to filtering with the filter, received symbols; known characteristics of the filter employed to predict the effect of the filtering of the symbol; and after being subjected to the filtering adjustment value is predicted to exceed the peak value of the symbol from the system to reduce peak signal output.
  48. 48.一种改进的扩频通信方法,包括:提供对应于组合在一起的多个分离的数据信道的数据符号;根据滤波器脉冲响应函数提供滤波操作;和在所述滤波操作之前执行峰值降低处理,所述峰值降低处理包括:利用与滤波器脉冲响应函数对应的滤波器系数值来提供预测的滤波输出;接收预测的滤波输出和预先确定的滤波输出限值,并根据预测的滤波输出超过限值的数量来确定峰值降低值;和将峰值降低值和数据符号组合,并提供峰值已调节的符号以便被滤波和由通信系统输出。 48. An improved method for spread spectrum communication, comprising: providing a plurality of corresponding combined separate data channel data symbols; filtering operation provided according to the filter impulse response function; and perform the filtering operation prior to a peak reduced process, the peak reducing process comprising: using a filter with impulse response filter coefficient values ​​to provide a function corresponding to the predicted filtered output; receiving the filtered output and the predicted filter output value determined in advance, and in accordance with the filtered output exceeds the predicted limit the number of values ​​to determine the peak reduction; and reduce the peak value and the data symbol combinations, and provides a peak regulated so as to be filtered and output symbols by the communication system.
  49. 49.如权利要求48所述的改进的扩频通信方法,还包括在滤波之前增加数据符号的采样率。 Improved spread spectrum communication method according to claim 49. 48, further comprising increasing the sampling rate of the data symbols prior to filtering.
  50. 50.如权利要求49所述的改进的扩频通信方法,其中以上采样滤波速率对上采样的数据符号进行滤波。 50. Improved spread spectrum communication method according to claim 49, wherein the filtering rate of the above sampling data symbols sampled filtered.
  51. 51.如权利要求48所述的改进的扩频通信方法,其中该峰值降低处理还包括延时数据符号,以便在所述组合期间以时间同步的方式在逐个符号的基础上接收峰值降低值和数据符号。 51. The improved method of spread spectrum communication according to claim 48, wherein the peak reducing process further comprises time-delay data symbols, so as to be time synchronized manner to reduce the peak value of the received symbol by symbol on the basis of the composition during and data symbols.
  52. 52.如权利要求49所述的改进的扩频通信方法,其中所述提供预测的滤波输出包括接收与在符号上间隔上的滤波器脉冲响应函数的样本对应的第一组滤波器系数和与在符号间间隔上的滤波器脉冲响应函数的样本对应的第二组滤波器系数。 52. Improved spread spectrum communication method according to claim 49, wherein said providing includes receiving predicted filtered output and the filter impulse in response to the first symbol interval on the set of filter coefficients corresponding to the sampling function and a in intersymbol filter impulse response of the second interval of the set of filter coefficients corresponding to the sampling function.
  53. 53.如权利要求49所述的改进的扩频通信方法,其中所述峰值降低处理包括提供基于符号上定时的峰值降低处理和提供基于符号间定时的峰值降低处理。 53. The improved method of spread spectrum communication according to claim 49, wherein said peak reducing process comprises providing a symbol timing based on the peak reducing process and provide reduced peak timing based on inter-symbol processing.
  54. 54.如权利要求48所述的改进的扩频通信方法,其中该组合包括相乘,并且其中该峰值降低值包括一个增益,该增益当它与数据符号相乘时提供一个已调节的符号,这将产生峰值降低的输出信号。 54. Improved spread spectrum communication method according to claim 48, wherein the composition comprises a multiplication, and wherein the reduced value comprises a peak gain, which is adjusted to provide a symbol when it is multiplied by the data symbols, this will produce an output signal of the peak reduction.
  55. 55.如权利要求48所述的改进的扩频通信方法,其中该组合包括相加,并且其中该峰值降低值包括一个值,该值当它被加到该符号上时提供一个已调节的符号,这将产生峰值降低的输出信号。 55. Improved spread spectrum communication method according to claim 48, wherein the composition comprises added, and wherein the reduced value comprises a peak value that a symbol has been adjusted when it is applied to the symbol this will produce an output signal of the peak reduction.
  56. 56.如权利要求48所述的改进的扩频通信方法,其中所述峰值降低处理还包括检测输入的预测滤波输出的量值,以及将输入的预测滤波输出的量值与信号输出峰值限值进行比较。 56. The improved method of spread spectrum communication according to claim 48, wherein the magnitude of the peak reducing process further comprises a signal output of the peak detector output filtering prediction input value, and a prediction of the filtered output of the input limit Compare.
  57. 57.如权利要求56所述的改进的扩频通信方法,其中该峰值降低处理还包括:只在输入的预测滤波输出的量值超过信号输出峰值限值时,才允许输出峰值降低值以便组合。 57. The improved method of spread spectrum communication according to claim 56, wherein the peak reducing process further comprising: input only when the magnitude of the prediction filter output signal output exceeds the peak limit value allowed in order to combine the output of the peak reduction .
  58. 58.如权利要求48所述的改进的扩频通信方法,其中该峰值降低处理还包括基于所述限值和预测的滤波输出实施峰值降低算法。 58. The improved method of spread spectrum communication according to claim 48, wherein the peak reducing process embodiment further includes a peak value of said filtered output reduction algorithm and Prediction.
  59. 59.如权利要求48所述的改进的扩频通信方法,其中所述提供对应于多个数据信道的数据符号包括:接收多个分离的数字数据信道;提供多个不同的数量等于分离的数字数据信道数量的扩展代码;将扩展代码中的每一个与分离的数字数据信道之一相乘,以提供多个扩频信道;以及将各个扩频信道组合在一起,以形成组合的数据符号。 59. The improved method of spread spectrum communication according to claim 48, wherein said data symbols are provided corresponding to the plurality of data channels comprising: receiving a plurality of separate digital data channels; equal to the number of a plurality of different separate digital the number of data channel spreading code; extend each one of the digital data is multiplied with channel separation code to provide a plurality of spread spectrum channel; and spreading the respective channels together, to form a data symbol combinations.
  60. 60.一种改进的扩频通信方法,包括:接收对应于多个组合在一起的分离的数据信道的数据符号;增加所述符号的采样速率并提供上采样符号;根据滤波器脉冲响应函数提供滤波操作,具有多个滤波器系数并与基于所述增加的采样率的定时对应;和在采用所述数据符号的所述滤波操作之前提供峰值降低处理,该峰值降低处理在多个峰值降低级中执行,每个峰值降低处理级包括利用与滤波器脉冲响应函数的一部分相对应的滤波器系数值来预测滤波操作对数据符号的影响,以提供预测的滤波输出,并对超过峰值限值的预测的滤波输出提供峰值降低处理,其中所述峰值降低处理在所述多级峰值降低处理之后提供峰值已调节的符号以便被滤波和输出。 60. An improved method for spread spectrum communication, comprising: receiving a combined corresponding to a plurality of separate data symbols of a data channel; increasing the sampling rate of the symbols and provides the symbol samples; filter impulse response function provided in accordance with filtering operation, and having a plurality of filter coefficients based on the timing corresponding to the increased sampling rate; and providing a peak prior to the filtering operation of the data symbols using the reduction process, reduction process to reduce the peak level in a plurality of peaks is performed, each stage comprises a peak reducing process part with a filter coefficient value corresponding to the filter impulse response function of the filtering operation to project the impact of data symbols to provide a prediction filter output exceeds the peak limit value and filter output to provide the predicted peak reducing process, wherein the peak reducing process after the reduction process to provide the multi-stage peak peak regulated so as to be filtered and output symbols.
  61. 61.如权利要求60所述的改进的扩频通信方法,其中以串行方式执行所述多级峰值降低处理。 Improved spread spectrum communication method according to claim 61. 60, performed in a serial manner wherein the peak value of the multi-stage reducing process.
  62. 62.如权利要求60所述的改进的扩频通信方法,其中以并行方式执行所述多级峰值降低处理。 Improved spread spectrum communication method according to claim 62. 60, wherein the parallel execution of the multi-stage peak reducing process.
  63. 63.如权利要求60所述的改进的扩频通信方法,其中每级峰值降低处理采用与符号速率偏离整数量的上采样定时的定时偏移相对应的滤波器系数。 63. The improved method of spread spectrum communication according to claim 60, wherein each stage peak reducing process using a symbol rate offset from an integer number of sampling timing offset filter coefficients corresponding to the timing.
  64. 64.如权利要求60所述的改进的扩频通信方法,其中每级峰值降低处理包括接收与滤波器脉冲响应函数的所述部分对应的滤波器系数并提供所述预测的滤波输出,以及根据预测的滤波输出计算峰值降低值。 64. The improved method of spread spectrum communication according to claim 60, wherein each stage comprises a peak reducing process receiving filter coefficients of the filter impulse response function corresponding to the portion and providing the filtered output predicted, and according to calculating the predicted filter output values ​​to reduce the peak.
  65. 65.如权利要求64所述的改进的扩频通信方法,其中每级峰值降低处理还包括接收计算的峰值降低值并将它们与输入符号进行组合。 65. The improved method of spread spectrum communication according to claim 64, wherein each stage further comprises a peak reducing process to reduce the peak value of the receiving computing and combining them with the input symbol.
  66. 66.如权利要求65所述的改进的扩频通信方法,其中每级峰值降低处理还包括延时输入符号和提供将被组合的延时的符号,以便峰值降低值和延时的符号在正确的定时上在逐个符号的基础上被组合。 66. The improved method of spread spectrum communication according to claim 65, wherein each stage further comprises a delay peak reducing process input symbols and provides combined symbols delay, so as to reduce the peak delay values ​​and symbols correctly the timing is a combination of a symbol by symbol basis.
  67. 67.如权利要求65所述的改进的扩频通信方法,其中所述组合包括相乘。 Improved spread spectrum communication method according to claim 67. 65, wherein said composition comprises multiplying.
  68. 68.如权利要求65所述的改进的扩频通信方法,其中所述组合包括相加。 Improved spread spectrum communication method according to claim 68. 65, wherein said combining comprises adding.
  69. 69.一种用于降低峰值信号输出值的方法,该方法适用于在输出通信信号之前采用预定脉冲响应函数提供符号滤波的通信方法中,包括:在滤波之前接收符号,并使用在多个采样点上与所述脉冲响应函数对应的滤波器系数来预测所述滤波对所述符号的影响;和调节被预测在经受所述滤波之后将超过峰值限值的符号。 69. A method of reducing the peak value of the output signal, the method suitable for using a predetermined pulse signal in response to a communication prior to the communication method output function provides the filtered symbol, comprising: receiving a symbol before the filter, and using a plurality of sampling the predicted impact point of the symbol on the filter with the impulse response filter coefficients corresponding to the function; and adjusting said predicted filtered after being subjected to exceed the peak limit sign.
  70. 70.如权利要求69所述的用于降低峰值信号输出值的方法,其中所述通信方法是扩频蜂窝通信方法。 70. A method as claimed in claim reduce the peak value for the output signal 69, wherein said communication method is a spread spectrum cellular communication method.
  71. 71.如权利要求69所述的用于降低峰值信号输出值的方法,其中调节符号包括计算峰值降低值以及组合峰值降低值与所述符号。 71. A method as claimed in claim reduce the peak value for the output signal 69, wherein the adjusting includes calculating a peak symbol values ​​and combinations of reducing the peak value and the reduced symbol.
  72. 72.如权利要求71所述的用于降低峰值信号输出值的方法,其中所述峰值降低值包括一个增益。 72. A method as claimed in claim reduce the peak value for the output signal 71, wherein said peak value comprises a gain reduction.
  73. 73.如权利要求72所述的用于降低峰值信号输出值的方法,其中所述组合包括相乘。 73. A method as claimed in claim reduce the peak value for the output signal 72, wherein said combining comprises multiplying.
  74. 74.如权利要求71所述的用于降低峰值信号输出值的方法,其中所述组合包括相加或相减。 74. A method as claimed in claim reduce the peak value for the output signal 71, wherein said composition comprises addition or subtraction.
  75. 75.如权利要求69所述的用于降低峰值信号输出值的方法,其中在采样定时和符号间采样定时上接收所述滤波器系数。 75. A method as claimed in claim reduce the peak value for the output signal 69, wherein said receiving filter coefficients between the sampling timing and sampling timing symbols.
  76. 76.如权利要求69所述的用于降低峰值信号输出值的方法,其中在多个偏移符号采样时间上接收所述滤波器系数。 76. A method as claimed in claim reduce the peak value for the output signal 69, wherein a plurality of offset symbols in a time of receiving said sampling filter coefficients.
  77. 77.如权利要求69的所述用于降低峰值信号输出值的方法,其中所述预测在多级中执行,每级接收表示不同的滤波输出定时的不同的滤波器系数。 77. The method of claim 69 reduce the peak value for the output signal, wherein said prediction performed in multiple stages, each stage receives a timing different from the output of filter coefficient different from the filter.
  78. 78.如权利要求77所述的用于降低峰值信号输出值的方法,其中以串行方式执行所述级。 78. A method as claimed in claim reduce the peak value for the output signal 77, wherein the stages perform a serial manner.
  79. 79.如权利要求77所述的用于降低峰值信号输出值的方法,其中以并行方式执行所述级。 79. A method as claimed in claim reduce the peak value for the output signal 77, wherein the stages carried out in parallel.
CN 01808283 2000-04-19 2001-02-28 System and method for peak power reduction in spread spectrum communications systems CN1284345C (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
US19851600 true 2000-04-19 2000-04-19
US21289200 true 2000-06-20 2000-06-20
US09746167 US6449302B2 (en) 2000-04-19 2000-12-22 System and method for peak power reduction in spread spectrum communications systems

Publications (2)

Publication Number Publication Date
CN1425242A true CN1425242A (en) 2003-06-18
CN1284345C true CN1284345C (en) 2006-11-08

Family

ID=27393889

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 01808283 CN1284345C (en) 2000-04-19 2001-02-28 System and method for peak power reduction in spread spectrum communications systems

Country Status (4)

Country Link
EP (1) EP1281262A4 (en)
KR (1) KR100466057B1 (en)
CN (1) CN1284345C (en)
CA (1) CA2406757C (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4287225B2 (en) 2003-09-18 2009-07-01 株式会社日立国際電気 Transmitter
JP4871813B2 (en) 2007-08-28 2012-02-08 株式会社日立製作所 Wireless communication device, wireless communication method and a peak suppression method
GB2483290B (en) * 2010-09-03 2015-07-22 Nvidia Technology Uk Ltd Transmitting a signal from a power amplifier
EP2940866A3 (en) * 2014-04-30 2016-01-27 Nxp B.V. Rf switch circuit

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5418813A (en) * 1993-12-06 1995-05-23 Motorola, Inc. Method and apparatus for creating a composite waveform
US6049535A (en) * 1996-06-27 2000-04-11 Interdigital Technology Corporation Code division multiple access (CDMA) communication system
JP3311950B2 (en) * 1996-12-19 2002-08-05 富士通株式会社 Code multiplexing wireless apparatus
JP3462388B2 (en) * 1998-04-28 2003-11-05 松下電器産業株式会社 Wireless communication device

Also Published As

Publication number Publication date Type
CA2406757C (en) 2004-03-23 grant
KR100466057B1 (en) 2005-01-13 grant
EP1281262A4 (en) 2008-09-10 application
CA2406757A1 (en) 2001-11-01 application
CN1425242A (en) 2003-06-18 application
EP1281262A1 (en) 2003-02-05 application

Similar Documents

Publication Publication Date Title
Gibson The mobile communications handbook
US6504862B1 (en) Method and apparatus for reducing the ratio of peak to average power in a Gaussian signal including a CDMA signal
US6757334B1 (en) Bit rate agile third-generation wireless CDMA, GSM, TDMA and OFDM system
US5422909A (en) Method and apparatus for multi-phase component downconversion
US6337606B1 (en) Digital communications modulator having a modulation processor which supports high data rates
US20060083320A1 (en) Demodulation of multiple signals
US6650688B1 (en) Chip rate selectable square root raised cosine filter for mobile telecommunications
US20040266369A1 (en) Methods and apparatus for controlling signals
US6674790B1 (en) System and method employing concatenated spreading sequences to provide data modulated spread signals having increased data rates with extended multi-path delay spread
US6141372A (en) Digital downconverter/despreader for direct sequence spread spectrum CDMA communications system
US20050163248A1 (en) Signal peak reduction circuit for non-constant envelope modulation signals
US20080247487A1 (en) Dynamic crest factor reduction system
Davidson et al. Design of orthogonal pulse shapes for communications via semidefinite programming
US7170952B2 (en) System and method for post filtering peak power reduction in communications systems
US6931053B2 (en) Peak power and envelope magnitude regulators and CDMA transmitters featuring such regulators
CN103634022A (en) Full-duplex transceiver in flat fading environment and method for canceling self-interference
EP0758823A2 (en) Spread spectrum communication system
CN1538650A (en) Multicarrier signal wave removing device and method
JP2004104162A (en) Amplitude limit apparatus
US20020176481A1 (en) Transmission timing control device, digital roll-off filter, and mobile radio terminal for digital radio communication
US6246697B1 (en) Method and system for generating a complex pseudonoise sequence for processing a code division multiple access signal
US20100029347A1 (en) Apparatuses and a method for reducing peak power in a transmitter of telecommunications systems
US7813433B2 (en) System and method for communicating data using symbol-based randomized orthogonal frequency division multiplexing (OFDM) with selected subcarriers turned on or off
US20040014436A1 (en) Amplitude limitation
US6449302B2 (en) System and method for peak power reduction in spread spectrum communications systems

Legal Events

Date Code Title Description
C06 Publication
C10 Request of examination as to substance
C14 Granted
ASS Succession or assignment of patent right

Owner name: POWERWAVE TECHNOLOGIES INC.

Free format text: FORMER OWNER: P-WAVE HOLDING LLC

Effective date: 20150506

Owner name: P-WAVE HOLDING LLC

Free format text: FORMER OWNER: POWERWAVE TCH INC.

Effective date: 20150506

Owner name: INTEL CORP .

Free format text: FORMER OWNER: POWERWAVE TECHNOLOGIES INC.

Effective date: 20150506

C41 Transfer of the right of patent application or the patent right