JP3715382B2 - Receiver - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a code division multiple access (CDMA) receiver.
[0002]
[Prior art]
As a mobile communication system capable of increasing user capacity and improving communication quality, a code division multiple access (hereinafter abbreviated as CDMA) system is adopted. This CDMA system is a multiple access system in which a specific code is assigned to each line, a modulated wave having the same carrier frequency is spread by this code, and the code is synchronized on the receiving side to identify a desired line. , Also called SSMA (Spread Spectrum Multiple Access) method. The CDMA system does not require an initial connection process, and has an advantage that communication can be performed directly for each call as long as the codes are determined in advance, and has excellent features such as confidentiality and interference resistance.
[0003]
By the way, in a communication system employing this CDMA system, in order to improve frequency utilization efficiency, communication is usually performed using the same carrier frequency in the service area (hereinafter referred to as a cell) of each base station, and voice The device has been devised to efficiently link the probability of occurrence (voice activity) to an increase in system capacity.
[0004]
Also, the user capacity of a CDMA system is limited by the total interference power of other users (other channels) generated in all cells, but the user capacity increases by positively removing and suppressing this interference. A receiving method that enables the above has been proposed. For example, optimal receivers, orthogonalized receivers, adaptive interference cancellation receivers, multistage receivers, etc. have been proposed as such reception methods, and among them, optimal receivers have the best characteristics for user multiplexing. Have.
[0005]
This optimum receiver generates a signal replica for all users who perform communication, and an information symbol (for example, BPSK (two-phase modulation) that minimizes the error signal power between the received signal and all user signal replicas. ) Method is a maximum likelihood sequence estimation method for estimating a combination of signals indicating transmission information such as {+1, −1}.
[0006]
[Problems to be solved by the invention]
As described above, the conventional optimum receiver has the best reception characteristics, but it must generate a received signal replica for all information symbol combinations of all users and obtain an error signal power from the received signal. Therefore, a huge amount of calculation is required for reception.
[0007]
In addition, in a conventional interference canceling receiver including an optimum receiver, a method corresponding to a system that uses voice activity (for example, a system that changes a bit rate by changing a duty cycle of a transmission signal) is considered. There wasn't. Furthermore, the conventional optimum receiver has not considered an efficient reception method for realizing diversity.
[0008]
Accordingly, an object of the present invention is to provide a receiving apparatus capable of reducing the amount of calculation at the time of reception. Another object of the present invention is to provide a receiving apparatus capable of realizing a system that can change the bit rate by changing the duty cycle of a transmission signal, and further realizing an efficient receiving method for realizing diversity.
[0009]
[Means for Solving the Problems]
In order to achieve the above object, the present invention receives and demodulates a code division multiple access (CDMA) system signal in which a plurality of wireless communication devices perform multiplex communication using different spreading codes within the same frequency band. In the receiving device,
Channel response estimation means for estimating the channel response received by the received signal for each spreading code, maximum likelihood sequence estimation only for some of the received signals, and the signal component of the determination result It comprises partial maximum likelihood sequence estimation means that removes the received signal and outputs it, and subsequent stage detection means that demodulates signals other than the part of the signal from the output signal from the maximum likelihood sequence estimation means. Further, the partial signal for performing partial maximum likelihood sequence estimation has the largest amplitude component among the time response of the transmission line response to each spreading code obtained by the transmission line response estimation means for each spreading code. Comparison was made and some of the higher-order products were selected from the comparison results in descending order of the amplitude component.
[0010]
As described above, in the present invention, the partial maximum likelihood sequence estimation means demodulates the signal of a part of the user from the received signal, removes the demodulated part from the received signal, and gives the remainder to the subsequent detection means as a residual signal. Then, the remainder is demodulated by the latter detection means. In the case of this system, when the residual signal is demodulated by the post-stage detection means, the residual signal does not include the user signal component demodulated by the partial maximum likelihood sequence estimation means. That is, the signal corresponding to the user demodulated by the partial maximum likelihood sequence estimation means is removed when the partial maximum likelihood sequence estimation means outputs the residual signal, and is no longer included. In the post-stage detection means, the demodulated user signal is not subject to demodulation, and the non-demodulation target signal is an interference signal. In the present invention, the demodulated user interference signal is included in the signal provided to the post-stage detection means. Since it is not included, the power of the interference component is reduced, and the subsequent detection means can obtain good reception characteristics. In addition, among the received signals, a plurality of users' signals are classified according to the signal strength, and first, maximum likelihood sequence determination is performed on the signals of the users of the group having a strong signal strength, and the signals of this group are determined from the received signals It is excluded, and the remaining is given to the subsequent detection means which is another maximum likelihood sequence estimation means, and the maximum likelihood sequence is determined.
[0011]
As a means for performing maximum likelihood sequence estimation, if the number of users to be judged is large, the scale of processing increases exponentially, and the processing time and the scale of processing hardware also increase in proportion to increase the cost. Invite. Also, depending on the scale, it cannot be realized with current processing hardware. However, this can be solved by dividing and processing as in the present invention.
[0012]
In addition, the receiving apparatus of the present invention uses partial maximum likelihood sequence estimation means for performing maximum likelihood sequence estimation including a state in which no signal is transmitted when performing demodulation.
In this case, the partial maximum likelihood sequence estimation means determines whether or not a signal is transmitted in addition to the information determination. Therefore, even in a system in which the duty cycle of a transmission signal changes in order to use voice activity, by determining whether or not a signal is transmitted in addition to information determination, there is no deterioration in characteristics. Optimal reception can be realized.
[0013]
Further, according to the present invention, the partial maximum likelihood sequence estimation means spreads all of the determination sequence compensation with a spreading code corresponding to each element of the sequence compensation, and provides a corresponding transmission line response to each spread signal. A reception signal replica compensation obtained by performing a convolution operation is generated, and a decision sequence candidate that minimizes the error power of the reception signal and the sum of the reception signal replica compensation of the partial signals is obtained as a reception result. Output it.
[0014]
In this way, by performing maximum likelihood sequence estimation using a received signal replica corresponding to the transmission path response of each signal, a bus diversity effect can be expected, and moreover, in a transmission path environment without a large level of delay signal, a plurality of Diversity of signals received by these antennas using delay elements with different times and combining them, treating them as new received signals, and performing optimal reception to obtain a diversity effect Can do.
[0015]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, a receiving method of a CDMA system according to the present invention will be described in detail with reference to the drawings.
FIG. 1 is a diagram for explaining the outline of a CDMA cellular system, in which 11 is a cell, 12 and 14 are CDMA base stations, and 13 and 14 are mobile terminals. The CDMA base stations 12 and 14 are arranged in a distributed manner, and in one base station, a range where radio waves from the base station can reach is a cell that is a service area of the base station.
[0016]
The CDMA base stations 12 and 14 can communicate with the mobile terminals 13 and 14 in the cell by performing wireless communication by the CDMA method. In order to prevent communication from being interrupted even if the mobile terminals 13 and 14 move between cells, a base station is installed in such a manner that a part of the cell overlaps with an adjacent cell. When the mobile terminals 13 and 14 are turned on, or moved from outside the service area of the CDMA cellular system, and when a location registration request is received from the CDMA base station, a specific control channel is set. The base station uses the base station cell to receive its own ID (identification code) to the CDMA base station, receive it at the base station, and receive it at a plurality of CDMA base stations. The location registration is made assuming that the mobile terminal is located inside.
[0017]
Location registration is performed by registering the mobile terminal ID and cell information in a location registration apparatus installed in a center that supervises the operation of the CDMA cellular system, and when communication with the mobile terminal becomes necessary By checking the location of the mobile terminal using the information of the location registration device, it grasps the current location and communicates with the mobile terminal via the base station serving the cell at the current location. become.
[0018]
In general, in a CDMA cellular system, a plurality of base stations are distributed and the radio wave coverage at each base station is defined as the service area of the base station, and each service area is referred to as a cell. The same carrier frequency is used for transmission and reception by code division multiple access. In order to continue communication between a mobile communication terminal and a base station, in a cell adjacent to a certain cell, the base station of that cell uses the carrier used by the mobile communication terminal. Communication is performed by CDMA using the same carrier frequency as the frequency.
[0019]
As described above, since the same carrier frequency is used in the plurality of adjacent cells 11, as shown in FIG. 1, a signal from the terminal device 131 belonging to the base station 12 of another cell is transmitted to another base station 14. It is mixed in the receiving device as an interference wave. Usually, these CDMA interference waves from the outside of the own cell behave in the same manner as noise in the receiving apparatus of the base station 14 because the code type, code timing, transmission path response state, etc. are unknown. FIG. 2 schematically shows signal components received by the base station in the frequency domain.
[0020]
Incidentally, there is an optimum receiver as a receiving method for reducing the influence of interference between other stations in the CDMA system. As described above, this reception method obtains a good user multiplexing characteristic by maximum likelihood sequence estimation, but has a drawback that the amount of calculation increases exponentially as the number of users increases. However, in the reception situation as shown in FIG. 2, the signal component of the user that is not large in power among the interference in the own cell, or the interference of the user in the own cell up to the same level as noise and other cell interference is Since the maximum likelihood sequence determination is not greatly affected, the calculation amount can be reduced without causing significant characteristic deterioration by removing the maximum likelihood determination from the compensation. The present invention takes advantage of this property.
[0021]
(Specific example 1)
FIG. 3 shows a specific example of the CDMA receiver according to the present invention using the above-described properties. In the figure, 31 is an RF receiver circuit, 32 is a code synchronization circuit, 33 is a transmission path response estimation circuit, 34 is a partial maximum likelihood sequence estimator, and 35 is a post-stage detection circuit.
[0022]
Among these, the RF receiving circuit 31 is a circuit that obtains a baseband signal from the signal received by the antenna, and the code synchronization circuit 32 performs synchronization acquisition and holding for each user's spreading code from the received signal. Circuit. The transmission path response estimation means 33 is a circuit that estimates the transmission path response received by each user's signal based on the baseband signal supplied from the RF receiver circuit 31 and the code timing from the code synchronization circuit 32.
[0023]
The partial maximum likelihood sequence estimation means 34 is a device that performs demodulation based on the baseband signal given from the RF receiving circuit 31 and the transmission path response estimation information from the transmission path response estimation means 33, Among the received signals, maximum likelihood sequence estimation is performed on a part of the signals, for example, signals for a predetermined number of users in order from the strongest received signal strength, and the received signal replica corresponding to the estimation result is received signal. The remaining signal (residual signal) subtracted from the signal is generated and given to the subsequent detection means 35.
[0024]
Subsequent detection means 35 is an apparatus for obtaining detection data by demodulating based on the signal (residual signal) from partial maximum likelihood sequence estimation means 34 and the signals from code synchronization circuit 32 and transmission line response estimation means 33. .
[0025]
This apparatus having such a configuration receives a transmission signal from a base station via an antenna. Then, the received transmission signal from the base station is sent to the RF receiving circuit 31 to be converted into a baseband signal. The baseband signal may be an analog signal or a digital signal. In the case of a digital signal, the RF receiving circuit 31 has an A / D (analog / digital) converter, and the processing stages after the RF receiving circuit 31 Then, necessary processing is performed by digital signal processing.
[0026]
The baseband signal converted by the RF receiving circuit 31 is supplied to the code synchronization circuit 32, the transmission path response estimation means 33, and the partial maximum likelihood sequence estimation means 34. Based on this baseband signal, the code synchronization circuit 32 acquires and holds synchronization for the spread code of each user from the received signal. Then, the code synchronization circuit 32 uses the obtained code timing to determine the channel response received by each user signal, the channel response estimation unit 33 for estimating the channel response, the partial maximum likelihood sequence estimation unit 34 for demodulation, and the subsequent detection. Provided to means 35.
[0027]
Based on this code timing, the transmission path response estimation means 33 processes the baseband signal and estimates the transmission path response received by each user's signal.
On the other hand, the partial maximum likelihood sequence estimation means 34 performs maximum likelihood sequence estimation on a predetermined number of users in order from a part of the received signals in the own cell, for example, signals having strong received signal strength, Further, the remaining signal (residual signal) obtained by subtracting the received signal replica corresponding to the maximum likelihood sequence estimation result from the received signal is generated and transmitted to the subsequent detection means 35. That is, some of the above-mentioned user signals in the baseband signal are demodulated and output as detection data. Also, a received signal replica is created for the demodulated signals of some users, and this received signal replica is subtracted from the baseband signal that is the original received signal r to obtain the remaining signal (residual signal (r And a demodulated replica of each user's signal))) is generated and provided to the post-detection means 35.
[0028]
As a result, the partial maximum likelihood sequence estimation means 34 demodulates the signals of some users having strong signal strength from the received signal, and the post-detection means 35 demodulates the signals of other users not demodulated. It will be. That is, the signals of other users that have not been demodulated by the partial maximum likelihood sequence estimation means 34 are included in the residual signal, and are provided to the subsequent detection means 35. Therefore, in the subsequent detection means 35, the residual signal Thus, the signal of an undemodulated user is demodulated. Then, the downstream detection means 35 outputs the demodulated user signal as detection data.
[0029]
As described above, in the present invention, the partial maximum likelihood sequence estimation means 34 demodulates the signal of a part of the user, and gives the remainder to the subsequent detection means 35 as a residual signal, which is demodulated here. This is a major feature of the invention. That is, when the residual signal is demodulated by the post-stage detection unit 35, the residual signal does not include the user signal component demodulated by the partial maximum likelihood sequence estimation unit 34. That is, the signal corresponding to the user demodulated by the partial maximum likelihood sequence estimation unit 34 is removed when the partial maximum likelihood sequence estimation unit 34 outputs the residual signal, and is no longer included. In the post-stage detection unit 35, the demodulated user signal is not subject to demodulation, and the signal not to be demodulated becomes an interference signal. In the present invention, however, the signal supplied to the post-stage detection unit 35 does not interfere with the demodulated user. Since no signal is included, the power of the interference component is reduced, and the post-detection means 35 can obtain good reception characteristics.
[0030]
An example of the signal processing procedure of the receiving apparatus of the present invention will be described using the flowchart of FIG. First, the receiving apparatus captures and holds the synchronization of the spreading code assigned to each user (step S1).
[0031]
In this case, the time when a new user starts communication is set as the start time. However, in the spreading code of the user who has already received, this process is performed in synchronization (the synchronization is performed regularly or periodically). Is equivalent). In addition, a signal that is irregularly mixed, such as a signal according to a control protocol before starting communication, can be excluded from this routine.
[0032]
Next, the channel response is estimated using the autocorrelation characteristics of each spread code of the received signal (step S2). In this process, the time, amplitude, and phase of the main wave and delay wave are measured using a known signal included in the transmission signal. In addition, when not using a delayed wave, it is also possible to simplify a receiver by measuring only the amplitude and phase of a main wave.
[0033]
Next, the transmission line response is corrected (step S3). The correction of the transmission line response here is a process of correcting the transmission line response estimated value obtained in the previous process by using the demodulation result. When the signal immediately after the known signal is received or in a system with little transmission line fluctuation Then it can be omitted.
[0034]
Next, the received signal strength for each user is detected using the obtained channel response estimation value for each user (spread code of each user), and the user numbers are rearranged in the order of received strength based on the detection result. (Step S4). As a result, assuming that the number of users is equal to K users (where K is 1 <L <K, and L and K are both integers), each spread code for K users has a strong received signal strength. They will be arranged in order from the one. Then, L users are selected in descending order of reception signal strength (step S5), and partial maximum likelihood sequence estimation is performed (step S6). As a result, partial maximum likelihood sequence estimation is performed for the first “L” user with strong reception strength among the spread codes for “K” users. Then, the spread codes for the remaining users are sent as residual signals to the subsequent detection means 35, and the subsequent detection means 35 demodulates the signals for the remaining "KL" users (step S7). .
[0035]
Thus, the received signal strength for each user is detected using the obtained channel response estimation value for each user (spread code of each user), and the user numbers are rearranged in the order of received strength (where, The received signal strength may be the signal power of the main wave, but as another specific example, the power integral value of the estimated transmission path response (delay profile) may be used) Demodulation processing is performed by the partial maximum likelihood sequence estimation means 34 which is a front-stage receiving unit, and demodulation of the remaining user spread codes is processed by the rear-stage detection means 35. As another specific example, maximum likelihood sequence estimation can be performed for a signal of a predetermined spreading code regardless of the signal strength, and the processing procedure can be simplified by this method.
[0036]
Further, when the number of users actually communicating is “L” or less, it is possible to demodulate all of them by the partial maximum likelihood sequence estimation means 34 regardless of the signal strength. Further, like the channel response estimation, this process can be performed for each symbol, or the same result can be used over a plurality of symbols.
[0037]
Next, the partial maximum likelihood sequence estimation means 34 receives “L” users selected in the previous step. In this process, demodulation is performed in the same manner as in the conventional optimum reception method, but the received signal replica obtained from the demodulation result, spreading code, transmission path response estimation result, and code timing is deleted from the received signal (remaining). By outputting the difference signal to the subsequent detector, it is possible to reduce the interference component given to other signals by the user demodulated in this process.
[0038]
Finally, the post-stage detection means 35 performs the process of step S7, demodulates the remaining user signal from the residual signal, and returns to the transmission path response correction process for the next symbol demodulation. As another specific example of the present invention, it is possible to simultaneously demodulate a plurality of symbols in time.
[0039]
As described above, in this specific example, among the received signals, a plurality of user signals are classified according to the signal strength, and first, maximum likelihood sequence determination is performed on the signals of the users in the group having the strong signal strength. The signal of the group is excluded from the received signal, and the remaining signal is given to another maximum likelihood sequence estimation means to perform maximum likelihood sequence determination.
[0040]
When the number of users to be judged is large, the maximum likelihood sequence estimation means increases the scale of processing exponentially, and the processing time and the scale of processing hardware also increase proportionally, resulting in high costs. Also, depending on the scale, it cannot be realized with current processing hardware. However, this can be solved by dividing and processing as in this specific example.
[0041]
(Specific example of partial maximum likelihood sequence estimation means 34)
Next, one specific example of the partial maximum likelihood sequence estimation means 34 of the present invention is shown in FIG. Reference numeral 40 in FIG. 4 denotes partial maximum likelihood sequence estimation means in this specific example. The partial maximum likelihood sequence estimation means 40 in this specific example is composed of a received power sorting means 41, a received signal replica generation means 42, partial sequence candidate generation and maximum likelihood sequence estimation means 43.
[0042]
Among these, the received power sorting means 41 sorts the user numbers in descending order of the signal strength (received power) based on the input from the transmission path response estimating means 33, and the upper L of them ( A function of selecting the upper L users). The received signal replica generation means 42 also transmits the transmission path response estimation result inputted from the transmission path response estimation means 33, the code timing information of each user, and the spread codes for the L users selected by the reception power sorting means 41. The received signal replica is generated from the partial sequence candidate generation and the decision sequence interpolation obtained from the maximum likelihood sequence estimation means 43.
[0043]
Further, the partial sequence candidate generation and maximum likelihood sequence estimation means 43 demodulates the spread codes for L users selected by the received power sorting means 41, outputs detection data for L users, and obtains it from the subtractor 44. And a function for outputting the difference signal to the subsequent detection means as a residual signal. The subtractor 44 is a circuit for obtaining a difference between the received signal from the RF receiving circuit and the received signal replica obtained by the received signal replica generating means 42.
[0044]
In such a configuration, the result obtained by the transmission path response estimation means 33 (sample value of the delay profile estimation value of the transmission path response received by each user) is input to the user sorting means 41, and the user sorting means 41 Sorts the user numbers in descending order of the signal strength (reception power) based on the input from the transmission path response estimation means 33, and the upper L of them, that is, the upper L having the strong signal strength. Select the user. Then, the user sorting means 41 notifies the user numbers for the selected L users to the partial sequence complement generation and maximum likelihood sequence estimation means 43. Further, the user sorting means 41 inputs the transmission path response estimation results for these selected L users to the reception replica generation means 42.
[0045]
In the subsequence complement generation and maximum likelihood sequence estimation means 43, signal symbols that can be transmitted by each user (± 1 for BPSK, (± 1 / √2, ± 1 / √2) for QPSK, and M-arySS). For example, the types of transmission sequences) are generated in combination for L users (decision sequence supplement), and all combinations are sequentially output to the received signal replica generation means 42.
[0046]
The reception signal replica generation means 42 receives the transmission path response estimation result inputted from the transmission path response estimation means 33, the decision sequence compensation, the spread code for the selected L users, and the code timing information of each user. Generate a signal replica. Then, the generated received signal replica is given to the subtractor 44.
[0047]
On the other hand, a reception signal is given to the subtractor 44 from the RF receiving circuit. The subtracter 44 receives an error (residual) between the reception signal from the RF receiving circuit and the reception signal replica from the reception signal replica generation means 42. Signal) is obtained by subtraction of both. Then, the subtractor 44 gives the obtained error (residual signal) to the partial sequence interpolation generation and maximum likelihood sequence estimation means 43.
[0048]
Then, the partial sequence complement generation and maximum likelihood sequence estimation means 43 calculates the power value (error power) again based on this.
Using this error power as a branch metric, maximum likelihood sequence estimation for a determination sequence candidate is performed. The partial maximum likelihood sequence estimation means 40 outputs the result of the maximum likelihood sequence estimation or the result of the previous time when the buses of the maximum likelihood sequence estimation are merged and the residual signal corresponding thereto.
[0049]
Here, the residual signal may be calculated anew at the time of series determination, or a memory is provided and stored in this memory until the judgment converges, and the contents of the memory are called according to the judgment result. Also good. As yet another specific example, the residual signal of each series compensation may be sequentially output to the subsequent detection means. In this case, the post-stage detection means sequentially demodulates the signals of the remaining users, stores the respective demodulation results in a memory corresponding to the sequence complement, and determines the sequence detection means when the series complement is determined. It is also possible to take a method of simultaneously determining the result. By adopting this method, the time required for demodulation can be greatly reduced.
[0050]
Further, as another specific example of the present invention, the partial maximum likelihood sequence estimation means prepares cross-correlation coefficients of spreading codes for all judgment system candidates and outputs an output signal from a correlator corresponding to each spreading code. It is also possible to perform maximum likelihood sequence estimation using
[0051]
In this case, calculation and determination of the error signal is performed at information symbol intervals, and residual signal generation is performed again using the determination result. By using this method, it is possible to reduce the processing time required for maximum likelihood sequence estimation.
[0052]
(Specific example 2)
Next, another specific example of the CDMA receiving apparatus of the present invention will be described with reference to FIG.
In the figure, 51 is an RF receiver circuit, 52 is a code synchronization circuit, 53 is a transmission line response estimation circuit, 54 is a partial maximum likelihood sequence estimator, and 55 and 56 are post-stage detection circuits.
[0053]
Among these, the RF receiving circuit 51 is a circuit that obtains a baseband signal from the signal received by the antenna, and the code synchronization circuit 52 is synchronized with the spread code of each user from the received signal (baseband signal). A circuit that captures and holds. The transmission path response estimation means 53 is a circuit that estimates the transmission path response received by each user signal based on the baseband signal given from the RF receiving circuit 51 and the code timing from the code synchronization circuit 52.
[0054]
The partial maximum likelihood sequence estimation means 54 is a device that performs demodulation based on the baseband signal given from the RF receiving circuit 51 and the transmission path response estimation information from the transmission path response estimation means 53, Among the received signals, maximum likelihood sequence estimation is performed on a part of the signals, for example, signals for a predetermined number of users in order from the strongest received signal strength, and the received signal replica corresponding to the estimation result is received signal. The remaining signal (residual signal) subtracted from the signal is generated and given to the subsequent detection means 55.
[0055]
Subsequent detection means 55 is an apparatus for obtaining detection data by demodulating based on the signal (residual signal) from partial maximum likelihood sequence estimation means 54 and the signals from code synchronization circuit 52 and transmission line response estimation means 53. . The downstream detection means 55 is a device that obtains detection data by demodulating a signal for L users with strong received power intensity, performs maximum likelihood sequence estimation from the residual signal, and receives a received signal replica corresponding to the estimation result, It has a function of generating a further remaining signal (final residual signal) subtracted from the residual signal and supplying it to the subsequent detection means 56. The post-stage detection unit 56 is a device that demodulates the final residual signal from the post-stage detection unit 55 to obtain detection data.
[0056]
This apparatus having such a configuration receives a transmission signal from a base station via an antenna. The received transmission signal from the base station is sent to the RF receiving circuit 51 to be converted into a baseband signal. The baseband signal converted by the RF receiving circuit 51 is given to the code synchronization circuit 52, the transmission path response estimation means 53, and the partial maximum likelihood sequence estimation means 54. Based on the baseband signal, the code synchronization circuit 52 acquires and holds synchronization for the spread code of each user from the received signal. Then, the code synchronization circuit 52 uses the obtained code timing to determine the channel response received by each user's signal, the channel response estimation unit 53 for estimating the channel response, the partial maximum likelihood sequence estimation unit 54 for demodulation, and the subsequent detection. The means 55 and 56 are given.
[0057]
On the other hand, the partial maximum likelihood sequence estimation means 54 is a signal corresponding to a predetermined number of users (for example, the number of L users) in order from a part of the received signals in the own cell, for example, signals having strong received signal strength. And the remaining signal (residual signal) obtained by subtracting the received signal replica corresponding to the maximum likelihood sequence estimation result from the received signal is generated and transmitted to the downstream detection means 55. In this manner, the signals for the L users as described above in the baseband signal are demodulated and output as detection data. Also, a received signal replica is created for the demodulated signals for L users, and this received signal replica is subtracted from the baseband signal that is the original received signal to generate the remaining signal (residual signal). Then, it is given to the subsequent detection means 55.
[0058]
As a result, the partial maximum likelihood sequence estimation means 54 demodulates the signals of some users having strong signal strength from the received signal, and the signals of other users not demodulated are detected by the subsequent detection means 55 and 56. It will be demodulated.
[0059]
That is, the signals of other users that have not been demodulated by the partial maximum likelihood sequence estimation means 54 are included in the residual signal, and are provided to the subsequent detection means 55. Thus, a predetermined number of undemodulated user signals with strong received power intensity are demodulated. Then, the downstream detection means 55 outputs the demodulated user signal as detection data. Further, the post-stage detection means 55 generates a replica of the demodulated user signal, subtracts it from the received residual signal, obtains it as a final residual signal, and gives this to the post-stage detection means 56. The signals of other users that have not been demodulated by the post-stage detection unit 55 are included in the final residual signal, and are supplied to the post-stage detection unit 55. An undemodulated user signal is demodulated. Then, the downstream detection means 56 outputs the demodulated user signal as detection data.
[0060]
As described above, the residual signal output from the partial maximum likelihood sequence estimation unit 54 is input to the subsequent stage detection unit 55. The residual signal output from the post-stage detection unit 55 is input to the post-stage detection unit 56. In this specific example, the inside of the subsequent stage detection means 55 and 56 is realized by the same configuration as the partial maximum likelihood sequence estimation means 54 that is a component of the previous stage. That is, in this specific example, the configuration of the partial maximum likelihood sequence estimation means (the latter detection means 55 and 56) at the rear stage is the same as that of the previous stage (the partial maximum likelihood sequence estimation means 54), but the sequence to be demodulated is In the partial maximum likelihood sequence estimation means 54, the groups for the L users having the strongest received power intensity are those from the “# 1” user to the “#L” user, whereas the latter detection means 55 receives the received power intensity. Are groups from “# L + 1” to “# 2L” users, which are groups for L users in the next order, and the downstream detection means 56 is a group for 1 to L users with the lowest received power intensity. From “Kn · L” user to “#K” user.
[0061]
In addition, when the latter stage detection means 55 and 56 cannot process all users, it is possible to connect another latter stage detection means. In this case, partial maximum likelihood sequence estimation means can be used as another post-stage detection means at the third stage (FIG. 16), or other methods can be used.
[0062]
In the partial maximum likelihood sequence estimation of FIG. 16, code timing acquisition is performed on the received signal (step S21), transmission path response estimation is performed (step S22), user numbers are rearranged in the order of reception intensity (step S23), and reception is performed. The signals for L users are selected in descending order of strength (step S24), replicas are generated for the sequence candidates (step S25), sequence candidates are selected (step S26), and the selected sequence candidates are output ( This is performed by processing such as step S26) and outputting the remaining signals to the subsequent stage (step S28).
[0063]
(Another specific example of the latter detection means)
A specific example of another subsequent detection means will be described with reference to FIG. In FIG. 6, 60 is a post-stage detection means. The post-stage detection means 60 includes a # L + 1 detector 61, a # L + 2 detector 62, a # L + 3 detector 63, and a #K detector 64. The # L + 1 detector 61, the # L + 2 detector 62, the # L + 3 detector 63, and the #K detector 64 are all configured by a matched filter or a correlator, and a matched filter corresponding to a spread code of an undemodulated user or Detection data is obtained by despreading the residual signal using a correlator. # L + 1 detector 61 is for L + 1 users, # L + 2 detector 62 is for L + 2 users, # L + 3 detector 63 is for L + 3 users, and #K detector 64 is for K users It corresponds to a code and despreads the signal of each corresponding user.
[0064]
In such a configuration, the residual signal output from the partial maximum likelihood sequence estimation means in the previous stage is despread using the matched filter 61 corresponding to the spreading code of the undemodulated user or the correlator.
[0065]
In addition, under the condition that there is no interference signal in the signal to be demodulated, the demodulated code of the non-demodulated user can be demodulated well by using a matched filter when demodulating the spread code of the non-demodulated user. can do. In this specific example, the residual signal output from the partial maximum likelihood sequence estimation means in the previous stage is demodulated in the subsequent stage. Therefore, since there is no signal that is not subject to demodulation in the residual signal, it can be said that there are few interference signals. Therefore, by using a matched filter as a detector, a spread code of an undemodulated user can be demodulated with an inexpensive configuration, and the cost of the system can be reduced.
[0066]
Here, as another specific example of the despreading means of the CDMA system using the orthogonal code as the spreading code, a high-speed Hadamard transformer can be used. Still another specific example of the latter-stage detection means will be described with reference to FIG.
[0067]
(Another specific example of the latter detection means)
7 includes a cross-correlation inverse characteristic matrix processing unit 71 that performs cross-correlation inverse characteristic matrix processing of spreading codes, an output of the cross-correlation inverse characteristic matrix processing unit 71, and a transmission path response estimator (transmission path response). A multiplier 72 for multiplying the output from the estimation means), and a discriminator 73 for receiving the multiplication output of the multiplier 72 and determining the sign of the output. There are a plurality of multipliers 72 and discriminators 73, each of which is for a user whose received power is in a specific order.
[0068]
This method is a decorrelator type reception method, and demodulates a spread code of a user other than the user (undemodulated user) demodulated by partial maximum likelihood sequence estimation using the cross-correlation inverse characteristic matrix processing means 71 of the spread code. To do.
[0069]
The demodulated output from the cross-correlation inverse characteristic matrix processing means 71 is subjected to phase correction using a multiplier 72 and subjected to sign determination using a discriminator 73.
(Another specific example of the latter detection means)
Still another specific example of the latter-stage detection means will be described with reference to FIG. This method uses a plurality of interference cancellation receivers. As shown in FIG. 8, in the latter-stage detection means 80 using this interference cancellation receiver, the residual signal from the partial maximum likelihood sequence estimation means, which is the previous stage, is input to the interference cancellation receiver 81, and the interference cancellation receiver In 81, first, the signal of the user ("# L + 1" user) having the largest signal strength among the undemodulated user channel response estimation values is despread and demodulated by the spreading code of the "# L + 1" user, The detection data of “# L + 1” user is obtained, and the transmission signal replica of “# L + 1” user is removed from the residual signal using the demodulation result, and is output to the next interference cancellation receiver 82 as the residual signal. . The interference cancellation receiver 82 is for a user (“# L + 2” user) having the largest signal strength among the remaining channel demodulation estimates of undemodulated users, and the interference cancellation receiver 82 uses the interference cancellation receiver 81. Is despread with the spreading code of the “# L + 2” user to demodulate the signal of “# L + 2” user, and using the demodulation result, the transmission signal replica of “# L + 1” user Are removed from the residual signal and output to the next interference cancellation receiver 83. Demodulation is performed by performing the same procedure for all undemodulated users.
[0070]
(One specific example of reception signal replica generation means)
FIG. 12 shows a specific example of the received signal replica generation means of the present invention. As shown in FIG. 12, received signal replica generation means 120 includes transmission path response estimation and sorting means 121, spreading sequence generation means 122, transmission path response memory 123, band limiting filter 124, convolution calculator 125, and reception signal replica memory 126. It consists of. The transmission path response memory 123, the band limiting filter 124, the convolution calculator 125, and the reception signal replica memory 126 are respectively provided for L systems.
[0071]
The transmission path response estimation / sorting means 121 performs transmission path response estimation and has a function of rearranging user numbers in descending order of received power intensity. The transmission path response memory 123 includes transmission path response estimation and sorting. This is a memory for holding the estimated channel response value estimated by the means 121.
[0072]
The spreading sequence generating means 122 generates a spreading sequence, and the band limiting filter 124 limits the spreading sequence output from the spreading sequence generating means 122 and outputs it, and is constituted by a low pass filter (LPF). ing. The convolution calculator 125 performs a convolution calculation process on the output of the band limiting filter 124 and the transmission path response estimated value held in the transmission path response memory 123, and the reception signal replica memory 126 holds the result.
[0073]
In such a configuration, the transmission channel response estimation and sorting means 121 obtains the transmission channel response estimation value and the signal strength. Then, the transmission path response estimation and sorting means 121 gives the obtained transmission path response estimated value to the transmission path response memory 123 and holds it there, and the user information selected based on the signal strength is sent to the spreading sequence generation means 122. Forward.
[0074]
The spread sequence generation means 122 generates a spread sequence of the user selected in accordance with this user information, and inputs these to the band limiting filter 124 to generate a signal corresponding to the spread sequence (band limited spread signal). obtain. The obtained band-limited spread signal and the transmission path response estimated value previously stored in the transmission path response memory 123 are convolved by the convolution calculator 125 and stored in the received signal replica memory 126 for each user. Is done.
[0075]
Thereby, the reception signal replica for each user is held in the reception signal replica memory 126. These signals are transferred to the outside of the reception signal replica generation means as necessary. It should be noted that the band limiting filter can be omitted as another means for simplifying this example.
[0076]
(Transmission path response estimation means)
FIG. 13 shows a specific example of the transmission path response estimation means of the present invention. As shown in FIG. 13, the transmission line response estimation means 130 of the present invention includes a matched filter 131, a multiplier 132, a shift register 133, a threshold value determination unit 134, a feedback weight coefficient memory 135, an adder 136, and a previous estimated value memory 137. , Switch 138.
[0077]
Description will be given focusing on the user “#k”. When the signal (baseband signal) output from the RF receiving circuit is supplied to the transmission path response estimation means 130 of the present invention, it is first input to the matched filter 131. The matched filter 131 obtains a correlation characteristic with the spreading code of the user “#k”. The output signal of the matched filter (correlation characteristic with the spread code of the user “#k”) is phase-corrected using the multiplier 132 based on the known signal sequence or the demodulation result, thereby obtaining a delay profile of the transmission line response. The obtained delay profile is sampled by being introduced into the shift register 133 at an appropriate timing by the switch 138.
[0078]
Each sample value in the shift register 133 is obtained by separately calculating the previous estimated value (previous transmission path response estimated value) stored in the previous estimated value memory 137, the feedback weight held in the feedback weight coefficient memory 135, and A weighted average is performed using an adder 136.
[0079]
Note that the threshold determination unit 134 is for removing a low-level signal and can be omitted. As another specific example, the threshold determination unit 134 can be connected to a memory output.
[0080]
Although the description has been given for the user “#k” here, channel response estimation for other users can be realized by using the coefficients of the matched filter and the reference or human demodulation result of the desired user. In addition, when estimation is performed using the demodulation result, the received signal is delayed by a time until there is a determination input by the delay unit.
[0081]
(Another example of transmission path response estimation means)
Another specific example of the transmission path response estimation means of the present invention will be described with reference to FIG. 14 includes a matched filter 141, multipliers 142 and 149, a transversal filter 143, a threshold determiner 144, a weight coefficient holder 145, an adder 146, a memory 147, a subtractor 148, and a multiplication. 1410 and switches 1411 and 1412.
[0082]
The received signal (baseband signal) output from the RF receiving circuit is given to the adder 146 and the weighted coefficient is given to the received signal at the current time calculated in the previous cycle by the weighting coefficient holding unit 145 to correct the weight. The adder 146 is added to add both.
[0083]
Then, the output of the adder 146 is given to the subtractor 148, where the interference wave component is removed. An interference wave component is input to the subtracter 148, and the output of the adder 146 is subtracted by the subtracter 148, thereby removing the interference wave component. It is not necessary to remove the interference signal for all users, and the interference wave component can be set to no input depending on the communication situation.
[0084]
The signal after the interference wave component removal is input to the matched filter 141, and a correlation characteristic with the spread code “#k” is obtained. The output signal of the matched filter 141 is multiplied by a complex conjugate signal of a known sequence or a demodulation result symbol through a delay element 1413 at an appropriate timing. This multiplication is performed by the multiplier 142. As a result of this multiplication, a delay profile signal is obtained.
[0085]
The delay profile signal obtained in this way is transferred to the memory 147 by the switch 1411 at an appropriate timing, and is sampled and held.
A signal (sample value) sampled and held in the memory 147 becomes a transmission line response coefficient, and is provided to the outside as a transmission line response coefficient. Further, each sample value in the memory 147 passes through the threshold value determination unit 144 independently and is subjected to threshold value determination, and those below the threshold value are cut. As a result, after the low-level component is deleted, it is input to the transversal filter 143 as a weighting coefficient.
[0086]
A spread signal is input to the transversal filter 143, and an output signal is obtained by performing processing as a transversal filter using this and the above-described weighting factor. The output signal of the transversal filter 143 is given to the multiplier 1410 together with a known sequence or a demodulation result symbol and multiplied, thereby generating an interference signal replica. In this way, the transmission path response estimation means 140 obtains a transmission path response coefficient and an interference signal replica.
[0087]
Here, the explanation has been given for the user “#k”, but the transmission path response estimation for other users is performed by combining the matched filter coefficients and the input interference signal and the reference or input demodulation result of the desired user. It can be realized by changing.
[0088]
When estimation is performed using the demodulation result, the received signal is delayed by a delay device until a determination input is received, and timing adjustment is performed.
(Specific example 3)
Another specific example of the present invention is shown in FIG. This example is an example for realizing diversity, and the receiving apparatus includes a plurality of antennas 178. Then, the antennas 178 are added by the adding means 177 after different times through the delay devices 176 having different delay times, supplied to the RF receiving circuit 171 and demodulated to become baseband signals. Note that the addition processing by the adding means 177 can be performed at an IF (intermediate frequency) signal stage or a baseband signal stage.
[0089]
The baseband signal converted by the RF reception circuit 171 is given to the code synchronization circuit 172, the transmission path response estimation means 173, and the partial maximum likelihood sequence estimation means 174. Based on the baseband signal, the code synchronization circuit 172 acquires and holds synchronization for the spread code of each user from the received signal. The code synchronization circuit 172 uses the obtained code timing to determine the channel response estimation unit 173 that estimates the channel response received by each user signal, the partial maximum likelihood sequence estimation unit 174 that performs demodulation, and the post-stage detection. Means 175 are provided.
[0090]
On the other hand, in partial maximum likelihood sequence estimation means 174, a part of the received signals in the own cell, for example, signals of a predetermined number of users (for example, the number of L users) in descending order of the received signal strength. And the remaining signal (residual signal) obtained by subtracting the received signal replica corresponding to the maximum likelihood sequence estimation result from the received signal is generated and transmitted to the downstream detection means 175. In this manner, the signals for the L users as described above in the baseband signal are demodulated and output as detection data. Also, a received signal replica is created for the demodulated signals for L users, and this received signal replica is subtracted from the baseband signal that is the original received signal to generate the remaining signal (residual signal). Then, it is given to the latter detection means 175.
[0091]
As a result, partial maximum likelihood sequence estimation means 174 demodulates the signals of some users having strong signal strength from the received signal, and signals of other users not demodulated are demodulated by post-stage detection means 175. It will be.
[0092]
(Maximum likelihood sequence estimation method)
A specific example of the maximum likelihood sequence estimation method of the present invention is shown in FIG.
In a CDMA system that uses voice activity based on duty ratio at the time of transmission or switching (VOX), the user in communication does not always send out radio waves, so the non-transmission state ("0" in the figure). The sequence estimation is performed by adding the indicated symbols). Thereby, since the signal in a non-transmission state can be accurately estimated, reception characteristics can be improved.
[0093]
(Another maximum likelihood sequence estimation method)
Another specific example of the maximum likelihood sequence estimation method of the present invention is shown in FIG.
In the maximum likelihood sequence estimation method of a CDMA system using VOX, the number of states increases, so the amount of signal processing increases accordingly.
[0094]
When the VOX switching pattern is a group of a plurality of symbols, a user in a non-transmission state is limited in state over a plurality of bits. Therefore, the sequence complement including the state of “0” which is a non-transmission state performs the maximum likelihood sequence estimation with a constraint condition of “0” in the next stage. By doing so, the amount of signal processing can be reduced.
[0095]
FIG. 10 is a state diagram of maximum likelihood sequence estimation in the case where there is a time difference between the multibus transmission path or the spread codes of each user. Although the state including “0” is not shown in this figure, it is also possible to perform sequence estimation including the “0” state as in FIGS. 9 and 10.
[0096]
(Specific example 4)
Another specific example of the CDMA receiver will be described. 18 and 19 show a specific example of the present invention. FIG. 18 is a block diagram of this receiving apparatus, which includes an RF receiving circuit 181, a code synchronization circuit 182, a transmission path response estimating means 183, a received signal replica generating means 184, and a partial sequence estimator 185. The RF receiving circuit 181 is a circuit that obtains a baseband signal from the signal received by the antenna, and the code synchronization circuit 182 captures and holds synchronization with respect to each user's spreading code from the received signal (baseband signal). It is a circuit to perform. The transmission path response estimation means 183 is a circuit that estimates the transmission path response received by each user's signal based on the baseband signal supplied from the RF receiving circuit 181 and the code timing from the code synchronization circuit 182.
[0097]
The received signal replica generation means 184 generates a replica signal of each user from the transmission path response of each user obtained by the transmission path response estimation means 182 and information on the spreading sequence of each user. Further, the partial sequence estimator 185 divides all users to be received into a plurality of groups as shown in FIG. 19, and performs sequence estimation. In this sequence estimation, all combinations of information symbols of users in the group are used as states, and survival path selection for sequence estimation is performed between the grouped states, and demodulation is performed.
[0098]
This apparatus having such a configuration receives a transmission signal from a base station via an antenna. Then, the received transmission signal from the base station is sent to the RF receiving circuit 181 to be converted into a baseband signal. The baseband signal converted by the RF reception circuit 181 is given to the code synchronization circuit 182, the transmission path response estimation means 183, and the partial maximum likelihood sequence estimation means 184.
[0099]
Based on the baseband signal, the code synchronization circuit 182 acquires and holds synchronization for the spread code of each user from the received signal. Then, the code synchronization circuit 182 gives the obtained code timing to the transmission path response estimation means 183 that estimates the transmission path response received by each user's signal and the partial maximum likelihood sequence estimation means 185 that performs demodulation.
[0100]
The transmission path response estimation means 183 processes the baseband signal based on the code timing and estimates the transmission path response received by each user's signal.
On the other hand, the received signal replica generation means 184 generates a replica signal of each user from the transmission path response of each user obtained by the transmission path response estimation means 182 and the spread sequence information of each user. Further, the partial sequence estimator 185 divides all users to be received into a plurality of groups as shown in FIG. 19, and performs sequence estimation. In this sequence estimation, all combinations of information symbols of users in the group are used as states, and a survival path selection for sequence estimation is performed between the grouped states to perform demodulation.
[0101]
Each state is equipped with a memory that holds decision sequence candidates and a memory that holds a residual signal obtained by subtracting the replica signal of the decision sequence compensation from the received signal, eliminating interference of the determined information symbols However, sequence estimation is performed. By using this reception method, good reception characteristics can be obtained with a small amount of calculation.
[0102]
(Specific example 5)
A specific example of another receiving apparatus will be described with reference to FIG.
As shown in FIG. 20, this receiving apparatus includes an RF receiving circuit 201, a code motive circuit 202, a transmission path response estimating means 203, a replica generating means 204, and partial sequence estimators 205, 206-1 to 206-n. The RF receiver circuit 201 corresponds to the RF receiver circuit 101 in FIG. 18, the code synchronization circuit 202 corresponds to the code synchronization circuit 182 in FIG. 18, and the transmission path response estimation means 203 corresponds to the transmission path response estimation means 183 in FIG. The replica generation unit 204 corresponds to the replica generation unit 184 in FIG. 18, and the partial sequence estimators 205 and 206-1 to 206-n correspond to the partial sequence estimator 185 in FIG.
[0103]
The receiver configuration of this specific example is basically the same as that shown in FIG. 18, but the partial sequence estimators are cascade-connected and have the same configuration as so-called multistage reception.
[0104]
At this time, the first-stage subsequence estimator 205 operates with the same algorithm as the above-described partial-sequence estimator 185, but the subsequent-stage subsequence estimator 206 after the cascade connection uses the result determined in the previous stage. Thus, sequence estimation is performed by constraining information symbols of users other than sequence estimation candidates.
[0105]
By adopting such a method, the reception characteristics can be further improved.
Although various specific examples have been described above, since these are described as independent configurations, the means for performing the same processing are described redundantly. However, when a receiving apparatus is realized by combining these, an overlapping processing system can be omitted. In addition, the above-described components of the receiving apparatus of the present invention do not need to be physically separated, and can be configured on a single or a plurality of signal processors.
[0106]
【The invention's effect】
As described above in detail, in the CDMA receiver according to the present invention, the number of states for maximum likelihood sequence estimation is limited according to the system operating state or the received signal power. Reduce the amount. In the reception system of the present invention, even in a system in which the duty cycle of a transmission signal changes in order to use voice activity, by determining whether a signal is transmitted in addition to information determination, Optimal reception can be realized without any characteristic degradation. Furthermore, in the receiving system of the present invention, a diversity effect can be expected by performing maximum likelihood sequence estimation using a received signal replica corresponding to the transmission path response of each signal, and in a transmission path environment where there is no large level delay signal. A diversity effect can be obtained by combining signals received by a plurality of antennas after being delayed by delay elements having different times, treating them as new received signals, and performing optimum reception.
[Brief description of the drawings]
FIG. 1 shows a CDMA cellular system.
FIG. 2 is a diagram showing a state of an interference signal in the frequency domain of a CDMA system.
FIG. 3 is a diagram showing a configuration example of a receiving apparatus that employs the receiving method of the present invention.
FIG. 4 is a diagram showing a configuration example of a partial maximum likelihood sequence estimator in the reception system of the present invention.
FIG. 5 is a diagram showing another configuration example of a receiving apparatus employing the receiving method of the present invention.
FIG. 6 is a diagram showing an example of the configuration of a post-stage detector in the reception system of the present invention.
FIG. 7 is a diagram showing another example of the configuration of the post-stage detector in the reception system of the present invention.
FIG. 8 is a diagram showing another example of the configuration of the post-stage detector in the reception system of the present invention.
FIG. 9 is an example of a state diagram showing a sequence estimation method for partial maximum likelihood sequence estimation according to the present invention (in the case of two users).
FIG. 10 is another example of a state diagram showing the sequence estimation method of partial maximum likelihood sequence estimation according to the present invention (in the case of two users).
FIG. 11 is another example of a state diagram showing the sequence estimation method of partial maximum likelihood sequence estimation according to the present invention (in the case of 2 users 1 symbol delay).
FIG. 12 is a diagram showing a configuration example of received signal replica generation means in the receiving system of the present invention.
FIG. 13 is a diagram showing a configuration example of a transmission path response estimation unit in the reception system of the present invention.
FIG. 14 is a diagram showing another configuration example of transmission path response estimation means in the reception system of the present invention.
FIG. 15 is a diagram showing an example of a reception processing procedure in the reception method of the present invention.
FIG. 16 is a diagram showing another example of a reception processing procedure in the reception method of the present invention.
FIG. 17 is a diagram showing another example of the reception scheme configuration of the present invention.
FIG. 18 is a diagram showing an example of the configuration of a receiving apparatus according to the present invention.
FIG. 19 is a diagram illustrating an example of a reception method according to the present invention. (-State transition diagram for 3 users in group, binary modulation)
FIG. 20 is a diagram showing another example of the configuration of the receiving apparatus of the present invention.
[Explanation of symbols]
11 ... cell
12, 14 ... CDMA base station
13, 15 ... Terminal device
31, 5, 1 ... RF receiving circuit
32, 52 ... Spreading code synchronization means
33, 53 ... Transmission path response estimation means
34, 40, 54, 55, 56 ... Partial maximum likelihood sequence estimation means
35, 60, 70.80 ... latter stage detection means
41. User sorting means
42. Receiving signal leveler generating means
43 ... Subsequence complement generation and maximum likelihood sequence estimation means
44, 148 ... subtractor
61 ... # L + 1 user signal receiver
62 ... # L + 2 user signal receiver
63 ... # L + 3 user signal receiver
64... #K user signal receiver
71 ... Decorrelator receiver
72. Multiplier
73 ... Determinator
81... “# L + 1” user signal receiver / remover
82... “# L + 2” user signal receiver / remover
83 ... "# K" user signal receiver
121 ... Transmission path response estimation and user sorting means
122. Spreading sequence generating means
123 ... Transmission path response information memory
124 ... Filter
125 ... Convolution calculator
126. Received signal replica memory
130, 140 ... Transmission path response estimation means
131, 141 ... # k correlator for user (or digital matched filter)
132, 142, 1410 ... Multiplier
133: Shift register
135.145 ... Return weight
136, 146 ... Adder
137, 147 ... memory
143 ... Transversal filter
149 ... Complex conjugate arithmetic unit
1411, 1412 ... switch
1413: Delay device
135.145 ... Return weight
136, 146 ... Adder
137, 147 ... memory
143 ... Transversal filter
149 ... Complex conjugate arithmetic unit
181, 201 RF receiver circuit
182, 202... Code synchronization circuit
183, 203 ... Transmission path response estimation and user sorting means
184, 204 ... Received signal replica generation means
185, 205 ... Subsequence estimation means
206-1 to 206-n ... Subsequence estimation means for the latter stage
1411, 1412 ... switch
1413: Delay device

Claims (14)

Receiving means for receiving a CDMA (Code Division Multiple Access) signal in which a plurality of user signals are multiplexed using different spreading codes;
Transmission path response estimation means for estimating a transmission path response received by each user signal included in the CDMA signal received by the reception means for each spreading code corresponding to each user signal;
A first partial maximum likelihood sequence for performing maximum likelihood sequence estimation on a predetermined number of user signals among a plurality of user signals included in the CDMA signal received by the receiving means, and demodulating the predetermined number of user signals An estimation means;
Generating means for generating a first residual signal obtained by removing the predetermined number of user signals from the CDMA signal received by the receiving means;
First demodulating means for demodulating user signals other than the predetermined number of user signals from the first residual signal;
A receiving apparatus comprising:   The first partial maximum likelihood sequence estimation means compares the largest amplitude component of the time response of the transmission line response to each spreading code obtained by the transmission line response estimation means for each spreading code, and The receiving apparatus according to claim 1, wherein the predetermined number of user signals are selected in descending order of the amplitude component from user signals.   The first partial maximum likelihood sequence estimation means adds power to the time response component of the transmission path response for each spreading code obtained by the transmission path response estimation means, and compares the result for each spreading code. The receiving apparatus according to claim 1, wherein the predetermined number of user signals are selected in descending order of the power component from user signals.   The number of user signals that are subjected to partial maximum likelihood sequence estimation by the first partial maximum likelihood sequence estimation means reduces the amount of computation of maximum likelihood sequence estimation that exponentially increases as the number of multiplexed user signals increases The receiving apparatus according to claim 1, wherein the receiving apparatus is defined for the purpose. The first partial maximum likelihood sequence estimation means includes:
Each decision sequence complement generated by combining a plurality of information symbols corresponding to each of the predetermined number of user signals is obtained by spreading using a spreading code corresponding to each of the predetermined number of user signals. For each of the plurality of spreading sequences, a reception signal replica compensation is generated by convolution calculation of the corresponding transmission line response, and the error power of the signal of the sum of the predetermined number of user signals and the reception signal replica compensation is The receiving apparatus according to claim 1, wherein the smallest determination sequence candidate is output as a demodulation result. The first partial maximum likelihood sequence estimation means includes:
Storage means for holding a received signal replica for at least one information symbol being compensated for a decision sequence, and multiplying an information symbol indicating a demodulation result of maximum likelihood sequence estimation and the received signal replica stored in the storage means The receiving apparatus according to claim 5 , further comprising:   2. The receiving apparatus according to claim 1, wherein the generating unit generates the first residual signal obtained by subtracting a sum of received signal replicas corresponding to the demodulation result from the CDMA signal.   The first demodulating means passes the first residual signal through a matched filter corresponding to each spreading code of each user signal other than the predetermined number of user signals, so that the signals other than the predetermined number of user signals are transmitted. 9. The receiving apparatus according to claim 8, wherein each user signal is demodulated.   2. The receiving apparatus according to claim 1, wherein the transmission path response estimation means includes storage means for holding a sampling value of the transmission path response of the estimation result.   It further comprises a plurality of antennas, different delay elements each having a delay time of one chip period or more of a spreading code for each antenna, and signal synthesis means for adding the outputs of the respective delay elements. The receiving device according to claim 1. It said first portion maximum likelihood sequence estimation means according to claim 5, wherein generating a plurality of decision sequence candidates including decision sequence candidates with information symbol indicating a state where the signal in the user signal is not transmitted Receiver. The first demodulating means includes
Second partial maximum likelihood sequence estimation means for performing maximum likelihood sequence estimation only on a predetermined number of user signals among a plurality of user signals included in the first residual signal and demodulating the predetermined number of user signals When,
Generating means for generating a second residual signal in which the predetermined number of user signals are removed from the first residual signal;
Second demodulation means for demodulating user signals other than the predetermined number of user signals from the second residual signal;
The receiving apparatus according to claim 1, further comprising:   The second maximum likelihood sequence estimation means generates a plurality of determination sequence candidates including a determination sequence candidate having an information symbol indicating a state in which no signal is transmitted by a user signal, and performs maximum likelihood sequence estimation. The receiving device according to claim 12, characterized in that: The first partial maximum likelihood sequence estimator is a maximum likelihood sequence estimator for user signals of the first group having the highest received power intensity among a plurality of user signals included in the CDMA signal received by the receiver. To demodulate
The second partial maximum likelihood sequence estimation means performs maximum likelihood sequence estimation on the second group of user signals having the next highest received power intensity among the plurality of user signals included in the first residual signal. 13. The receiving apparatus according to claim 12, wherein the receiving apparatus demodulates.

Images