An equalizer for implementing multipath diversity in a spread spectrum SC-FDE system

Abstract

The application provides an equalizer for realizing multipath diversity in a spread spectrum SC-FDE system, which comprises the following steps: channel estimation is performed by using a UW word vector of each SC-FDE symbol; first noise reduction is performed on a channel impulse response estimated by using each SC-FDE symbol in a Doppler variation domain; the result after the noise reduction is converted to a time domain to perform second noise reduction; and Rake reception is performed on a data block of a sending end by using the time domain impulse response after the noise reduction, so as to perform coherent combination on multipath information in all received data blocks. The application realizes coherent combination of multipath in a spread spectrum single carrier frequency domain equalization SC-FDE system by improving a channel estimation method and combining the advantages of Rake reception, instead of passive combination, so that the SC-FDE system can obtain better performance when passing through a multipath fading channel in a low signal-to-noise ratio scene.

Description

Technical Field

[0001] This invention belongs to the field of communication transmission technology, and specifically relates to an equalizer for implementing multipath diversity in a spread spectrum SC-FDE system. Background Technology

[0002] SC-FDE technology can overcome the peak-to-average power ratio problem of OFDM systems without increasing complexity, and can achieve performance similar to OFDM systems. It can also be combined with a variety of technologies. Systems that combine spread spectrum technology with SC-FDE technology can improve the anti-interference performance of the system and mainly operate in low signal-to-noise ratio scenarios.

[0003] Currently, the commonly used equalization methods in SC-FDE systems include frequency domain zero-forcing (ZF) equalization, frequency domain minimum mean square error (MMSE) equalization, and decision feedback equalization (DFE), but these methods all have certain drawbacks.

[0004] The ZF equalization algorithm is derived from the peak distortion criterion. The peak distortion criterion refers to minimizing inter-symbol interference (ISI) at the equalizer output in the worst-case scenario. The impulse response {hn} of the discrete-time linear filter and the equalizer impulse response {wn} can be cascaded into an equivalent filter {qn}. {qn} can be considered as a convolution of {wn} and {hn}. By designing the tap coefficients of the equivalent filter {qn}, ISI can be completely eliminated. The resulting zero-forcing equalizer's transfer function W(z) is the inverse filter of the linear filter model H(z), satisfying:

[0005]

[0006] The zero-forcing equalization algorithm is very simple to implement, but the ZF algorithm does not take into account the presence of noise. As a result, the noise signal may be amplified after using the zero-forcing equalizer, which will further degrade the system performance in low signal-to-noise ratio scenarios.

[0007] The MMSE (Minimum Mean Square Error) equalization algorithm addresses the shortcomings of ZF equalization by minimizing the total power of residual ISI symbol interference and additive noise at the equalizer output. It treats the mean square error at a frequency point within an SC-FDE symbol at the receiver as a function of the value Wl of the equalizer at that corresponding frequency point. To minimize the mean square error, taking the partial derivative of the mean square error expression with respect to Wl yields the expression Wl for the MMSE equalizer at a given frequency point:

[0008]

[0009] MMSE equalization can reduce the impact of noise on the received signal to some extent, but some inter-symbol interference remains, and there is potential for further performance optimization.

[0010] DFE equalization algorithm uses the traditional MMSE equalization to detect the received symbol, and judges to obtain the estimation value of the transmitted symbol, and the estimation value of the transmitted symbol can be used to estimate the ISI of the current received symbol and perform iterative cancellation. The DFE algorithm can eliminate the inter-symbol interference term in the received signal through the iterative process, but the DFE equalization is a nonlinear equalizer, and the computer complexity is high, and the hardware implementation is difficult. SUMMARY

[0011] In order to solve the above problems existing in the prior art, the application provides an equalizer for realizing multipath diversity in a spread spectrum SC-FDE system. The technical problem to be solved by the application is solved by the following technical scheme:

[0012] An equalizer for realizing multipath diversity in a spread spectrum SC-FDE system is applied to a receiving end, and the equalizer for realizing multipath diversity in the spread spectrum SC-FDE system performs the following processes:

[0013] S100, a pilot is taken out from a combination block to obtain a first UW block, channel estimation is performed on the first UW block and a second UW block of a transmitting end to obtain an estimation value of a channel coefficient vector of each SC-FDE symbol;

[0014] The first UW block and the second UW block include all UW word vectors of all SC-FDE symbols.

[0015] S200, Doppler variation domain noise reduction is performed on the estimation value of the channel coefficient vector to obtain a first time of noise-reduced channel coefficient vector;

[0016] S300, time domain noise reduction is performed on the first time of noise-reduced channel coefficient vector to obtain a second time of noise-reduced channel coefficient vector;

[0017] S400, the second time of noise-reduced channel coefficient vector is used to determine a time domain impulse response;

[0018] S500, the time domain impulse response is used to perform Rake reception on a data block to perform coherent combination on multipath information in all received data blocks.

[0019] Advantages:

[0020] 1. For the low signal-to-noise ratio scene in which the spread spectrum SC-FDE system works, the channel estimation result is subjected to noise reduction processing in the Doppler variation domain and the time domain, respectively. The application first performs a first time of noise reduction operation in the Doppler variation domain, converts the noise-reduced result to the time domain by using IFFT transformation, and then performs a second time of noise reduction processing in the time domain. The operation steps are simple, the channel estimation accuracy can be effectively improved, and the influence of noise in subsequent multipath combination is reduced.

[0021] 2、The application applies Rake combination to SC-FDE system, firstly utilizes UW block to complete channel estimation, and carries out multiple denoising to channel estimation result, carries out Rake reception to data block according to time domain impulse response obtained after multiple denoising, so as to carry out coherent combination of multipath information to all data blocks. The coherent combination of the application is different from the classic passive combination, and can make SC-FDE system obtain better bit error rate performance in the face of multipath fading channel, especially deep fading channel.

[0022] The application will be further described in detail below in combination with the drawings and embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 is a schematic diagram of the spread spectrum SC-FDE system provided by the application;

[0024] Figure 2 is a schematic diagram of the realization process of the equalizer for realizing multipath diversity in the spread spectrum SC-FDE system provided by the application;

[0025] Figure 3 is a principle block diagram of Rake reception provided by the application;

[0026] Figure 4 is a MSE curve diagram of channel impulse response when MCS combination is QPSK-0.25-16 under the LOS scenario provided by the application;

[0027] Figure 5 is a comparison diagram of bit error rate performance when MCS combination is QPSK-0.25-16 under the LOS scenario provided by the application;

[0028] Figure 6 is a MSE curve diagram of channel impulse response when MCS combination is QPSK-0.25-4 under the LOS scenario provided by the application;

[0029] Figure 7 is a comparison diagram of bit error rate performance when MCS combination is QPSK-0.25-4 under the LOS scenario provided by the application. DETAILED DESCRIPTION

[0030] The application will be further described in detail below in combination with the drawings and embodiments.

[0031] Rake receiver can realize high performance communication in multipath environment. In point-to-point communication system, Rake receiver can overcome frequency selective fading of channel. Since direct sequence spread spectrum link in multi-user environment is interference limited, using processing that can reduce interference will improve communication quality or system capacity of link. In mobile communication system, there are two main interference sources, one is interference caused by other users, this kind of interference can be overcome by power control, multi-user detection and good cross-correlation characteristics of spread spectrum sequence; the other interference comes from multipath phenomenon. Rake receiver has tap structure, this unique structure can reduce multipath interference, and even use multipath phenomenon to improve quality of received signal to some extent. Combining Rake reception with spread spectrum SC-FDE system can realize effective use of multipath information, and compared with traditional equalization algorithm, better bit error rate performance and low computer complexity can be obtained.

[0032] The present application aims at combining Rake reception technology with spread spectrum SC-FDE system in the scene of low signal-to-noise ratio and frequency selective fading channel, realizing multipath coherent combining of SC-FDE system, and improving anti-multipath ability and bit error rate performance of system.

[0033] The present application provides an equalizer for realizing multipath diversity in spread spectrum SC-FDE system, which is applied to receiving end of spread spectrum SC-FDE system, and the reference Figure 1 , Figure 1 is a principle block diagram of the equalizer for realizing multipath diversity in spread spectrum SC-FDE system. In Figure 1 , the upper subgraph is a processing schematic diagram of sending end. The lower subgraph is a processing schematic diagram of receiving end. The present application replaces traditional frequency domain equalization processing procedure with Rake reception in receiving end, and the Rake reception processing procedure is equalizer implementation procedure of the present application.

[0034] Spread spectrum SC-FDE system mainly works in low signal-to-noise ratio scene, when using UW word vector in each SC-FDE symbol to perform channel estimation, estimation result is greatly influenced by noise, and in this scene, it is necessary to perform noise reduction processing on estimated channel response, and the present application selects to perform twice noise reduction processing on estimated channel response in Doppler variation domain and time domain.

[0035] In Figure 1The sending end encodes and interleaves modulation of information bits to be sent by itself to obtain a plurality of modulation symbols, spreads the plurality of modulation symbols according to a known spreading sequence to obtain a spreading symbol vector of each modulation symbol, combines the spreading symbol vector according to the data length of an SC-FDE symbol to obtain a data part of the SC-FDE symbol, performs frame processing on the data part to obtain a subframe, and transmits the subframe to a receiving end through a channel; the frame received by the receiving end is a target frame. The subframe is transmitted through the channel, and the channel is greatly affected by noise, resulting in a difference between data of the receiving end and the sending end. Each subframe contains a plurality of SC-FDE symbols, and each SC-FDE symbol contains a spreading symbol vector corresponding to a plurality of modulation symbols.

[0036] The spreading sequence known by the sending end is represented as C = [c1, c2, c3,..., cV], V is the number of spreading symbols, and the vector expression of the jth spreading modulation symbol is xij = xij·C = [c1xij, c2xij, c3xij,..., cVxij], i represents the ith SC-FDE symbol, and the length of data in one SC-FDE symbol is N. Then frame processing is performed to obtain a subframe, and one frame contains a plurality of SC-FDE symbols, and each SC-FDE symbol includes a plurality of modulation symbols; then the data part in the ith SC-FDE symbol can be represented as:

[0037] di = [xi1, xi2, xi3,.., xij,... xiq];

[0038] Wherein q represents the number of spreading symbols in one SC-FDE symbol, N = q·V is satisfied, the UW block in the ith SC-FDE symbol can be represented as: ui = [ui(1), ui(2), ui(3),..., ui(M)], wherein M is the length of the UW word, the CP is the same as the content of the UW block, the CP of the ith SC-FDE symbol is represented as cpi, and is placed in front of the UW block. The data part is added with the UW block and the CP to obtain the ith SC-FDE symbol, which can be represented as: Si = [ui, di, cpi].

[0039] The SC-FDE symbol of the subframe is sent to the receiving end through the channel, and the time domain impulse response of the channel experienced by the ith SC-FDE symbol in an ideal case is:

[0040]

[0041] Each path has a channel coefficient hi,p (p = 1, 2, 3,..., l) and a time delay np (p = 1, 2, 3,..., l), and the maximum time delay is Lmax. The vector representation of the channel impulse response of the ith SC-FDE symbol is hi = [hi(1), hi(2),..., hi(Lmax)].

[0042] The ith SC-FDE symbol of the sending end can be expressed as:

[0043] Si=[si(1),si(2),si(3),...,si(j),...,si(N+2M)];

[0044] For one symbol si(j)∈Si, the receiving end can receive the received signal ri(j) of the symbol si(j) under the influence of the time-domain impulse response in an ideal case as follows:

[0045]

[0046] The UW block in the ith SC-FDE symbol of the sending end can be expressed as:

[0047] ui=[ui(1),ui(2),ui(3),...,ui(j),...,ui(M)];

[0048] In combination with Figure 1 and Figure 2 , the equalizer for realizing multipath diversity in the spread spectrum SC-FDE system executes the following process:

[0049] S100, the pilot is taken out from the combined block to obtain a first UW block, the first UW block and a second UW block of the sending end are subjected to channel estimation to obtain an estimated value of a channel coefficient vector of each SC-FDE symbol;

[0050] The first UW block and the second UW block include all UW word vectors of all SC-FDE symbols.

[0051] The receiving end of the present application receives a target frame; the extra pilots in the target frame are removed, and then the target frame is subjected to serial-parallel conversion to obtain a combined block; the combined block includes a data block, a first UW block and a CP block.

[0052] In an alternative embodiment of the present application, S100 includes:

[0053] Wherein, represents the UW word vector corresponding to the ith SC-FDE symbol of the receiving end, hi represents the vector of the channel impulse response of the ith SC-FDE symbol in an ideal case, ui represents the UW word vector corresponding to the ith SC-FDE symbol of the sending end, ui=[ui(1),ui(2),ui(3),...,ui(M)]; nu,i is the noise vector received by the UW block part in the ith SC-FDE symbol;

[0054] S120, performing FFT transform on the Uw word vector of the receiving end to obtain a frequency domain response value at a jth frequency point position

[0055] This step uses FFT transform to convert the received first Uw block and the second Uw block of the sending end to the frequency domain, and the frequency domain response value at the jth frequency point position is:

[0056]

[0057] Wherein, Hi(j) is the frequency domain response value of the frequency domain response vector corresponding to the time domain impulse response vector of the channel at the jth frequency point position, Ui(j) is the frequency domain response value of the second Uw block of the sending end at the jth frequency point position, Nui(j) represents the value at the jth frequency point position in the frequency domain noise vector obtained by performing FFT transform on the noise vector corresponding to the ith SC-FDE symbol, and the value range of j is 1 to M, and M is the word length of the Uw block;

[0058] S130, performing point division operation on the frequency domain response value to obtain an estimated value of the frequency domain response vector at the frequency point position j

[0059] The value of the frequency domain response vector at the jth frequency point position obtained by performing M-point FFT transform on ui=[ui(1), ui(2), ui(3),..., ui(M)], and the estimated value of the estimated channel frequency domain response vector at the frequency point position j is obtained by using point division operation Indicated as:

[0060]

[0061] Wherein, The result of dividing Ui(j) in the above formula is the error source of the frequency domain response estimation value.

[0062] S140, the estimated values at the M frequency point positions are combined to form a frequency domain response vector Indicated as:

[0063]

[0064] S150, performing IFFT transform of the corresponding point number on the frequency domain response vector to obtain an estimated value of the channel coefficient vector Indicated as:

[0065]

[0066] ​S200, Doppler domain noise reduction is performed on the estimated value of the channel coefficient vector to obtain a first noise-reduced channel coefficient vector;

[0067] In one specific embodiment of the application, S200 comprises:

[0068] S210, for any subframe, values at the same frequency point position in channel coefficient vectors corresponding to P SC-FDE symbols of the subframe are extracted to form a set Φ.

[0069] If a subframe contains P SC-FDE symbols, P channel coefficient vectors can be estimated, and values at the same frequency point position in the estimated P channel coefficient vectors can be extracted to obtain channel coefficient sets at the same delay, denoted as:

[0070] Φ = {Φ 1, Φ 2, Φ 3,..., Φ M}, Φ j = [h 1 (j), h 2 (j), h 3 (j),..., h i (j),..., h P (j)];

[0071] wherein h i (j) represents a coefficient at frequency point position j in the estimated channel coefficient vector of the i th SC-FDE symbol, and i ranges from 1 to P;

[0072] S220, P-point FFT transformation is performed on each item in the set Φ to obtain a Doppler spectrum of the channel coefficient at frequency point position j;

[0073] wherein the Doppler spectrum ranges from [-fs' / 2, fs' / 2], fs' is a sampling rate of channel estimation, and fs is a sampling rate of the SC-FDE symbol;

[0074] A total of N+2M symbols are contained in one SC-FDE symbol, and thus:

[0075]

[0076] The number of points of the Doppler spectrum is P, and thus the minimum resolution interval of the Doppler spectrum is:

[0077]

[0078] Channel coefficients at the same delay in different SC-FDE symbols are affected by the maximum Doppler, and the distribution range of the channel coefficients in the Doppler spectrum is mainly determined by the frequency dispersion bandwidth, which can be considered as Bd=2fm, wherein fm represents the maximum Doppler frequency offset, and thus it can be considered that the useful signal is mainly distributed in [-0.5Bd, 0.5Bd] in the Doppler spectrum.

[0079] According to the range of the Doppler spectrum, the frequency dispersion bandwidth Bd, and the minimum resolution interval fmin of the Doppler spectrum, the number of effective spectral lines distributed in the frequency dispersion bandwidth can be determined, and the distribution interval [-B'd, B'd] of the effective data in the Doppler spectrum, i.e., the effective information interval, can be determined according to the minimum resolution interval and the size of the frequency dispersion bandwidth, where B'd is expressed as follows:

[0080]

[0081] S230, performing noise reduction processing on the Doppler spectrum to set the spectral lines outside the effective information interval to zero to obtain a noise reduction result, and performing IFFT transformation on the noise reduction result to obtain an inverse transformation result;

[0082] This step only retains the spectral lines in the Doppler spectrum that are distributed in the interval [-B'd, B'd], and sets the spectral lines not in the interval to zero to obtain a new Doppler spectrum.

[0083] S240, recombining the values at the same positions in the set of inverse transformation results to obtain a channel coefficient vector after the first noise reduction The value range of j is 1 to M.

[0084] This step performs P-point IFFT transformation on the obtained new spectral line set to obtain a corrected channel coefficient set at the same time delay in different SC-FDE symbols, and repeats the sub-steps S220 to S230 on the channel coefficient set Φi of different time delays i to complete the first noise reduction processing on the estimated channel impulse response, thereby obtaining a channel coefficient vector after the first noise reduction of the i-th SC-FDE symbol The channel coefficient vector after the first noise reduction of the i-th SC-FDE symbol is which can be expressed as:

[0085] S300, performing time domain noise reduction on the channel coefficient vector after the first noise reduction to obtain a channel coefficient vector after the second noise reduction.

[0086] In one specific embodiment of the present application, S300 includes:

[0087] S310, performing zero setting processing on the part of the channel coefficient vector after the first noise reduction that exceeds the maximum time delay Lmax, and sorting the channel coefficients in the maximum time delay Lmax according to the modulus values from large to small to obtain a sorting result

[0088] Given that the number of multipaths is l, the part of the channel coefficient vector after the first noise reduction that exceeds the maximum time delay Lmax is set to zero, and the channel coefficients in the maximum time delay Lmax are sorted according to the modulus values from large to small to obtain: Given that the number of multipaths is l, the part of the channel coefficient vector after the first noise reduction that exceeds the maximum time delay Lmax is set to zero, and the channel coefficients in the maximum time delay Lmax are sorted according to the modulus values from large to small to obtain:​

[0089] {|hi'(q1)|, |hi'(q2)|, |hi'(q3)|,..., |hi'(q3)|,..., |hi'(qLmax)|};

[0090] Wherein, each modulus corresponding to the number is: {q1, q2, q3,..., qLmax};

[0091] S320, according to each modulus corresponding to the number, the modulus from big to small in the sorting result, the first l items are reserved, and the second time after the noise reduction channel coefficient vector is obtained

[0092] This step records the original number corresponding to each modulus: {q1, q2, q3,..., qLmax}, according to the number sorting, only the value of the item with the same item in the original channel coefficient vector is reserved to obtain the new channel coefficient vector after the second time after the noise reduction The value of the channel coefficient vector at the frequency point position j is:

[0093]

[0094] The spread spectrum SC-FDE system mainly works in a low signal-to-noise ratio environment, and the noise has a greater influence on the channel estimation result, so it is necessary to carry out noise reduction processing on the channel response estimated by using the UW block. In the present application, the received UW block is first divided by the local UW block to estimate the frequency domain response of the channel, and the estimated frequency domain response is subjected to IFFT transformation to obtain the time domain impulse response of the estimated channel. The channel coefficients at the same time delay in different SC-FDE symbols are concentrated together to perform FFT transformation to obtain the distribution of the Doppler spectrum. The channel coefficients at the same time delay in different SC-FDE symbols are affected by the maximum Doppler, and the distribution range of the Doppler spectrum is mainly determined by the frequency dispersion bandwidth. According to the assumption in the Jakes channel model, the frequency dispersion bandwidth Bd=2fm, so it can be considered that the useful signal part is mainly distributed in the range of [-0.5Bd, 0.5Bd] in the Doppler spectrum. The minimum resolution fmin of the Doppler spectrum can be determined by the sampling rate fs' of the SC-FDE symbol and the number P of SC-FDE symbols transmitted therefrom, and thus the effective interval of the channel coefficient distribution can be determined. The spectral lines not in the effective interval are zeroed to complete the noise reduction in the Doppler variation domain. The coefficients after the noise reduction are subjected to IFFT transformation to convert to the time domain to obtain the channel coefficient vector after the first time after the noise reduction of the i-th SC-FDE symbol Only the channel coefficient vector obtained The first l items with the largest modulus in the maximum time delay are reserved, and the rest of the frequency point positions are zeroed to obtain the channel coefficient vector after the second time after the noise reduction ​​

[0095] S400, determining the time-domain impulse response by using the channel coefficient vector after the second noise reduction;

[0096] The data part in the i-th SC-FDE symbol of the sending end can be expressed as:

[0097] di=[di(1),di(2),di(3),...,di(j),...,di(N)];

[0098] For one symbol di(j)∈di, the symbol received by the receiving end is is expressed as:

[0099]

[0100] The channel impulse response obtained by using the UW block estimation and noise reduction is The present application considers that according to The received data part is subjected to Rake reception in the time domain, and the equalized data is subjected to despreading demodulation decoding and other processing, so that the sending signal can be restored.

[0101] The channel coefficient vector obtained by using the UW block estimation and noise reduction is:

[0102]

[0103] wherein Lmax is the maximum time delay, and the corresponding channel impulse response is The present application considers that according to The time-domain impulse response is obtained by performing equal conversion on the channel coefficient vector after the second noise reduction. is expressed as:

[0104]

[0105] wherein hi',p is the channel coefficient vector at the frequency point position corresponding to the time delay np, p=0,1,...,l, and δ(n-np) represents an impulse function, n represents a variable, np represents a time delay, and l is a positive integer.

[0106] In the application scenario of the present application, the meaning of the variable n represents time, and the property of the impulse function is that the value at the corresponding time delay np is 1, and the values at other frequency point positions are 0.

[0107] For one data symbol di(j) of the sending end, the receiving end can receive the corresponding symbol The expression is as follows:

[0108]

[0109] Here, the application considers the estimated channel impulse response close to the actual channel impulse response h(n), i.e.:

[0110]

[0111] The application uses the time-domain expression of the received symbol to obtain the delay information [n1, n2,.., nl] of each path and the amplitude information [|hi,1|, |hi,2|,.., |hi,l|] corresponding to each path, and uses the estimated time-domain impulse response to approximate the actual time-domain impulse response h(n) and perform Rake combining on the received data part.

[0112] S500, using the time-domain impulse response to perform Rake reception on the data block, to coherently combine the multipath information in all the received data blocks.

[0113] In a specific embodiment of the application, S500 includes:

[0114] Using the time-domain impulse response to perform Rake reception on the data block received by the receiving end, to coherently combine the symbols according to the delay, amplitude and phase of the multipath, to obtain the combined data block

[0115] wherein the symbol at the jth frequency point position in the ith SC-FDE symbol in the data block received by the receiving end is expressed as:

[0116]

[0117] wherein hi,p represents the value of the ideal channel coefficient vector hi at the np corresponding frequency point position, di(j-np) represents the data symbol obtained after np time delay on the data symbol di(j) at the jth frequency point position in the ith SC-FDE symbol, and represents the noise suffered by the data symbol di(j).

[0118] According to the time-domain expression of , there are a total of l paths, and the amplitude information of each path is [|hi,1|, |hi,2|,.., |hi,l|], so a combination method can be selected for combination. Referring to Figure 3 , Figure 3 is a principle block diagram of Rake reception.

[0119] The combined data block is expressed as:

[0120]

[0121] wherein ap represents a coherent combining coefficient, represents a data block received by a receiving end,

[0122] The present application can select the MRC combining, EGC combining or selective combining corresponding to the relevant combining coefficient according to the time delay, amplitude and phase of the multi-path, and coherently combines the symbols to obtain the combined data block

[0123]

[0124] The MRC combining signal utilizes the information from all the received branches, the weighting coefficient of each branch matches the channel coefficient of each branch, and the MRC combining corresponding coherent combining coefficient is represented as:

[0125]

[0126] wherein hi represents the value of the ith path in the time domain impulse response obtained after the above-mentioned twice noise reduction, hi is a complex number containing amplitude and phase information, hp* represents the new value obtained by taking the conjugate of the value of the pth path in the time domain impulse response obtained after the above-mentioned twice noise reduction, taking the conjugate here will not change the phase, which is equivalent to the amplitude being unchanged, and the phase is taken in the opposite value, which is to compensate the phase information.

[0127] The EGC combining signal also utilizes the information from all the received branches, but it is different from the MRC combining in that each branch is weighted with equal gain, and the EGC combining corresponding coherent combining coefficient is represented as:

[0128]

[0129] The selective combining only considers the signal of one path at any given time, and the selected path is the path with the maximum energy in the multi-path, and the selective combining corresponding coherent combining coefficient is represented as:

[0130]

[0131] wherein ap selects ap1, ap2 or ap2, and hp represents the value of the pth path in the time domain impulse response obtained after twice noise reduction, hp is a complex number containing amplitude and phase information.

[0132] The coherent combining coefficient is determined by the combining mode, and the received data block is subjected to time delay compensation and weighted summation to obtain the Rake combined data block represented as:

[0133]

[0134] wherein ap represents a coherent combining coefficient, and the coherent combining coefficient ap (p=q1, q2, q3,..., ql) is determined by the combining mode.

[0135] The data block after Rake combining at the receiving end After despreading, demodulation, decoding and other operations, the initial bit information transmitted by the transmitting end can be obtained.

[0136] The simulation results of the spreading scheme of the equalizer for realizing multipath diversity proposed in the application and the spreading scheme based on the commonly used frequency domain equalization (MMSE equalization) in the SC-FDE system are compared as follows:

[0137]

[0138] Figure 4 For the LOS channel scenario, when the modulation and coding combination is QPSK-0.25-16, the curves of the MSE (mean square error) of the channel response after two times of noise reduction and the channel response without noise reduction in the target scheme varying with SNR (signal-to-noise ratio). From the figure, Figure 4 It can be seen that the MSE of the estimated value of the channel impulse response can represent the accuracy of the channel estimation, and from the figure, it can be seen that the two times of noise reduction of the Doppler transform domain noise reduction and the time domain noise reduction can effectively improve the accuracy of the channel estimation, thereby improving the BER performance of the system.

[0139] Figure 5 For the LOS channel scenario, when the modulation and coding combination is QPSK-0.25-16, the BER performance comparison chart of the target scheme and the baseline scheme. From the figure, Figure 5 It can be seen that when the MCS combination is QPSK-0.25-16, the spreading scheme based on Rake reception using the channel estimation scheme without noise reduction has a performance improvement of about 4.5 dB compared with the spreading scheme based on MMSE equalization, and the spreading scheme based on Rake reception after the Doppler transform domain noise reduction of the channel estimation scheme has a performance improvement of nearly 1.5 dB compared with the spreading scheme based on Rake reception using the channel estimation scheme without noise reduction.

[0140] Figure 6 For the LOS channel scenario, when the modulation and coding combination is QPSK-0.25-4, the curves of the MSE (mean square error) of the channel response after two times of noise reduction and the channel response without noise reduction in the target scheme varying with SNR (signal-to-noise ratio). From the figure, Figure 6 It can be seen that the MSE of the estimated value of the channel impulse response can represent the accuracy of the channel estimation, and from the figure, it can be seen that the two times of noise reduction of the Doppler transform domain noise reduction and the time domain noise reduction can effectively improve the accuracy of the channel estimation, thereby improving the BER performance of the system.

[0141] Figure 7 The figure is the BER performance comparison between the target scheme and the baseline scheme when the MCS combination is QPSK-0.25-4 in the LOS channel scenario. Figure 7 As can be seen from the figure, when the MCS combination is QPSK-0.25-4, the performance of the spread spectrum scheme based on Rake reception using the channel estimation scheme without noise reduction is improved by about 3.6 dB compared with the spread spectrum scheme based on MMSE equalization, and the performance of the spread spectrum scheme based on Rake reception after the Doppler transform domain noise reduction of the channel estimation scheme is improved by nearly 1.8 dB compared with the spread spectrum scheme based on Rake reception using the channel estimation scheme without noise reduction.

[0142] In the description of the present application, the meaning of "a plurality of" is two or more than two, unless otherwise explicitly and specifically limited.

[0143] Although the present application is described herein in conjunction with various embodiments, other variations of the disclosed embodiments can be understood and effected by those skilled in the art in the practice of the claimed application, from the description, drawings and appended claims. In the claims, the word "comprising" does not exclude other components or steps, and the indefinite article "a" or "an" does not exclude a plurality.

[0144] The above is a further detailed description of the present application in conjunction with specific preferred embodiments, and the specific implementation of the present application cannot be limited to these descriptions. For those of ordinary skill in the art to which the present application belongs, a number of simple deductions or substitutions can be made without departing from the concept of the present application, and all of them should be regarded as falling within the protection scope of the present application.

Claims

1. An equalizer for implementing multipath diversity in a spread spectrum SC-FDE system, characterized by, The equalizer for realizing multipath diversity in the spread spectrum SC-FDE system is applied to a receiving end and performs the following process: S100, taking out pilots from a combined block to obtain a first UW block, performing channel estimation on the first UW block and a second UW block of a sending end to obtain an estimated value of a channel coefficient vector of each SC-FDE symbol; The first UW block and the second UW block include all UW word vectors of all SC-FDE symbols; S200, performing Doppler variation domain noise reduction on the estimated value of the channel coefficient vector to obtain a first time of noise-reduced channel coefficient vector; S300, performing time domain noise reduction on the first time of noise-reduced channel coefficient vector to obtain a second time of noise-reduced channel coefficient vector; S400, determining a time domain impulse response by using the second time of noise-reduced channel coefficient vector; S500, performing Rake reception on a data block by using the time domain impulse response to perform coherent combination on multipath information in all received data blocks; S500 includes: The time-domain impulse response is used to make Rake reception on the data block received by the receiving end, to make coherent combination on the symbol according to the time delay, amplitude and phase of the multipath , and obtain the combined data block ; Wherein, in the data block received by the receiving end, the symbol at the jth frequency point position in the ith SC-FDE symbol is represented as: , wherein, the symbol at the jth frequency point position in the ith SC-FDE symbol is represented as:​​ ； wherein, denotes an ideal channel coefficient vector In the value at the corresponding frequency position, is the channel coefficient vector after the second noise reduction after a time delay the value at the corresponding frequency position, denotes the data symbol at the th frequency position in the th SC-FDE symbol after a time delay, denotes the data symbol subjected to noise; , denotes the time delay, is a positive integer; Merged data block is represented as: ； wherein, denotes a coherent combining coefficient, denotes a data block received at the receiving end, ; The symbols are combined coherently based on the time delay, amplitude and phase of the multipath to obtain a combined data block comprising: According to the time delay, amplitude and phase of the multipath, the corresponding correlation combining coefficients of MRC combining, EGC combining or selective combining are selected to make coherent combining of the symbols to obtain the combined data block .

2. The equalizer for achieving multipath diversity in a spread spectrum SC-FDE system according to claim 1, wherein, The receiving end is configured to: receive a target frame; remove extra pilots in the target frame, and then perform serial-parallel conversion to obtain a combined block; the combined block includes a data block, a first UW block and a CP block.

3. The equalizer for achieving multipath diversity in a spread spectrum SC-FDE system according to claim 2, wherein, The target frame is sent by a sending end through a channel, and the sending end performs encoding, interleaving and modulation on information bits to be sent to obtain a plurality of modulation symbols before sending a subframe, performs spreading on the plurality of modulation symbols according to a known spreading sequence to obtain a spreading symbol vector of each modulation symbol, and combines the spreading symbol vector according to a data length of an SC-FDE symbol to obtain a data part of the SC-FDE symbol; and performs framing processing on the data part to obtain the subframe, which is sent to the receiving end through the channel; Each subframe contains a plurality of SC-FDE symbols, and each SC-FDE symbol contains a spreading symbol vector corresponding to a plurality of modulation symbols.

4. The equalizer for realizing multipath diversity in a spread spectrum SC-FDE system according to claim 3, characterized in that, The known spreading sequence is represented as , is the number of spreading symbols, and the vector representation of the th modulated symbol after spreading is , represents the th SC-FDE symbol, and the length of data in one SC-FDE symbol is .

5. The equalizer for achieving multipath diversity in a spread spectrum SC-FDE system according to claim 4, wherein, S100 includes: S110, taking out pilots from the combined block to obtain a first UW block; the first UW block includes all UW word vectors of all SC-FDE symbols, and each UW word vector is expressed as: ； in, Indicates the first... The UW word vector corresponding to each SC-FDE symbol In the ideal case, the first The vector of the channel impulse response of each SC-FDE symbol. Indicates the first... The UW word vector corresponding to each SC-FDE symbol ; For the first The noise vector received by the UW block in each SC-FDE symbol; S120, performing FFT transform on the UW word vector of the receiving end to obtain the frequency domain response value at the first frequency point position , which is represented as: ； wherein, is a frequency domain response value of a frequency domain response vector corresponding to a time domain impulse response vector of the channel at a frequency point position, is a frequency domain response value of the second UWB block of the transmitting end at a frequency point position, represents a value of a frequency domain noise vector obtained by performing FFT transform on a noise vector corresponding to the SC-FDE symbol at a frequency point position, the value range of , is a word length of the UWB block; S130, to the frequency domain response value point division operation, the estimated value of the frequency domain response vector at the frequency point position is expressed as:​ ； wherein represents with the result of dividing S140, the estimated values at the frequency bin positions form a frequency domain response vector is expressed as: ； S150, performing an IFFT transform on the frequency domain response vector to obtain an estimate of the channel coefficient vector is expressed as: 。 6. The equalizer for realizing multipath diversity in a spread spectrum SC-FDE system according to claim 5, characterized in that, S200 includes: S210, for any subframe, taking out the values at the same frequency position in the channel coefficient vector corresponding to the total number of SC-FDE symbols of the subframe, and composing a set , denoted as: ​ ， ； wherein, denotes the channel coefficient vector estimated by the th SC-FDE symbol, denotes the coefficient of the channel coefficient vector estimated by the th SC-FDE symbol at the frequency point position ; S220, performing FFT transform on each item in the set to obtain Doppler spectrum of channel coefficients at frequency point position ​​ wherein the Doppler spectrum has a range of , is a sampling rate for channel estimation, is a sampling rate for SC-FDE symbols; S230, performing noise reduction processing on the Doppler spectrum to set spectrum lines outside an effective information interval to zero to obtain a noise reduction result, and then performing IFFT transformation on the noise reduction result to obtain an inverse transformation result; S240, recombine the values at the same positions in the sets of inverse-transformed results to obtain a first denoised channel coefficient vector , the value range of the first denoised channel coefficient vector is 1 to .​ 7. The equalizer for achieving multipath diversity in a spread spectrum SC-FDE system according to claim 4, wherein, S300 includes: S310, the channel coefficient vector after the first noise reduction exceeds the maximum delay zero processing, and then sorting the channel coefficients within the maximum delay from large to small to obtain a sorting result, denoted as: ； wherein each modulus corresponds to a number as follows: ; S320, according to the number corresponding to each modulus, the sorting result is reserved from large to small modulus value of the first K Item, get the second noise reduction after the channel coefficient vector , expressed as: 。 8. The equalizer for realizing multipath diversity in a spread spectrum SC-FDE system according to claim 7, characterized in that, S400 includes: channel coefficient vector after second noise reduction performing the conversion, to obtain a time-domain impulse response is expressed as ； wherein denotes a is an impulse function, denotes a variable.

9. The equalizer for realizing multipath diversity in a spread spectrum SC-FDE system according to claim 8, characterized in that, wherein, The coherent combination coefficient corresponding to the MRC combination is expressed as: ； in, The first term in the time-domain impulse response obtained after two denoising steps represents the second term. Strip diameter value, It is a complex number that contains both amplitude and phase information. The first term in the time-domain impulse response obtained after two denoising steps represents the second term. The value of the stripe diameter is taken as the new value obtained by conjugation; The coherent combination coefficient corresponding to the EGC combination is expressed as: ； The coherent combination coefficient corresponding to the selection formula combination is expressed as: ； in, choose , or , The first term in the time-domain impulse response obtained after two denoising steps represents the second term. The value of the strip diameter.

Images