WO2009012630A1 - Récepteur et procédé pour l'élimination d'interférences multi-utilisateur basé sur l'égalisation de domaine de fréquence associée à un domaine de temps - Google Patents

Récepteur et procédé pour l'élimination d'interférences multi-utilisateur basé sur l'égalisation de domaine de fréquence associée à un domaine de temps Download PDF

Info

Publication number
WO2009012630A1
WO2009012630A1 PCT/CN2007/003567 CN2007003567W WO2009012630A1 WO 2009012630 A1 WO2009012630 A1 WO 2009012630A1 CN 2007003567 W CN2007003567 W CN 2007003567W WO 2009012630 A1 WO2009012630 A1 WO 2009012630A1
Authority
WO
WIPO (PCT)
Prior art keywords
frequency domain
signal
sequence
user
length
Prior art date
Application number
PCT/CN2007/003567
Other languages
English (en)
Chinese (zh)
Inventor
Guoping Xu
Yu Xin
Yunfei Huang
Qun Wei
Xin Zhang
Dacheng Yang
Original Assignee
Zte Corporation
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 Zte Corporation filed Critical Zte Corporation
Publication of WO2009012630A1 publication Critical patent/WO2009012630A1/fr

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
    • H04L25/03006—Arrangements for removing intersymbol interference
    • H04L25/03159—Arrangements for removing intersymbol interference operating in the frequency domain
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/69—Spread spectrum techniques
    • H04B1/707—Spread spectrum techniques using direct sequence modulation
    • H04B1/7097—Interference-related aspects
    • H04B1/7103—Interference-related aspects the interference being multiple access interference
    • H04B1/7105—Joint detection techniques, e.g. linear detectors
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/69—Spread spectrum techniques
    • H04B1/707—Spread spectrum techniques using direct sequence modulation
    • H04B1/7097—Interference-related aspects
    • H04B1/7103—Interference-related aspects the interference being multiple access interference
    • H04B1/7107—Subtractive interference cancellation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
    • H04L25/03006—Arrangements for removing intersymbol interference
    • H04L25/03178—Arrangements involving sequence estimation techniques
    • H04L25/03312—Arrangements specific to the provision of output signals
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/69—Spread spectrum techniques
    • H04B1/707—Spread spectrum techniques using direct sequence modulation
    • H04B1/7097—Interference-related aspects
    • H04B1/7103—Interference-related aspects the interference being multiple access interference
    • H04B1/7107—Subtractive interference cancellation
    • H04B1/71072—Successive interference cancellation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
    • H04L25/03006—Arrangements for removing intersymbol interference
    • H04L2025/0335—Arrangements for removing intersymbol interference characterised by the type of transmission
    • H04L2025/03375—Passband transmission
    • H04L2025/03414—Multicarrier

Definitions

  • the present invention relates to the field of mobile communications, and in particular to a CDMA system with a cyclic prefix (CP) and a normal CDMA system without a cyclic prefix, frequency domain equalization of uplink multipath channels combined with time domain multi-user (packet) serial interference cancellation Receiver and receiving method, and interference removing unit ICU in the receiver.
  • CP cyclic prefix
  • packet serial interference cancellation Receiver and receiving method
  • interference removing unit ICU in the receiver.
  • Cyclic prefix method (CP method: Cyclic Prefix Method);
  • Zero-padding method Zero Padding Method
  • the three frequency domain equalization methods of CP/ZP/OC are implemented by frequency domain equalizers, which are based on the time domain convolution corresponding to the frequency domain multiplication.
  • the processing methods used are all
  • the data is transformed into the frequency domain using a fast Fourier transform FFT module, then subjected to frequency domain equalization, and then transformed back into the time domain using the IFFT module.
  • the main differences between them are:
  • the CP and ZP methods use redundant data to naturally divide data into frequency domain equalization. Data segment, and since there is a complete correspondence between time domain convolution and frequency domain multiplication in the two methods, there is no error in the frequency domain equalization process;
  • the first two methods all need to add redundant data. That is to say, in order to complete the frequency domain equalization of the CP method and the ZP method, it is necessary to adjust the frame structure of the CDMA system; and the frequency domain equalization by the OC method is not required. By adjusting the frame structure, it is not necessary to reduce the transmission efficiency of the system by adding redundant data, and the data of the existing CDMA system can be directly received.
  • the technical problem to be solved by the present invention is to provide a receiver and a receiving method based on frequency domain equalization combined with time domain multi-user interference deletion, which is used for accelerating multi-user interference deletion while reducing the complexity of channel equalization.
  • the convergence speed of the algorithm in order to quickly and efficiently guarantee the quality of signal reception, while providing an interference cancellation unit for the receiver.
  • the present invention provides an interference deletion unit, which includes a frequency domain equalization subunit connected in series, an interference deletion subunit, and a decision module, where:
  • the frequency domain equalization subunit receives the residual signal E g , p of the current iteration level of the user and the reconstructed signal of the previous iterative stage of the user, and sums the two and then performs fast Fourier transform into the frequency domain signal. , then perform frequency domain equalization;
  • the subunit interference cancellation unit receives the signal from frequency domain equalization on the frequency domain sub-equalization, by inverse fast Fourier transform to obtain time-domain signal sequence user ⁇ ⁇ , said signal sequence ⁇ ⁇ ⁇ performed after despreading
  • Nonlinear processing yields signal sequence 2:
  • the non-linearly processed signal sequence ⁇ when the current level is not the last iteration level. Reconstructed to generate the reconstructed signal i gp of the user level, and the reconstructed signal is output to the next iteration level of the user. Interference deleting unit, the reconstructed signal is further subjected to 1 ⁇ +1 £ ⁇ + 1 ⁇ - operation, and the obtained residual signal E ⁇ +1 is output to the interference deleting unit of the next user's iteration level;
  • the decision module is configured to receive the non-linearly processed signal sequence and perform a hard decision.
  • the frequency domain equalization subunit includes a serially connected input terminal, a fast Fourier transformer, and a frequency domain equalizer, wherein the input end receives the residual signal of the user's current iteration level and the user previous one.
  • the iterative-level reconstructed signal ⁇ , ⁇ , and the sum of the two is sent to the fast Fourier transformer to transform to the frequency domain signal, and then the frequency domain equalizer performs frequency domain equalization processing, and the frequency domain equalized processed signal
  • the sequence is output to the interference deletion subunit.
  • the interference deletion subunit includes a fast Fourier inverse transformer, a despreading module, a nonlinear function module, a reconstruction module, and a second output, which are sequentially connected in series, and further includes connecting the nonlinear function module and the judgment a third output end of the module, and a first output end respectively connected to the input end and the second output end, wherein
  • the fast Fourier inverse transformer performs fast Fourier transform on the signal in the frequency domain equalization process to obtain a user signal sequence ⁇ , and the signal sequence is despread by the despreading module and then sent to the nonlinear function module to obtain a non-
  • the linearly processed signal sequence ⁇ when the current level is the last iteration stage, the nonlinearly processed signal sequence; ⁇ is output to the decision module through the third output terminal for decision; In the iterative stage, the non-linearly processed signal sequence is output to the reconstruction module to generate the reconstructed signal I g , p of the user level, and the reconstructed signal is output to the user's next iteration through the second output.
  • the nonlinear function module processes the user signal sequence Z by using a nonlinear function relationship, that is, a user after frequency domain equalization and despreading for constant amplitude modulation of BPSK, QPSK or 8PSK
  • a nonlinear function relationship that is, a user after frequency domain equalization and despreading for constant amplitude modulation of BPSK, QPSK or 8PSK
  • DEC (.) means hard judgment
  • Arg (.) e (- ⁇ , ⁇ is to calculate the principal value of a complex number
  • the decision module is configured to perform a hard decision on the signal sequence ⁇ that has undergone nonlinear processing in the last iteration, and determine the user's data as a symbol on the constellation point.
  • the frequency domain equalizer uses different frequency domain equalization processing methods for different code division multiple access signals when performing frequency domain equalization processing:
  • the cyclic prefix method is used for frequency domain equalization processing
  • the zero-padding method is used for frequency domain equalization processing
  • the frequency domain equalization processing is performed by using an overlap shear method or an adaptive overlap shear method.
  • the frequency domain equalizer performs frequency domain equalization processing on the received signal by using the determined guard interval after adaptively setting the guard interval of the overlap-shear method for the existing code division multiple access system.
  • the protection interval of the adaptive setting overlap shear method is:
  • the transmitting end transmits the known data of the frequency domain equalizer, and the frequency domain equalizer receives the data, calculates the data error, and determines the position and length of the guard interval of the frequency domain equalization.
  • the frequency domain equalizer receives a continuous pilot sequence transmitted by the base station, that is, a base station having a length greater than a channel delay, and the pilot sequence is known data of the receiving end;
  • the frequency domain equalizer calculates the systematic error sequence e of the frequency domain equalization; and detects from the two ends of the e to the middle direction, and if the error of a certain segment is greater than the specified threshold, the corresponding segment needs to be protected. Interval, and continue to detect the adjacent segments on the inside; otherwise stop detecting; How long is the guard interval required?
  • the present invention further provides a receiver based on frequency domain equalization combined with time domain multi-user interference deletion, comprising at least one interference deleting unit, wherein the interference deleting unit comprises an input connected in series, a fast Fourier transformer, and a frequency domain equalization. , a fast Fourier inverse transformer, a despreading module, a nonlinear function module, a reconstruction module, a second output, further comprising a decision module connected to the non-linear function module through a third output, and the input a first output end connected to the second output end, wherein:
  • the input end receives the residual signal E of the user's current iteration stage and the reconstructed signal I g of the previous iterative stage of the user, and sends the sum of the two to the fast Fourier transformer to transform into the frequency domain signal.
  • the frequency domain equalizer performs frequency domain equalization, and then transforms into the time domain by the fast Fourier inverse transformer to obtain a user signal sequence z g , p .
  • the signal sequence is despread by the despreading module and then sent to the nonlinear function module to obtain a non-
  • the nonlinearly processed signal sequence p is output to the decision module through the third output for decision, when the current level is not the last iteration level the nonlinear processed signal sequence is output to the reconstruction module generates present user of this reconstructed signal level of the reconstructed signal Ig, p is output to the user the next iteration of the present stage interference removing unit by a second output terminal,
  • the non-linear function module processes the user signal sequence Z w by using a nonlinear function relationship, that is, a user data sequence after frequency domain equalization and despreading for constant amplitude modulation of BPSK, QPSK or 8PSK.
  • a nonlinear function relationship that is, a user data sequence after frequency domain equalization and despreading for constant amplitude modulation of BPSK, QPSK or 8PSK.
  • DEC (.) means hard judgment
  • Arg(.) e (- ⁇ , ⁇ ) is the main value angle for calculating a complex number.
  • the value f + ⁇ . [g + log 10 (dish + 77,)] , where , and % are two constants used to control the condition: 1/14 ⁇ ⁇ ⁇ 1/2.
  • the frequency domain equalizer performs frequency domain equalization processing on the existing code division multiple access system signal by using an overlap shear method or an adaptive overlap shear method when performing frequency domain equalization processing.
  • the guard interval of the overlap-cut method is set, the received signal is subjected to frequency domain equalization processing by using the determined guard interval, wherein the guard interval of the adaptive setting overlap-shear method is:
  • the transmitting end transmits the known data of the frequency domain equalizer, and the frequency domain equalizer receives the data, calculates the data error, and determines the position and length of the guard interval of the frequency domain equalization.
  • the frequency domain equalizer receives a continuation pilot sequence that is sent by the base station and is longer than the channel delay, and the pilot sequence is known data of the receiving end;
  • the frequency domain equalizer calculates a systematic error sequence e of the frequency domain equalization; and detects from the two ends of the e to the middle direction, and if the error of a certain segment is greater than a specified threshold, the guard interval is required on the corresponding segment. And continue to detect the inner adjacent segments; otherwise stop the detection; determine the length of the guard interval before and after each.
  • the present invention further provides a method for adaptively setting a guard interval of an overlap shearing method, including:
  • the transmitting end sends the known data of the receiving end, the receiving end receives the data, calculates the data error, and determines the position and length of the guard interval of the frequency domain equalization. Further, the method specifically includes the following steps:
  • the transmitting end that is, the base station, transmits a continuous pilot sequence whose length is greater than the channel delay, and the pilot sequence is known data of the receiving end;
  • the receiving end that is, the frequency domain equalizer receives the pilot sequence, and calculates a systematic error sequence e of the frequency domain equalization
  • the present invention further provides a method for adaptively setting a guard interval of an overlap shear method, wherein if the signal to noise ratio is less than a set signal to noise ratio threshold, the length of the guard interval is set to 0; otherwise, the sender transmits and receives
  • the terminal knows the data, the receiving end receives the data, calculates the data error, and determines the position and length of the guard interval of the frequency domain equalization. Further, the method includes the following steps:
  • the transmitting end that is, the base station, transmits a continuous pilot sequence whose length is greater than the channel delay, and the pilot sequence is known data of the receiving end;
  • the receiving end that is, the frequency domain equalizer receives the pilot sequence, and calculates a systematic error sequence e of the frequency domain equalization
  • the segments are respectively detected. If the error of a certain segment is greater than the specified threshold, the guard interval is required on the corresponding segment, and the inner adjacent segment is continuously detected; otherwise, the detection is stopped; Determine the length of the guard interval required before and after each.
  • the present invention further provides a receiving method based on frequency domain equalization combined with time domain multi-user interference deletion, including:
  • the receiver receives the signal, performs frequency domain equalization processing on the received signal, and then transforms the frequency domain equalized processed signal back into the time domain for despreading, and performs nonlinear processing on the time domain signal by using a nonlinear function relationship.
  • the iterative stage decides the signal, otherwise it feeds back to the receiver to continue the interference removal iteration process until the final iteration level is reached.
  • the receiving method is further divided into the following steps:
  • Each interference deleting unit of the receiver receives the residual signal of the current iteration stage of the user and the reconstructed signal of the previous iterative stage of the user, and sums the two and then performs fast Fourier transform into the frequency domain signal, and then performs frequency conversion. Domain equalization
  • the user signal sequence Z g>p is obtained by inverse fast Fourier transform to the time domain, and the signal sequence is despread and then nonlinearly processed to obtain the signal sequence i g , p :
  • the nonlinearly processed signal sequence p is output to the decision module for decision;
  • -1 is:
  • DEC (.) means hard judgment
  • Arg(.) e (-; ⁇ ,; ⁇ ) is the main value angle for calculating a complex number.
  • the decision module When the interference cancellation unit is the last iteration stage, the decision module performs a hard decision on the non-linearly processed signal sequence 3 ⁇ 4 gp, and determines the user's data as a symbol on the constellation point.
  • the step of performing frequency domain equalization is to use a frequency domain equalizer.
  • the received signal is frequency-transmitted by using the determined guard interval.
  • Domain equalization processing wherein the guard interval of the adaptive setting overlap shear method is: if the signal to noise ratio is less than the set signal to noise ratio threshold, the length of the guard interval is set to 0;
  • the transmitting end transmits the known data of the frequency domain equalizer, and the frequency domain equalizer receives the data, calculates the data error, and determines the position and length of the guard interval of the frequency domain equalization.
  • the frequency domain equalization module receives a continuation pilot sequence that is sent by the base station and has a length greater than a channel delay, and the pilot sequence is known data at the receiving end; and then calculates a systematic error of the frequency domain equalization. Sequence e ; and from the two ends of e to the middle direction, the segment is separately detected. If the error of a segment is greater than the specified threshold, the guard interval is required on the corresponding segment, and the adjacent segment is continuously detected; otherwise, the detection is stopped. ; Determine the length of the guard interval required before and after each. Is a user group.
  • the receiver and the method based on frequency domain equalization combined with time domain multi-user interference deletion according to the present invention can improve the signal receiving effect to remove the interference signal because of the functions of frequency domain equalization and time domain interference deletion.
  • the present invention also provides an adaptive overlapping cut reception method. Since the frequency domain equalization can compensate the multipath channel with lower computational complexity, after analyzing the systematic error of the OC method in the frequency domain equalization, a frequency domain equalization method with accuracy exceeding the OC method is provided. : The AOC method maintains compatibility with existing CDMA systems. At the same time, the present invention designs a nonlinear function to perform nonlinear processing on the user data provided to the reconstruction module, thereby reducing the occurrence of error diffusion. Theoretical analysis and simulation have proved that the receiver we designed has better performance, low complexity and good industrial applicability. BRIEF abstract
  • FIG. 1 is a structural diagram of a receiver system according to the present invention.
  • FIG. 2 is a structural diagram of an interference deletion unit (lCU( g , p) ) in the present invention
  • Figure 3 is a schematic diagram showing the amplitude distribution of the improved pedestrian A channel error sequence
  • Figure 4 is a schematic diagram showing the amplitude distribution of the on-board A channel error sequence
  • Figure 5 is a schematic diagram showing the selection sequence of the detection sequence protection area
  • Figure 6 shows the bit error rate performance of the AOC frequency domain equalization combined serial interference cancellation system.
  • the present invention proposes a receiver technology based on frequency domain equalization technology combined with time domain nonlinear feedback (packet) serial multiuser interference cancellation. After the compensation of the multipath channel is completed by using the frequency domain equalization technology, in the process of performing (packet) serial multiuser interference deletion in the time domain, the decision mode of the user data and the reliability of the detection signal are determined by using a nonlinear function. Establishing a reasonable connection will help improve reception performance.
  • the frequency domain equalization technique here needs to adopt the CP method for frequency domain equalization for the CDMA system with additional CP; if the CDMA system for ZP, the ZP method for frequency domain equalization is needed.
  • the invention theoretically deduces the detection protection interval of the traditional OC method, and obtains an important conclusion: If the OC method protection interval is adaptively changed according to the actual situation of the channel, the system error of the OC method frequency domain equalization can be reduced, and the system error can be improved. The BER performance of the system. Therefore, for the existing CDMA system, the OC method or the adaptive overlap-shear method (AOC method: Adaptive Overlap-cut Method) is used in the frequency domain equalization, and the BER performance of the AOC frequency domain equalization is better than the OC method. The frequency domain equalization is close to the bit error rate performance of the CP method in the frequency domain equalization.
  • AOC method Adaptive Overlap-cut Method
  • the reliability of the user's detection data in each (packet) serial interference removal iteration level is different.
  • the (packet) serial interference deletion algorithm converges, the higher the level of iteration, the higher the credibility of the user's detection data.
  • a nonlinear function is established to closely link the decision mode of the user's test data with the credibility of the data. In this way, the serial (packet serial) interference cancellation algorithm is quickly converged, and the receiver can obtain better detection performance with fewer iterations of the (packet) serial interference cancellation algorithm.
  • Serial interference cancellation and packet serial interference deletion only differ in the case of packet serial interference deletion, the user must The noise ratio is grouped. First, the user group with the strongest signal-to-noise ratio is detected, and then the groups are detected in order according to the magnitude of the signal-to-noise ratio. All users in the group use parallel interference deletion. When detecting a group, the reconstructed data of all users in the previous group is deleted, so serial interference deletion is performed between the groups. Parallel interference cancellation within the group can reduce system delay, and serial interference cancellation between groups can minimize the interference of strong users to weak users. If there is only one user in each packet in the packet serial interference cancellation algorithm, then it becomes serial interference cancellation reception. Therefore, serial interference cancellation is a special case of packet serial interference cancellation. Since the serial interference cancellation and the packet serial interference removal structure are very similar and the principle is basically the same, the present invention describes the receiver by taking serial interference deletion as an example.
  • ICU Interference Cancellation Unit
  • the interference cancellation unit modules ICU (g , p ) in Figure 1 there are two inputs: the residual signal E g , p and the reconstructed signal for this user (group) in the previous iteration ⁇ (here
  • the subscript g represents the gth interference deletion level; and the subscript represents the first user or the user group, because grouping according to the signal to noise ratio may result in a group of multiple users or one user, so the uniform is used here. representative).
  • the residual E u is equal to the received signal of the system.
  • ICU ( ⁇ has three outputs: I, is the reconstructed signal provided to the next iteration of the user (group), and provides the residual signal to the next user (group) of this level of iteration, if the detection data of this level of iteration is to be To make a decision, the third output P is used.
  • FIG. 2 is a detailed block diagram of a module ICU ( ⁇ . It should be noted that if there are multiple users in the ICU (the user group to be processed by the ⁇ module), these users will perform parallel processing.
  • Each ICU ( ⁇ ) includes an input connected in series, a fast Fourier transformer, a frequency domain equalizer, a fast Fourier inverse transformer, a despreading module, a nonlinear function module, a reconstruction module, a second output, and The third output end connecting the nonlinear function module and the decision module, and the first output end connected to the input end and the second output end respectively, wherein:
  • the input end receives the residual signal E gip of the current iteration stage of the user and the reconstructed signal 1 £ _ 1>/; of the previous iterative stage of the user, and sends the sum of the two to the fast Fourier transformer to convert to the frequency
  • the frequency domain equalizer performs frequency domain equalization, and then transforms into the time domain by the fast Fourier inverse transformer to obtain a user signal sequence Z g , p , and the signal sequence is despread and sent by the despreading module.
  • the nonlinear function module obtains the nonlinearly processed signal sequence. When the current stage is the last iteration stage, the nonlinearly processed signal sequence i gp is output to the decision module through the third output terminal for decision.
  • the input residual signal E g , p When performing signal processing, first, the input residual signal E g , p will be added to the reconstructed signal I g _ ⁇ of the user (group) currently to be detected in the previous iteration. Transforming to the frequency domain by the FFT module, for different CDMA systems (CP_CDMA, ZP-CDMA or existing CDMA systems), using the CP method, the ZP method, and the OC method or the AOC method for frequency domain equalization, respectively, and then The data in the frequency domain equalization is transformed back to the time domain by the IFFT module to obtain the signal sequence Z g , p .
  • CDMA Code Division Multiple Access
  • ZP-CDMA ZP-CDMA or existing CDMA systems
  • the user's data is processed using a nonlinear function to obtain the processed signal sequence: ⁇ p .
  • the feedback of the user's data is linked to the detected credibility by the processing of the nonlinear function, so that the entire serial iterative algorithm converges at a faster rate.
  • the set iteration level is level 6, then after the end of the level 6 iterative process, before the system finally outputs the user's data, a hard decision is made on the data of each user obtained from the last iteration of the interference deletion, that is, the user is
  • the data is judged to be a symbol on the constellation point. For example, in BPSK, 5.5 is judged to be + 1, - 2.2 is judged to be -1; for example, in QPSK, 1.01 + 1.05i is judged to be 1+i Constellation point.
  • the CP method and the ZP method can completely realize the transition from the channel matrix to the cyclic structure, so the bit error rate performance is good, but it is necessary to add redundant data to the CDMA system, thereby reducing the spectrum utilization rate of the system. If the CDMA system to which the CP or the ZP is attached is used, the frequency domain equalization in the present invention correspondingly adopts the CP method frequency domain equalization and the ZP method frequency domain equalization.
  • the OC method is a computational method that approximates linear convolution to circular convolution. Due to this similarity, the bit error rate performance is slightly poor.
  • the present invention proposes an adaptive overlapping shear frequency domain equalization receiver with improved performance and low complexity for the CDMA downlink channel.
  • the frequency domain equalization of a CDMA system is a chip level channel equalization process.
  • Discrete multipath channels can be characterized by a tapped delay line model.
  • the signal received by the mobile station can be expressed as:
  • ⁇ is the channel matrix formed by the channel impulse response of (M+iL-1)xM dimension, and n is Gaussian white noise.
  • the output of the frequency domain equalizer can be obtained as follows:
  • X is the estimated value of the frequency domain of the transmitted signal
  • c ⁇ c ⁇ ,..., ⁇ ] is the equalization coefficient of the frequency domain equalizer, which can be based on the Zero-Forcing (ZF) criterion or the minimum mean square error (Minimum) Mean Square Error, MMSE) guidelines are obtained.
  • Y 3 ⁇ 4, which is the Fourier transform of the received signal. From the correspondence between the circular convolution and the FFT, the frequency domain samples of the received sequence are multiplied by the equalization coefficients to complete the frequency domain equalization calculation. Finally, the IFFT is used to transform the equalized sequence back into the time domain for despreading and decision.
  • the system error distribution characteristics of the OC method are analyzed below, and an adaptive overlapping shear receiver receiving method is proposed based on the error distribution characteristics.
  • the former represents the block-and-comb signal of all users sent by the CDMA system, after Represents the impulse response of a discrete multipath channel.
  • rL- ⁇ A: - 1 This gives the circular convolution from r.
  • the first -1 values of the -1 values and the linear convolution between X and h are not the same.
  • ⁇ ⁇ — ⁇ will be equal to the Lth value of the linear convolution to the Nth value.
  • the sequence is the same. Let us estimate how many phase changes in each of the N members in each of them can be kept within ⁇ /2. Since the numerator and denominator of the formula (11) have a similar form, it is sufficient to estimate the phase change of the molecular portion. Referring to formula (11), the molecular part of ⁇
  • Figure 3 and Figure 4 are the distributions of the OC method error e of the ITU M.1225 modified pedestrian A channel and the vehicle A channel, respectively.
  • the transmission data is randomly generated, and the difference between the linear convolution and the cyclic convolution of the data and the channel is used to obtain ⁇ - ⁇ , ⁇ . ⁇ ], and the distribution of the error is calculated by the equation (9).
  • the value of z is large according to equation (15).
  • equation (10) when calculating a, when it is close to 0 or N-1, the adjacent phase weighting coefficient ⁇ -* ⁇ "" will be smaller, which results in continuous in equation (10).
  • the z-phases are close to the ⁇ -addition.
  • the N terms of a are divided into
  • the ⁇ - phase in each segment is similar, so the summation and the modulus in the segment are large. If the sum between the segments cannot be canceled, the energy of the error sequence e is concentrated at both ends; otherwise, the error at both ends may be small.
  • the length of protection required to accurately estimate the frequency domain equalization of the AOC method is preferably based on the time-division band that is less interfered, and is generally used in the downlink channel of the existing CDMA system. Both sets the time-divided frequency band. This time-division band is exactly the same for users in the same cell, so the interference is small (ignoring the interference of the neighboring cell).
  • the time division pilot band in the downlink channel can be used to estimate the length of the AOC frequency domain equalization protection based on equations (8) and (9), and then fed back to the base station.
  • AOC method frequency domain equalization is also applicable to the reception of downlink signals in CDMA systems. This is because the multi-user signals received by the mobile station experience the same multipath channel in the downlink channel, so the frequency domain equalization can eliminate the inter-symbols simultaneously. Interference (ISI: Inter-Symbol Interference) and Multi-Access Interference (MAI).
  • ISI Inter-Symbol Interference
  • MAI Multi-Access Interference
  • the additional work required is the setting of the protection length that the mobile station feeds back to the base station.
  • the estimation of the protection length required for the frequency domain equalization of the uplink signal AOC method can also be directly performed by the base station. It can be seen from equation (8) that an accurate estimate of the required guard length can be obtained based on the pilot band of the code division as long as other data of the detected position is known.
  • the detection is performed from the opposite ends of e to the middle direction.
  • the amplitude mean l , l of the first segment before and after is calculated separately, and if the mean is greater than the threshold ⁇ (which is a multiple of 1 for the threshold, it can be directly set to an integer as needed), then This interval requires a guard interval; otherwise it is not needed.
  • the outer section needs protection, then The same method is applied to the adjacent segments on the inside to determine whether it is necessary to increase the length of the protection; otherwise, the detection of the adjacent segments on the inside is stopped.
  • the maximum protection length before and after setting is 4, which is determined according to equation (17).
  • the non-linear function module used in Figure 2 is to correlate the fed back user data with the credibility of the currently detected user data.
  • the higher the number of serial interference deletion stages the higher the credibility of the user data after despreading in the ICU module. And this degree of trust and the number of iterations is a complex nonlinear relationship.
  • DEC (.) indicates a hard decision
  • ⁇ (. ) £ ( ⁇ ) is the main value of a complex number.
  • % e ⁇ ( ⁇ is determined by the iteration series and the signal to noise ratio Threshold, which is modeled as follows:
  • g /(g,SM?) (19)
  • g is the level of the interference removal iteration, which is the signal-to-noise ratio of the user.
  • the performance of a receiver based on frequency domain equalization combined with time domain multiuser interference cancellation is illustrated by an example.
  • the established simulation system is as follows: The normal CDMA system without additional CP or ZP is back channel, uncoded, the RF carrier frequency is 2 GHz, and the channel adopts the vehicle A channel model specified by M.1225, BPSK modulation. The chips are transmitted in time slots, each time slot is 2048 chips, and the chip rate is 1.2288 Mchip/s. The ideal channel estimate is used in the simulation.
  • Frequency domain equalization can compensate for multipath channels with low computational complexity.
  • the systematic error of the OC method for frequency domain equalization provides a frequency domain equalization method with accuracy exceeding the OC method:
  • the AOC method maintains compatibility with existing CDMA systems.
  • Theoretical analysis and simulation have proved that the receiver designed by us has better performance, low complexity and good industrial applicability.
  • the present invention is applicable to a CDMA system, and particularly to a conventional CDMA system without a cyclic prefix, which utilizes a frequency domain equalization method with an accuracy exceeding the OC method: AOC method, which reduces the computational complexity of multipath channel compensation, and can maintain the existing CDMA system.
  • AOC method which reduces the computational complexity of multipath channel compensation, and can maintain the existing CDMA system.
  • the reconstruction module of the interference removal unit performs nonlinear processing on the user data, which reduces the occurrence of error diffusion.
  • the receiver of the invention has better performance, low complexity and good industrial applicability.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Power Engineering (AREA)
  • Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)

Abstract

La présente invention concerne un récepteur et un procédé de réception, servant pour éliminer les interférences multi-utilisateur en fonction d'une égalisation de domaine de fréquence associée à un domaine de temps. Le récepteur reçoit les signaux, réalise le traitement d'égalisation de domaine de fréquence sur les signaux reçus, puis convertit les signaux après l'égalisation de domaine de fréquence de retour au domaine de temps pour les diffuser et réaliser le traitement non-linéaire sur les signaux du domaine de temps à l'aide d'une relation de fonction non-linéaire, décide des signaux à la dernière étape itérative, sinon les renvoie au récepteur pour réaliser, de manière séquentielle, le traitement itératif d'élimination des interférences, jusqu'à ce qu'ils parviennent à la dernière étape itérative. L'invention peut être compatible avec la présente technologie CDMA, réduit la complexité de la compensation de canal multi-parcours, diminue les occasions itératives d'élimination d'interférences et évite les erreurs à répartir.
PCT/CN2007/003567 2007-07-25 2007-12-12 Récepteur et procédé pour l'élimination d'interférences multi-utilisateur basé sur l'égalisation de domaine de fréquence associée à un domaine de temps WO2009012630A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN2007101297551A CN101192846B (zh) 2007-07-25 2007-07-25 一种自适应设置重叠剪切法的保护间隔的方法
CN200710129755.1 2007-07-25

Publications (1)

Publication Number Publication Date
WO2009012630A1 true WO2009012630A1 (fr) 2009-01-29

Family

ID=39487637

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2007/003567 WO2009012630A1 (fr) 2007-07-25 2007-12-12 Récepteur et procédé pour l'élimination d'interférences multi-utilisateur basé sur l'égalisation de domaine de fréquence associée à un domaine de temps

Country Status (2)

Country Link
CN (1) CN101192846B (fr)
WO (1) WO2009012630A1 (fr)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021030951A1 (fr) * 2019-08-16 2021-02-25 Qualcomm Incorporated Planification de communication air-sol
TWI723900B (zh) * 2020-06-16 2021-04-01 英業達股份有限公司 決定連續時間線性等化器設定值之方法
CN116436739A (zh) * 2023-06-08 2023-07-14 西南交通大学 一种信道估计方法、装置、设备及可读存储介质

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105227287B (zh) * 2014-06-17 2018-12-25 华为技术有限公司 数据传输同步方法及装置
CN104778150B (zh) * 2015-03-27 2018-12-14 华为技术有限公司 一种频域处理方法及装置
CN106953817A (zh) * 2017-03-23 2017-07-14 电子科技大学 60GHz毫米波通信系统中重叠剪切均衡器的实现方法
CN108429709A (zh) * 2018-03-20 2018-08-21 中山大学 一种无线时变信道下sc-fde系统的信道均衡方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1574820A (zh) * 2003-06-18 2005-02-02 三星电子株式会社 时域同步正交频分复用接收装置及其均衡方法
CN1617530A (zh) * 2004-12-06 2005-05-18 山东大学 一种选频方式的单载波分块传输方法
CN1972509A (zh) * 2006-11-17 2007-05-30 凯明信息科技股份有限公司 基于串行干扰消除的多小区信道估计方法

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1574820A (zh) * 2003-06-18 2005-02-02 三星电子株式会社 时域同步正交频分复用接收装置及其均衡方法
CN1617530A (zh) * 2004-12-06 2005-05-18 山东大学 一种选频方式的单载波分块传输方法
CN1972509A (zh) * 2006-11-17 2007-05-30 凯明信息科技股份有限公司 基于串行干扰消除的多小区信道估计方法

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021030951A1 (fr) * 2019-08-16 2021-02-25 Qualcomm Incorporated Planification de communication air-sol
TWI723900B (zh) * 2020-06-16 2021-04-01 英業達股份有限公司 決定連續時間線性等化器設定值之方法
CN116436739A (zh) * 2023-06-08 2023-07-14 西南交通大学 一种信道估计方法、装置、设备及可读存储介质
CN116436739B (zh) * 2023-06-08 2023-09-05 西南交通大学 一种信道估计方法、装置、设备及可读存储介质

Also Published As

Publication number Publication date
CN101192846B (zh) 2011-11-30
CN101192846A (zh) 2008-06-04

Similar Documents

Publication Publication Date Title
JP4609625B2 (ja) 干渉キャンセル方法および装置
JP4189477B2 (ja) Ｏｆｄｍ（直交周波数分割多重）適応等化受信方式及び受信機
US7215704B2 (en) Adaptive equalizing apparatus and program therefor
US8295383B2 (en) Receiving apparatus and mobile communication system
US8811145B2 (en) Detection method and apparatus for multiple-input multiple-output single carrier block transmission system
US7974372B2 (en) OFDM receiving method and receiving apparatus
US7212569B1 (en) Frequency domain decision feedback equalizer
JP4854094B2 (ja) 無線受信機、無線通信システムおよび無線受信方法
Zemen et al. Iterative joint time-variant channel estimation and multi-user detection for MC-CDMA
Adachi et al. Introduction of frequency-domain signal processing to broadband single-carrier transmissions in a wireless channel
US20040132416A1 (en) Equalisation apparatus and methods
EP1903732A2 (fr) Détermination au domaine fréquentielle de coefficients d'égalisateur pour radio MIMO/MISO
US9768844B2 (en) Double iterative MIMO receiver
US8325588B2 (en) Iterative reception method and iterative receiver
WO2009012630A1 (fr) Récepteur et procédé pour l'élimination d'interférences multi-utilisateur basé sur l'égalisation de domaine de fréquence associée à un domaine de temps
WO2009084121A1 (fr) Traitement de signal pour système de communications sans fil multi-secteur et son procédé
AU2007233563A1 (en) Channel estimation for rapid dispersive fading channels
JP2005065197A (ja) 高精度タイミング再生を備えたｍｉｍｏ−ｏｆｄｍ受信方式および受信機
US7720134B2 (en) Frequency-domain multi-user access interference cancellation and nonlinear equalization in CDMA receivers
US7450652B2 (en) Apparatus and method for receiving signals in an OFDM communication system
JP2009049491A (ja) 受信装置、受信方法及びプログラム
Suyama et al. A MIMO-OFDM receiver employing the low-complexity turbo equalization in multipath environments with delay difference greater than the guard interval
WO2005043851A1 (fr) Appareil et procede pour le precodage d'un signal de multiporteuse
Li et al. Frequency domain adaptive detectors for SC-FDE in multiuser DS-UWB systems based on structured channel estimation and direct adaptation
Li et al. Frequency-domain adaptive detectors for single-carrier frequency-domain equalisation in multiuser direct-sequence ultra-wideband systems based on structured channel estimation and direct adaptation

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: 07845917

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: 07845917

Country of ref document: EP

Kind code of ref document: A1