System and method for channel estimation in communication based on multicarrier modulation

Abstract

Channel estimation mechanisms in FBMC or other multi-carrier modulation-based communication channels providing multiple subcarriers include transmitting pilot symbols on non-adjacent (e.g., alternating) sub-channels based on Zadoff-Chu or other constant-amplitude zero autocorrelation sequences. In multi-user scenarios, pilot symbols from each user equipment are transmitted on the same selected sub-channels but undergo cyclic shifting. These mechanisms avoid the need for guard bands between pilot symbols for each user equipment and the need for noise correlation matrices.

Description

Technical Field

[0001] This invention relates to channel estimation in multi-carrier modulation-based communications, such as FBMC / OQAM. Background Technology

[0002] Accurate channel estimation (CE) plays a crucial role in realizing the full potential of mMIMO, such as improved spectral efficiency (SE) and resilience against challenging channel conditions. Unlike Orthogonal Frequency Division Multiplexing (OFDM), FBMC / OQAM only exhibits real-field orthogonality. Therefore, the corresponding signals suffer from inherent interference, which is detrimental to their reliability and encourages the use of alternative channel estimation techniques. Furthermore, typical channel estimation methods for mMIMO-FBMC systems require significant training overhead, especially when there are many users or when flat fading channel conditions exist in each subcarrier band.

[0003] Due to the loss of complex orthogonality, the channel estimation process in FBMC / OQAM systems differs from that in conventional OFDM systems. Most current channel estimation methods rely on frequency domain models employing the Interference Approximation (IAM) scheme and assume that the symbol period is significantly larger than the maximum channel delay spread, as in the work of C. Lele, P. Siohan, R. Legouable, and J.-P. Javaudin. 2007 IEEE International Symposium on Power Line Communications and Its Applications This is discussed in the article titled "Preamble-based channel estimation techniques for OFDM / OQAM over the powerline" published in 2007, pp. 59–64 (LELE et al.). Alternative methods include time-domain models, where frequency-domain pilots are used to estimate the time-domain channel impulse response (CIR) without constraints on the symbol interval length, as discussed in, for example, in the following article: by Kong, D. Qu and T. Jiang in I EEE Transactions on Signal Processing The article titled "Time domain channel estimation for OQAM-OFDM systems: Algorithms and performance bounds," published in Volume 62, Issue 2, pp. 322–330, 2014 (KONG et al.); by P. Singh and K. Vasudevan in Wireless Personal CommunicationsThe article titled "Time domain channel estimation for MIMO-FBMC / OQAM systems" was published in Volume 108, Issue 10, 2019 (SINGH et al.); or by M. Caus and AI Perez-Neira in IEEE Transactions on Signal Processing The article titled “Transmitter-receiver designs for highly frequency selective channels in MIMO FBMC systems” was published in Volume 60, Issue 12, pp. 6519–6532, 2012 (CAUS et al.).

[0004] The basis of these two fundamental schemes and their inherent drawbacks are discussed in the following chapters.

[0005] Frequency domain model Considering a time-invariant channel, the received signal can be represented as follows:

[0006] in, It's CIR. It is the additive white Gaussian noise (AWGN) term, and It is a transmitted signal.

[0007] By assuming that the preamble has a time index =0, and since sufficient GB are inserted between them to avoid interference from data symbols, the result is obtained through the preceding equation at the . The received pilot signal on the subcarrier.

[0008] in, The inherent interference is represented as follows:

[0009] The last equation is expressed in the first... Complex channel frequency response (CFR) at subcarriers. It is the total number of paths. It is the impulse response of the prototype filter (PF), and It is in the The demodulation noise at the location is given as follows: Considering the frequency domain model in equation (2), the channel estimation based on the interference approximation method is given by the following equation: As illustrated by LELE et al., interference approximation methods exhibit favorable channel estimation performance when the channel delay spread is significantly smaller than the symbol interval. However, if the channel delay spread is not sufficiently small compared to the symbol interval, the use of interference approximation methods will lead to a significant deterioration in the accuracy of channel estimation.

[0010] In their paper, “Novel preamble-based channel estimation for OFDM / OQAM systems,” published at the 2009 IEEE International Conference on Communications, IEEE, 2009, pp. 1–6, J. Du and S. Signell improved the interference approximation scheme by considering the transmission of purely real and fictitious pilot values. This enhancement is particularly interesting for short PFs, as their short duration allows for complete separation of pilots from data. A study titled “2dB better than CP-OFDM with OFDM / OQAM for preamble-based channel estimation,” published at the 2008 IEEE International Conference on Communications, IEEE, 2008, pp. 1302–1306, by C. Lele, P. Siohan, and R. Legouable, shows that when using a channel estimation technique based on the interference approximation method, FBMC / OQAM with a short PF is 2.3 dB better than OFDM and 0.3 dB better than FBMC / OQAM with a long PF. In by C. Lélé, J.-P. Javaudin, R. Legouable, A. Skrzypczak, and P. Siohan European Transactions on Telecommunications A similar finding was reported in the article titled “Channel estimation methods for preamble-based OFDM / OQAM modulations” published in Volume 19, Issue 7, pp. 741–750, 2008. However, techniques based on interference approximation methods suffer from high peak-to-average power ratios (PAPRs), making them susceptible to nonlinear distortions caused by high-power amplifiers (HPAs), as described by E. Kofidis, D. Katselis, A. Rontogiannis, and S. Theodoridis. Signal processingThis is discussed in the article titled “Preamble-based channel estimation in OFDM / OQAM systems: A review” published in Volume 93, Issue 7, pp. 2038–2054, 2013.

[0011] Time-domain model As an alternative to the frequency domain model, a time domain model for channel estimation has been studied.

[0012] Figure 1 This schematically presents the basis of time-domain channel estimation known in the prior art for FBMC.

[0013] As shown, the transmitted frames are represented as a symbol matrix, where the corresponding symbols are distinguished by the frequency at which they are transmitted (on one axis) and the time sequence of their continuous transmission (on another axis).

[0014] Pilot symbols transmitted for the purpose of performing channel estimation as discussed herein are represented by black-filled circles, data symbols belonging to the channel payload are represented by shaded circles, and blank circles represent the transmission of zero symbols.

[0015] Thus, the first set of transmissions 110 in the time series is used to transmit pilot symbols at each frequency. A series of transmissions 120 exist in the time series for transmitting zero-value symbols at each frequency. Finally, data symbols 130 are transmitted starting at each frequency.

[0016] In subcarrier index The demodulation pilot symbol at time symbol 𝑛 can be written as: According to KONG et al., the received pilot vector It can be represented as follows: in, It is a vector containing channel taps. It is a noise vector, and yes Matrix, which is in the first position row and number The entries in the column are defined as: With the addition of filtering operations, the orthogonality of FBMC / OQAM in the real domain leads to the correlation of noise vector elements in (7).

[0017] Therefore, the conventional least-squares channel estimation method for OFDM systems proposed by AS Ahmed, MM Hamdi, MS Abood, AM Khaleel, M. Fathy, and SH Khaleefah in their article “Channel estimation using LS and MMSE channel estimation techniques for MIMO-OFDM systems” published at the 2022 International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA), 2022, pp. 1–6 [AHMED et al.] cannot be easily applied to KONG et al. To address this issue, the authors in KONG et al. included a channel noise correlation matrix in the least-squares method, as follows: Together with: The authors, KONG et al., have shown that applying the scheme of AHMED et al. outperforms the frequency domain interference approximation method. However, in multi-user channel estimation scenarios, an additional guard band is needed to separate the pilot signals of different users, which is detrimental to SE.

[0018] To avoid guard band interpolation between pilots of different users, the authors of an article published by H. Hosseiny, A. Farhang, and B. Farhang-Boroujeny at ICC 2020 - IEEE International Conference on Communications (ICC), 2020, pp. 1–7, propose an alternative PS that relies on interleaving user pilots in time and frequency. According to this method, each user employs… Instead Pilot, among which, Corresponding to the user's channel length, three FBMC guard symbols are inserted between the pilot and data symbols to shield the pilot from interference. While spectrally efficient due to the absence of guard symbols, this method introduces inherent interference due to the correlation experienced between estimates of different user channels. To avoid performance loss, inherent interference needs to be explicitly considered when applying AHMED et al.'s method during the estimation process. However, this leads to higher computational complexity. Furthermore, the authors of this paper simply assume an exponentially decaying channel in their simulation. Therefore, the performance may differ for more realistic channels.

[0019] The aim is to develop a channel estimation scheme for multi-carrier modulation / communication to address at least some of these drawbacks. Summary of the Invention

[0020] According to a first aspect of the present invention, a method is provided for performing channel characteristic estimation for one or more users in a multi-carrier modulation-based communication channel providing multiple subcarriers, comprising: allocating multiple subcarriers for transmission of complex pilot symbols for each user, wherein no two adjacent subcarriers are allocated in this manner; transmitting complex pilot symbols from each user on each allocated subcarrier; receiving the pilot symbols on each allocated subcarrier; and performing frequency-domain-based analysis of the communication channel based on the received pilot symbols to obtain an estimated channel.

[0021] In the development of the first aspect, frequency-domain-based analysis of the communication channel based on the received pilot symbols includes analysis based on linear channel estimators.

[0022] In the development of the first aspect, the pilot symbols reflect a constant amplitude zero autocorrelation sequence.

[0023] In the development of the first aspect, the pilot symbols for each user in the sequence on each subchannel undergo a cyclic shift with a predetermined increment relative to the previous user in the sequence, such that the pilot symbols from each user on each allocated subcarrier are interleaved in both the time and frequency domains.

[0024] In the development of the first aspect, the planned increment ( i At least equal to: Where L is the sequence length, U is the number of users, and m is the index of the corresponding subcarrier.

[0025] In the development of the first aspect, the number of consecutive subcarriers not allocated for transmitting pilot symbols from each user after each subcarrier has been assigned to transmit pilot symbols from each user is determined as a function of the channel fading characteristics of all users.

[0026] In the development of the first aspect, the method includes an additional step of filtering the pilot symbols using a prototype filter having an overlap factor of one.

[0027] According to a second aspect of the present invention, a method is provided to support channel characteristic estimation at a user equipment in a multi-carrier modulation-based communication channel providing multiple subcarriers, the method comprising: receiving a carrier allocation, the carrier allocation allocating multiple subcarriers for transmission of a user's pilot data structure, wherein no two adjacent subcarriers are allocated in this manner; and transmitting pilot symbols on each allocated subcarrier for frequency-domain-based analysis of the communication channel based on the received pilot symbols.

[0028] According to a third aspect of the present invention, a method is provided for performing channel characteristic estimation for one or more users at a base station in a multi-carrier modulation-based communication channel providing multiple subcarriers, the method comprising: allocating multiple subcarriers for transmission of pilot data structures for each user, wherein no two adjacent subcarriers are allocated in this manner; transmitting the carrier allocation to each user; receiving pilot symbols from each user on each allocated subcarrier; and performing frequency-domain-based analysis of the communication channel based on the received pilot symbols.

[0029] According to a fourth aspect of the present invention, a linear modulation-based communication system is provided, which is adapted to perform channel characteristic estimation for one or more users in a multi-carrier modulation-based communication channel providing multiple subcarriers. The system is adapted to allocate multiple subcarriers for transmitting pilot data symbols for each user, wherein no two adjacent subcarriers are allocated in this manner, transmit pilot symbols from each user on each allocated subcarrier, receive pilot symbols on each allocated subcarrier, and perform frequency domain-based analysis of the communication channel based on the received pilot symbols.

[0030] According to a fifth aspect of the present invention, a user equipment is provided that supports channel characteristic estimation in a multi-carrier modulation-based communication system providing multiple subcarriers, the user equipment being adapted to receive a carrier allocation that allocates multiple subcarriers for the transmission of the pilot symbols, wherein no two adjacent subcarriers are allocated in this way, and a pilot symbol is transmitted on each allocated subcarrier for frequency-domain analysis of the communication channel based on the received pilot symbols.

[0031] According to a sixth aspect of the present invention, a base station is provided for performing channel characteristic estimation for one or more users in a multi-carrier modulation-based communication system providing multiple subcarriers, the base station being adapted to allocate multiple subcarriers for transmitting pilot symbols, wherein no two adjacent subcarriers are allocated in this manner, transmitting the carrier allocation to each user, receiving pilot symbols from each user on each allocated subcarrier, and performing frequency-domain-based analysis of the communication channel based on the received pilot symbols.

[0032] In any of the fourth, fifth, or sixth aspects of development, the prototype filter of the transmitter or receiver has an overlap factor of one.

[0033] According to a seventh aspect of the present invention, a computer program comprising instructions, which, when executed by a computer, cause the computer to perform the method of any one of the first, second, or third aspects.

[0034] According to a seventh aspect of the present invention, a computer-readable medium comprising instructions, which, when executed by a computer, cause the computer to perform the method of any one of the first, second, or third aspects. Attached Figure Description

[0035] The invention will be better understood through the following description of several exemplary embodiments for illustrative purposes only, along with their accompanying drawings, in which various features and advantages will become apparent: Figure 1 This schematically illustrates the basis of time-domain channel estimation known in the prior art for FBMC; Figure 2 The steps of the method according to the embodiments are presented; Figure 3 The illustration illustrates the basis of time-domain channel estimation for FBMC according to a single user embodiment; Figure 4a The illustration shows the basic principle of assigning pilot symbols to subcarriers in a multi-user environment according to an embodiment; Figure 4b The illustration illustrates the basis of time-domain channel estimation for FBMC in a multi-user scenario according to a multi-user embodiment; Figure 5 A method performed at a user equipment according to an embodiment is presented; Figure 6 A method performed at a base station according to an embodiment is presented; Figure 7 The implementation of the first transmitter is shown; Figure 8An alternative transmitter implementation is shown; Figure 9 A computer suitable for performing the methods according to the embodiments is presented; and Figure 10 A base station 910 is presented, which incorporates a computer suitable for performing the methods according to the embodiments. Detailed Implementation

[0036] To address these issues, a preamble-based pilot structure (PS) is proposed, which reduces training overhead by interleaving user pilots in time and frequency while reintroducing the possibility of applying conventional OFDM channel estimation techniques, particularly least squares (LS) methods in the context of single-user (SU) and multi-user (MU) scenarios. Simulations performed on 5G QuaDRiga channels demonstrate that the proposed PS improves spectral efficiency while achieving state-of-the-art performance in both preamble-based MU-mMIMO and SU systems. Furthermore, compared to available solutions in the literature that require large guard bands (GBs) to separate pilots for different users, the proposed method significantly improves spectral efficiency while allowing the application of conventional channel estimation techniques in OFDM.

[0037] Figure 2 The steps of the method according to the embodiments are presented.

[0038] Specifically, a method is provided for performing channel characteristic estimation for one or more users in a multi-carrier modulation-based communication channel providing multiple subcarriers. The embodiments are extended to arbitrary multi-carrier systems, such as, for example, FBMC / OQAM waveforms or OFDM, in which case out-of-band power leakage is reduced, addressing one of the channel estimation problems for OFDM systems.

[0039] The method begins at step 200 and proceeds to step 210, where multiple subcarriers are allocated for the transmission of pilot symbols for each user, wherein no two adjacent subcarriers are allocated in this manner. It can be noted that this allocation may be fixed. The pilot symbols may, for example, reflect a constant amplitude zero autocorrelation sequence, such as the Zadoff-Chu sequence. Although any such sequence can be used, the Zadoff-Chu sequence is advantageous in providing a low peak-to-average power ratio (PAPR).

[0040] The method then proceeds to step 220: transmitting pilot symbols from each user on each allocated subcarrier, and then receiving pilot symbols on each allocated subcarrier at step 230. Then, before terminating at step 250, at step 240, a frequency-domain-based analysis of the communication channel is performed based on the received pilot symbols to obtain an estimated channel.

[0041] Frequency-domain-based analysis of a communication channel based on received pilot symbols can include, for example, analysis based on linear channel estimators, such as least squares or least mean square error analysis.

[0042] Figure 3 The illustration illustrates the basis for time-domain channel estimation for FBMC according to an embodiment.

[0043] As described above, at step 210, multiple subcarriers are allocated for the transmission of pilot symbols for each user (in this embodiment, a single user), wherein no two adjacent subcarriers are allocated in this way.

[0044] This is Figure 3 As shown in the diagram, it can be compared with... Figure 1 Compared with the existing technology situation.

[0045] As shown, the transmitted frames are represented as a symbol matrix, where the corresponding symbols are distinguished by the frequency at which they are transmitted (on one axis) and the time sequence of their continuous transmission (on another axis).

[0046] Pilot symbols transmitted for the purpose of performing the channel estimation discussed herein are represented by black-filled circles, data symbols belonging to the channel payload are represented by shaded circles, and blank circles represent the transmission of zero symbols.

[0047] Thus, the first group of transmissions in the time series is used to transmit pilot symbols at the selected frequencies. Specifically, subbands 311, 313, and 315 are selected for pilot symbol transmission, while subbands 312, 314, and 316 transmit zero values. A series of transmissions 320 follow in the time series, in which zero-value symbols are transmitted at each frequency. Finally, data symbols 330 are transmitted in accordance with... Figure 1 The same manner described begins at each frequency. For a filtered multicarrier waveform, there exists a set of zeros to be inserted; however, as discussed herein, contrary to prior art solutions, in certain embodiments a short prototype filter can be used, such that only one guard band symbol needs to be inserted.

[0048] Development support will now be available. Figure 2 The underlying analysis of the operability of the method.

[0049] The Zadoff-Chu (ZC) sequence is used in 5G NR wireless systems and was developed by D. Chu. IEEE Transactions on Information TheoryOne of the pilot sequences described in the article entitled "Polyphase codes with good periodic correlation properties (corresp.)" published in Volume 18, Issue 4, pp. 531–532, 1972, possesses constant amplitude zero autocorrelation (CAZAC) properties, which are preserved even when processed by IFFT / FFT because the output samples retain the Zadoff-Chu sequence. Furthermore, its low peak-to-average power ratio makes it an attractive choice for channel estimation. The following discussion will employ the Zadoff-Chu sequence as the chosen constant amplitude zero autocorrelation sequence.

[0050] Meanwhile, for systems such as FBMC that exhibit orthogonality only in the real domain, it is impossible to use complex-valued pilots to load adjacent subcarriers, because the latter must be shielded from interference.

[0051] To solve this problem, according to Figure 2 The method considers that complex pilots can be assigned to even-numbered frequency indices while keeping adjacent FBMC subcarrier positions idle or inactive. Therefore, if If it is a vector containing complex-valued elements generated from the ZC sequence, then the transmitted pilot signal... It can be expressed mathematically as follows:

[0052] Together with:

[0053] Based on the proposed pilot symbol scheme, complex-valued pilot sequences can still be used in multi-carrier modulation-based communication channels, such as FBMC / OQAM, and channel estimation techniques considered for OFDM. However, a sufficient number of guard symbols must be introduced between training and data symbols to ensure their overall isolation, resulting in a data rate loss.

[0054] In fact, the required number of FBMC guard symbols depends on the overlap factor of the prototype filter under consideration, which can suggest either a long or short prototype filter. According to a particular embodiment, a short prototype filter, such as the near-perfect reconstruction 1 (NPR1) short prototype filter, can be employed because only one FBMC symbol is needed as a guard interval to completely isolate the pilot from the data, resulting in the minimum data rate loss described in the article entitled “Design and Evaluation of a Novel ShortPrototype Filter for FBMC / OQAM Modulation” published by J. Nadal, CA Nour, and A. Baghdadi in IEEE Access, Vol. 6, pp. 19610–19625, 2018 (NADAL et al.). Other prototype filters can be used, and any short prototype filter offers the benefit of requiring only one guard band symbol. However, NPR1 is preferred in other respects due to its improved performance.

[0055] Apply equation (1) to Figure 3 The pilot symbol scheme is represented in the figure, and by employing a filtering stage in the time domain, the received pilot symbols can be represented by equation (7), where the matrix The following changes have been made:

[0056] In equation (8), the even-numbered subcarrier index is defined. and the Entries on the line .

[0057] As previously mentioned, noise correlation is typically included in LS channels due to the lack of complex orthogonality. However, utilizing... Figure 3 The embodiment employs an odd number of subcarriers as a guard band and a pilot symbol scheme with short filters. The correlation between noise components can be calculated as follows: in, It is the noise variance.

[0058] By making changes to the variables The previous equation can be rewritten as follows: in, ,and ,in, .

[0059] Since the NPR1 PF has a short duration, the interference only affects adjacent FBMC symbols. This means that the value of 𝑞 is limited to -1 or 1. In addition, the proposed pilot structure includes guard bands as adjacent FBMC symbols, making the correlation between noise components negligible in equation (14) (i.e., Therefore, the assumption of uncorrelated noise is valid, and analytical methods based on conventional linear channel estimators, such as those used in OFDM systems, can be applied to obtain channel estimates without considering the noise correlation matrix (as required in the prior art time-domain methods discussed above), thereby reducing complexity.

[0060] Therefore, equation (9) simplifies to conventional LS channel estimation as follows: In fact, higher than The estimated CIR has a focus on several first samples. Most of the power is thus divided into two parts as follows:

[0061] Therefore, it is advantageous to denoise the channel estimates above. Denoising is the process of eliminating noise from a noisy signal. It can be observed that the estimated channel in (15) is corrupted by noise. Therefore, the CIR can be considered a noisy signal, and denoising it can lead to an improved CIR estimate. Denoising can be performed by pruning the CIR coefficients using a threshold. Therefore, we It has been applied Point-inverse discrete cosine transform type IV (IDCT4). Next, we will only consider the first... One tap is used, and the remainder is filled with zeros. Mathematically, this is transformed as follows:

[0062] in, It is the channel delay spread in the sample.

[0063] therefore, Figure 2 The method may include an optional step of denoising the estimated channel. It should be noted that conventional channel estimation techniques are not based on channel denoising. In this embodiment, channel denoising can be implemented to remove inherent interference from neighboring users. As a result, inherent interference terms will not be included in the channel estimation, significantly reducing complexity.

[0064] In fact, some of the channel taps included in the first part of the previous equation may not be significant. Therefore, in order to accurately select effective channel taps, we can first estimate the noise power as follows: Then, important channel taps can be selected according to the rules:

[0065] in, yes In the tap index Power at the location.

[0066] The channel estimate can then be obtained using the Discrete Cosine Transform Type IV (DCT4) with equation (19) as follows: in, ,as well as .

[0067] Note that the above equation only provides a channel estimate for even-numbered subcarrier indices, since odd-numbered subcarriers are idle. Therefore, channels on odd-numbered subcarriers can be used, for example, in the work of H. Akima. J. ACM Volume 17, Issue 4, pp. 589–602, October 1970 [Online] (Able to access from...) https: / / doi.org / 10.1145 / 321607.321609 The interpolation can be performed using the method proposed in the article titled "A new method of interpolation and smooth curve fitting based on local procedures" published in [Journal Name], or other methods that may be conceived by those skilled in the art.

[0068] although Figure 3 Alternative subcarriers allocated for the transmission of pilot data structures for each user are shown, but it will be appreciated that other sequences in which two adjacent subcarriers are not allocated are also envisioned; for example, each r-th subband could be allocated, where r is an arbitrary integer. Furthermore, the spacing between the allocated subbands need not be constant, as long as no two adjacent subbands are allocated.

[0069] Based on the channel fading characteristics of all users, such as given the Doppler effect (slow or fast fading), determine the number of consecutive subcarriers not allocated for transmitting pilot symbols from each user after each subcarrier allocated for transmitting pilot symbols from each user.

[0070] In practice, in response to detected changes in the channel fading characteristics of all users, the number of consecutive subcarriers not allocated for transmitting pilot symbols from each user can be adjusted over time for continuous transmission: after each subcarrier allocated for transmitting pilot symbols from each user.

[0071] The preceding discussion addressed a simple scenario involving a single user across all channels. Meanwhile, Figure 3 The same approach applies to multi-user scenarios, as will be considered now.

[0072] To ensure the orthogonality of pilot signals for multiple users in the uplink, symbol protection is typically required in existing technologies. However, most available solutions in the literature are not efficient in terms of spectrum usage, as discussed above.

[0073] According to an embodiment, the frequency resources for pilot signals for different users are mutually inclusive, so that each frequency slot is used by all users.

[0074] More specifically, on each subchannel, the pilot symbol for each user in the sequence undergoes a predetermined incremental subcarrier shift relative to the previous user in the sequence, such that the pilot symbols from each user are interleaved in the frequency domain on each allocated subcarrier.

[0075] Figure 4a The illustration shows the basic principle of assigning pilot symbols to subcarriers in a multi-user environment according to an embodiment.

[0076] As in Figure 4a As shown, a set of pilot symbols 401a-e is defined for the first user, and a corresponding set of pilot symbols 402a-e is defined for, for example, the second user. The separation distance defines a predetermined increment between the subcarriers 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, and 422 assigned to the corresponding user by the corresponding pilot symbol ae.

[0077] Here, in general, it is shown that a specified minimum distance should always exist between pilots of different users. This distance may not necessarily be uniform for all pilots, although this may be the case in some embodiments, such as those based on cyclic shifts discussed below. However, this separation distance is at least as large as the channel delay spread in all cases.

[0078] As shown, the following distribution is adopted:

[0079] Based on this, as shown in the figure, the minimum interval distance is two.

[0080] This can be achieved through various shifting mechanisms. Possible solutions may include the following: A specific sequence of multiplications is performed on the pilot set for each user. This operation ensures that the resulting pilot set for each user becomes orthogonal to the pilot sets belonging to all other users. In simpler terms, this process ensures that pilot signals used by different users are separated and do not interfere with each other.

[0081] Another approach involves developing an interleaving scheme that carefully reorganizes the positions of pilot signals from different users. This process preferably ensures that the time-frequency resources allocated to the pilot signals of different users are mutually inclusive. In other words, an element such as '401e' from a second user can be located on the same frequency subcarrier as an element such as '401d' from a first user, as in... Figure 4a As illustrated in the example. However, the preferred condition is that the corresponding elements of different users (such as between '402e' and '401e') consistently maintain a minimum spacing distance equal to the delay spread of the channel upwards.

[0082] According to a preferred embodiment, the user The pilot signal is Cyclic shift of the pilot signal. Mathematically, this is translated as follows:

[0083] Together with: Make: Where mod is the modulo operator. The length of a constant-amplitude zero-autocorrelation sequence, and Extend the average latency for all users.

[0084] The absence of guard symbols between user pilots leads to inherent inter-user interference. This latter presence translates into inter-user dependency on the information used to compute channel estimation. Consequently, channel estimation must take into account the inter-user interference caused by… Represented from the user To users The inherent interference.

[0085] Therefore, among them It is the total number of users, on the subcarrier. Time Index The demodulated pilot signal can be mathematically represented as: in, It is the length L u The uth user CIR.

[0086] For ease of notation, the above equation can be represented in the following vector form: in: and

[0087] in, yes Matrix, where its th row and number The column is defined as: Together (M / 2) x L j’ A matrix, wherein the m-th row and l-th column are defined as: Using equation (15) and ignoring the noise correlation discussed earlier, the LS channel estimate can be obtained as follows: in, .

[0088] In fact, the proposed pilot symbol scheme for multi-user scenarios shifts the inherent interference caused by user pilots to the noise component. In other words, the estimated channel in equation (27) can be divided into two parts: the CIR component and the noise plus inherent interference component. According to this definition, equation (16) is modified as follows:

[0089] Next, by using equation (17) for the first The user's estimated impulse response is denoised so that the second part can be ignored, which modifies equation (19) to: in, It refers to the first one defined in (22). The cyclic shift applied by the user.

[0090] Therefore, the inherent disturbances in MU CE can be ignored, and thus, equation (25) simplifies to a diagonal matrix as follows:

[0091] As a result, the CE for MU is reduced to SU-1, where the channel estimate of the MU for the t user can be calculated by using equations (17), (19) and (20).

[0092] Figure 4b The illustration illustrates the basis of time-domain channel estimation for FBMC in a multi-user scenario according to an embodiment.

[0093] As described above, at step 210, multiple subcarriers are allocated for the transmission of pilot symbols for each user, wherein no two adjacent subcarriers are allocated in this manner.

[0094] In light of the foregoing discussion, Figure 2 The method is suitable for multi-user scenarios. It involves subjecting the pilot symbols for each user in the sequence on each sub-channel to a predetermined incremental cyclic shift relative to the previous user in the sequence, so that the pilot symbols from each user on each allocated subcarrier are interleaved in the time and frequency domains.

[0095] This is illustrated in Figure 4, which can be compared with... Figure 3 Compared to single-user scenarios.

[0096] As explained above, the predetermined increment ( i ) can be equal to , where L is the sequence length, U is the number of users, and m is the index of the corresponding subcarrier.

[0097] As clear in the foregoing discussion, the proposed pilot structure eliminates the need for guard bands between user pilots. This is advantageous compared to most pilot structures in the literature, as FBMC / OQAM typically uses guard bands to prevent interference from adjacent users, or alternatively, if they do not insert guard bands between user pilots, they incorporate inherent interference terms into channel estimation techniques, leading to increased complexity compared to OFDM.

[0098] Compared to existing pilot structures, this embodiment allows the use of conventional OFDM channel estimation techniques without the need to include inherent interference terms.

[0099] It can be noted that, Figure 2 The proposed method is from the perspective of the entire system. This method can also be considered from the perspective of the user device.

[0100] Figure 5 A method performed at a user equipment according to an embodiment is presented.

[0101] Specifically, Figure 5A method is shown to support channel characteristic estimation at a user's user equipment in a multi-carrier modulation-based communication channel that provides multiple subcarriers.

[0102] As shown, the method begins at step 500 and proceeds to step 510, where carrier allocation assigns multiple subcarriers for transmitting the user's pilot data structure, wherein no two adjacent subcarriers are thus allocated for reception. The method then proceeds to step 520, where pilot symbols are transmitted on each allocated subcarrier for frequency-domain-based analysis of the communication channel based on the received pilot symbols. The method then terminates at step 530. The method can be modified as necessary to incorporate relevant... Figure 2 As described in Figure 4, any variation.

[0103] It can be noted that, Figure 2 The method presented is from the perspective of the entire system. This method can also be considered from the perspective of the base station.

[0104] Figure 6 A method performed at a base station according to an embodiment is presented.

[0105] Specifically, Figure 6 A method is shown to support channel characteristic estimation at a user's user equipment in a multi-carrier modulation-based communication channel that provides multiple subcarriers.

[0106] As shown, the method begins at step 600 and proceeds to step 610, where multiple subcarriers are allocated for transmitting pilot data structures for each user, wherein no two adjacent subcarriers are allocated in this manner. Then, at step 620, the method continues to transmit carrier allocations to each user. At step 630, the base station receives pilot symbols from each user on the allocated subcarriers, and at step 640, before terminating at step 650, performs a frequency-domain-based analysis of the communication channel based on the received pilot symbols. The method can be modified as necessary to incorporate information regarding… Figure 2 As described in Figure 4, any variation.

[0107] Although embodiments have been described according to the method, embodiments are also implemented in hardware. As an example, embodiments based on two standard FBMC hardware architectures are presented in the following figures.

[0108] Figure 7 A first exemplary transmitter implementation according to an embodiment is shown.

[0109] As shown, Figure 7The implementation includes a processing channel, which consists of an inverse fast Fourier transform (IFFT) unit 751 and a filtering stage 761.

[0110] During the transmission of the pilot signal, the value indicated by the selected constant amplitude zero autocorrelation sequence is provided from memory 721.

[0111] In addition to these conventional components, the processing channel also includes an upsampling module 731, which amplifies each signal according to a factor q. Values ​​are received from the upsampling module 731, and its output is then processed by the cyclic shift module 741 according to the foregoing discussion.

[0112] The outputs of the two Inverse Fast Fourier Transform (IFFT) blocks 751 are used to process the subcarriers according to the following equation:

[0113] in, M is the complex CAZAC sequence element at subcarrier index m and time slot n, and M is the total number of available subcarriers.

[0114] Filtering stage 761 corresponds to the filtering operation performed on the pilot symbols. Specifically, the filtering operation is described as follows:

[0115] in, It is the impulse response of a prototype filter with an overlap factor K and a length KM.

[0116] The filtering operation 761 can be viewed as a digital filter, and its design represents an important trade-off between performance and system complexity.

[0117] When using a short filter with an overlap factor of 1 for the filtering stage, the latter can be treated as a windowing operation: the output of the IFFT is simply multiplied by the impulse response of the prototype filter. Therefore, the hardware complexity introduced by the filtering stage is limited.

[0118] Figure 8 A second exemplary transmitter implementation according to an embodiment is shown.

[0119] As shown, Figure 8 The implementation methods include a finite impulse response (FIR) filter 851 and an inverse fast Fourier transform block 861.

[0120] During the transmission of the pilot signal, the value indicated by the selected constant amplitude zero autocorrelation sequence is provided from memory 821.

[0121] In addition to these conventional components, the processing channel also includes an upsampling module 831, which amplifies each signal by a factor q, receives values ​​from the corresponding upsampling module 831, and its output is then processed by the corresponding cyclic shift module 841 according to the foregoing discussion.

[0122] The original concept of this design was to shift the filter stage into the frequency domain. At least for long filters, the hardware complexity should be higher than that of the first implementation. In fact, an IFFT of size L = KM is required. However, for short filters (K = 1), the IFFT size is the same as the first implementation.

[0123] A corresponding arrangement can be provided on the receiver side.

[0124] Therefore, a communication system based on linear modulation can be provided, for example, based on... Figure 7 or Figure 8 The implementation method is suitable for performing channel characteristic estimation for one or more users in a multi-carrier modulation-based communication channel that provides multiple subcarriers. It is suitable for allocating multiple subcarriers for transmitting pilot data symbols for each user, wherein no two adjacent subcarriers are allocated in this way, transmitting pilot symbols from each user on each allocated subcarrier, receiving pilot symbols on each allocated subcarrier, and performing frequency domain-based analysis of the communication channel based on the received pilot symbols.

[0125] Similarly, a method based on, for example Figure 7 or Figure 8 The user equipment is configured to support channel characteristic estimation in a multi-carrier modulation-based communication system that provides multiple subcarriers. The user equipment is adapted to receive carrier allocations that allocate multiple subcarriers for the transmission of pilot symbols, wherein no two adjacent subcarriers are allocated in this way, and to transmit pilot symbols on each allocated subcarrier in order to perform frequency-domain analysis of the communication channel based on the received pilot symbols.

[0126] Alternatively, a method based on, for example Figure 7 or Figure 8 A base station is configured to perform channel characteristic estimation for one or more users in a multi-carrier modulation-based communication system that provides multiple subcarriers. The base station is adapted to allocate multiple subcarriers for the transmission of pilot symbols, wherein no two adjacent subcarriers are allocated in this manner, the carrier allocation is sent to each user, pilot symbols from each user are received on each allocated subcarrier, and frequency-domain-based analysis of the communication channel is performed based on the received pilot symbols.

[0127] Thus, channel estimation mechanisms are provided in FBMC or other multi-carrier modulation-based communication channels that provide multiple subcarriers, including transmitting pilot symbols on non-adjacent (e.g., alternating) sub-channels based on Zadoff-Chu or other constant-amplitude zero autocorrelation sequences. In multi-user scenarios, pilot symbols from each user equipment are transmitted on the same selected sub-channels but undergo cyclic shifting. These mechanisms avoid the need for guard bands between pilot symbols for each user equipment and the need for noise correlation matrices.

[0128] The disclosed methods can take the form of a completely hardware embodiment (e.g., an FPGA), a completely software embodiment (e.g., to control a system according to the invention), or an embodiment including both hardware and software elements. Software embodiments include, but are not limited to, firmware, resident software, microcode, etc. The invention can take the form of a computer program product accessible from a computer-usable or computer-readable medium, which provides program code for use by or in conjunction with a computer or instruction execution system. The computer-usable or computer-readable medium can be any means capable of containing, storing, transmitting, propagating, or sending a program for use by or in conjunction with an instruction execution system, apparatus, or device. The medium can be an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system (or apparatus or device) or a propagation medium.

[0129] These methods and processes can be implemented using computer applications or services, application programming interfaces (APIs), libraries and / or other computer program products or any combination of these entities.

[0130] The method of the present invention can be executed by a properly programmed general-purpose computer or computer system (possibly including a computer network), storing a suitable program in non-volatile form on a computer-readable medium such as a hard disk, solid-state drive, or CD-ROM, and executing the program using its microprocessor and memory.

[0131] Figure 9 A computer suitable for performing the methods according to the embodiments is presented.

[0132] exist Figure 9In this computer, a central processing unit (CPU) 901 is included, which can execute the method steps described above and simultaneously run an executable program, i.e., a set of computer-readable instructions, stored in a memory device such as RAM 902a or ROM 902b, or hard disk drive 931, DVD / CD drive 932, or remotely stored. Furthermore, one or more computer files defining a spatial relationship diagram, as well as computer files including a virtual space and objects defined therein, can also be stored on one or more of the memory devices 902a, 902b, 931, 932, or remotely stored.

[0133] This invention is not limited to the form of a computer-readable medium on which the computer-readable instructions of the present invention are stored. For example, instructions and files can be stored on a CD, DVD, FLASH memory, RAM, ROM, PROM, EPROM, EEPROM, hard disk, or any other information processing device (such as a server or computer) that communicates with a computer. Programs can be stored on the same memory device or on different memory devices.

[0134] Furthermore, computer programs suitable for performing the methods of the present invention may be provided as utilities, background daemons, or components or combinations thereof that execute in conjunction with a CPU 901 and an operating system (such as Microsoft XP, Microsoft Windows 10, UNIX, Solaris, LINUX, Apple MAC-OS, and other systems known to those skilled in the art).

[0135] CPU 901 can be a Xenon processor from Intel or a Ryzen processor from AMD, or it can be other processor types, such as a Freescale ColdFire, IMX, or ARM processor from NXP Semiconductors. Alternatively, the CPU can be a processor such as one from Intel Corporation, or it can be implemented on an FPGA, ASIC, PLD, or using discrete logic circuitry, as will be recognized by those skilled in the art. Furthermore, the CPU can be implemented as multiple processors that work cooperatively to execute computer-readable instructions of the processes described above.

[0136] The computer may include a network interface 920, such as an Intel Ethernet PRO network interface card from Intel Corporation, for communicating with networks such as a local area network (LAN) 974, a wide area network (WAN), or the Internet 975. The method can be implemented remotely via a web application, for example, on a remote server 776.

[0137] The computer may also include a display controller 910, such as an NVIDIA GeForce RTX graphics adapter from NVIDIA Corporation, for interacting with a display 911. A general-purpose I / O interface 903 interfaces with a keyboard 912 and pointing devices 913 (such as a ball, mouse, touchpad, etc.). The display, keyboard, touch-sensitive surface, and pointing devices, together with the display controller and I / O interface, form a graphical user interface used by the user to provide input commands. Commands may also be issued via a network connection, as described below.

[0138] The disk controller 930 connects the HDD 931 and DVD / CD 932 to the communication bus 920, which can be ISA, EISA, VESA, PCI or similar, for interconnecting all components of the computer.

[0139] For the sake of brevity, descriptions of the general characteristics and functions of displays, keyboards, pointing devices, display controllers, disk controllers, network interfaces, and I / O interfaces are omitted in this document, as these characteristics are known.

[0140] Figure 9 The system can implement operations belonging to a base station or any user equipment. Therefore, multiple such systems can be provided to operate in a coordinated manner to perform the corresponding parts of the described methods and operations.

[0141] As an example, Figure 10 A base station 910 is presented, which incorporates a computer suitable for performing the methods according to the embodiments.

[0142] As in Figure 10 As shown, computer 900 includes information about Figure 9 The subset of components described is suitable for such an implementation context. Specifically, the computer includes a central processing unit (CPU) 901 that can execute the method steps described above while running an executable program, i.e., a set of computer-readable instructions, stored in a memory device 902 that communicates with an I / O interface 903.

[0143] The computer may include a network interface 920 for communicating with networks such as a local area network (LAN) 974, a wide area network (WAN), or the Internet 975. The operation of this method can be performed remotely via a web application, for example, on a remote server 976.

[0144] It should be understood that the configurations and / or methods described herein are exemplary in nature, and these particular embodiments or examples should not be considered limiting, as many variations are possible. The specific routines or methods described herein may represent one or more of any number of processing strategies. Therefore, the various actions shown and / or described may be performed in the shown and / or described order, in a different order, in parallel, or omitted. Similarly, the order of the above processes may be changed.

[0145] The above embodiments are given as non-limiting examples of embodiments of the present invention. They do not limit the scope of the invention as defined by the appended claims in any way.

Claims

1. A method for performing channel characteristic estimation for one or more users in a multi-carrier modulation-based communication channel providing multiple subcarriers, the method comprising: Multiple subcarriers are allocated for the transmission of complex pilot symbols for each user, wherein no two adjacent subcarriers are allocated in this manner. The complex pilot symbols from each user are transmitted on each assigned subcarrier. The pilot symbols are received on each assigned subcarrier, and A frequency-domain-based analysis of the communication channel is performed based on the received pilot symbols to obtain an estimated channel.

2. The method according to claim 1, wherein, The frequency-domain-based analysis of the communication channel based on the received pilot symbols includes analysis based on a linear channel estimator.

3. The method according to any of the preceding claims, wherein, The pilot symbols reflect a constant amplitude zero autocorrelation sequence.

4. The method according to any of the preceding claims, wherein, On each of the sub-channels, the pilot symbols for each user in the sequence undergo a predetermined incremental cyclic shift relative to the previous user in the sequence, such that the pilot symbols from each user are interleaved in both the time and frequency domains on each assigned subcarrier.

5. The method according to claim 4, wherein, The predetermined increment ( i At least equal to: ， Where L is the sequence length, U is the number of users, and m is the index of the corresponding subcarrier.

6. The method according to any of the preceding claims, wherein, After each subcarrier is allocated for transmitting pilot symbols from each user, the number of consecutive subcarriers not allocated for transmitting pilot symbols from each user is determined as a function of the channel fading characteristics of all users.

7. The method according to any of the preceding claims, further comprising the step of filtering the pilot symbols with a prototype filter having an overlap factor of one.

8. A method for supporting channel characteristic estimation at a user's user equipment in a multi-carrier modulation-based communication channel providing multiple subcarriers, the method comprising: The system receives a carrier allocation, which allocates multiple subcarriers for transmitting the user's pilot data structure, wherein no two adjacent subcarriers are allocated in this manner. Pilot symbols are transmitted on each assigned subcarrier for frequency-domain-based analysis of the communication channel based on the received pilot symbols.

9. A method for performing channel characteristic estimation for one or more users at a base station in a multi-carrier modulation-based communication channel providing multiple subcarriers, the method comprising: Multiple subcarriers are allocated for the transmission of pilot data structures for each user, wherein no two adjacent subcarriers are allocated in this manner. The carrier allocation is sent to each of the users. Pilot symbols from each user are received on each of the allocated subcarriers, and Frequency-domain-based analysis of the communication channel is performed based on the received pilot symbols.

10. A linear modulation-based communication system suitable for performing channel characteristic estimation for one or more users in a multi-carrier modulation-based communication channel providing multiple subcarriers, the system being adapted to: Multiple subcarriers are allocated for the transmission of pilot data symbols for each user, wherein, No two adjacent subcarriers were allocated in this way. Pilot symbols from each user are transmitted on each assigned subcarrier. The pilot symbols are received on each assigned subcarrier, and Frequency-domain-based analysis of the communication channel is performed based on the received pilot symbols.

11. A user equipment supporting channel characteristic estimation in a multi-carrier modulation-based communication system providing multiple subcarriers, the user equipment being adapted to: Receive carrier allocation, wherein the carrier allocation allocates multiple subcarriers for the transmission of pilot symbols, wherein, No two adjacent subcarriers were allocated in this way. Pilot symbols are transmitted on each assigned subcarrier for frequency-domain-based analysis of the communication channel based on the received pilot symbols.

12. A base station for performing channel characteristic estimation for one or more users in a multi-carrier modulation-based communication system providing multiple subcarriers, the base station being adapted to: Multiple subcarriers are allocated for the transmission of pilot symbols, wherein, No two adjacent subcarriers were allocated in this way. The carrier allocation is sent to each of the users. Pilot symbols from each user are received on each of the allocated subcarriers, and Frequency-domain-based analysis of the communication channel is performed based on the received pilot symbols.

13. The system according to any one of claims 10 to 12, wherein, The prototype filter of the transmitter or the receiver has an overlap factor of 1.

14. A computer program comprising instructions that, when executed by a computer, cause the computer to perform the method according to any one of claims 1 to 9.

15. A computer-readable storage medium comprising instructions that, when executed by a computer, cause the computer to perform the method according to any one of claims 1 to 10.

Images