JP6223374B2 - Receiver - Google Patents

Description

  The present invention relates to a receiving apparatus that receives a differential space-time encoded signal.

  Diversity technology is used as a technology for suppressing the influence of fading and the like during wireless signal transmission. For example, a space-time block code (STBC) that can obtain a diversity effect by transmitting a signal according to a specific rule using a plurality of transmission antennas has a simple process on the reception side. For various reasons, it is used in various systems.

  In a system to which STBC is applied, propagation path estimation is necessary on the receiving side. On the other hand, Non-Patent Document 1 discloses differential space-time block coding (DSTBC) that eliminates the need for channel estimation by performing differential coding in units of STBC blocks. Yes. In a system to which DSTBC is applied, it is possible to obtain a diversity effect while making propagation path estimation on the receiving side unnecessary. For example, in DSTBC using two-antenna transmission, two symbols are configured as one block, differential encoding is performed using data of two blocks, and reception data of two blocks is differentially decoded on the receiving side. To demodulate the data.

  On the other hand, in a general wireless communication system, it is necessary to provide a means for the receiving side to synchronize by a correlation calculation or the like using a signal such as a synchronization word (SW) or a unique word. For example, on the transmitting side, the receiving side transmits an SW sequence having a known signal pattern, and on the receiving side, a correlation calculation is performed between the known SW sequence and the received signal, and a frame or symbol is detected by a technique such as peak detection. Extract the timing.

  In wireless communication, it is known that various interferences can cause performance degradation. For example, in a system to which broadband transmission is applied, it may happen that the signal is distorted due to the frequency selectivity of the propagation path and cannot be demodulated correctly. This phenomenon occurs due to delayed waves in a multipath environment, and the distortion varies depending on the characteristics of the propagation path, that is, the number of delay waves, the phase relationship between each delay wave, and the magnitude. Therefore, measures are taken by using a multicarrier transmission scheme such as equalization processing and OFDM (Orthogonal Frequency Division Multiplexing) that narrows the signal band. In particular, the equalization process generally estimates the preceding wave and delay wave of the received signal, adjusts the phase, etc., and combines them to effectively use the delay wave and improve performance. However, in general, a pilot sequence is required for estimation of the preceding wave and the delayed wave. A method for obtaining a tap coefficient for a filter to be set in an equalizer corresponding to STBC based on a propagation path estimated value has been studied in Non-Patent Document 2. Non-Patent Document 2 discloses a method of applying a linear equalizer and a decision feedback equalizer. In addition, for a propagation path that fluctuates at high speed, tracking that performs propagation path tracking based on a determination result of a data series section other than the known series is used. Tracking with respect to the STBC signal is disclosed in Non-Patent Document 3, for example.

  However, a method for making the equalization technique described in Non-Patent Document 2 and the tracking technique described in Non-Patent Document 3 correspond to a DSTBC signal that is a differentially encoded signal by DSTBC has not been sufficiently studied. In differential spatio-temporal modulation used in DSTBC, even when the constellation of a signal to be transmitted is QPSK (Quadrature Phase Shift Keying), the number of candidate points exceeds four (see Non-Patent Document 4, Patent Document 1, etc.). Therefore, when applying the propagation path estimation that tracks the propagation path based on the equalization output or the decision feedback equalizer (DFE), there are many candidate points and signal points between the candidate points. The distance between them becomes smaller. Therefore, there has been a problem that the performance of channel estimation and decision feedback equalization is degraded due to erroneous determination. In addition, the method of re-encoding the decision sequence after differential decoding and using it for tracking, etc. has a problem of error propagation in which, when a part of the differential encoded sequence is wrong, an error is propagated to the subsequent sequence. The application of re-encoding has been difficult. Further, a technique for selectively receiving either the preceding wave or the delayed wave by beam forming is also conceivable. For example, when the preceding wave is selectively received, the delay wave is received with a null. However, when the preceding wave and the delayed wave are overlapped, it is difficult to receive only the preceding wave. Also, it is not easy to cope with a propagation path that fluctuates at high speed.

Japanese Unexamined Patent Publication No. 2014-12083

V. Tarokh and H.M. Jafarkhani, “A Differential Detection Scheme for Transmit Diversity,” IEEE Journal on Selected Areas in Communications, Vol. 18, pp 1169-1174, July 2000. W. H. Gerstacker, F.A. Obernosterer, R.A. Schober, A.M. T. T. Lehmann, A.M. Lampe, and P.M. Gunreben, "Equalization concepts for Alamouti's space-time block code," IEEE Trans. Commun. , Vol. 52, no. 7, pp. 1178-1190, July 2004. N. Nededov and G. P. Mattellini, “Evaluation of potential transmission diversity schemes with iterative receivers in EDGE,” Proc. IEEE PIMRC, vol. 5, pp. 2087-2091, Sept. 2002. X. Shao and J.H. Yuan, “A new differential space-time block coding scheme,” Proc. ICCS 2002, vo. 1, pp. 183-187, Nov. 2002.

  The problem of error propagation in the method of re-encoding the determination sequence after differential decoding and using it for tracking, etc. is that the determination sequence before differential decoding is used for tracking instead of the determination sequence after differential decoding. It is possible to avoid it. However, in this case, there is another problem that erroneous determination increases because the constellation is complicated, that is, the distance between signal points is reduced. When the number of misjudgments increases, the quality of tracking and feedback signals deteriorates, and it is difficult to accurately realize channel tracking and decision feedback equalization that are essential in an environment with high-speed propagation path fluctuations. As described above, conventionally, there has been a problem that it is difficult to solve both the problem of error propagation and the problem of reception performance degradation in an environment with high-speed propagation path fluctuation.

  The present invention has been made in view of the above, and an object of the present invention is to obtain a receiving apparatus with improved receiving performance.

  In order to solve the above-described problems and achieve the object, the present invention provides a propagation path estimation means for estimating a propagation path based on a known signal included in a received signal, and a differential space-time coded signal. An equalization unit that equalizes the received signal based on the estimation result by the propagation path estimation unit, and a block determination that determines the signal after equalization by the equalization unit in units of blocks configured by a plurality of symbols Means. The propagation path estimation means updates the propagation path estimation result based on the determination result in the block determination means.

  According to the present invention, there is an effect that it is possible to realize a receiving apparatus with improved receiving performance.

FIG. 10 illustrates a configuration example of a receiving device according to the first embodiment; FIG. 9 is a diagram illustrating a configuration example of a receiving device according to the second embodiment.

  Hereinafter, a receiving apparatus according to an embodiment of the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited to the embodiments.

Embodiment 1 FIG.
First, signals received by the receiving apparatus according to the present invention will be described. The receiving apparatus according to the first embodiment and the second embodiment to be described later receives a signal differentially encoded by a differential space-time coding system (hereinafter, DSTBC system).

  The differential encoding using DSTBC can be expressed by the following equation (1).

  Here, C (k) is a matrix corresponding to the differential encoding result in a certain block k. S (k) is a mapping result, that is, a matrix corresponding to the signal before differential encoding. For example, in QPSK transmission, the element of S (k) is composed of a general QPSK modulation point, its sign inverted, and a complex conjugate. In differential encoding, the output C (k) of the next block is calculated using its own output C (k−1). For example, assuming that DSTBC transmission is performed using two antennas, the block is configured in units of two symbols, and the matrix has a size of 2 × 2. Hereinafter, the description will be made assuming a two-antenna DSTBC. In this example, the matrices S (k) and C (k) are configured as in the following equation (2).

The matrix element of C (k) corresponds to the output waveform of each antenna when transmitted with two antennas and two symbols (corresponding to time in DSTBC). For example, the matrix element of two symbols constituting a block Send c (1, k) and c (2, k) with two antennas in the previous symbol, and send -c ^* (2, k) and c ^* (1, k) with the next symbol. means. In the following explanation, the signal is associated with the symbol and the antenna as described above, but in actuality, which element is transmitted by which antenna at which time (symbol timing) in the block is uniquely limited. Not. For example, the order of signals transmitted in two symbols may be reversed. As described above, the differential encoding calculation result is appropriately assigned to space (antenna) and time (symbol timing), subjected to frequency conversion, filtering, etc., modulated onto a desired carrier wave, and transmitted from the antenna. On the other hand, the mapping signal matrix S (k) is similarly arranged for each block k so that two symbols s (1, k) and s (2, k) correspond to the respective complex conjugates.

  Hereinafter, a receiving apparatus according to the first embodiment that receives a DSTBC signal that is a signal differentially encoded by the above-described DSTBC method will be described. In the first embodiment, as an example, a case where a receiving apparatus receives signals with two antennas will be described. In addition, it is assumed that the opposite apparatus that is the transmission source of the DSTBC signal received by the receiving apparatus transmits signals using two antennas. The number of antennas is not limited to this.

  FIG. 1 is a diagram illustrating a configuration example of a receiving device 100 according to the first embodiment. The receiving apparatus 100 includes antennas 1-1 and 1-2 that receive radio signals, RF units 2-1 and 2-2 that convert radio signals received by the antennas into baseband signals, and transmission of received signals. Propagation path estimation units 3-1 and 3-2 for estimating the state of the propagation path between the original opposite device and the own apparatus, and baseband signals input from the RF units 2-1 and 2-2 The linear equalizer 4 that is a linear equalizer that generates a baseband signal that is compensated for the influence received in the propagation path by executing linear processing and the signal output from the linear equalizer 4 in units of blocks A block determination unit 5 for determination, a differential space-time decoding unit 6 that differentially decodes a determination result in the block determination unit 5, and an error that performs error correction processing on the differential decoding result in the differential space-time decoding unit 6 A correction decoding unit 7.

  Next, the operation of receiving apparatus 100 will be described. The RF units 2-1 and 2-2 perform processing similar to that of a general single carrier reception RF unit, specifically, down conversion processing, band limitation processing, and the like on the received signal in the radio frequency band. , Convert to baseband signal. The RF units 2-1 and 2-2 output a symbol-synchronized 1-times oversampled signal or several-times oversampled signals as baseband signals, respectively. The output signal from the RF unit 2-1 is input to the propagation path estimation unit 3-1 and the linear equalization unit 4, and the output signal from the RF unit 2-2 is transmitted to the propagation path estimation unit 3-2 and the linear equalization unit 4 Is input.

  The propagation path estimators 3-1 and 3-2, which are propagation path estimation means, extract a sync word, which is a known signal, from the baseband signals output from the RF sections 2-1 and 2-2, and extract the extracted sync word In other words, the delay wave is included based on the sync word in the state affected by the propagation path and the sync word generated in the receiving apparatus 100, that is, the sync word in the state not affected by the propagation path. Perform propagation path estimation. For example, in the propagation path estimation process assuming 2-antenna transmission and a 3-tap propagation path, the propagation path estimation units 3-1 and 3-2 each estimate 6 elements. Here, the propagation path estimation is performed by a method similar to a known propagation path estimation technique for STBC and other single carrier equalization techniques. The estimation results in the propagation path estimation units 3-1 and 3-2 are output to the linear equalization unit 4. Here, when the channel estimation units 3-1 and 3-2 receive the determination result from the block determination unit 5 described later, the channel estimation units 3-1 and 3-2 update the channel estimation result based on the received determination result.

  The linear equalization unit 4, which is equalization means, generates tap coefficients of the space-time filter for equalization based on the estimation results received from the propagation path estimation units 3-1 and 3-2, and the generated taps The output signals from the RF units 2-1 and 2-2 are equalized using a space-time filter in which coefficients are set. The tap coefficient is generated in accordance with, for example, a ZF (Zero Forcing) standard and a MMSE (Minimum Mean Square Error) standard in the same manner as a known equalizer for STBC. For example, when generating the filter coefficient according to the MMSE standard, the filter coefficient is generated so that the least square error between the reference signal and the received signal is minimized. In this embodiment, since the signals transmitted from the two antennas are received and equalized, the linear equalization unit 4 removes intersymbol interference from the signals received from the signals transmitted from the respective transmission antennas. The equalized signal becomes the equalization result. That is, two systems of equalization results are obtained. The linear equalization unit 4 combines the two systems of equalization results obtained by the equalization processing, and outputs the result to the block determination unit 5 as a baseband signal after equalization. Since the original information is one stream in DSTBC, the linear equalization unit 4 restores the signal of one transmission branch by equalization processing, or the first symbol of the two symbols constituting the block. It is only necessary that two signals can be restored by equalization processing, and the equalization result of one transmission branch or the equalization result of two signals of the first symbol of two symbols may be output as an equalized baseband signal. .

  The block determination unit 5 serving as a block determination unit determines the equalized baseband signal input from the linear equalization unit 4 in units of one block. That is, the signal point pattern that can be transmitted is compared with the baseband signal after equalization to determine which signal point pattern is transmitted. Specifically, it is determined that the nearest signal point has been transmitted. Here, one block is a unit in which the opposite apparatus that is the signal transmission source performs differential encoding. In this embodiment, the opposite device that is a signal transmission source uses two antennas, and one block is composed of two symbols. The block determination unit 5 calculates the distance from the baseband signal after equalization for all combinations of signals that can be transmitted in two symbols, and the candidate having the closest distance from the baseband signal after equalization is transmitted. Judge that For example, a transmission signal candidate (one of the combinations that c (1, k) and c (2, k) can take) and the square error of the baseband signal after equalization are calculated, and 2 symbols (1 block) ) Is determined that the transmission signal candidate having the smallest sum of squared errors is transmitted in the block. The block determination unit 5 determines the distance between the memory storing all signal point patterns that can be transmitted and the signal output from the linear equalization unit 4 and each signal pattern stored in the memory. This can be realized by a processor that determines the signal pattern calculated and received.

  In the conventional determination process, the signal after differential decoding is determined for each symbol, whereas the block determination unit 5 according to the first embodiment performs determination at a stage before differential decoding. Are distributed over many points. However, for example, a signal obtained by differentially encoding a signal modulated with BPSK (Binary Phase Shift Keying) or QPSK with DSTBC is related to the first symbol and the second symbol of the block. If the eye is determined, the constellation of the second symbol is limited to a range subject to certain restrictions. Also, depending on the initial values of mapping and differential encoding, the first symbol and the second symbol may be linked in a one-to-one relationship. That is, by selecting the block determination candidates in consideration of the above-described interlocking relationship, the number of candidate points is reduced and the distance between the signal points is increased on average as compared with the conventional case of determining one symbol at a time. The determination accuracy can be improved. Actually, signals s (1, k), s (2, k) and initial values c (1, k), c (2, k) before differential encoding are determined at the design stage of the radio. In many cases, combinations of candidate points c (1, k) and c (2, k) that can be taken in advance are obtained in a simulation and the combination pattern is used as a determination candidate in each block of the determination unit. Good. The distance between a block candidate related to a differentially encoded signal, that is, a combination of two symbol transmission signal candidates and another block candidate is the case where differential space-time encoding processing is performed in a unitary matrix. This is the same as the distance before differential encoding. Therefore, it can be avoided that the distance between the signal points becomes small. When DSTBC is applied to BPSK and QPSK, differential encoding is performed using a unitary matrix. Therefore, by performing block determination processing for determining one block, that is, two symbols, the above-mentioned equalized signal On the other hand, higher determination performance can be obtained as compared with the case where determination is performed in symbol units as in the prior art.

  The determination result in the block determination unit 5 is input to the differential space-time decoding unit 6 and the propagation path estimation units 3-1 and 3-2.

  The propagation path estimators 3-1 and 3-2 use the determination result input from the block determination unit 5 as a tracking transmission signal pattern, perform tracking processing to follow the time variation of the propagation path, and Update the estimate. The tracking process is performed using an algorithm such as LMS (Least Mean Square) or RLS (Recursive Least Square). Here, when the sync word sequence is arranged at a place other than the head of the frame and it is necessary to decode a past signal and a future signal from the sync word, respectively, the former signal is traced back in time. Tracking may be performed for the latter signal in the direction of time tracking. When tracking back in time, the received signal sequence is once buffered, and the tracking is started after receiving data such as a sync word sequence as a starting point of processing. The channel estimation value updated by the tracking process is input to the linear equalization unit 4, and the linear equalization unit 4 updates the equalization filter tap coefficient using the input channel estimation value. With the above configuration, in transmission of a DSTBC signal including a known sequence such as a sync word, the transmitted signal sequence can be restored even in a situation where there is a time variation of the propagation path and frequency selectivity. .

  The differential space-time decoding unit 6 performs differential space-time decoding on the determination result input from the block determination unit 5 and outputs the differential space-time decoding result to the error correction decoding unit 7. The differential space-time decoding unit 6 performs differential decoding using an input value corresponding to the previous block, and performs processing for extracting the original QPSK mapping sequence by differential decoding, for example, in QPSK transmission.

  The process of the error correction decoding unit 7 can be a reception process (including a demapping process) similar to that of a general transmission system to which DSTBC is not applied. In a system that does not use error correction coding, a hard decision process is performed on the differential space-time decoding result, and the decision result is output. In FIG. 1, the output of the block determination unit 5 is input to the differential space-time decoding unit 6. However, since the output of the block determination unit 5 has already been output by the signal corrected by the determination process, The dynamic space-time decoding unit 6 cannot generate soft decision data effective for the error correction decoding unit 7. For this reason, the signal before block determination, that is, the output signal from the linear equalization unit 4 is configured to be input to the differential space-time decoding unit 6, and the differential space-time decoding unit 6 performs soft decision and performs soft decision. Data may be generated so that the error correction decoding unit 7 can perform error correction decoding processing based on the soft decision data.

  As described above, the receiving apparatus 100 according to the first embodiment performs the determination on the equalization result in the linear equalization unit 4 for each block, and uses this determination result to follow the time variation of the propagation path. Tracking processing was performed. As a result, it is possible to avoid the occurrence of error propagation that becomes a problem when the determination result after differential decoding is re-encoded and used for tracking. In addition, it is possible to suppress erroneous determination due to the narrowness of the distance between signal points by performing the determination processing necessary for tracking processing before differential decoding, compared with the case where determination is performed for each symbol. Performance can be improved. That is, tracking performance and equalization performance can be improved as compared with the case where determination is performed for each symbol.

Embodiment 2. FIG.
FIG. 2 is a diagram illustrating a configuration example of the receiving device 100a according to the second embodiment. The receiving apparatus 100a has a configuration in which the linear equalization unit 4 of the receiving apparatus 100 according to the first embodiment shown in FIG. 1 is replaced with a DFE equalizing filter unit 11. Since the configuration other than the DFE equalization filter unit 11 is the same as that of the first embodiment, the same reference numerals as those of the receiving device 100 of the first embodiment are given and description thereof is omitted.

  The DFE equalization filter unit 11 is a decision feedback equalizer (DFE: Decision Feedback Equalizer).

  The DFE equalization filter unit 11 includes an FF filter 12 that performs a feedforward filtering process, an FB filter 13 that performs a feedback filtering process, a filtering result output from the FF filter 12, and a filtering result output from the FB filter 13. And an adding unit 14 for adding.

  The operation of the receiving device 100a will be described. The output signal from the RF unit 2-1 is input to the propagation path estimation unit 3-1, and the FF filter 12 of the DFE equalization filter unit 11, and the output signal from the RF unit 2-2 is transmitted to the propagation path estimation unit 3-2. And input to the FF filter 12 of the DFE equalization filter unit 11. Propagation path estimation values, which are estimation results in the propagation path estimation units 3-1 and 3-2, are output to the FF filter 12 and the FB filter 13 of the DFE equalization filter unit 11. The FF filter 12 and the FB filter 13 generate filter tap coefficients based on the propagation path estimation values output from the propagation path estimation units 3-1 and 3-2. The tap coefficient is generated by, for example, the ZF norm and MMSE norm as well as the DFE tap coefficient corresponding to the known STBC.

  The FF filter 12 generates tap coefficients generated based on the propagation path estimation values output from the propagation path estimation sections 3-1 and 3-2 with respect to the signals input from the RF sections 2-1 and 2-2. Use for filtering. The filtered signal, that is, the filtering result for the input signal from the RF unit 2-1 and the filtering result for the input signal from the RF unit 2-2 are combined and then input to the adding unit 14.

  The FB filter 13 filters the determination result input from the block determination unit 5 using the tap coefficient generated based on the channel estimation values output from the channel estimation units 3-1 and 3-2. I do. The filtered signal is input to the adding unit 14.

  The adding unit 14 adds, that is, combines the signal input from the FF filter 12 and the signal input from the FB filter 13, and outputs the combined signal to the block determination unit 5.

  In the feedback filtering process of DFE, it is necessary to feed back the determination result. In receiving apparatus 100a of the second embodiment, instead of re-encoding the determination sequence after differential decoding, the determination result in block determination unit 5 is fed back to DFE equalization filter unit 11. The DFE equalization filter unit 11 performs filtering on the determination result fed back by the FB filter 13. The determination result in the block determination unit 5 is also used as a transmission signal pattern for tracking in the propagation path estimation units 3-1 and 3-2, similarly to the reception device 100 of the first embodiment.

  The processing after differential space-time decoding unit 6 is the same as that of receiving apparatus 100 of the first embodiment. Further, when error correction decoding using soft decision data is performed, the input to the block determination unit 5, that is, the output of the DFE equalization filter unit 11 may be provided to the differential space-time decoding unit 6.

  In Embodiments 1 and 2 described above, the number of reception antennas is two, but the number of reception antennas is not limited to this. When the number of reception antennas is other than two, reception apparatuses 100 and 100a according to Embodiments 1 and 2 require the same number of RF units and propagation path estimation units as the number of reception antennas. In the receiving apparatus 100, a signal from each RF unit and a propagation path estimation value from each propagation path estimation unit are input to the linear equalization unit 4, and the linear equalization unit 4 is input from each RF unit. Equalization processing is performed on the signal, and the equalization results are combined and output to the block determination unit 5. In the receiving apparatus 100a, the output signal from each RF unit and the propagation path estimation value output from each propagation path estimation unit are input to the FF filter 12, and the FF filter 12 filters the signal input from each RF unit. Process. Further, the propagation path estimation value output from each propagation path estimation unit is input to the FB filter 13, and the FB filter 13 performs a filtering process on the determination result input from the block determination unit 5.

  As described above, the receiving apparatus 100a according to the second embodiment includes the DFE equalization filter unit 11 that is a decision feedback equalizer, and performs determination on the equalization result in the DFE equalization filter unit 11 for each block. The determination result is fed back to the DFE equalization filter unit 11 and the tracking process for following the time variation of the propagation path is performed using the determination result. As a result, it is possible to avoid the occurrence of error propagation that becomes a problem when the determination result after differential decoding is re-encoded and used for tracking. In addition, it is possible to suppress erroneous determination due to the narrowness of the distance between signal points by performing the determination processing necessary for tracking processing before differential decoding, compared with the case where determination is performed for each symbol. Performance can be improved. That is, tracking performance and equalization performance can be improved as compared with the case where determination is performed for each symbol.

  The configuration described in the above embodiment shows an example of the contents of the present invention, and can be combined with another known technique, and can be combined with other configurations without departing from the gist of the present invention. It is also possible to omit or change the part.

1-1, 1-2 antenna, 2-1, 2-2 RF unit, 3-1, 3-2 propagation path estimation unit, 4 linear equalization unit, 5 block determination unit, 6 differential space-time decoding unit, 7 Error correction decoding unit, 11 DFE equalization filter unit, 12 FF filter, 13 FB filter, 14 addition unit.

Claims (3)

Propagation path estimation means for estimating a propagation path based on a known signal included in the received signal;
Equalizing means for equalizing the received signal, which is a differential space-time encoded signal, based on an estimation result by the propagation path estimating means;
Block determination means for determining the signal after equalization by the equalization means in units of blocks composed of a plurality of symbols;
With
The propagation path estimation means updates the propagation path estimation result based on the determination result in the block determination means,
A receiving apparatus.   The receiving apparatus according to claim 1, wherein the equalizing means is a linear equalizer. The equalizing means includes
A feedforward filter that performs filtering on the differential space-time encoded signal using a tap coefficient based on an estimation result by the propagation path estimation unit;
A feedback filter that performs filtering on the determination result using a tap coefficient based on an estimation result by the propagation path estimation unit;
An adder that adds the signal output from the feedforward filter and the signal output from the feedback filter to output to the block determination means;
The receiving apparatus according to claim 1, further comprising:

Images