WO2009056044A1 - Procédé et dispositif de suppression de valeur de pic, et dispositif de production de signaux de suppression de valeur de pic de référence - Google Patents

Procédé et dispositif de suppression de valeur de pic, et dispositif de production de signaux de suppression de valeur de pic de référence Download PDF

Info

Publication number
WO2009056044A1
WO2009056044A1 PCT/CN2008/072786 CN2008072786W WO2009056044A1 WO 2009056044 A1 WO2009056044 A1 WO 2009056044A1 CN 2008072786 W CN2008072786 W CN 2008072786W WO 2009056044 A1 WO2009056044 A1 WO 2009056044A1
Authority
WO
WIPO (PCT)
Prior art keywords
peak
signal
amplitude
rate
time domain
Prior art date
Application number
PCT/CN2008/072786
Other languages
English (en)
French (fr)
Inventor
Yong Jiang
Original Assignee
Huawei Technologies Co., Ltd.
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
Application filed by Huawei Technologies Co., Ltd. filed Critical Huawei Technologies Co., Ltd.
Publication of WO2009056044A1 publication Critical patent/WO2009056044A1/zh

Links

Classifications

    • 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

Definitions

  • the invention relates to a peak cancellation method, a peak cancellation device and a reference peak cancellation signal generation device.
  • the application is filed on October 26, 2007, and the application number is 200710176420.5, and the invention name is "peak cancellation method, peak cancellation device and The priority of the Chinese Patent Application of the Reference Peaking Signal Generating Device, the entire contents of which is incorporated herein by reference.
  • Technical field is “peak cancellation method, peak cancellation device and The priority of the Chinese Patent Application of the Reference Peaking Signal Generating Device, the entire contents of which is incorporated herein by reference.
  • the present invention relates to communication technologies, and more particularly to a peak cancellation method, a peak cancellation device, and a reference peak elimination signal generating device. Background technique
  • Orthogonal Frequency Division Multiplexing (OFDM) technology is an effective broadband transmission technology, which has been widely applied to Digital Audio Broadcast (DAB) and digital video broadcasting ( Digital Video Broadcast, hereinafter referred to as DVB), High Definition Television (HDTV) and many other fields.
  • DVB Digital Audio Broadcast
  • HDTV High Definition Television
  • PAPR peak-to-average power ratio
  • a peak cancellation waveform of an approximate ideal pulse of M times the sampling rate is pre-stored or generated in the field, that is, a Kernel signal with a M times the sampling rate.
  • the peak cancellation method using the Kernel signal is usually performed at a rate of 4 times or more, and the time domain signal is 4 times or 4 times.
  • the above-mentioned time domain signal of the sampling rate, the Kernel signal is a Kernel signal of 4 times or more than the sampling rate.
  • Embodiments of the present invention provide a peak cancellation method, a peak cancellation device, and a reference peak elimination signal generation device, which can reduce processing complexity, shorten delay, and improve work efficiency when performing peak cancellation.
  • a peak cancellation method including: receiving a time-domain signal of a 1 ⁇ rate;
  • the 1 time rate time domain signal is peaked using the peak elimination signal corresponding to the peak point.
  • a peak cancellation apparatus including: a prediction module, configured to predict a peak point of a time domain signal of a M times over a sampling rate according to a time domain signal of a 1 time rate Characteristic information;
  • a peak cancellation module configured to select a reference peak elimination signal according to the characteristic information of the peak point, and perform circumferential shift, amplitude adjustment, and phase rotation on the selected reference peak elimination signal to generate a cancellation corresponding to the peak point A peak signal, and the peak-down signal corresponding to the peak point is used to peak the 1 time rate time domain signal.
  • the feature information of the peak point of the time domain signal of the M times over the sample rate is predicted according to the time domain signal of the 1 time rate, and the reference peak elimination signal is selected according to the feature information, and the selected reference peak elimination signal is circled.
  • the peak cancellation is performed at the M times the sampling rate, the buffer capacity requirement of the buffer unit is reduced, the workload of the corresponding processing is reduced, the complexity is low, and the delay is small, thereby effectively improving the peak value.
  • the working efficiency of the cancellation device improves its working performance.
  • a reference peaking signal generating apparatus including:
  • a weight setting module configured to set a weight for each subcarrier according to a peak weighting factor of each subcarrier in the orthogonal frequency division multiplexing symbol, and generate a frequency domain weight signal
  • a zero insertion module configured to perform orthogonal frequency division multiplexing subcarrier mapping on the frequency domain weight signal, and continuously insert (M-1) xN zeros in the frequency domain weight signal;
  • the frequency-time conversion module is configured to perform frequency-domain to time-domain transformation on the frequency-domain weight signal after the interpolation, and obtain a quasi-reference peak-sampling signal with M times of the same;
  • An amplitude adjustment module configured to perform amplitude adjustment on the quasi-reference peak-sampling signal, normalize the highest amplitude, and obtain a normalized M-fold-like quasi-reference peak-sampling signal
  • a grouping module configured to group the normalized M times the sample rate of the quasi-reference peaking signal, and divide the normalized M times the quasi-reference peaking signal into different starting The reference peaking signal of the M points of the phase.
  • another reference peaking signal generating apparatus including:
  • a weight setting module configured to set weights for each subcarrier according to a peak weighting factor of each subcarrier in the orthogonal frequency division multiplexing symbol, to generate a frequency domain weight signal
  • a frequency-time conversion module configured to perform frequency-domain to time-domain transformation on the phase-adjusted frequency-domain weight signal to generate a quasi-reference peak-down signal
  • the amplitude adjustment module performs amplitude adjustment on the quasi-reference peak-sampling signal, and normalizes the highest amplitude to obtain a reference peak-canceling signal.
  • a further reference peaking signal generating apparatus including:
  • a weight setting module configured to set a weight for each subcarrier according to a peak weighting factor of each subcarrier in the orthogonal frequency division multiplexing symbol, and generate a frequency domain weight signal
  • a frequency-time conversion module configured to perform frequency domain to time domain transformation on the frequency domain weight signal, to obtain a quasi-reference peak-down signal of 1 time rate
  • a zero insertion filtering module configured to obtain a quasi-reference peak-sampling signal with a M-fold over-rate rate from the quasi-reference peak-sampling signal of the 1 ⁇ rate;
  • An amplitude adjustment module configured to perform amplitude adjustment on the quasi-reference peak-sampling signal of the M-fold sampling rate, and normalize the highest amplitude
  • a grouping module configured to group the normalized M times the sample rate of the quasi-reference peaking signal, and divide the normalized M times the sample rate of the quasi-reference peaking signal into different The reference peaking signal of the M points of the initial phase.
  • a further reference peaking signal generating apparatus including:
  • a weight setting module configured to perform peak elimination weighting of each subcarrier in the symbol according to orthogonal frequency division multiplexing Sub, setting weights for each subcarrier to generate a frequency domain weight signal
  • a frequency-time conversion module configured to perform frequency domain to time domain transformation on the frequency domain weight signal, to obtain
  • An amplitude adjustment module configured to perform amplitude adjustment on the 1x rate quasi-reference peaking signal, and normalize the highest amplitude
  • the multi-phase filtering module is configured to perform multi-phase filtering of different phases of the pseudo-reference peak-sampling signal with a magnitude normalized amplitude to generate a reference peak-canceling signal.
  • the reference peak-sampling signal generating apparatus provided by the above embodiment of the present invention can generate M N-point reference peak-sampling signals according to a quasi-reference peak-down signal of 1 time rate, so as to perform circumferential shift, amplitude adjustment and phase rotation after 1 ⁇ speed
  • the time domain signal of the rate is used for peak elimination.
  • Embodiment 1 is a flowchart of Embodiment 1 of a peak cancellation method according to the present invention
  • Embodiment 2 is a flowchart of Embodiment 2 of a peak cancellation method according to the present invention
  • Embodiment 3 is a flowchart of Embodiment 1 of generating an MP-Kernel signal according to the present invention
  • FIG. 4 is a diagram showing an example of an amplitude of an MP-Kernel signal at four 1024 points used in the peak elimination process according to an embodiment of the present invention
  • Figure 5 is a partial enlarged view of an example of the amplitude of the MP-Kernel signal of Figure 4;
  • Embodiment 6 is a flowchart of Embodiment 2 of generating an MP-Kernel signal according to the present invention
  • FIG. 7 is a flowchart of Embodiment 3 of generating an MP-Kernel signal according to the present invention.
  • Embodiment 8 is a flowchart of Embodiment 4 of generating an MP-Kernel signal according to the present invention.
  • Embodiment 9 is a flow chart of Embodiment 1 of the first embodiment of the present invention for predicting the peak value of the time domain signal of the M times over the sample rate, and obtaining and recording the characteristic information of the peak point;
  • FIG. 10 is a flowchart of a second embodiment of the present invention for predicting a peak value of a time domain signal of a M-timed sample rate, and obtaining and recording characteristic information of a peak point;
  • Figure 11 is a peak point of a time domain signal for predicting an M-fold over sample rate according to the present invention, which is obtained and recorded.
  • the characteristic information of the peak point is the flowchart of the third embodiment
  • Embodiment 13 is a schematic structural diagram of Embodiment 1 of a reference peak-sampling signal generating apparatus according to the present invention.
  • Embodiment 14 is a schematic structural diagram of Embodiment 2 of a reference peak-sampling signal generating apparatus according to the present invention.
  • Embodiment 15 is a schematic structural diagram of Embodiment 3 of a reference peak-sampling signal generating apparatus according to the present invention.
  • Embodiment 4 of a reference peak-sampling signal generating apparatus according to the present invention
  • FIG. 17 is a schematic structural view of a first embodiment of a peak cancellation device according to the present invention.
  • Embodiment 18 is a schematic structural diagram of Embodiment 1 of a prediction module according to the present invention.
  • Embodiment 19 is a schematic structural diagram of Embodiment 2 of a prediction module according to the present invention.
  • Embodiment 3 of a prediction module according to the present invention is a schematic structural diagram of Embodiment 3 of a prediction module according to the present invention.
  • 21 is a schematic structural diagram of an embodiment of a peak cancellation module according to the present invention.
  • FIG. 22 is a schematic structural view of a second embodiment of a peak cancellation device according to the present invention.
  • FIG. 23 is a schematic structural view of a third embodiment of a peak cancellation device according to the present invention.
  • FIG. 24 is a schematic structural view of a fourth embodiment of a peak cancellation device according to the present invention.
  • Embodiment 25 is a schematic structural view of Embodiment 5 of a peak cancellation device according to the present invention. detailed description
  • the peak point of the time-domain signal of the M-fold over-sample rate is predicted according to the N-time time domain signal of the 1x rate, and the position, amplitude and phase information of the peak point are included according to the characteristic information of each peak point, from M
  • the reference peak-sampling signal at point N that is, a multi-phase Kernel (Multi-Phase Kernel, hereinafter referred to as MP-Kernel) signal selects a suitable MP-Kernel signal for circumferential shift, amplitude adjustment, and phase rotation, and then The generated peak elimination signal corresponding to the peak point is superimposed with the signal of the 1 time rate time domain to achieve peak elimination of the time domain signal of 1 time rate.
  • MP-Kernel multi-phase Kernel
  • Step 101 Receive a time-domain signal of a 1 time rate.
  • Step 102 predict, according to the received time-domain signal of the 1 ⁇ rate, a peak point of the time domain signal of the M times the sampling rate, and obtain and record characteristic information of the peak point, including position information of the peak point, and a circumferential shift. Information, amplitude phase information.
  • Step 103 Select an appropriate MP-Kemel signal according to the characteristic information of each peak point, and perform circumferential shift, amplitude adjustment and phase rotation on the selected MP-Kemel signal to generate a peak-suppressing signal corresponding to each peak point.
  • Step 104 The peak signal of the received 1 ⁇ rate is peaked by using the peak elimination signal corresponding to each peak point.
  • the feature information of the peak point of the time domain signal of the M times over the sample rate is predicted according to the time domain signal of the 1 time rate, the reference peak elimination signal is selected according to the feature information, and the selected reference peak elimination signal is circumferentially shifted and amplitude
  • the adjustment and the phase rotation are performed, and then the peak-down signal corresponding to the peak point is used to peak the 1 time rate of the time domain signal, thereby realizing the peak cancellation of the signal at 1 times the sampling rate, and reducing the buffering
  • the buffer capacity requirement of the unit reduces the workload of the corresponding processing, the complexity is low, and the delay is small, thereby effectively improving the working efficiency of the peak cancellation device and improving the working performance.
  • the signal after the peak elimination can also be detected, that is, the amplitude of the peak-down signal, and whether the amplitude of the peak-slip signal is less than a preset amplitude threshold, or the cancellation Whether the number of peaks of the peak signal reaches a preset number of times threshold, if the judgment result is true, the peak elimination signal is directly output; otherwise, for the peak elimination signal, the process returns to step 102.
  • the amplitude threshold may be set according to the maximum peak allowed by the communication system, for example: the amplitude threshold may be set to the maximum allowed peak value or slightly smaller than the maximum allowed peak value, for example: set to 0.95 times the maximum allowable peak value. ;
  • the threshold can be set according to the empirical value, for example: The number of peaks can be any one of 2 to 32, and the number of peaks is not necessarily one, for example: 2 to 32 An integer in .
  • each subcarrier acquired from the base station may be used.
  • the peak elimination weight factor is used to obtain the MP-Kemel signal online, and the peak elimination weighting factor includes modulation coding scheme (hereinafter referred to as MC S ), constellation mapping mode, data importance degree, transmission power, channel condition and reception. Factors such as quality requirements.
  • MC S modulation coding scheme
  • the MP-Kemel signal can be pre-generated and stored for use in step 103 according to the fixed typical MCS information, which is also called offline determination. MP-Kemel signal. In order to meet the requirements of different protocols, more than one set of MP-Kemel signals can be obtained and stored offline.
  • the corresponding MP- The Kemel signal is also different, so at least the corresponding two sets of MP-Kemel signals need to be generated and stored.
  • Embodiment 2 of the peak cancellation method of the present invention includes the following steps:
  • Step 201 Generate an MP-Kemel signal according to a peak elimination weighting factor of each subcarrier.
  • Step 202 Receive and cache a time-domain signal of 1 time rate.
  • Step 203 predict a peak point of the time domain signal of the M times over the sampling rate according to the time domain signal of the 1 ⁇ rate, and obtain and record the feature information of each peak point, including the position information of the peak point and the circumferential shift information. , amplitude phase information.
  • step 202 to the step 203 may be performed simultaneously with the step 201, or may be performed before the step 201.
  • Step 204 Select an appropriate MP-Kemel signal from the generated MP-Kemel signal according to the feature information of each peak point, and perform circumferential shift, amplitude adjustment, and phase rotation, respectively, on the selected MP-Kemel signal, and generate and peak values.
  • the corresponding peak elimination signal is
  • Step 205 superimpose the peak elimination signal corresponding to each peak point with the buffered 1 time rate time domain signal, and perform peak elimination on the 1 time rate time domain signal in the buffer to obtain a peak elimination signal.
  • Step 206 Detect a peak amplitude and a number of peaks of the peak elimination signal.
  • Step 207 it is determined whether the peak amplitude of the peak elimination signal is less than a preset amplitude threshold, or whether the number of peak elimination times reaches a preset number of times threshold, if the determination result is true, then step 208 is performed; Otherwise, the peaking signal is buffered, and then for the peaking signal, step 203 is returned.
  • the peak elimination signal is buffered and the peak is cancelled again to ensure the peak elimination effect.
  • Step 208 directly outputting the received signal after the peak elimination.
  • the method of setting the amplitude threshold or the number threshold in step 207 may refer to step 206.
  • the MP-Kernel signal can be generated by the method of any of the embodiments of FIG. 3 and FIG. 6 to FIG. 8, and step 201 is implemented.
  • FIG. 3 it is a flowchart of Embodiment 1 of generating an MP-Kernel signal according to the present invention, which includes the following steps:
  • Step 301 Set weights for each subcarrier according to a peak elimination weighting factor of each subcarrier in the OFDM symbol, and generate a frequency domain weight signal.
  • the peak elimination weight factors include MCS, constellation mapping mode, data importance degree, transmission power, channel condition and reception quality requirement, which can be obtained from the base station.
  • the weights of the respective subcarriers can be obtained by the following formula:
  • the weight decreases as the order corresponding to the constellation mapping mode increases, that is, when the order corresponding to the constellation mapping mode is low, the weight value is used. Can be set larger, otherwise the opposite.
  • QPSK Quadrature phase-shift keying
  • QAM 16 Quadrature Amplitude Modulation
  • 64QAM three kinds of constellation mapping methods are sequentially increased.
  • the corresponding weight settings should be reduced in turn, for example, the weights can be set to 0.70, 0.31, and 0.15, respectively.
  • the weight decreases as the encoding rate of the subcarrier increases, that is, when the encoding rate of the subcarrier is low, the weight is weighted.
  • the value can be set larger, otherwise it is the opposite.
  • the weight of the 1/3 rate coding mode can be set higher than the weight of the 1/2 rate coding mode, and the weights of the 1/3 rate coding mode and the 1/2 rate coding mode can be set to 0.2 and respectively. 0.12.
  • the method of setting the weights of other peak elimination weighting factors is similar to the method of setting the weights of the encoding method and the constellation mapping method.
  • the minimum processing unit for peak elimination weight setting may be each subcarrier, or multiple subcarriers of the same or similar situation.
  • the weight of each subcarrier can also be set according to the frequency domain vector amplitude error (Error Vector Magnitude, hereinafter referred to as EVM).
  • EVM Error Vector Magnitude
  • Step 302 Perform OFDM subcarrier mapping on the frequency domain weight signal, and continuously insert (M-1) zeros in the frequency domain weight signal.
  • Step 303 Perform frequency domain to time domain transform on the frequency domain weight signal after zero insertion, for example: inverse fast Fourier transform (hereinafter referred to as IFFT), and generate a Kernel signal of the MxN point, also called: M times the quasi-reference peaking signal.
  • IFFT inverse fast Fourier transform
  • Step 304 Align the reference peak-sampling signal to perform amplitude adjustment, and normalize the highest amplitude to facilitate amplitude adjustment of the reference peak-sampling signal during the peak cancellation process to obtain a normalized M-fold-like quasi-reference. Peak elimination signal.
  • Step 305 grouping the normalized M-fold over-sample rate quasi-reference peak-sampling signals, and dividing the Kernel signal of the ⁇ point into M-point peak-off signals MP-Kernel of different starting phases, that is, :
  • the reference peaking signal which can store the MP-Kernel signal.
  • Kernel signal of the defect can be divided by:
  • FIG. 4 it is an example of the amplitude of four 1024-point MP-Kemel signals in which the 4096-point Kernel signal is divided according to the above method in the process of peak elimination in the present invention
  • FIG. 4 A partial enlarged view of the amplitude map of the MP-Kemel signal, where ' +, the point indicates Kernel ⁇ ' ⁇ ' point indicates Kernel 2 , '*, the point indicates Kerne", '., and the point indicates Kernel 4 .
  • a flowchart of Embodiment 2 of generating an MP-Kemel signal according to the present invention includes the following steps:
  • Step 401 Set a weight for each subcarrier according to a peak weighting factor of each subcarrier in the OFDM symbol, and generate a frequency domain weight signal. For a specific setting method, refer to step 301.
  • Step 402 Perform phase adjustment on the frequency domain weight signal by using each frequency domain phase adjustment factor.
  • the frequency domain phase adjustment factors of different initial phases can be set as follows:
  • w( , k) represents the phase adjustment factor on the kth subcarrier of the mth initial phase signal
  • Step 403 Perform frequency domain to time domain transform on the phase adjusted subcarrier, for example: IFFT, to generate a quasi-reference peak elimination signal.
  • Step 404 Align the reference peak elimination signal to perform amplitude adjustment, and normalize the highest amplitude to obtain a reference peak elimination signal MP-Kernel.
  • Step 405 Store the amplitude-adjusted MP-Kemel signal.
  • a flowchart of Embodiment 3 of generating an MP-Kemel signal according to the present invention includes the following steps:
  • Step 501 According to a peaking weighting factor of each subcarrier in the OFDM symbol, for each subcarrier The wave sets the weight and generates a frequency domain weight signal. For specific setting methods, refer to step 301.
  • Step 502 Perform frequency domain to time domain transform on the frequency domain weight signal, for example: IFFT, to generate a 1x rate quasi-reference peaking signal.
  • Step 503 Obtain a quasi-reference peak-sampling signal with a M-fold over-sampling rate from a quasi-reference peak-down signal of a 1x rate, for example: inserting (M-1) zeros after each sample point by interpolation filtering And perform filtering processing.
  • Step 504 Perform amplitude adjustment on the quasi-reference peaking signal of the M times the sampling rate, and normalize the highest amplitude.
  • Kernel i [Kernel (i) Kernel (i + M) ⁇ Kernel (i + (N _ 1) ⁇ M)], for the quasi-reference peak-sampling signal of the M-fold over-sampling rate Grouping, generating a reference peaking signal MP-Kemel signal, storing the MP-Kemel signal.
  • a flowchart of Embodiment 4 of generating an MP-Kemel signal according to the present invention includes the following steps:
  • Step 601 Set a weight for each subcarrier according to a peak elimination weighting factor of each subcarrier in the OFDM symbol, and generate a frequency domain weight signal.
  • Step 301 For the specific setting method, refer to step 301.
  • Step 602 Perform frequency domain to time domain transform on the frequency domain weight signal, for example: IFFT, to generate a 1x rate quasi-reference peaking signal.
  • Step 603 Perform amplitude adjustment on the quasi-reference peaking signal of 1 time rate, and normalize the highest amplitude.
  • Step 604 Using a polyphase filter with preset filter coefficients, respectively performing multiphase filtering of different phases on the quasi-reference peaking signal of 1 time rate to generate MP-Kemel.
  • the polyphase filtering for 0 phase can be omitted, and the signal before filtering can be directly maintained, that is: for the peak with phase 0
  • the signal can be directly replaced by a quasi-reference peak-down signal with a magnitude normalized amplitude.
  • the 64-order root raised cosine filter performs multiphase filtering.
  • the filter coefficients can be set as follows:
  • the root-lifting low-pass filter coefficients of M times over the sample rate can be generated by the FIRRCOS function in the Maltab software produced by Mathworks, USA.
  • 63 is the third group of polyphase filter coefficients
  • the fourth, 8, 12, 60, 64 are the fourth group of polyphase filter coefficients, and the amplitude of the specific filter coefficients can be appropriately adjusted as needed. At this time, filtering using the first group of polyphase filter coefficients can be omitted.
  • Step 605 Store the filtered MP-Kernel signal.
  • any one of the embodiments of FIG. 9 to FIG. 11 can be used to predict the feature information of the peak point of the time domain signal of the M times the sample rate according to the received time domain signal of the first rate, and implement step 102. Or step 203.
  • FIG. 9 is a flowchart of the first embodiment of the present invention, which is used to predict the peak value of the time-domain signal of the M-timed sample rate, and obtains and records the characteristic information of the peak point, which specifically includes the following steps:
  • Step 701 Perform interpolation filtering on the buffered 1 time rate time domain signal to obtain a time domain signal with an M times over the sampling rate.
  • Step 702 Calculate the square of the amplitude or amplitude of each sample point of the time domain signal of the M times the sample rate. Since each sample point is identified by a complex number, the amplitude is the time domain signal of the M times the sample rate. The sum of the square of the real part and the square of the imaginary part is re-opened, and the square of the amplitude is the sum of the square of the real part of the time-domain signal of the sample rate and the square of the imaginary part of the sample.
  • Step 703 The square of the amplitude or amplitude of each sample point according to the time domain signal of the M times the sampling rate. The peak point of the time domain signal of the M times over the sample rate is detected.
  • the peak value of the time-domain signal of the M-fold over-rate rate can be predicted by: pre-setting the square of the amplitude threshold ThresholdO or ThresholdO, ie: ThresholdO 2 , the amplitude threshold can be the system PAPR threshold, and the amplitude is greater than ThresholdO or A sample point having a magnitude larger than ThresholdO 2 and greater than the square of the amplitude or amplitude of the adjacent sample point is determined as the peak point.
  • Step 704 Calculate, according to the position information of the peak point, a sequence index m of the reference peak-suppressing signal MP-Kernel signal corresponding to each peak point. , the number of bits required for circumferential shift ⁇ ⁇ . , amplitude phase adjustment factor /.
  • the position information of the peak point, the circumferential shift information, that is, the position information of the circumferential shift, and the amplitude phase information, that is, the amplitude phase adjustment factor information constitute characteristic information of the peak point.
  • the feature information of the peak point can be stored for subsequent use.
  • FIG. 10 it is a flowchart of the second embodiment of the present invention for predicting the peak value of the time domain signal of the ⁇ times the sample rate, and calculating and recording the characteristic information of the peak point, which specifically includes the following steps:
  • Step 801 Obtain a time-domain signal that is multiplied by the time rate signal of the buffered 1 time rate.
  • the time domain signal can be obtained by multiplying the sample rate by inserting (M-1) zeros after each sample point of the 1 time rate time domain signal.
  • Step 802 Perform coarse filtering on the time-domain signal of the ⁇ times sample rate, that is, use a filter with a lower order, for example, a filter of 16th order, for filtering.
  • Step 803 Calculate the amplitude or amplitude of each sample point of the time domain signal after the coarse filtering Square, that is: the sum of the square of the real part and the square of the imaginary part, or the sum of the square of the real part and the square of the imaginary part.
  • Step 804 the selected amplitude is greater than (/X Threshold ⁇ ) or the square of the amplitude is greater than
  • Step 805 Perform fine filtering on the peak point, that is, filter with a higher order than the filter used in step 802, for example, a 64-order filter.
  • Step 806 Select, according to the filtering result of the quasi-peak point, the peak point of the time domain signal from the quasi-peak point. For the method of specifically selecting the peak point, refer to step 703.
  • Step 807 Calculate, according to the position information of the peak point, a sequence index m of the reference peak-suppressing signal MP-Kemel signal corresponding to the peak point. , the number of bits required for circumferential shift ⁇ ⁇ . , amplitude phase adjustment factor /. . For a specific calculation method, reference may be made to step 704.
  • the second embodiment firstly performs low-order filtering on the time-domain signal of the sample rate, and initially detects possible peak points, and then performs fine filtering on the possible peak points, for example: The 16th-order coarse filtering, and then the 64th or 128th order fine filtering, does not perform high-order filtering for each sample point, but only performs high-order filtering on the possible peak points determined after coarse filtering, which simplifies information processing.
  • the complexity reduces the amount of information processing, shortens the detection time of the feature information of the peak point, and improves the work efficiency.
  • FIG. 11 it is a flowchart of the third embodiment of the present invention for predicting the peak value of the time domain signal of the sample rate, and calculating and recording the characteristic information of the peak point, which specifically includes the following steps:
  • Step 901 Obtain a time-domain signal that is multiplied by the time-domain signal of the buffered 1 time rate. Specifically, by inserting (M-1) zeros after each sample point of the 1 time rate time domain signal by interpolation filtering, the time domain signal of the ⁇ times the sample rate can be obtained.
  • Step 902 Perform coarse filtering on the M-time over-rate rate time domain signal, that is, use a filter with a lower order, for example, a 16-order filter to perform filtering.
  • Step 903 Calculate the square of the amplitude or amplitude of each sample point, that is, the sum of the square of the real part and the square of the imaginary part, and the sum of the root number or the square of the real part and the square of the imaginary part.
  • Step 905 Cache the quasi-peak information, including the position information of the quasi-peak point and the real part and the imaginary part value of the quasi-peak point after the coarse filtering; or the quasi-peak information may not be cached, and the subsequent steps are directly performed.
  • Step 906 according to the position information of the quasi-peak point, align the peak point to perform supplementary precision filtering, for example: if the coarse filtering is 16th-order filtering, the effect of 64-order filtering is required, and the 64-16-order tap that is not calculated by the coarse filtering is calculated.
  • the coefficient is obtained to obtain the compensation value of the real part and the imaginary part of the quasi-peak point after the supplementary precision filtering.
  • Step 907 superimposing the real part and the imaginary part value of the quasi-peak point obtained by the buffer coarse filtering with the compensation value obtained by filtering the supplementary precision to obtain more accurate peak information, so as to further select the true peak point from the subsequent steps.
  • the peak point
  • Step 908 according to the more accurate peak information obtained in step 907, selecting a peak point from the quasi-peak point, the peak point can be selected by: presetting the amplitude threshold 73 ⁇ 4re oW0 and its square, ie: ThresholdQ 1 , the amplitude It may be the amplitude corresponding to the PAPR threshold of the system, and the amplitude of the sample larger than the threshold of the adjacent sample point is determined as the peak point, or the square of the amplitude is greater than the Threshold Q 2 , and is larger than the square of the amplitude of the adjacent sample point. The sample point is determined as the peak point.
  • Step 909 calculating an MP-Kernel signal corresponding to the peak point according to the position information of the peak point.
  • the serial number index the number of bits that need to be shifted circumferentially. , amplitude phase adjustment factor /. .
  • the fine filter can omit the operation corresponding to the coarse filter in the coarse filter, and the second embodiment shown in FIG. In comparison, less information is processed and implementation is simpler.
  • step 104 or step 204 the specific method for peak-removing the received time-domain signal of 1 time rate by using the peak-reduction signal corresponding to the peak point of the ⁇ multiplication rate is as follows:
  • the sequence index m of the peak point MP-Kemel signal calculated according to each of the above embodiments. Select the appropriate MP-Kemel signal from the M MP-Kemel signals, and then according to the characteristic information of each peak point, that is, the number of bits i required for the circumferential shift. And amplitude phase adjustment factor f. Performing corresponding circumferential shift, amplitude adjustment and phase rotation on the selected MP-Kemel signal, generating a peak-suppressing signal corresponding to each peak point, and then superimposing with the buffered 1 time-rate time domain signal, The received 1 time rate of the time domain signal is peaked. In the specific implementation, one or more peaks can be cancelled each time.
  • the appropriate MP-Kemel signal can be selected from the M MP-Kemel signals by the following method:
  • a d + 1 MP-Kemel signal in the -Kemel signal as a reference peaking signal of the peak point of the time-domain signal of the M-fold rate that is, a suitable MP-Kemel signal
  • d is the M
  • the number of sample points separated by the peak point of the time domain signal of the sample rate and the time domain signal of the adjacent 1 time rate, d 0, 1 , . . . , ⁇ -1.
  • one or more peak points can be eliminated each time. It is also possible to eliminate all peak points, and the specific number of peak points to be eliminated each time can be combined with the PAPR threshold and the number of peaks. For example, for an OFDM system with a single carrier 1024-point IFFT, when the PAPR threshold is 7 dB, if four peaks are eliminated in total, it can be specified that 8 peak points are eliminated each time.
  • the above embodiment of the present invention uses the N-point MP-Kernel for peak cancellation at a rate of 1 times, compared with the conventional scheme of peak cancellation using the Kernel signal of the defect at the M-fold sampling rate.
  • the peak effect has not decreased, as explained below:
  • the M-time over-sampling rate of the traditional scheme can be expressed as follows:
  • PMN(k) For the reserved subcarrier (Tone Reservation, hereinafter referred to as TR) scheme, in the reserved subcarrier position, the value is not 0, and the remaining points are all 0; for Peak Cancelling (hereinafter referred to as PC) scheme, On the data subcarriers, all are 1 and are 0 on the remaining subcarriers.
  • PC Peak Cancelling
  • AWPC Adaptive Weight Peak Cancelling
  • the peak elimination process is the use of 1 ⁇ 2 ⁇ different circumferential shift, amplitude and phase adjusted signals, and the process of superimposing with the original signal, and ⁇ ) is the position of the peak point, that is, the Kernel signal when the peak point is cancelled ;?
  • the above superposition process only causes the change of the first N/2 data of W and the last N/2 data.
  • ⁇ . ⁇ For the amplitude and phase adjustment coefficients, for the frequency domain signal corresponding to the N-point Kernel signal at the original 1x rate, the new N-point Kernel after the corresponding phase rotation, e N is the N-point Kernel signal at the original 1x rate.
  • the coefficient of the circumferential shift, the new Kernel signal after the phase rotation is called the MP-Kemel signal.
  • the MP-Kernel signal is different from the Kernel signal of the sample rate.
  • the Kernel signal is sampled for the starting interval M, and the process of peak elimination using MP-Kernel is the same as the traditional frequency-domain generation using the Kenrel signal using the M-fold sample rate, that is, the present invention is
  • the technique of peak cancellation using the N-point MP-Kernel at 1x rate is the same as the conventional scheme of peak cancellation using the Kernel signal of the defect at the M-fold rate. .
  • FIG. 13 is a schematic structural diagram of Embodiment 1 of a reference peak-sampling signal generating apparatus according to the present invention, which can be used to implement the embodiment shown in FIG. 3.
  • the embodiment of the reference peak-sampling signal generating apparatus includes:
  • a weight setting module configured to set a weight for each subcarrier according to a peak weighting factor of each subcarrier in the OFDM symbol, and generate a frequency domain weight signal
  • a zero insertion module configured to perform OFDM subcarrier mapping on the frequency domain weight signal, and continuously insert (M-1) zeros in the frequency domain weight signal;
  • the frequency-time conversion module is configured to perform frequency-domain to time-domain transformation on the frequency-domain weight signal after the interpolation, and obtain a quasi-reference peak-sampling signal with M times of the same;
  • the amplitude adjustment module is configured to adjust the amplitude of the reference peak-sampling signal, and normalize the highest amplitude to obtain a normalized M-fold-like quasi-reference peak-sampling signal;
  • a grouping module configured to group the normalized M times the sample rate of the quasi-reference peaking signal, and divide the normalized M times the quasi-reference peaking signal into M points of different starting phases The base peaking signal at point N.
  • FIG. 14 is a schematic structural diagram of Embodiment 2 of a reference peak-sampling signal generating apparatus according to the present invention. It can be used to implement an embodiment as shown in FIG. 6, the embodiment of the reference peaking signal generating apparatus comprising:
  • a weight setting module configured to set weights for each subcarrier according to a peak cancellation weighting factor of each subcarrier in the OFDM symbol, to generate a frequency domain weight signal
  • a frequency-time conversion module configured to perform frequency-domain to time-domain transformation on the phase-adjusted frequency-domain weight signal to generate a quasi-reference peak-down signal
  • the amplitude adjustment module adjusts the amplitude of the reference peak elimination signal to normalize the highest amplitude to obtain a reference peak elimination signal.
  • FIG. 15 is a schematic structural diagram of Embodiment 3 of a reference peak-sampling signal generating apparatus according to the present invention, which may be used to implement the embodiment shown in FIG. 7.
  • the embodiment of the reference peak-sampling signal generating apparatus includes:
  • a weight setting module configured to set a weight for each subcarrier according to a peak weighting factor of each subcarrier in the OFDM symbol, and generate a frequency domain weight signal
  • a frequency-time conversion module configured to perform frequency domain to time domain transformation on a frequency domain weight signal, to obtain a quasi-reference peak-down signal of 1 time rate
  • a zero-insertion filter module for obtaining a quasi-reference peak-sampling signal with a M-fold over-sample rate from a quasi-reference peak-down signal of 1 ⁇ rate;
  • the amplitude adjustment module is configured to perform amplitude adjustment on the quasi-reference peak-sampling signal of the M-fold sampling rate, and normalize the highest amplitude; a grouping module, configured to group the normalized M times the sample rate of the quasi-reference peaking signal, and divide the normalized M times the sample rate of the quasi-reference peaking signal into M of different starting phases A baseline extinction signal at point N.
  • FIG. 16 is a schematic structural diagram of Embodiment 4 of the reference peak-sampling signal generating apparatus of the present invention, which can be used to implement the embodiment shown in FIG. 8.
  • the embodiment of the reference peak-sampling signal generating apparatus includes:
  • a weight setting module configured to set a weight for each subcarrier according to a peak weighting factor of each subcarrier in the OFDM symbol, and generate a frequency domain weight signal
  • a frequency-time conversion module configured to perform frequency domain to time domain transformation on a frequency domain weight signal, to obtain a quasi-reference peak-down signal of 1 time rate
  • the amplitude adjustment module is configured to perform amplitude adjustment on the quasi-reference peak-sampling signal of 1 ⁇ rate, and normalize the highest amplitude;
  • the multi-phase filtering module is configured to perform multi-phase filtering of different phases of the quasi-reference peak-spinning signal with a magnitude normalized amplitude to generate a reference peak-canceling signal.
  • the polyphase filtering module may be composed of a plurality of filters of different filter coefficients.
  • the first storage module may be included for storing the reference peak-sampling signal finally generated by the reference peak-sampling signal generating means for subsequent selection and use.
  • FIG. 17 is a schematic structural diagram of Embodiment 1 of a peak cancellation device according to the present invention, which can be used to implement the embodiment shown in FIG. 1.
  • the embodiment of the peak cancellation device includes sequentially connected:
  • a prediction module configured to predict characteristic information of a peak point of the time domain signal of the M times the sampling rate according to the time domain signal of the 1 time rate;
  • the peak cancellation module is configured to select a reference peak elimination signal according to the characteristic information of the peak point, and perform circumferential shift, amplitude adjustment and phase rotation on the selected reference peak elimination signal to generate a peak elimination signal corresponding to the peak point, and The 1 time rate time domain signal is peaked by the peak elimination signal corresponding to the peak point.
  • FIG. 18 it is a schematic structural diagram of Embodiment 1 of the prediction module of the present invention, which can be used to implement the embodiment shown in FIG. 9.
  • the prediction module embodiment includes sequentially connected:
  • the interpolation filtering unit is configured to perform interpolation filtering on the time-domain signal of the 1 ⁇ rate to obtain a time domain signal with a M times over the sampling rate;
  • a first calculating unit configured to calculate a square of the amplitude or amplitude of each sample point of the time domain signal of the M times the sample rate
  • a detecting unit configured to detect a position of a peak point of the time domain signal of the M times over the sampling rate according to the square of the amplitude or amplitude of each sample point of the time domain signal of the M times over the sampling rate;
  • a second calculating unit configured to calculate, according to the position information of the peak point, a sequence index of the reference peak-spinning signal corresponding to the peak point, a number of bits of the circumferential shift, and an amplitude phase adjustment factor, where the feature information of the peak point includes a peak point Position information, sequence number index, number of bits of the horizontal shift and amplitude phase adjustment factor:
  • the detecting unit and the second calculating unit are respectively connected to the peak cancellation module to provide characteristic information of the peak point required.
  • FIG. 19 it is a schematic structural diagram of Embodiment 2 of the prediction module of the present invention, which can be used to implement the embodiment shown in FIG. 10.
  • the prediction module embodiment includes:
  • a zero insertion unit for interpolating a time rate signal of 1 time rate into a time domain signal of M times the sampling rate
  • a first filtering unit configured to perform coarse filtering on the time domain signal of the M times the sampling rate
  • a first calculating unit configured to calculate a square of the amplitude or amplitude of each sample point of the time domain signal of the M times the sample rate after the coarse filtering
  • a first selection unit for selecting a magnitude greater than (; x 73 ⁇ 4r ⁇ AoW 0) or a square of magnitude greater than
  • a second filtering unit configured to align the peak points for fine filtering
  • a second selecting unit configured to select, according to a result of the fine filtering, a peak point of the time domain signal of the M times the sampling rate from the quasi-peak points;
  • a second calculating unit configured to calculate, according to the position information of the peak point, a sequence index of the reference peak-spinning signal corresponding to the peak point, a number of bits of the circumferential shift, and an amplitude phase adjustment factor, where the feature information of the peak point includes a peak point Position information, sequence number index, number of bits of the horizontal shift and amplitude phase adjustment factor:
  • the second selection unit and the second calculation unit are respectively connected to the peak cancellation module to provide characteristic information of peak points required.
  • FIG. 20 it is a schematic structural diagram of Embodiment 3 of the prediction module of the present invention, which can be used to implement the embodiment shown in FIG. 11.
  • the prediction module embodiment includes:
  • a zero insertion unit for interpolating a time rate signal of 1 time rate into a time domain signal of M times the sampling rate
  • a first filtering unit configured to perform coarse filtering on the time domain signal of the M times the sampling rate
  • a first calculating unit configured to calculate a square of the amplitude or amplitude of each sample point of the time domain signal of the M times the sample rate after the coarse filtering
  • a first selection unit for selecting a magnitude greater than (; x73 ⁇ 4r ⁇ AoW0) or a square of magnitude greater than
  • a third filtering unit configured to align the peak points for supplementary precision filtering
  • a third calculating unit configured to superimpose the value of the quasi-peak point obtained by the coarse filtering and the compensation value of the real part and the imaginary part obtained by the supplementary precision filtering;
  • a third selecting unit configured to select, according to the superposition result, a peak point of the time domain signal that is multiplied by the sampling rate from the quasi-peak point;
  • a second calculating unit configured to calculate, according to position information of the peak point, a sequence index of the reference peak-spinning signal corresponding to the peak point, a number of bits of the circumferential shift, and an amplitude phase adjustment factor, where the characteristic information of the peak point includes a peak value
  • the third selection unit and the second calculation unit are respectively connected to the peak cancellation module to provide characteristic information of peak points required.
  • the prediction module embodiment shown in FIG. 18 to FIG. 20 may further include a second storage module, configured to store feature information of a peak point, and the peak cancellation module is connected to the second storage module, from the second storage module. Select the feature information of the appropriate peak point.
  • the second The storage module is respectively connected to the detecting unit and the second calculating unit, and stores the feature information of the corresponding peak point to the second storage module for selection by the peak cancellation module; in the embodiment shown in FIG. 19, the second selecting unit and The second computing unit is respectively connected to the second storage module, and stores the feature information of the corresponding peak point to the second storage module for selection by the peak cancellation module; in the embodiment shown in FIG. 20, the third selection unit and the The second computing unit is respectively connected to the second storage module, and stores the feature information of the corresponding peak point to the second storage module for selection by the peak cancellation module.
  • FIG. 21 is a schematic structural diagram of an embodiment of a peak cancellation module according to the present invention, which can be used in corresponding steps in the embodiment shown in FIG. 1 and FIG. 2.
  • the embodiment of the peak cancellation module includes sequential connections:
  • a selection unit configured to be connected to the prediction module, configured to select a reference peak cancellation signal according to the feature information of the peak point generated or further stored by the prediction module in the foregoing embodiment, where the reference peak cancellation signal may be obtained by using the reference cancellation provided by any of the foregoing embodiments.
  • the peak signal generating device is generated online, and may be generated and stored in advance;
  • a circumferential shifting unit configured to perform circumferential shifting on the selected reference peaking signal according to the characteristic information of the peak point
  • the amplitude and phase adjustment unit is configured to perform amplitude adjustment and phase rotation on the selected reference peak elimination signal according to the characteristic information of the peak point, and generate a peak elimination signal corresponding to the peak point;
  • the peak elimination unit is configured to perform peak elimination on the received original 1 time rate time domain signal by using a peak elimination signal corresponding to the peak point.
  • the selecting unit may include:
  • the first selecting unit is configured to select, from each of the M reference peak-spinning signals, the d + 1 reference peak-sampling signal as a reference peak-sampling signal of the peak point of the M-time over-time-rate time domain signal.
  • FIG. 22 it is a schematic structural diagram of a second embodiment of a peak cancellation device according to the present invention.
  • the peak cancellation device of the embodiment further includes: a cache module, and a prediction module and a peak cancellation, respectively.
  • the module is connected to receive and buffer the received time rate signal of 1 time rate, and the prediction module is configured to predict the peak value of the time domain signal of the M times the sampling rate according to the time domain signal buffered at the rate of 1 times buffered in the buffer module. Feature information of the point.
  • the cache module may be connected to the interpolation filtering unit or the zero insertion unit in the prediction module, and input a time-domain signal of 1 ⁇ rate; the cache module may be connected to the peak elimination unit in the peak cancellation module, and provided thereto Time domain signal at 1 rate of peak to be peaked.
  • FIG. 23 is a schematic structural diagram of Embodiment 3 of a peak cancellation device according to the present invention, which can be used to implement the embodiment shown in FIG. 2, and the peak cancellation device of the embodiment is based on Embodiment 2, Includes:
  • a detection module connected to the peak cancellation module or the peak elimination unit thereof, for detecting a peak amplitude and a peak elimination number of the peak elimination signal outputted by the peak cancellation module;
  • the judging module is respectively connected to the detecting module and the buffer module, and is configured to determine whether the peak amplitude is less than a preset amplitude threshold, or whether the number of peak erasures reaches a preset number threshold, and if the judgment result is true, the peak is directly output. Signal; otherwise, the peak elimination signal is sent to the cache module for the next peak elimination.
  • the reference peak-reduction signal required in the embodiment shown in Figs. 17 to 23 of the present invention may be generated and stored in advance, or may be generated in real time online.
  • the reference peak cancellation signal generating apparatus according to any one of the above embodiments of the present invention may be provided in the embodiment of the above-described peak cancellation apparatus according to the present invention, and the grouping in the reference peak elimination signal generating apparatus
  • the module, the amplitude adjustment module, the multi-phase filtering module or the first storage module is connected to the peak cancellation module or the selected unit or the first selection unit therein, for providing a reference peak elimination signal required by the peak cancellation module, such as FIG. 24 is a schematic structural view of Embodiment 4 of the peak cancellation device of the present invention.
  • the first memory module may be disposed in the embodiment of the peak cancellation device of the present invention to store a reference peak cancellation signal required by the pre-generated peak cancellation module, such as Figure 25 is a schematic diagram showing the structure of the fifth embodiment of the peak cancellation device of the present invention. Figure.
  • the peak cancellation method or device provided by the above embodiments of the present invention is also applicable to other multi-carrier communication systems that need to reduce the PAPR or reduce the crest factor (hereinafter referred to as CF), and are also applicable to the multi-carrier OFDM system. Peak suppression.
  • the method includes the steps of the foregoing method embodiments; and the foregoing storage medium includes: a medium that can store program codes, such as a ROM, a RAM, a magnetic disk, or an optical disk.
  • the feature information of the peak point of the time domain signal of the M times over the sample rate is predicted according to the time domain signal of the 1 time rate, and the reference peak elimination signal is selected according to the feature information, and the selected reference peak elimination signal is circled. Shift, amplitude adjustment and phase rotation, and then use the generated peak-reduction signal corresponding to the peak point to peak the 1 time rate of the time domain signal, achieving peak cancellation of the signal at 1 times the sample rate.
  • the peak cancellation is performed at the M times the sampling rate, and the buffer capacity requirement of the buffer unit is reduced without affecting the performance of the PAPR suppression, thereby reducing the workload of the corresponding processing, which is complicated.
  • the low degree and low time delay effectively improve the working efficiency of the peak cancellation device and improve its working performance.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Transmitters (AREA)
  • Digital Transmission Methods That Use Modulated Carrier Waves (AREA)
  • Synchronisation In Digital Transmission Systems (AREA)

Description

峰值对消方法、 峰值对消装置与基准消峰信号生成装置 本申请要求于 2007年 10月 26日提交中 国专利局、 申请号为 200710176420.5、 发明名称为 "峰值对消方法、 峰值对消装置与基准消峰信 号生成装置" 的中国专利申请的优先权, 其全部内容通过引用结合在本申请 中。 技术领域
本发明涉及通信技术, 尤其是一种峰值对消方法、 峰值对消装置与基准 消峰信号生成装置。 背景技术
正交频分复用 ( Orthogonal Frequency Division Multiplexing , 以下简称: OFDM )技术是一种有效的宽带传输技术, 己被广泛应用到数字音频广播 ( Digital Audio Broadcast, 以下简称: DAB ) 、 数字视频广播 ( Digital Video Broadcast, 以下简称: DVB ) 、 高清晰度电视( High Definition Television, 以下简称: HDTV )等诸多领域。 然而, 在釆用 OFDM技术的系统中, 由于信 号的峰值功率较高, 从而使得信号的峰值功率与平均功率的比值, 即: 峰均 功率比( Peak-to-Average Power Ratio , 以下简称: PAPR )较高, 这增加了系 统实现的难度和成本。
为了降低 OFDM通信系统中信号的峰值功率, 现有技术中, 预先存储或 者现场生成一个 M倍过釆样速率的近似理想脉冲的峰值对消波形, 也即: M 倍过釆样速率的 Kernel信号。现有技术利用 Kernel信号进行峰值对消的方案在 消峰过程中, 通常都需要在 4倍或者 4倍以上的过釆样速率下进行, 这就要求 消峰时时域信号为 4倍或者 4倍以上过釆样速率的时域信号, Kernel信号为 4倍 或者 4倍以上过釆样速率的 Kernel信号。 在实现本发明的过程中, 发明人发现现有技术进行峰值对消时, 至少存 在如下问题:
现有技术利用 Kenrel信号进行峰值对消的技术方案在消峰过程中, 需要 在 M倍过釆样速率下进行, 进行相应处理的复杂度较高, 并且延时较大, 从 而降低了工作效率。 发明内容
本发明实施例提供一种峰值对消方法、峰值对消装置与基准消峰信号 生成装置, 实现在进行峰值对消时降低处理复杂度、 缩短时延, 提高工作效 率。
根据本发明实施例的第一个方面, 提供的一种峰值对消方法, 包括: 接收 1倍速率的时域信号;
根据所述 1倍速率的时域信号,预测 M倍过釆样速率的时域信号的峰值 点的特征信息;
根据所述峰值点的特征信息, 选取基准消峰信号, 并对选取的基准消峰 信号进行圓周移位、 幅度调整与相位旋转, 生成与所述峰值点对应的消峰信 号;
利用与所述峰值点对应的所述消峰信号对所述 1倍速率的时域信号进行 消峰。
根据本发明实施例的第二个方面, 提供的一种峰值对消装置, 包括: 预测模块, 用于根据 1倍速率的时域信号, 预测 M倍过釆样速率的时域 信号的峰值点的特征信息;
峰值对消模块, 用于根据所述峰值点的特征信息, 选取基准消峰信号, 并对选取的基准消峰信号进行圓周移位、 幅度调整与相位旋转, 生成与所述 峰值点对应的消峰信号, 并利用与所述峰值点对应的所述消峰信号对所述 1 倍速率的时域信号进行消峰。 本发明实施例根据 1倍速率的时域信号来预测 M倍过釆样速率的时域 信号的峰值点的特征信息, 根据该特征信息选取基准消峰信号, 对选取的基 准消峰信号进行圓周移位、 幅度调整与相位旋转, 然后利用生成的与峰值点 对应的消峰信号对 1倍速率的时域信号进行消峰, 实现了在 1倍过釆样速率下 对信号的峰值对消, 与现有技术在 M倍过釆样速率下进行峰值对消相比, 降 低了对緩存单元的緩存容量要求, 减少了相应处理的工作量, 复杂度低, 延 时小, 从而有效改善了峰值对消装置的工作效率, 提高了其工作性能。
根据本发明实施例的第三个方面, 提供的一种基准消峰信号生成装置, 包括:
权重设置模块, 用于根据正交频分复用符号内各个子载波的消峰权重因 子, 为各个子载波设定权重, 生成频域权重信号;
插零模块, 用于对所述频域权重信号进行正交频分复用子载波映射, 在 所述频域权重信号中连续插入(M-1 ) xN个零;
频时转换模块, 用于对插零后的频域权重信号进行频域到时域的变换, 得到 M倍过釆样的准基准消峰信号;
幅度调整模块, 用于对所述准基准消峰信号进行幅度调整, 将最高幅度 归一化, 得到归一化的 M倍过釆样的准基准消峰信号;
分组模块, 用于对所述归一化的 M倍过釆样速率的准基准消峰信号进行 分组, 将所述归一化的 M倍过釆样的准基准消峰信号划分成不同起始相位的 M个 N点的基准消峰信号。
根据本发明实施例的第四个方面,提供的另一种基准消峰信号生成装置, 包括:
权重设置模块, 用于根据正交频分复用符号内各个子载波的消峰权重因 子, 对各个子载波设定权重, 生成频域权重信号;
相位调整模块, 用于分别利用由
Figure imgf000006_0001
设定的各频域相位调整因子对所述的频域权重信号进行相位调整,其中 , w( ,k)表示第 m个初始相位的信号第 k个子载波上的相位调整因子, m = 1 , M
2, · .. , M;
频时转换模块, 用于对相位调整后的频域权重信号进行频域到时域的变 换, 生成准基准消峰信号;
幅度调整模块, 对所述准基准消峰信号进行幅度调整, 将最高幅度归一 化, 得到基准消峰信号。
根据本发明实施例的第五个方面,提供的又一种基准消峰信号生成装置, 包括:
权重设置模块, 用于根据正交频分复用符号内各个子载波的消峰权重因 子, 为各个子载波设定权重, 生成频域权重信号;
频时转换模块, 用于对所述频域权重信号进行频域到时域的变换, 得到 1倍速率的准基准消峰信号;
插零滤波模块,用于由所述 1倍速率的准基准消峰信号获得 M倍过釆样 速率的准基准消峰信号;
幅度调整模块,用于对所述 M倍过釆样速率的准基准消峰信号进行幅度 调整, 将最高幅度归一化;
分组模块,用于对所述归一化的 M倍过釆样速率的准基准消峰信号进行 分组, 将所述归一化的 M倍过釆样速率的准基准消峰信号划分成不同起始相 位的 M个 N点的基准消峰信号。
根据本发明实施例的第六个方面,提供的再一种基准消峰信号生成装置, 包括:
权重设置模块, 用于根据正交频分复用符号内各个子载波的消峰权重因 子, 为各个子载波设定权重, 生成频域权重信号;
频时转换模块, 用于对所述频域权重信号进行频域到时域的变换, 得到
1倍速率的准基准消峰信号;
幅度调整模块, 用于对所述 1倍速率的准基准消峰信号进行幅度调整, 将最高幅度归一化;
多相滤波模块,用于对幅度归一化的 1倍速率的准基准消峰信号进行不同 相位的多相滤波, 生成基准消峰信号。
本发明上述实施例提供的基准消峰信号生成装置, 可以根据 1倍速率的 准基准消峰信号生成 M个 N点的基准消峰信号, 以便进行圓周移位、 幅度调 整与相位旋转后 1倍速率的时域信号进行消峰使用。 附图说明
图 1为本发明峰值对消方法实施例一的流程图;
图 2为本发明峰值对消方法实施例二的流程图;
图 3为本发明生成 MP-Kernel信号实施例一的流程图;
图 4为本发明实施例在消峰过程中釆用的 4个 1024点时的 MP-Kernel 信号一个幅度示例图;
图 5为图 4中 MP-Kernel信号的幅度示例图的局部放大图;
图 6为本发明生成 MP-Kernel信号实施例二的流程图;
图 7为本发明生成 MP-Kernel信号实施例三的流程图;
图 8为本发明生成 MP-Kernel信号实施例四的流程图;
图 9为本发明预测 M倍过釆样速率的时域信号的峰值点,求取并记录峰 值点的特征信息实施例一的流程图;
图 10为本发明预测 M倍过釆样速率的时域信号的峰值点, 求取并记录 峰值点的特征信息实施例二的流程图;
图 11为本发明预测 M倍过釆样速率的时域信号的峰值点, 求取并记录 峰值点的特征信息实施例三的流程图;
图 12-1至图 12-4为 M = 4, N = 1024时, 分别选取第 1-4个 MP-Kernel 信号的峰值点的分布示意图;
图 13为本发明基准消峰信号生成装置实施例一的结构示意图;
图 14为本发明基准消峰信号生成装置实施例二的结构示意图;
图 15为本发明基准消峰信号生成装置实施例三的结构示意图;
图 16为本发明基准消峰信号生成装置实施例四的结构示意图;
图 17为本发明峰值对消装置实施例一的结构示意图;
图 18为本发明预测模块实施例一的结构示意图;
图 19为本发明预测模块实施例二的结构示意图;
图 20为本发明预测模块实施例三的结构示意图;
图 21为本发明峰值对消模块实施例的结构示意图;
图 22为本发明峰值对消装置实施例二的结构示意图;
图 23为本发明峰值对消装置实施例三的结构示意图;
图 24为本发明峰值对消装置实施例四的结构示意图;
图 25为本发明峰值对消装置实施例五的结构示意图。 具体实施方式
本发明实施例根据 1倍速率的 N点时域信号预测出 M倍过釆样速率时域 信号的峰值点, 根据各个峰值点特征信息, 包括峰值点的位置、 幅度与相位 信息, 从 M个 N点的基准消峰信号, 即: 多相 Kernel ( Multi-Phase Kernel , 以下简称: MP-Kernel )信号中选取一个合适的 MP-Kernel信号, 进行圓周移 位、 幅度调整与相位旋转, 然后将生成的与峰值点对应的消峰信号与 1倍速 率时域的信号进行叠加, 实现对 1倍速率的时域信号的消峰。
如图 1 所示, 为本发明峰值对消方法实施例一的流程图, 其包括以下步 骤: 步骤 101 , 接收 1倍速率的时域信号。
步骤 102, 根据接收到的 1倍速率的时域信号, 预测 M倍过釆样速率的 时域信号的峰值点, 求取并记录峰值点的特征信息, 包括峰值点的位置信息、 圓周移位信息、 幅度相位信息。
步骤 103 , 根据各个峰值点特征信息, 选取合适的 MP-Kemel信号, 并对 选取的 MP-Kemel信号进行圓周移位, 幅度调整与相位旋转, 生成与各个峰 值点对应的消峰信号。
步骤 104,利用各个峰值点对应的消峰信号对接收到的 1倍速率的时域信 号进行消峰。
根据 1倍速率的时域信号来预测 M倍过釆样速率的时域信号的峰值点的 特征信息, 根据该特征信息选取基准消峰信号, 对选取的基准消峰信号进行 圓周移位、 幅度调整与相位旋转, 然后利用生成的与峰值点对应的消峰信号 对 1倍速率的时域信号进行消峰, 实现了在 1倍过釆样速率下对信号的峰值 对消, 降低了对緩存单元的緩存容量要求, 减少了相应处理的工作量, 复杂 度低, 延时小, 从而有效改善了峰值对消装置的工作效率, 提高了其工作性 能。
在对 1倍速率的时域信号进行消峰后, 还可以检测消峰后的信号, 即: 消 峰信号的幅度, 并判断消峰信号的幅度是否小于预先设定的幅度阈值, 或者 该消峰信号的消峰次数是否达到预先设定的次数阈值, 若判断结果为真, 则 直接输出消峰信号; 否则, 针对该消峰信号, 返回执行步骤 102。
其中, 可以根据通信系统所允许的最大峰值设定幅度阈值, 例如: 可将 该幅度阈值设定为允许的最大峰值或略小于允许的最大峰值, 例如: 设定为 允许的最大峰值的 0.95倍; 可以根据经验值设定次数阈值, 例如: 其中的消 峰次数可以为 2到 32中的任一个整数, 每次消峰的个数也不一定为 1个, 例如: 也可以是 2到 32中的某一个整数。 在图 1所示实施例的步骤 103之前, 可以根据从基站获取的各个子载波 的消峰权重因子在线求取 MP-Kemel信号, 其中的消峰权重因子包括调制编 码方式( modulation coding scheme , 以下简称: MC S )、 星座映射方式、 数据 重要程度、 发射功率、 信道情况以及接收质量要求等因子。 另外, 为了实现 简单,节省在线求取 MP-Kemel信号所需的时间,也可以根据固定典型的 MCS 信息, 预先生成 MP-Kemel信号并存储, 以供步骤 103使用, 该方法也称离 线求取 MP-Kemel信号。 为了适应不同协议的要求, 可以离线求取并存储一 组以上的 MP-Kemel信号,例如:对于 IEEE 802.16e等协议,由于奇数 OFDM 符号与偶数 OFDM符号上的导频位置不同, 相应的 MP-Kemel信号也不同, 因此, 至少需要生成并存储相应的两组 MP-Kemel信号。
如图 2所示, 为本发明峰值对消方法实施例二的流程图, 其包括以下步 骤:
步骤 201 , 根据各个子载波的消峰权重因子生成 MP-Kemel信号。
步骤 202, 接收并緩存 1倍速率的时域信号。
步骤 203 , 根据 1倍速率的时域信号, 预测 M倍过釆样速率的时域信号 的峰值点, 并求取、 记录各个峰值点的特征信息, 包括峰值点的位置信息、 圓周移位信息、 幅度相位信息。
另外, 该步骤 202〜步骤 203也可以与步骤 201同时执行, 也可以先于步 骤 201执行。
步骤 204 ,根据各个峰值点的特征信息,从生成的 MP-Kemel信号中选取 合适的 MP-Kemel信号, 分别对选取的 MP-Kemel信号进行圓周移位, 幅度 调整与相位旋转, 生成与各峰值点对应的消峰信号。
步骤 205,将与各峰值点对应的消峰信号与緩存的 1倍速率的时域信号进 行叠加, 对緩存中的 1倍速率的时域信号进行消峰, 得到消峰信号。
步骤 206, 检测消峰信号的峰值幅度与消峰次数。
步骤 207,判断消峰信号的峰值幅度是否小于预先设定的幅度阈值,或者 消峰次数是否达到预先设定的次数阈值, 若判断结果为真, 则执行步骤 208; 否则, 緩存该消峰信号, 然后针对该消峰信号, 返回执行步骤 203。
在消峰信号的峰值幅度与消峰次数均达不到预先设定的阈值条件时, 将 该消峰信号进行緩存并重新进行消峰, 保证了消峰效果。
步骤 208 , 直接输出消峰后的接收信号。
具体地, 步骤 207中设置幅度阈值或次数阈值的方法可以参考步骤 206。 可以通过图 3、 图 6至图 8任一实施例的方法来生成 MP-Kernel信号, 实现步骤 201。 如图 3所示, 为本发明生成 MP-Kernel信号实施例一的流程 图, 其包括以下步骤:
步骤 301 , 根据 OFDM符号内各个子载波的消峰权重因子, 对各个子载 波设定权重, 生成频域权重信号。 其中的消峰权重因子包括 MCS、 星座映射 方式、 数据重要程度、 发射功率、 信道情况以及接收质量要求等因子, 可以 从基站获取。
具体地, 可通过下述公式获得各个子载波的权重:
Wj = ^ Wj^ 或 = ]""[ Wj^ 其中, 为第 j个子载波的消峰权重( ≥0 ); 为第 j个子载波在第 i个消峰权重因子下的权重。
以星座映射方式为例说明消峰权重因子的权重设定方法, 权重随着星座 映射方式对应的阶数升高而降低, 即: 当星座映射方式对应的阶数较低时, 则将权重值可设大一些, 否则反之。 例如: 对于正交移相调制 (Quadrature phase-shift keying,以下简称: QPSK )、 16正交调幅调制( Quadrature Amplitude Modulation, 以下简称: QAM ) 以及 64QAM三种星座映射方式对应阶数依次 升高, 则相应的权重设置应该依次变小, 例如可将权重依次设为 0.70、 0.31 和 0.15。
以子载波的编码速率为例说明消峰权重因子的权重设定方法, 权重随着 子载波的编码速率的升高而降低, 即当子载波的编码速率较低时, 则将权重 值可设大一些, 否则反之。 例如, 1/3速率的编码方式的权重就可以设得比 1/2 速率的编码方式的权重高一些 ,可将 1/3速率编码方式和 1/2速率编码方式的权 重分别设为 0.2和 0.12。
其它消峰权重因子的权重的设定方法与编码方式和星座映射方式的权重 的设定方法类似。 消峰权重设定的最小处理单位可以是每个子载波, 也可能 是情况相同或相似的多个子载波。
上面描述了根据该子载波上的各个消峰权重因子来设定各个子载波的权 重的方法。 另外, 还可根据频域矢量幅度误差(Error Vector Magnitude, 以下 简称: EVM )来设置各个子载波的权重。假设子载波 1到 N的频域 EVM分别为 小于等于 EVMl5 . . ., EVMN,则相应的各个子载波的权重 Wl, . . . , WN 以按照下 述公式求出:
wl κ¾ wN
EVM, ~ EVM2 ~ . ~ EVMN
步骤 302 , 对频域权重信号进行 OFDM子载波映射, 在频域权重信号中连 续插入(M- 1 ) χΝ个零。
步骤 303 , 对插零后的频域权重信号进行频域到时域的变换, 例如: 反向 快速傅立叶变换( inverse fast Fourier transform, 以下简称: IFFT ) , 生成 MxN 点的 Kernel信号, 也称: M倍过釆样的准基准消峰信号。
步骤 304、 对准基准消峰信号进行幅度调整, 将最高幅度归一化, 以方便 在峰值对消过程中对基准消峰信号进行幅度调整, 得到归一化的 M倍过釆样 的准基准消峰信号。
步骤 305 , 对归一化的 M倍过釆样速率的准基准消峰信号进行分组, 将 ΜχΝ 点的 Kernel 信号划分成不同起始相位的 M 个 N 点的消峰信号 MP-Kernel , 也即: 基准消峰信号, 可以存储该 MP-Kernel信号。
具体地, 可以通过下式来划分 ΜχΝ点的 Kernel信号:
Kernel i = [Kernel (i) Kernel (i + M) · · · Kernel (i + (N - 1) · )] 其中, KerndiJ")为归一化的 M倍过釆样速率的准基准消峰信号的第 j个点, Kemeli表示对应于第 i个起始相位的 MP-Kemel信号, 其中, i、 j为正整数, i = 1 ,2, ..· ,M, j=l ,2 , ... ,Μ ΧΝ , {Kemeli}即为 MP-Kemel。
如图 4所示,为本发明在消峰过程中依照上式所述方法将 4096点的 Kernel 信号划分成的 4个 1024点的 MP-Kemel信号的幅度示例图; 图 5所示为图 4 中 MP-Kemel信号的幅度图的局部放大图, 其中, ' +,点表示 Kernel^ 'ο'点表 示 Kernel2 , '*,点表示 Kerne" , '.,点表示 Kernel4
如图 6所示, 为本发明生成 MP-Kemel信号实施例二的流程图, 其包括 以下步骤:
步骤 401 , 根据 OFDM符号内各个子载波的消峰权重因子, 为各个子载 波设定权重, 生成频域权重信号, 具体的设定方法可以参考步骤 301。
步骤 402 , 分别利用各频域相位调整因子对频域权重信号进行相位调整。 具体地, 可通过如下方式设定不同初始相位的频域相位调整因子:
Figure imgf000013_0001
其中, w( ,k)表示第 m个初始相位的信号第 k个子载波上的相位调整因
M
子。
步骤 403 , 对相位调整后的子载波进行频域到时域的变换, 例如: IFFT, 生成准基准消峰信号。
步骤 404、 对准基准消峰信号进行幅度调整, 将最高幅度归一化, 得到基 准消峰信号 MP-Kernel。
步骤 405 , 存储幅度调整后的 MP-Kemel信号。
如图 7所示, 为本发明生成 MP-Kemel信号实施例三的流程图, 其包括 以下步骤:
步骤 501 , 根据 OFDM符号内各个子载波的消峰权重因子, 为各个子载 波设定权重, 生成频域权重信号。 具体的设定方法可以参考步骤 301。
步骤 502 , 对频域权重信号进行频域到时域的变换, 例如: IFFT, 生成 1 倍速率的准基准消峰信号。
步骤 503 , 由 1倍速率的准基准消峰信号获得 M倍过釆样速率的准基准 消峰信号, 例如: 通过内插滤波的方法, 在各釆样点后插入(M- 1 )个零, 并进行滤波处理。
步骤 504 , 对 M倍过釆样速率的准基准消峰信号进行幅度调整, 将最高 幅度归一化。
步骤 505 , 通过
Kernel i = [Kernel (i) Kernel (i + M) ·■■ Kernel (i + (N _ 1) · M)], 对归一 ^匕的 M倍过釆样速率的准基准消峰信号进行分组, 生成基准消峰信号 MP-Kemel 信号, 存储该 MP-Kemel信号。
其中, Kernel (J表示 ΜχΝ点 Kernel信号的第 j个点, Kemeli表示对应于第 i 个起始相位的子 Kernel信号,其中, i、j为正整数, i = 1 ,2, ... ,M, j=l,... ,Μ ΧΝ , {Kemeli}即为 MP-Kemel。
如图 8所示, 为本发明生成 MP-Kemel信号实施例四的流程图, 其包括 以下步骤:
步骤 601 , 根据 OFDM符号内各个子载波的消峰权重因子, 为各个子载 波设定权重, 生成频域权重信号。 具体的设定方法可以参考步骤 301。
步骤 602 , 对频域权重信号进行频域到时域的变换, 例如: IFFT, 生成 1 倍速率的准基准消峰信号。
步骤 603 ,对 1倍速率的准基准消峰信号进行幅度调整,将最高幅度归一 化。
步骤 604 ,釆用预先设置好滤波器系数的多相滤波器,分别对 1倍速率的 准基准消峰信号进行不同相位的多相滤波, 生成 MP-Kemel。 对于 0相位的 多相滤波可以省略, 直接保持滤波前的信号即可, 即: 对于相位为 0的消峰 信号可以直接利用幅度归一化后的 1倍速率的准基准消峰信号代替。
对于多相滤波器的要求, 下边以根升余弦滤波为例进行说明, 假设釆用
64阶根升余弦滤波器进行多相滤波, 例如: 可以通过如下方法设置滤波器系 数:
可以通过美国 Mathworks公司生产的 Maltab软件中的 FIRRCOS函数, 生成 M倍过釆样速率的根升余低通滤波器系数, 例如: M=4时的 64阶滤波 器, 可以通过 firrcos(64, 0.25, 0.1 , 4,,rolloff,,, sqrt,)设置其滤波器系数, 通过 该方式可以生成 65个滤波器系数, 然后将该 65个滤波器系数分成 M = 4组 多相滤波器系数, 则其中第 1 , 5, 9, 61 , 65个系数为第一组多相滤波 系数, 第 2, 6, 10, 58, 62为第二组多相滤波系数, 第 3 , 7, 11 , 59, 63为第三组多相滤波系数, 第 4, 8, 12, 60, 64为第四组多相滤 波系数, 具体滤波器系数的幅度可根据需要适当调整。 此时, 利用第一组多 相滤波系数进行滤波可以省略。
步骤 605 , 存储滤波后的 MP-Kernel信号。
另外, 可以通过图 9至图 11任一实施例的方法, 来根据接收到的 1倍速 率的时域信号, 预测 M倍过釆样速率的时域信号的峰值点的特征信息, 实现 步骤 102或步骤 203。
如图 9所示, 为本发明预测 M倍过釆样速率的时域信号的峰值点, 求取 并记录峰值点的特征信息实施例一的流程图, 其具体包括以下步骤:
步骤 701 , 对緩存的 1倍速率的时域信号进行内插滤波, 得到 M倍过釆 样速率的时域信号。
步骤 702,计算 M倍过釆样速率的时域信号各釆样点幅度或幅度的平方, 由于各釆样点以复数标识, 其幅度即 M倍过釆样速率的时域信号各釆样点实 部平方与虚部平方之和再开根号, 其幅度的平方即 M倍过釆样速率的时域信 号各釆样点实部平方与虚部平方之和。
步骤 703 ,根据 M倍过釆样速率的时域信号各釆样点幅度或幅度的平方, 检测 M倍过釆样速率时域信号的峰值点。
其中, 可以通过如下方式预测 M倍过釆样速率时域信号的峰值点: 预先 设定幅度门限 ThresholdO或 ThresholdO的平方, 即: ThresholdO2 , 该幅度门限可 以是系统 PAPR门限, 将幅度大于 ThresholdO或幅度平方大于 ThresholdO2 , 且 大于相邻釆样点幅度或幅度的平方的釆样点判定为峰值点。
步骤 704 ,根据峰值点的位置信息,计算与各峰值点对应的基准消峰信号 MP-Kernel信号的序号索引 m。 ,需要圓周移位的位数 ζ·。,幅度相位调整因子 /。, 其中, 峰值点的位置信息, 圓周移位信息, 即: 圓周移位的位数信息, 与幅 度相位信息, 即: 幅度相位调整因子信息, 构成峰值点的特征信息。 可以存 储该峰值点的特征信息以供后续使用。
假设峰值点的位置为 i, 其对应的 OFDM符号为 x(i) , x(i)为复数, 则可 以通过如下方法求取各因子 。、 。与 /。: m0 = (i - l)%M + l
,0 =L(,'- 1) / M」 + 1 f0 =
Figure imgf000016_0001
其中, 为门限调整因子, Λ≤1 , 例如: 取 = 0.99 , 这样可以减少 由于峰值位置估计精度不高引起的同一峰值多次对消的现象。
如图 10所示, 为本发明预测 Μ倍过釆样速率的时域信号的峰值点, 求取 并记录峰值点的特征信息实施例二的流程图, 其具体包括以下步骤:
步骤 801 , 由緩存的 1倍速率的时域信号获得 Μ倍过釆样速率时域信号。 具体地, 可以通过在 1倍速率的时域信号各釆样点后插入(M-1 )个零, 来获 得 Μ倍过釆样速率时域信号。
步骤 802 , 对 Μ倍过釆样速率时域信号进行粗滤波, 即: 釆用阶数较低的 滤波器, 例如: 16阶的滤波器, 进行滤波。
步骤 803 , 计算粗滤波后 Μ倍过釆样速率时域信号各釆样点幅度或幅度的 平方, 即: 实部平方与虚部平方之和开根号或实部平方与虚部平方之和。 步骤 804 , 选择幅度大于 (/ X Threshold θ)或幅度的平方大于
( x 7¾m^oW0)2的釆样点作为可能的峰值点, 也即: 准峰值点, 其中, 为准 峰值点的门限调整因子, ; ≤1 , 例如: 取; = 0.75 , 这样可以保证不会漏掉 可能的峰值点; 另外, 也可以这对不同的消峰次数时取不同的值, 此时, 也可以不必要求; ≤ 1 ,例如:第一次消峰时取; = 1.25 ,第二次消峰时取; = 1 , 第三次消峰时取; = 0.75 , 这样可以先消除幅度较高的峰值点, 避免或减少 发生峰值再生的可能性。
步骤 805 , 对准峰值点进行精滤波, 即: 釆用与步骤 802釆用的滤波器相 比阶数较高的滤波器, 例如: 64阶的滤波器, 进行滤波。
步骤 806, 根据准峰值点的滤波结果, 从准峰值点中选择 Μ倍过釆样速率 时域信号的峰值点。 具体选择峰值点的方法可以参考步骤 703。
步骤 807, 根据峰值点的位置信息, 计算该峰值点对应的基准消峰信号 MP-Kemel信号的序号索引 m。 ,需要圓周移位的位数 ζ·。,幅度相位调整因子 /。。 具体的计算方法可以参考步骤 704。
相对于图 9所示的实施例, 该实施例二先对 Μ倍过釆样速率时域信号低 阶滤波, 初步检测可能的峰值点, 再针对可能的峰值点进行精滤波, 例如: 先进行 16阶的粗滤波, 再进行 64阶或者 128阶的精滤波, 不对每一个釆样点 都进行高阶滤波, 而仅仅对粗滤波后判定的可能的峰值点进行高阶滤波, 简 化了信息处理的复杂度, 减少了信息处理量, 缩短了峰值点的特征信息的检 测时间, 提高了工作效率。
如图 11所示, 为本发明预测 Μ倍过釆样速率的时域信号的峰值点, 求取 并记录峰值点的特征信息实施例三的流程图, 其具体包括以下步骤:
步骤 901 , 由緩存的 1倍速率的时域信号获得 Μ倍过釆样速率时域信号。 具体地, 可以通过内插滤波的方式, 在 1倍速率的时域信号各釆样点后插入 ( M-1 )个零, 来获得 Μ倍过釆样速率时域信号。 步骤 902, 对 M倍过釆样速率时域信号进行粗滤波, 即: 釆用阶数较低 的滤波器, 例如: 16阶的滤波器, 进行滤波。
步骤 903 , 计算各釆样点幅度或幅度的平方, 即: 实部平方与虚部平方之 和再开根号或实部平方与虚部平方之和。
步骤 904 , 选择幅度大于 {γ χ Threshold 0)或幅度的平方大于 ( x 7¾m^oW0)2的釆样点作为可能的峰值点, 即: 准峰值点, 其中, 为准峰 值点的门限调整因子, y≤l , 例如: 取 = 0.75 , 这样可以保证不会漏掉可 能的峰值点; 另外, ; κ也可以在不同的消峰次数时取不同的值, 此时, 也可 以不必要求 ≤ 1 , 例如: 第一次消峰时取 = 1.25 , 第二次消峰时取 y = 1 , 第三次消峰时取 = 0.75 , 这样可以先消除幅度较高的峰值点, 避免或减少 发生峰值再生的可能性。
步骤 905 , 緩存准峰值信息, 包括该准峰值点的位置信息与粗滤波之后该 准峰值点的实部与虚部值; 也可以不緩存准峰值信息, 而直接进行后续步骤。
步骤 906, 根据准峰值点的位置信息, 对准峰值点进行补充精度的滤波, 例如: 若粗滤波为 16阶滤波, 需达到 64阶滤波的效果, 计算粗滤波未计算的 64-16阶抽头系数, 从而获得补充精度滤波后的准峰值点实部与虚部的补偿 值。
步骤 907,将緩存的粗滤波得到的准峰值点的实部与虚部值与补充精度的 滤波得到的补偿值叠加, 获得更加精确的峰值信息, 以便后续步骤中进一步 从准峰值点中选择真正的峰值点。
步骤 908, 根据步骤 907中获得的更加精确的峰值信息, 从准峰值点中选 择峰值点,可以通过如下方式选择峰值点: 预先设定幅度阈值 7¾re oW0及其 平方, 即: ThresholdQ1 , 该幅度可以是系统 PAPR门限对应的幅度, 将幅度大 于 ThreshoMQ , 且大于相邻釆样点幅度的釆样点判定为峰值点, 或将幅度平方 大于 Threshold Q2 , 且大于相邻釆样点幅度平方的釆样点判定为峰值点。
步骤 909, 根据峰值点的位置信息, 计算该峰值点对应的 MP-Kernel信号 的序号索引 , 需要圓周移位的位数。, 幅度相位调整因子 /。。 具体的计算 方法可以参考步骤 704。
如果精滤波器釆用的滤波系数的截短等于或近似等于粗滤波器的滤波 系数, 则精滤波器可以省去在粗滤波器进行粗滤波对应的运算, 与图 10所示 的实施例二相比, 处理的信息量更少, 实现更简单。
根据步骤 104或步骤 204 , 利用与 Μ倍过釆样速率的峰值点对应的消峰信 号, 对接收到的 1倍速率的时域信号进行消峰的具体方法如下:
根据上述各实施例计算出的峰值点 MP-Kemel信号的序号索引 m。, 从 M 个 MP-Kemel信号中选取合适的 MP-Kemel信号, 然后根据各峰值点的特征信 息, 即: 需要圓周移位的位数 i。和幅度相位调整因子 f。, 对选择出的合适的 MP-Kemel信号进行相应的圓周移位、 幅度调整与相位旋转, 生成与各峰值 点对应的消峰信号, 然后与緩存的 1倍速率的时域信号进行叠加, 对接收到 的 1倍速率的时域信号进行消峰。 具体实现时, 可以每次对消一个或多个峰。
具体可以通过如下方法从 M个 MP-Kemel信号中选取合适的 MP-Kemel 信号:
由于 M倍过釆样速率的时域信号的峰值点可能落在原 1倍速率的釆样点 上, 也可能落在原 1倍速釆样点之间的新 M-1个釆样点上, 即: m0 = (i - l)%M + l ) , 根据图 9至图 1 1任一实施例的方法预测出的 M倍过釆样 速率的时域信号的峰值点的位置信息选取 M个 MP-Kemel信号中的第 d + 1个 MP-Kemel信号作为该 M倍过釆样速率的时域信号的峰值点的基准消峰信号, 也即合适的 MP-Kemel信号, 其中, d为该 M倍过釆样速率的时域信号的峰值 点与前面相邻 1倍速率的时域信号釆样点相隔的釆样点的个数, d=0, 1 , . . . ,Μ- 1。
图 12-1、 12-2、 12-3与 12-4为 Μ = 4 , Ν = 1024时, 分别选取第 1-4个 MP-Kemel信号的峰值点的分布示意图, 其中, 'o'表示 1倍速率釆样点, ' +, 表示 M倍过釆样速率后增加的点。
对 1倍速率的时域信号进行消峰时, 每次可以消除一个或多个峰值点, 也可以消除全部峰值点, 每次所要消除峰值点的具体数目可以结合 PAPR门 限与消峰次数综合考虑。 例如: 对于单载波 1024点 IFFT的 OFDM系统, 在 PAPR门限为 7dB时, 如果共计进行四次消峰, 则可以规定每次消除 8个峰 值点。
本发明上述实施例在 1倍速率下利用 N点 MP-Kernel进行峰值对消的技术 方案, 与在 M倍过釆样速率下利用 ΜχΝ点的 Kernel信号进行峰值对消的传统 方案相比, 消峰效果并没有下降, 具体说明如下:
传统方案的 M倍过釆样速率 Kernel信号的第 k个样点可表达如下:
画—\ -^mk
PMN(k) =
Figure imgf000020_0001
函 对于预留子载波(Tone Reservation, 以下简称: TR) 方案, 在预留子载 波位置 ,.不为 0, 其余点均为 0; 对于峰值对消(Peak Cancelling, 以下简称: PC)方案, 在数据子载波上 ,.均为 1, 在其余子载波上均为 0; 对于自适应 权重峰值对消 (Adaptive Weight Peak Cancelling, 以下简称: AWPC )方案, 根据各个子载波的具体情况设置消峰权重, 无论以上哪种方案, 当 N/2<i<=M xN-N/2时, 都有 ν,.=0。 消峰过程即为利用; ½ν的不同圓周移位、 幅度与相位调整后的信号, 与 原信号进行叠加的过程,设^)为峰值点的位置,也即:对消该峰值点时 Kernel 信号;?層需要圓周移位的位数, 令^ = Mz'。 + w。其中 。 =0, 1, N-1, m0 =0, 1, ..., M-l。 则对;?層进行圓周移位、 幅度与相位调整后的信号可以表 示为: PMf N (k = ao -eJ 。 · P , τ^)
MN-l .2 (k+T0)
jd0
= a0 -e MN
=0
MN-l .2πίτ .2mk
a0. e
Figure imgf000021_0001
根据上式,从频域观察, 上述叠加过程仅仅引起了 W 的前 N/2个数据以 及最后的 N/2 个数据的变化。 其中, α。· 。为幅度与相位调整系数, 对原始 1 倍速率下 N点 Kernel信号对应的频域信号 ,.进行相应相位旋转后的 新 N点 Kernel, e N 为原始 1倍速率下 N点 Kernel信号进行。位圓周移位的系 数, 将相位旋转后的新 Kernel信号称为 MP-Kemel信号。 显然, 在 M倍过釆样 速率下, 共存在 M个 MP-Kemel信号, 分别对应的相位旋转矢量为 e ^「, 其 中, i = 0, 1, ·.., N/2-1, (Μ-1)Ν+Ν/2, ·.., MN-l, m0=0,l,...,M-l。 根据 m0 取值的不同, 将 MP-Kemel信号分别记作 Kemel。, Kernelj, Kernel^, 则 MP-Kemel可以表示为:
N-1 2 m • 2mk
J-
Pmo(k) =∑ w; · e MN e N
when i < N / 2
其中 i :
i + (M -\)-N when i≥N 12
这样, 对 进行 N点的 IFFT变换, 即可得到对应 m。时的 N点的
Figure imgf000021_0002
MP-Kemel信号, 当 = ^^^+1¾)时, 有:
1 2m(M xi0 + m0 )
PMN (M x i0 + m0 ) = X Wie ~ ~~
MN _ 1 . 2 ^m o . 2 通过以上推导可以看出 , MP-Kernel信号就是 Μ倍过釆样速率的 Kernel信 号以不同 m。为起点间隔 M进行抽样的 Kernel信号, 并且利用 MP-Kernel进行消 峰的过程与传统的利用 M倍过釆样速率的 Kenrel信号进行消峰后对频域产生 的影响相同, 也即本发明在 1倍速率下利用 N点 MP-Kernel进行峰值对消的技 术方案, 与现有技术在 M倍过釆样速率下利用 ΜχΝ点的 Kernel信号进行峰值 对消的传统方案相比, 消峰效果相同。
如图 13所示,为本发明基准消峰信号生成装置实施例一的结构示意图, 其可用于实现如图 3所示的实施例, 该基准消峰信号生成装置实施例包括 依次连接的:
权重设置模块, 用于根据 OFDM符号内各个子载波的消峰权重因子, 为 各个子载波设定权重, 生成频域权重信号;
插零模块, 用于对频域权重信号进行 OFDM子载波映射, 在频域权重信 号中连续插入 ( M-1 ) χΝ个零;
频时转换模块, 用于对插零后的频域权重信号进行频域到时域的变换, 得到 M倍过釆样的准基准消峰信号;
幅度调整模块, 用于对准基准消峰信号进行幅度调整, 将最高幅度归一 化, 得到归一化的 M倍过釆样的准基准消峰信号;
分组模块, 用于对归一化的 M倍过釆样速率的准基准消峰信号进行分 组,将归一化的 M倍过釆样的准基准消峰信号划分成不同起始相位的 M个 N 点的基准消峰信号。
如图 14所示, 为本发明基准消峰信号生成装置实施例二的结构示意图, 其可用于实现如图 6所示的实施例, 该基准消峰信号生成装置实施例包括依 次连接的:
权重设置模块, 用于根据 OFDM符号内各个子载波的消峰权重因子, 对 各个子载波设定权重, 生成频域权重信号;
相位调整模块, 用于分别利用由
Figure imgf000023_0001
设定的各频域相位调整因子对的频域权重信号进行相位调整, 其中, w(£L,k)表示第 m个初始相位的信号第 k个子载波上的相位调整因子, m = 1 , M
2, · .. , M;
频时转换模块, 用于对相位调整后的频域权重信号进行频域到时域的变 换, 生成准基准消峰信号;
幅度调整模块, 对准基准消峰信号进行幅度调整, 将最高幅度归一化, 得到基准消峰信号。
如图 15所示, 为本发明基准消峰信号生成装置实施例三的结构示意图, 其可用于实现如图 7所示的实施例, 该基准消峰信号生成装置实施例包括依 次连接的:
权重设置模块, 用于根据 OFDM符号内各个子载波的消峰权重因子, 为 各个子载波设定权重, 生成频域权重信号;
频时转换模块, 用于对频域权重信号进行频域到时域的变换, 得到 1倍 速率的准基准消峰信号;
插零滤波模块, 用于由 1倍速率的准基准消峰信号获得 M倍过釆样速率 的准基准消峰信号;
幅度调整模块,用于对 M倍过釆样速率的准基准消峰信号进行幅度调整, 将最高幅度归一化; 分组模块, 用于对归一化的 M倍过釆样速率的准基准消峰信号进行分 组,将归一化的 M倍过釆样速率的准基准消峰信号划分成不同起始相位的 M 个 N点的基准消峰信号。
如图 16所示, 为本发明基准消峰信号生成装置实施例四的结构示意图, 其可用于实现如图 8所示的实施例, 该基准消峰信号生成装置实施例包括依 次连接的:
权重设置模块, 用于根据 OFDM符号内各个子载波的消峰权重因子, 为 各个子载波设定权重, 生成频域权重信号;
频时转换模块, 用于对频域权重信号进行频域到时域的变换, 得到 1倍 速率的准基准消峰信号;
幅度调整模块, 用于对 1倍速率的准基准消峰信号进行幅度调整, 将最 高幅度归一化;
多相滤波模块, 用于对幅度归一化的 1倍速率的准基准消峰信号进行不 同相位的多相滤波, 生成基准消峰信号。 具体地, 该多相滤波模块可以由多 个不同滤波系数的滤波器组成。
在上述基准消峰信号生成装置实施例一至四中, 都可以包括第一存储模 块, 用于存储由基准消峰信号生成装置最终生成的基准消峰信号, 以供后续 选取使用。
如图 17所示,为本发明峰值对消装置实施例一的结构示意图,其可用于 实现如图 1所示的实施例, 该峰值对消装置实施例包括依次连接的:
预测模块, 用于根据 1倍速率的时域信号, 预测 M倍过釆样速率的时域 信号的峰值点的特征信息;
峰值对消模块, 用于根据峰值点的特征信息, 选取基准消峰信号, 并对 选取的基准消峰信号进行圓周移位、 幅度调整与相位旋转, 生成与峰值点对 应的消峰信号, 并利用与峰值点对应的消峰信号对 1倍速率的时域信号进行 消峰。 如图 18所示,为本发明预测模块实施例一的结构示意图,其可用于实现 如图 9所示的实施例, 该预测模块实施例包括依次连接的:
内插滤波单元, 用于对 1倍速率的时域信号进行内插滤波, 获得 M倍过 釆样速率的时域信号;
第一计算单元,用于计算 M倍过釆样速率的时域信号各釆样点幅度或幅 度的平方;
检测单元,用于根据 M倍过釆样速率的时域信号各釆样点幅度或幅度的 平方, 检测 M倍过釆样速率的时域信号的峰值点的位置;
第二计算单元, 用于根据峰值点的位置信息, 通过下式计算峰值点对应 的基准消峰信号的序号索引、 圓周移位的位数与幅度相位调整因子, 峰值点 的特征信息包括峰值点的位置信息、 序号索引、 圓周移位的位数与幅度相位 调整因子:
m0 = (i - l)%M + l
,0 =L(,'- 1) / M」 + 1 f0 = (|x(z')| - λ Threshold 0)
Figure imgf000025_0001
其中, 为序号索引, z。为圓周移位的位数, /。为幅度相位调整因子, i 为 M倍过釆样速率的时域信号的峰值点的位置, x(i) 为峰值点的位置 i对应 的 OFDM符号, ThresholdO为预先设定的幅度门限, 为门限调整因子。
上述预测模块实施例一中, 检测单元与第二计算单元分别与峰值对消模 块连接, 以提供其需要的峰值点的特征信息。
如图 19所示,为本发明预测模块实施例二的结构示意图,其可用于实现 如图 10所示的实施例, 该预测模块实施例包括依次连接的:
插零单元,用于将 1倍速率的时域信号内插为 M倍过釆样速率的时域信 号;
第一滤波单元, 用于对 M倍过釆样速率的时域信号进行粗滤波; 第一计算单元,用于计算粗滤波后 M倍过釆样速率的时域信号各釆样点 幅度或幅度的平方;
第一选择单元, 用于选择幅度大于 (; x 7¾r^AoW 0)或幅度的平方大于
( x 7¾m^oW 0)2的釆样点作为准峰值点,其中, 为准峰值点的门限调整因子, r≤i ;
第二滤波单元, 用于对准峰值点进行精滤波;
第二选择单元, 用于根据精滤波的结果, 从准峰值点中选择 M倍过釆样 速率的时域信号的峰值点;
第二计算单元, 用于根据峰值点的位置信息, 通过下式计算峰值点对应 的基准消峰信号的序号索引、 圓周移位的位数与幅度相位调整因子, 峰值点 的特征信息包括峰值点的位置信息、 序号索引、 圓周移位的位数与幅度相位 调整因子:
m0 = (i - l)%M + l
,0 =L(,'- 1) / M」 + 1 f0 = (|x(z')| - λ Threshold 0)
Figure imgf000026_0001
其中, 为序号索引, z。为圓周移位的位数, /。为幅度相位调整因子, i 为 M倍过釆样的时域信号的峰值点位置, x(i) 为峰值点位置 i对应的 OFDM 符号, ThresholdO为预先设定的幅度门限, /1为门限调整因子。
上述预测模块实施例二中, 第二选择单元与第二计算单元分别与峰值对 消模块连接, 以提供其需要的峰值点的特征信息。
如图 20所示,为本发明预测模块实施例三的结构示意图,其可用于实现 如图 11所示的实施例, 该预测模块实施例包括依次连接的:
插零单元,用于将 1倍速率的时域信号内插为 M倍过釆样速率的时域信 号;
第一滤波单元, 用于对 M倍过釆样速率的时域信号进行粗滤波; 第一计算单元,用于计算粗滤波后 M倍过釆样速率的时域信号各釆样点 幅度或幅度的平方;
第一选择单元, 用于选择幅度大于 (; x7¾r^AoW0)或幅度的平方大于
( x7¾m^oW0)2的釆样点作为准峰值点,其中, 为准峰值点的门限调整因子, γ <\
第三滤波单元, 用于对准峰值点进行补充精度滤波;
第三计算单元 , 用于将粗滤波得到的准峰值点的值与补充精度滤波得到 的实部与虚部的补偿值叠加;
第三选择单元, 用于根据叠加结果, 从准峰值点中选择 Μ倍过釆样速率 的时域信号的峰值点;
第二计算单元, 用于根据峰值点的位置信息, 通过下式计算该峰值点对 应的基准消峰信号的序号索引、 圓周移位的位数与幅度相位调整因子, 峰值 点的特征信息包括峰值点的位置信息、 序号索引、 圓周移位的位数与幅度相 位调整因子:
m0 =(i-l)%M + l
,0=L(,'- 1)/M」 + 1 f0 = (|x(z')| -λχ Threshold 0)
Figure imgf000027_0001
其中, 为序号索引, z。为圓周移位的位数, /。为幅度相位调整因子, i 为 M倍过釆样的时域信号的峰值点位置, x(i) 为峰值点位置 i对应的 OFDM 符号, ThresholdO为预先设定的幅度门限, /1为门限调整因子。
上述预测模块实施例三中, 第三选择单元与第二计算单元分别与峰值对 消模块连接, 以提供其需要的峰值点的特征信息。
另外, 图 18至 20所示的预测模块实施例中,还可以包括第二存储模块, 用于存储峰值点的特征信息, 峰值对消模块与该第二存储模块连接, 从该第 二存储模块中选取合适的峰值点的特征信息。 在图 18所示的实施例中, 第二 存储模块分别与检测单元及第二计算单元连接, 向第二存储模块存储相应的 峰值点的特征信息, 以供峰值对消模块选取; 在图 19所示的实施例中, 第二 选择单元及第二计算单元分别与第二存储模块连接, 向第二存储模块存储相 应的峰值点的特征信息, 以供峰值对消模块选取; 在图 20所示的实施例中, 第三选择单元及第二计算单元分别与第二存储模块连接, 向第二存储模块存 储相应的峰值点的特征信息, 以供峰值对消模块选取。
如图 21所示,为本发明峰值对消模块实施例的结构示意图,其可用于如 图 1与图 2所示实施例中的相应步骤, 该峰值对消模块实施例包括依次连接 的:
选取单元, 与预测模块连接, 用于根据上述实施例中预测模块生成或进 一步存储的峰值点的特征信息, 选取基准消峰信号, 该基准消峰信号可以通 过上述任一实施例提供的基准消峰信号生成装置在线生成, 也可以预先生成 并存储;
圓周移位单元, 用于根据峰值点的特征信息, 对选取的基准消峰信号进 行圓周移位;
幅相调整单元, 用于根据峰值点的特征信息, 对选取的基准消峰信号, 对圓周移位后的基准消峰信号进行幅度调整与相位旋转, 生成与峰值点对应 的消峰信号;
消峰单元, 用于利用与峰值点对应的消峰信号对接收到的原始 1倍速率 的时域信号进行消峰。
其中, 选取单元可以包括:
判断单元, 用于判断 M倍过釆样速率的时域信号的峰值点与前面相邻 1 倍速率的时域信号釆样点相隔的釆样点的个数 d, d=0,l,...,M-l ;
第一选取单元, 用于为各个峰值点从 M个基准消峰信号中选取第 d + 1 个基准消峰信号作为该 M倍过釆样速率的时域信号的峰值点的基准消峰信 号。 如图 22所示,为本发明峰值对消装置实施例二的结构示意图,该实施例 的峰值对消装置在实施例一的基础上, 还包括: 緩存模块, 分别与预测模块、 峰值对消模块连接, 用于接收并緩存接收到的 1倍速率的时域信号, 预测模 块用于根据緩存模块中緩存的 1倍速率的时域信号,预测 M倍过釆样速率的 时域信号的峰值点的特征信息。 具体地, 緩存模块可以与预测模块中的内插 滤波单元或插零单元连接, 向其输入 1倍速率的时域信号; 緩存模块可以与 峰值对消模块中的消峰单元连接, 向其提供待消峰的 1倍速率的时域信号。
如图 23所示,为本发明峰值对消装置实施例三的结构示意图,其可用于 实现如图 2所示的实施例, 该实施例的峰值对消装置在实施例二的基础上, 还包括:
检测模块, 与峰值对消模块或其中的消峰单元连接, 用于检测峰值对消 模块处理后输出的消峰信号的峰值幅度与消峰次数;
判断模块, 分别与检测模块及緩存模块连接, 用于判断峰值幅度是否小 于预先设定的幅度阈值, 或者消峰次数是否达到预先设定的次数阈值, 若判 断结果为真, 则直接输出消峰信号; 否则, 将消峰信号送入緩存模块进行下 一次的消峰。
本发明图 17至图 23所示实施例中所需的基准消峰信号可以预先生成并 存储, 也可以在线实时生成。 针对在线实时生成基准消峰信号的情况, 可以 在本发明上述峰值对消装置实施例中设置本发明上述任一实施例提供的基准 消峰信号生成装置, 该基准消峰信号生成装置中的分组模块、 幅度调整模块、 多相滤波模块或第一存储模块与峰值对消模块或其中的选取单元或其中的第 一选取单元连接,用于提供峰值对消模块所需的基准消峰信号,如图 24所示, 为本发明峰值对消装置实施例四的结构示意图。
针对预先生成并存储基准消峰信号的情况, 可以在本发明上述峰值对消 装置实施例中设置第一存储模块即可, 以存储预先生成的峰值对消模块所需 的基准消峰信号, 如图 25所示, 为本发明峰值对消装置实施例五的结构示意 图。
本发明上述各实施例提供的峰值对消方法或装置同样适用于其它需要降 低 PAPR或者降低峰值因子( Crest Factor, 以下简称: CF )的多载波通信系统 , 也同样适用于多载频 OFDM系统的峰值抑制。
本领域普通技术人员可以理解: 实现上述方法实施例的全部或部分步骤 可以通过程序指令相关的硬件来完成, 前述的程序可以存储于一计算机可读 取存储介质中, 该程序在执行时, 执行包括上述方法实施例的步骤; 而前述 的存储介质包括: ROM, RAM, 磁碟或者光盘等各种可以存储程序代码的介 质。
本发明实施例根据 1倍速率的时域信号来预测 M倍过釆样速率的时域信 号的峰值点的特征信息, 根据该特征信息选取基准消峰信号, 对选取的基准 消峰信号进行圓周移位、 幅度调整与相位旋转, 然后利用生成的与峰值点对 应的消峰信号对 1倍速率的时域信号进行消峰, 实现了在 1倍过釆样速率下对 信号的峰值对消, 与现有技术在 M倍过釆样速率下进行峰值对消相比, 在不 会影响 PAPR抑制的性能的前提下, 降低了对緩存单元的緩存容量要求, 减少 了相应处理的工作量, 复杂度低, 延时小, 从而有效改善了峰值对消装置的 工作效率, 提高了其工作性能。
最后所应说明的是: 以上实施例仅用以说明本发明的技术方案, 而非对 本发明作限制性理解。 尽管参照上述较佳实施例对本发明进行了详细说明, 本领域的普通技术人员应当理解: 其依然可以对本发明的技术方案进行修改 或者等同替换, 而这种修改或者等同替换并不脱离本发明技术方案的精神和 范围。

Claims

权 利 要 求
1、 一种峰值对消方法, 其特征在于, 包括:
接收 1倍速率的时域信号;
根据所述 1倍速率的时域信号,预测 M倍过釆样速率的时域信号的峰值 点的特征信息, 其中, M为过釆样速率的倍数, 取大于 1的整数;
根据所述峰值点的特征信息, 选取基准消峰信号, 并对选取的基准消峰 信号进行圓周移位、 幅度调整与相位旋转, 生成与所述峰值点对应的消峰信 号;
利用与所述峰值点对应的所述消峰信号对所述 1倍速率的时域信号进行 消峰。
2、根据权利要求 1所述的峰值对消方法, 其特征在于, 所述根据 1倍速 率的时域信号, 预测 M倍过釆样速率的时域信号的峰值点特征信息包括: 对所述 1倍速率的时域信号进行内插滤波,获得 M倍过釆样速率的时域 信号;
计算所述 M倍过釆样速率的时域信号各釆样点幅度或幅度的平方; 根据所述 M倍过釆样速率的时域信号各釆样点幅度或幅度的平方,选择 幅度大于 Threshold 0或幅度的平方大于 Threshold Q2的釆样点作为 M倍过釆样 速率的时域信号的峰值点;
根据所述峰值点的位置信息, 通过下式计算所述峰值点对应的基准消峰 信号的序号索引、 圓周移位的位数与幅度相位调整因子, 所述峰值点的特征 信息包括所述峰值点的位置信息、 所述序号索引、 所述圓周移位的位数与所 述幅度相位调整因子:
m0 = (i - \)%M + \
,0 =L(,'- 1) / M」 + 1 f0 = (|x(z')| - λ χ Threshold 0) 其中, 为序号索引, Z。为圓周移位的位数, /。为幅度相位调整因子, i 为所述 M倍过釆样速率的时域信号的峰值点的位置, x(i) 为峰值点的位置 i 对应的正交频分复用符号, ThresholdO为预先设定的幅度门限, 为门限调整 因子。
3、根据权利要求 1所述的峰值对消方法, 其特征在于, 所述根据 1倍速 率的时域信号, 预测 M倍过釆样速率的时域信号的峰值点的特征信息包括: 将 1 倍速率的时域信号内插为 M倍过釆样速率的时域信号并进行粗滤 波;
计算粗滤波后 M倍过釆样速率的时域信号各釆样点幅度或幅度的平方; 选择幅度大于 (/ X Threshold 0)或幅度的平方大于 (/ x Threshold θ)2的釆样 点作为准峰值点, 其中, 为准峰值点的门限调整因子, y≤l ;
对所述准峰值点进行精滤波;
才艮据 滤波的结果选择幅度大于 Threshold^或幅度的平方大于 Threshold^2的准峰值点作为 M倍过釆样速率的时域信号的峰值点;
根据所述峰值点的位置信息, 通过下式计算所述峰值点对应的基准消峰 信号的序号索引、 圓周移位的位数与幅度相位调整因子, 所述峰值点的特征 信息包括峰值点的位置信息、 序号索引、 圓周移位的位数与幅度相位调整因 子:
m0 = (i - l)%M + l
,0 =L(,'- 1) / M」 + 1 f0 = (|x(z')| - λ Threshold 0)
Figure imgf000032_0001
其中, 为序号索引, z。为圓周移位的位数, /。为幅度相位调整因子, i 为所述 M倍过釆样的时域信号的峰值点位置, x(i) 为峰值点位置 i对应的正 交频分复用符号, ThresholdO为预先设定的幅度门限, /1为门限调整因子。
4、根据权利要求 1所述的峰值对消方法, 其特征在于, 所述根据 1倍速 率的时域信号, 预测 M倍过釆样速率的时域信号的峰值点的特征信息包括: 将 1倍速率的时域信号内插为 M倍速率的时域信号并进行粗滤波; 计算粗滤波后 M倍过釆样速率的时域信号各釆样点幅度或幅度的平方; 选择幅度大于 (/ X Threshold 0)或幅度的平方大于 (/ x Threshold θ)2的釆样 点作为准峰值点, 其中, 为准峰值点的门限调整因子, < 1;
对所述准峰值点进行补充精度滤波;
将粗滤波得到的准峰值点的实部与虚部值与补充精度滤波得到的补偿值 叠力口;
才艮据叠加结果,选择幅度大于 7¾rra/wW0或幅度的平方大于 Threshold Q2的 准峰值点作为 M倍过釆样速率的时域信号的峰值点;
根据所述峰值点的位置信息, 通过下式计算该峰值点对应的基准消峰信 号的序号索引、 圓周移位的位数与幅度相位调整因子, 所述峰值点的特征信 息包括峰值点的位置信息、 序号索引、 圓周移位的位数与幅度相位调整因子: m0 = (i - l)%M + l
,0
Figure imgf000033_0001
X i)
f0 = ( x(i) - λ χ Threshold 0) x
Figure imgf000033_0002
其中, 为序号索引, z。为圓周移位的位数, /。为幅度相位调整因子, i 为所述 M倍过釆样的时域信号的峰值点位置, x(i) 为峰值点位置 i对应的正 交频分复用符号, ThresholdO为预先设定的幅度门限, /1为门限调整因子。
5、 根据权利要求 2、 3或 4所述的峰值对消方法, 其特征在于, ≤1。
6、根据权利要求 1所述的峰值对消方法, 其特征在于, 所述根据峰值点 的特征信息, 选取基准消峰信号, 并对选取的基准消峰信号进行圓周移位、 幅度调整与相位调整具体为:
根据所述峰值点特征信息中的序号索引选取基准消峰信号, 并利用所述 峰值点特征信息中的圓周移位的位数与幅度相位调整因子信息, 对选取的基 准消峰信号进行圓周移位、 幅度调整与相位调整。
7、根据权利要求 6所述的峰值对消方法, 其特征在于, 所述根据峰值点 的特征信息, 选取基准消峰信号包括:
根据所述 M倍过釆样速率的时域信号的峰值点位置选取 M个基准消峰 信号中的第 d + 1个基准消峰信号作为该 M倍过釆样速率的时域信号的峰值 点的基准消峰信号, 其中, d为该 M倍过釆样速率的时域信号的峰值点与前 面相邻 1倍速率的时域信号釆样点相隔的釆样点的个数, d=0,l,...,M-l。
8、根据权利要求 1所述的峰值对消方法, 其特征在于, 所述利用与所述 峰值点对应的消峰信号对所述 1倍速率的时域信号进行消峰之后, 还包括: 检测消峰信号的峰值幅度与消峰次数, 并判断所述峰值幅度是否小于预 先设定的幅度阈值, 或者消峰次数是否达到预先设定的次数阈值, 若判断结 果为真, 则直接输出所述消峰信号; 否则, 针对该消峰信号, 执行所述根据 所述 1倍速率的时域信号,预测 M倍过釆样速率的时域信号的峰值点的特征 信息的步骤。
9、 根据权利要求 1、 2、 3、 4、 6、 7或 8所述的峰值对消方法, 其特征 在于, 还包括: 生成基准消峰信号。
10、 根据权利要求 9所述的峰值对消方法, 其特征在于, 所述生成基准 消峰信号包括:
根据正交频分复用符号内各个子载波的消峰权重因子, 为各个子载波设 定权重, 生成频域权重信号;
对所述频域权重信号进行正交频分复用子载波映射, 在所述频域权重信 号中连续插入(M-1 ) χΝ个零, 其中, N为 1倍速率的时域信号的长度; 对插零后的频域权重信号进行频域到时域的变换, 得到 M倍过釆样的准 基准消峰信号;
对所述准基准消峰信号进行幅度调整, 将最高幅度归一化, 得到归一化 的 M倍过釆样的准基准消峰信号; 对所述归一化的 M倍过釆样速率的准基准消峰信号进行分组, 将所述归 一化的 M倍过釆样的准基准消峰信号划分成不同起始相位的 M个 N点的基 准消峰信号。
11、 根据权利要求 9所述的峰值对消方法, 其特征在于, 所述生成基准 消峰信号包括:
根据正交频分复用符号内各个子载波的消峰权重因子, 对各个子载波设 定权重, 生成频域权重信号;
通过下式设定不同的初始相位的频域相位调整因子:
波上的相位调整因
Figure imgf000035_0001
子, m = 1 , 2, · .. , M;
分别利用各频域相位调整因子对所述的频域权重信号进行相位调整; 分别对相位调整后的频域权重信号进行频域到时域的变换, 生成准基准 消峰信号;
对所述准基准消峰信号进行幅度调整, 将最高幅度归一化, 得到基准消 峰信号。
12、 根据权利要求 9所述的峰值对消方法, 其特征在于, 所述生成基准 消峰信号包括:
根据正交频分复用符号内各个子载波的消峰权重因子, 为各个子载波设 定权重, 生成频域权重信号;
对所述频域权重信号进行频域到时域的变换, 得到 1倍速率的准基准消 峰信号;
由所述 1倍速率的准基准消峰信号获得 M倍过釆样速率的准基准消峰信 对所述 M倍过釆样速率的准基准消峰信号进行幅度调整, 将最高幅度归 一化;
对所述归一化的 M倍过釆样速率的准基准消峰信号进行分组, 将所述归 一化的 M倍过釆样速率的准基准消峰信号划分成不同起始相位的 M个 N点 的基准消峰信号。
13、 根据权利要求 9所述的峰值对消方法, 其特征在于, 所述生成基准 消峰信号包括:
根据正交频分复用符号内各个子载波的消峰权重因子, 为各个子载波设 定权重, 生成频域权重信号;
对所述频域权重信号进行频域到时域的变换, 得到 1倍速率的准基准消 峰信号;
对所述 1倍速率的准基准消峰信号进行幅度调整, 将最高幅度归一化; 对幅度归一化的 1倍速率的准基准消峰信号进行不同相位的多相滤波, 生成基准消峰信号。
14、 根据权利要求 10或 12所述的峰值对消方法, 其特征在于, 所述对 所述归一化的 M倍过釆样速率的准基准消峰信号进行分组具体为:
通 过 公 式 Kernel i = [Kernel (i) Kernel (i + M )■■■ Kernel (i + (N _ 1) · M)], 对所述归一 化的 M倍过釆样速率的准基准消峰信号进行分组, 其中, Kernel IJ)为)、 3—化 的 M倍过釆样速率的准基准消峰信号的第 j个点, 611½11为对应于第 i个起 始相位的基准消峰信号, 其中, i、 j为正整数, i = l,2, ... ,M, j=l,2, ... ,Μ ΧΝ。
15、 一种基准消峰信号生成装置, 其特征在于, 包括:
权重设置模块, 用于根据正交频分复用符号内各个子载波的消峰权重因 子, 为各个子载波设定权重, 生成频域权重信号;
插零模块, 用于对所述频域权重信号进行正交频分复用子载波映射, 在 所述频域权重信号中连续插入(M- l ) χΝ个零, 其中, M为过釆样速率的倍 数, M取大于 1的整数, N为 1倍速率的时域信号的长度;
频时转换模块, 用于对插零后的频域权重信号进行频域到时域的变换, 得到 M倍过釆样的准基准消峰信号;
幅度调整模块, 用于对所述准基准消峰信号进行幅度调整, 将最高幅度 归一化, 得到归一化的 M倍过釆样的准基准消峰信号;
分组模块,用于对所述归一化的 M倍过釆样速率的准基准消峰信号进行 分组, 将所述归一化的 M倍过釆样的准基准消峰信号划分成不同起始相位的 M个 N点的基准消峰信号。
16、根据权利要求 15所述的基准消峰信号生成装置, 其特征在于, 还包 括:
第一存储模块, 用于存储所述基准消峰信号。
17、 一种基准消峰信号生成装置, 其特征在于, 包括:
权重设置模块, 用于根据正交频分复用符号内各个子载波的消峰权重因 子, 对各个子载波设定权重, 生成频域权重信号;
相位调整模块, 用于分别利用由
Figure imgf000037_0001
设定的各频域相位调整因子对所述的频域权重信号进行相位调整,其中 , 其中, M为过釆样速率的倍数, M取大于 1的整数, N为 1倍速率的时域信号 的长度, W ( , k)表示第 m个初始相位的信号第 k个子载波上的相位调整因子,
M
m = 1 , 2, · .. , M;
频时转换模块, 用于对相位调整后的频域权重信号进行频域到时域的变 换, 生成准基准消峰信号;
幅度调整模块, 对所述准基准消峰信号进行幅度调整, 将最高幅度归一 化, 得到基准消峰信号。
18、根据权利要求 17所述的基准消峰信号生成装置, 其特征在于, 还包 括:
第一存储模块, 用于存储所述基准消峰信号。
19、 一种基准消峰信号生成装置, 其特征在于, 包括:
权重设置模块, 用于根据正交频分复用符号内各个子载波的消峰权重因 子, 为各个子载波设定权重, 生成频域权重信号;
频时转换模块, 用于对所述频域权重信号进行频域到时域的变换, 得到 1倍速率的准基准消峰信号;
插零滤波模块, 用于由所述 1倍速率的准基准消峰信号获得 M倍过釆样 速率的准基准消峰信号, 其中, M为过釆样速率的倍数, M取大于 1的整数; 幅度调整模块, 用于对所述 M倍过釆样速率的准基准消峰信号进行幅度 调整, 将最高幅度归一化;
分组模块,用于对所述归一化的 M倍过釆样速率的准基准消峰信号进行 分组, 将所述归一化的 M倍过釆样速率的准基准消峰信号划分成不同起始相 位的 M个 N点的基准消峰信号, N为 1倍速率的时域信号的长度。
20、根据权利要求 19所述的基准消峰信号生成装置, 其特征在于, 还包 括:
第一存储模块, 用于存储所述基准消峰信号。
21、 一种基准消峰信号生成装置, 其特征在于, 包括:
权重设置模块, 用于根据正交频分复用符号内各个子载波的消峰权重因 子, 为各个子载波设定权重, 生成频域权重信号;
频时转换模块, 用于对所述频域权重信号进行频域到时域的变换, 得到 1倍速率的准基准消峰信号;
幅度调整模块, 用于对所述 1倍速率的准基准消峰信号进行幅度调整, 将最高幅度归一化; 多相滤波模块, 用于对幅度归一化的 1倍速率的准基准消峰信号进行不 同相位的多相滤波, 生成基准消峰信号。
22、根据权利要求 21所述的基准消峰信号生成装置, 其特征在于, 还包 括:
第一存储模块, 用于存储所述基准消峰信号。
23、 一种峰值对消装置, 其特征在于, 包括:
预测模块, 用于根据 1倍速率的时域信号, 预测 M倍过釆样速率的时域 信号的峰值点的特征信息, 其中, M为过釆样速率的倍数, M取大于 1的整 数;
峰值对消模块, 用于根据所述峰值点的特征信息, 选取基准消峰信号, 并对选取的基准消峰信号进行圓周移位、 幅度调整与相位旋转, 生成与所述 峰值点对应的消峰信号, 并利用与所述峰值点对应的所述消峰信号对所述 1 倍速率的时域信号进行消峰。
24、根据权利要求 23所述的峰值对消装置, 其特征在于, 所述预测模块 包括:
内插滤波单元, 用于对所述 1倍速率的时域信号进行内插滤波, 获得 M 倍过釆样速率的时域信号;
第一计算单元,用于计算所述 M倍过釆样速率的时域信号各釆样点幅度 或幅度的平方;
检测单元,用于根据所述 M倍过釆样速率的时域信号各釆样点幅度或幅 度的平方, 检测所述 M倍过釆样速率的时域信号的峰值点的位置;
第二计算单元, 用于根据所述峰值点的位置信息, 通过下式计算所述峰 值点对应的基准消峰信号的序号索引、圓周移位的位数与幅度相位调整因子, 所述峰值点的特征信息包括所述峰值点的位置信息、 所述序号索引、 所述圓 周移位的位数与所述幅度相位调整因子:
m0 = (i - l)%M + l
Figure imgf000040_0001
其中, 为序号索引, z。为圓周移位的位数, /。为幅度相位调整因子, i 为所述 M倍过釆样速率的时域信号的峰值点的位置, x(i) 为峰值点的位置 i 对应的正交频分复用符号, ThresholdO为预先设定的幅度门限, 为门限调整 因子。
25、根据权利要求 23所述的峰值对消装置, 其特征在于, 所述预测模块 包括:
插零单元,用于将 1倍速率的时域信号内插为 M倍过釆样速率的时域信 号;
第一滤波单元, 用于对所述 M倍过釆样速率的时域信号进行粗滤波; 第一计算单元,用于计算粗滤波后 M倍过釆样速率的时域信号各釆样点 幅度或幅度的平方;
第一选择单元, 用于选择幅度大于
Figure imgf000040_0002
或幅度的平方大于
( x 7¾m^oW 0)2的釆样点作为准峰值点,其中, 为准峰值点的门限调整因子, γ < \
第二滤波单元, 用于对所述准峰值点进行精滤波;
第二选择单元, 用于根据精滤波的结果, 选择幅度大于 7¾re 。W0或幅 度的平方大于 Threshold Q2的准峰值点作为 M倍过釆样速率的时域信号的峰值 点;
第二计算单元, 用于根据所述峰值点的位置信息, 通过下式计算所述峰 值点对应的基准消峰信号的序号索引、圓周移位的位数与幅度相位调整因子, 所述峰值点的特征信息包括峰值点的位置信息、 序号索引、 圓周移位的位数 与幅度相位调整因子:
m0 = (i - l)%M + l
Figure imgf000041_0001
其中, 为序号索引, z。为圓周移位的位数, /。为幅度相位调整因子, i 为所述 M倍过釆样的时域信号的峰值点位置, x(i) 为峰值点位置 i对应的正 交频分复用符号, ThresholdO为预先设定的幅度门限, /1为门限调整因子。
26、根据权利要求 23所述的峰值对消装置, 其特征在于, 所述预测模块 包括:
插零单元,用于将 1倍速率的时域信号内插为 M倍过釆样速率的时域信 号;
第一滤波单元, 用于对所述 M倍过釆样速率的时域信号进行粗滤波; 第一计算单元,用于计算粗滤波后 M倍过釆样速率的时域信号各釆样点 幅度或幅度的平方;
第一选择单元, 用于选择幅度大于 (/ x TAr^AoA O)或幅度的平方大于
( x 7¾m^oW 0)2的釆样点作为准峰值点,其中, 为准峰值点的门限调整因子, γ < \
第三滤波单元, 用于对所述准峰值点进行补充精度滤波;
第三计算单元 , 用于将粗滤波得到的准峰值点的实部与虚部值与补充精 度滤波得到的补偿值叠加;
第三选择单元, 用于根据叠加结果, 选择幅度大于 7¾re oW0或幅度的 平方大于 Threshold^1的准峰值点作为 M倍过釆样速率的时域信号的峰值点; 第二计算单元, 用于根据所述峰值点的位置信息, 通过下式计算该峰值 点对应的基准消峰信号的序号索引、 圓周移位的位数与幅度相位调整因子, 所述峰值点的特征信息包括峰值点的位置信息、 序号索引、 圓周移位的位数 与幅度相位调整因子:
m0 = (i - l)%M + l ,0 =L(,' - 1) /M」 + 1 f = (|x(z')| - λ χ Threshold 0)
Figure imgf000042_0001
其中, 为序号索引, z。为圓周移位的位数, /。为幅度相位调整因子, i 为所述 M倍过釆样的时域信号的峰值点位置, x(i) 为峰值点位置 i对应的正 交频分复用符号, ThresholdO为预先设定的幅度门限, /1为门限调整因子。
27、 根据权利要求 24、 25或 26所述的峰值对消装置, 其特征在于, 还 包括:
第二存储模块, 用于存储所述峰值点的特征信息。
28、根据权利要求 23至 26任意一项所述的峰值对消装置,其特征在于, 所述峰值对消模块包括:
选取单元, 用于根据所述峰值点的特征信息, 选取基准消峰信号; 圓周移位单元, 用于根据所述峰值点的特征信息, 对选取的基准消峰信 号进行圓周移位;
幅相调整单元, 用于根据所述峰值点的特征信息, 对选取的基准消峰信 号, 对圓周移位后的基准消峰信号进行幅度调整与相位旋转, 生成与所述峰 值点对应的消峰信号;
消峰单元, 用于利用与所述峰值点对应的所述消峰信号对所述 1倍速率 的时域信号进行消峰。
29、根据权利要求 28所述的峰值对消装置, 其特征在于, 所述选取单元 包括:
判断单元,用于判断所述 M倍过釆样速率的时域信号的峰值点是否落在 所述 1倍速的时域信号的釆样点上;
第一选取单元,用于根据所述 M倍过釆样速率的时域信号的峰值点位置 选取 M个基准消峰信号中的第 d + 1个基准消峰信号作为该 M倍过釆样速率 的时域信号的峰值点的基准消峰信号, 其中, d为该 M倍过釆样速率的时域 信号的峰值点与前面相邻 1倍速率的时域信号釆样点相隔的釆样点的个数, d=0,l,...,M-l。
30、 根据权利要求 28所述的峰值对消装置, 其特征在于, 还包括: 緩存模块 , 用于緩存所述 1倍速率的时域信号;
所述预测模块用于根据所述緩存模块中緩存的 1倍速率的时域信号, 预 测 M倍过釆样速率的时域信号的峰值点的特征信息。
31、 根据权利要求 30所述的峰值对消装置, 其特征在于, 还包括: 检测模块, 用于检测消峰信号的峰值幅度与消峰次数;
判断模块, 用于判断所述峰值幅度是否小于预先设定的幅度阈值, 或者 消峰次数是否达到预先设定的次数阈值, 若判断结果为真, 则直接输出所述 消峰信号; 否则, 将所述消峰信号送入所述緩存模块。
32、根据权利要求 30所述的峰值对消装置, 其特征在于, 还包括如权利 要求 19至 22任意一项所述的基准消峰信号生成装置, 用于生成所述峰值对 消模块所需的基准消峰信号。
33、 根据权利要求 30所述的峰值对消装置, 其特征在于, 还包括: 第一存储模块, 用于存储所述峰值对消模块所需的基准消峰信号。
PCT/CN2008/072786 2007-10-26 2008-10-22 Procédé et dispositif de suppression de valeur de pic, et dispositif de production de signaux de suppression de valeur de pic de référence WO2009056044A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200710176420.5 2007-10-26
CN2007101764205A CN101420404B (zh) 2007-10-26 2007-10-26 峰值对消方法、峰值对消装置与基准消峰信号生成装置

Publications (1)

Publication Number Publication Date
WO2009056044A1 true WO2009056044A1 (fr) 2009-05-07

Family

ID=40590552

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2008/072786 WO2009056044A1 (fr) 2007-10-26 2008-10-22 Procédé et dispositif de suppression de valeur de pic, et dispositif de production de signaux de suppression de valeur de pic de référence

Country Status (2)

Country Link
CN (1) CN101420404B (zh)
WO (1) WO2009056044A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111343119A (zh) * 2018-12-18 2020-06-26 深圳市中兴微电子技术有限公司 一种数据处理方法及装置、计算机可读存储介质
CN111490957A (zh) * 2020-03-10 2020-08-04 熊军 一种时域生成前导序列的方法及装置

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102223337B (zh) * 2010-04-16 2014-04-16 华为技术有限公司 基准对消信号生成方法和装置
CN103178806B (zh) * 2011-12-23 2015-11-04 中国科学院声学研究所 一种一维数据的包络提取方法及系统
CN103312482B (zh) * 2012-03-14 2016-12-14 中兴通讯股份有限公司 下行基带信号生成方法及装置、基站
CN104519004B (zh) * 2013-09-26 2018-01-16 中国科学院上海高等研究院 Ngb‑w系统的预留子载波位置图案的形成方法
US9893928B2 (en) * 2013-10-22 2018-02-13 Telefonaktiebolaget Lm Ericsson (Publ) Method and apparatus for peak to average power ratio reduction
CN105282081A (zh) * 2014-06-27 2016-01-27 中兴通讯股份有限公司 一种载波频偏估计的方法及装置
CN109309542B (zh) * 2018-10-11 2021-03-23 西北工业大学 一种基于时域过采样的正交信分复用水声通信方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6175551B1 (en) * 1997-07-31 2001-01-16 Lucent Technologies, Inc. Transmission system and method employing peak cancellation to reduce the peak-to-average power ratio
US20030133433A1 (en) * 1999-06-02 2003-07-17 Cimini Leonard Joseph Method and system for reduction of peak-to-average power ratio of transmission signals comprising overlapping waveforms
US20070153932A1 (en) * 2000-03-28 2007-07-05 Cimini Leonard J Jr Ofdm communication system and method having a reduced peak-to-average power ratio
CN101035105A (zh) * 2007-04-06 2007-09-12 西安电子科技大学 基于ifft/fft的预留子载波降低ofdm系统峰均功率比的方法和装置

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060029158A1 (en) * 2002-11-29 2006-02-09 Dietmar Lipka Amplitude peak cancellation
CN101060344B (zh) * 2006-04-19 2010-05-12 大唐移动通信设备有限公司 时分同步码分多址系统中的中频消峰方法和系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6175551B1 (en) * 1997-07-31 2001-01-16 Lucent Technologies, Inc. Transmission system and method employing peak cancellation to reduce the peak-to-average power ratio
US20030133433A1 (en) * 1999-06-02 2003-07-17 Cimini Leonard Joseph Method and system for reduction of peak-to-average power ratio of transmission signals comprising overlapping waveforms
US20070153932A1 (en) * 2000-03-28 2007-07-05 Cimini Leonard J Jr Ofdm communication system and method having a reduced peak-to-average power ratio
CN101035105A (zh) * 2007-04-06 2007-09-12 西安电子科技大学 基于ifft/fft的预留子载波降低ofdm系统峰均功率比的方法和装置

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111343119A (zh) * 2018-12-18 2020-06-26 深圳市中兴微电子技术有限公司 一种数据处理方法及装置、计算机可读存储介质
CN111343119B (zh) * 2018-12-18 2022-10-14 深圳市中兴微电子技术有限公司 一种数据处理方法及装置、计算机可读存储介质
CN111490957A (zh) * 2020-03-10 2020-08-04 熊军 一种时域生成前导序列的方法及装置
CN111490957B (zh) * 2020-03-10 2023-06-16 北京睿信丰科技有限公司 一种时域生成前导序列的方法及装置

Also Published As

Publication number Publication date
CN101420404B (zh) 2011-06-01
CN101420404A (zh) 2009-04-29

Similar Documents

Publication Publication Date Title
WO2009056044A1 (fr) Procédé et dispositif de suppression de valeur de pic, et dispositif de production de signaux de suppression de valeur de pic de référence
JP5522605B2 (ja) Ofdm受信機
KR100923892B1 (ko) 고속 푸리어 변환 트위들 승산
JP4557320B2 (ja) 直交周波数分割多重化(ofdm)方式を用いるデジタル通信システムにおける適応チャンネル等化器
US7936851B2 (en) Channel equalization
US7711059B2 (en) Low noise inter-symbol and inter-carrier interference cancellation for multi-carrier modulation receivers
JP5281078B2 (ja) 受信装置、集積回路、デジタルテレビ受像機、受信方法、及び受信プログラム
JP2008537655A (ja) Ofdmシステムでの高速フーリエ変換処理
US7602853B2 (en) Method and apparatus for channel estimation
EP1866794A1 (en) Partial fft processing and demodulation for a system with multiple subcarriers
JPH11163771A (ja) Ofdm受信機のための等化方法と等化器
US20040252772A1 (en) Filter bank based signal processing
US8259786B2 (en) Channel estimation filter for OFDM receiver
CN100539568C (zh) 多载波数据接收方法、多载波调制装置及多载波调制系统
US7113559B2 (en) Efficient methods for filtering to avoid inter-symbol interference and processing digital signals having large frequency guard bands
WO2008106857A1 (fr) Procédé, dispositif pour réduire la valeur de crête d&#39;un signal et dispositif de transmission
CN107276923B (zh) 抗混叠的信道估计装置、方法以及接收机
US20090110044A1 (en) Method and Apparatus for Deciding a Channel Impulse Response
JP2003101503A (ja) Ofdm用等化装置およびofdm用等化方法
US7970068B2 (en) Mobile channel estimation for DBV-T COFDM demodulator
JP5640640B2 (ja) 受信装置、受信方法、およびプログラム
US7340000B1 (en) Decision feedback equalizer in an OFDM WLAN receiver
US20060146690A1 (en) Methods, circuits and computer program products for estimating frequency domain channel in a DVB-T receiver using transform domain complex filtering
JP5433012B2 (ja) チャネル推定回路を有する受信機
Rahimi et al. Oversampled perfect reconstruction DFT modulated filter banks for multi-carrier transceiver systems

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 08844638

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 08844638

Country of ref document: EP

Kind code of ref document: A1