JP3957511B2 - Path detection device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a path detection apparatus in mobile communication.
[0002]
[Prior art]
In a multiple propagation path environment, a wave transmitted from one transmission station passes through a plurality of propagation paths, and arrives at a reception station as a plurality of arrival waves having different delay times. In the direct carrier code sequence multiple access (DS-CDMA) system, it is possible to perform reception using multipath positively by performing Rake combining. A path detection device is used to select a path for performing such Rake synthesis.
[0003]
FIG. 4 is a block diagram schematically showing a conventional path detection apparatus.
In the figure, reference numeral 2 denotes a matched filter, which inputs a complex baseband signal obtained by orthogonally demodulating a code spread received signal with a reference frequency signal, in other words, a received I-phase signal and a received Q-phase signal. Then, despreading is performed by correlating with the replica of the spreading code. The despread received signal is output to the phase compensation / Rake combiner 3 and the path detector 7 for each chip period of the spread code.
The phase compensation and Rake combining unit 3 matches the correlation output timings of the symbols of the plurality of paths included in the despread received signal, and multiplies them by the complex conjugate of the channel estimation value, thereby combining them. Perform phase compensation and maximum ratio synthesis.
Specifically, the channel estimation of each path is performed by estimating the channel at the time of the data symbol of each path from the measured values of the amplitude and phase of the pilot symbol in each path.
[0004]
The path detection device 21 detects the individual incoming waves (paths) by measuring the power delay profile of the received signal, and controls the phase compensation and the Rake combiner 3 so as to Rake combine only the desired signal path. To do.
The path detection device 21 includes a power measuring device 22, a path selector 7, and a threshold setting device 23.
The power meter 22 measures the received power level of the despread signal. The output is a power delay profile.
The threshold setting unit 23 estimates the average power Pave of the thermal noise component (including the interference component) from the power delay profile, and sets a threshold for the path selection unit 7 based on the estimated value.
The path selection unit 7 selects, from among a plurality of correlation value outputs of the power delay profile, one that exceeds the threshold set in the threshold setting unit 23 as an effective path including the desired wave signal, and the phase Output to the compensation and Rake synthesis unit 3.
[0005]
FIG. 5 is a configuration diagram of a frame format of a transmission signal.
1 frame is N_FEach slot consists of a number N inserted at the beginning_PPilot symbols P followed by N_DData symbols.
Hereinafter, a conventional path detection method will be described in detail.
As a conventional first path detection method, the path detection apparatus 21 shown in FIG. 4 performs based on the pilot symbol power delay profile. By using a plurality of pilot symbols in one slot, in-phase addition is performed to improve path detection accuracy.
[0006]
FIG. 6 is a schematic explanatory diagram of a power delay profile for explaining the conventional first path detection method. One cycle (n chips) of the spread code is taken on the horizontal axis, and the reception level at the measurement timing (sample timing) for each chip cycle is taken on the vertical axis. The number of measurement timings in one spreading code period is a number equal to the spreading factor (the number of spreading code chips). When oversampling is performed, the number is further oversampling times.
(1) At each measurement timing, the received signals of Np pilot symbols are added in phase.
That is, the received I signals of Np pilot symbols are added (I = I₁+ I₂+ ... + I_Np) And squared (I²) And the received Q signals of Np pilot symbols are added (Q = Q₁+ Q₂+ ... + Q_Np) And squared (Q²) Plus (I²+ Q²) Is the received power level at each measurement timing described above.
(2) From the n measurement timings in the power delay profile, the remaining (n−), except the m received power levels with the largest received power level (m is the number of paths contributing to Rake reception, the number of fingers). m) The received power levels are averaged over time, and this is used as the average thermal noise power level Pave.
(3) Using a predetermined value ΔP, set Pave × ΔP as a threshold value. By making a path larger than this threshold value an effective path, the received signal of the path with only thermal noise power is not Rake combined.
[0007]
In mobile communication, the number of multipaths arriving at the receiver changes according to the communication status, so in the method of fixing the number of paths contributing to Rake reception to m, the average thermal noise power level Pave is accurate. It cannot be measured. Therefore, the effective path estimation accuracy deteriorates.
In particular, when there are more multipaths than m, since the path of the desired wave signal is included in the calculation of Pave, Pave becomes larger than actual. Since ΔP is generally set to a fixed value, as a result, the value of Pave × ΔP increases, and it may be determined that a path that should originally be valid is invalid.
[0008]
Although not shown, the following method is known as Japanese Patent Laid-Open No. 10-336072 as the second conventional technique for path detection. Path detection is performed based on the power delay profile of the pilot symbol.
The minimum received signal power Smin and the maximum received signal power Smax are detected from the delay profile. A threshold Δnoise [dB] is set for the minimum received signal power Smin, and a threshold ΔRake [dB] is set for the maximum received signal power Smax.
Of the L signal components, the received signal power S (l) (1 ≦ l ≦ L) is
S (l) ≧ max {Smin + Δnoise, Smax−ΔRake}
A signal component satisfying the condition is selected.
[0009]
By setting the threshold Δnoise, the received signal level at the measurement timing of only thermal noise (including interference components) is not Rake combined. If the multipath has an effective SINR (Signal to Interference Noise Ratio), Rake synthesis is performed so that even a temporarily reduced path can be used for Rake synthesis by setting the threshold ΔRake.
However, if the propagation environment changes, the reception level of the desired wave and the thermal noise level change greatly. However, since ΔRake and Δnoise described above are fixed values, if these are not properly selected, path detection is not properly performed.
[0010]
As a third conventional technique for path detection, the following method is known from Japanese Patent Laid-Open No. 2000-78110.
(1) The received signal level is averaged by performing cyclic integration on the power delay profile of the received signal including the pilot signal, which is directly obtained from the output of the matched filter 2.
(2) An average value and a standard deviation are obtained from the power delay profile obtained in (1), and a threshold value is obtained by adding the average value to a value obtained by applying a predetermined weight to the standard deviation. A path whose average received signal power is larger than this threshold is selected as an effective path.
[0011]
However, when the average value and the standard deviation are obtained from the power delay profile, since it is obtained from all received signals, a desired wave signal of an effective path is also included. Therefore, the average value and standard deviation of the reception level vary depending on the magnitude of the reception power of the desired wave signal. However, since the values originally desired are the average value and standard deviation of thermal noise (including interference components), it is difficult to estimate an appropriate threshold value.
[0012]
The path detection device 21 in the above description is for DS-CDMA.
Recently, a multicarrier direct sequence code division multiple access (MC / DS-CDMA) scheme has been studied for the fourth generation mobile communication system, for example, Shinsuke Hara, "Overview of Multicarrier CDMA" , IEEE Communication Magazine Dec.1997, p.126-133, etc.
This MC / DS-CDMA uses a plurality of subcarriers that are orthogonal to each other, distributes transmission data to each subcarrier channel and modulates each subcarrier, and each modulation signal is the same or different spreading code. The signal is spread in series, and the spread modulated signal is combined into a transmission signal. Also in this MC / DS-CDMA, a path detection device is required to perform Rake combining in order to effectively use multipath.
Each modulated signal may be spread with a different spreading code sequence.
[0013]
FIG. 7 is a block configuration diagram showing a schematic configuration of a transmitter and a receiver used in the MC / DS-CDMA communication system. FIG. 7A shows a block configuration, and FIG. 7B shows a transmission symbol sequence.
In order to simplify the explanation, the primary modulation is BPSK (Binary Phase Shift Keying), and the receiving side uses only the received I signal and does not consider the phase fluctuation.
On the transmission side in FIG. 7A, 31 is a serial-parallel converter. As shown in FIG. 7 (b), the transmission symbol sequence also includes pilot symbols, and the transmission symbols are distributed with the total number of subcarriers Nc as one unit, and the transmission symbol sequence of each subcarrier channel is multiplied by a multiplier 32.₀, 32₁, ...... 32_Nc-1Output to the spreading code sequence c₀(T), c₁(T), ..., c_Nc-1Multiply by (t) and code spread for each subcarrier channel.
Reference numeral 33 denotes an IFFT (Fast Inverse Fourier Transform) unit, which is a multiplier 34 corresponding to the subcarrier channel.₀, 34₁, ...... 34_Nc-1And the code-spread transmission symbol for each subcarrier channel with each orthogonal carrier cos (2πf₀t), cos (2πf₁t), ..., cos (2πf_Nc-1Multiply with t). Each multiplication output is synthesized by the synthesizer 35 to become a transmission signal s (t).
[0014]
On the other hand, on the receiving side, the received signal r (t) is input to an FFT (Fast Fourier Transform unit) 36 and a multiplier 37.₀, 37₁, ..., 37_Nc-1Each orthogonal carrier cos (2πf₀t), cos (2πf₁t), ..., cos (2πf_Nc-1Multiplying by t), the received signal of each subcarrier channel is output. This received signal is multiplied by a multiplier 38.₀, 38₁, ......, 38_Nc-1, Spread code replica c₀(T), c₁(T), ..., c_Nc-1Low-pass filter (LPF) 39 is multiplied by (t) and despread.₀, 39₁, ..., 39_Nc-1The unnecessary frequency component is removed through the determination unit 40.₀, 40₁, ..., 40_Nc-12 is binarized corresponding to the primary modulation BPSK.
Judgment device 40₀, 40₁, ..., 40_Nc-1In the parallel-to-serial (P / S) converter 5, the same received data bit string as that of the transmission data bit string is output.
[0015]
If attention is paid to the individual subcarrier channels of MC / DS-CDMA described above, it is the same as conventional DS-CDMA. Therefore, the conventional Rake combining and path detection device described above can be used as it is for each subcarrier channel.
However, as described above, the conventional path detection device has a problem. In addition, there has been a demand for a path detection apparatus that effectively uses MC / DS-CDMA multicarrier.
[0016]
[Problems to be solved by the invention]
The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a path detection device with high effective path detection accuracy.
It is another object of the present invention to provide a path detection apparatus suitable for MC / DS-CDMA with high effective path detection accuracy by effectively using multicarriers.
[0017]
According to the first aspect of the present invention, there is provided a path detection device for detecting an effective path including a desired wave signal according to a reception level measurement value of a power delay profile of a received direct spread signal, wherein the reception level A cumulative probability calculation means for calculating the cumulative probability of the measured value;Among the calculated cumulative probabilities, first to Nth having predetermined cumulative probabilities from the first to the Nth (N is an integer of 2 or more) in a region not including the predetermined desired signal. A predetermined reception level is detected, and a threshold value for determining a region including the desired wave signal and a region not including the desired wave signal is set based on the first to Nth predetermined reception levels.Threshold setting means and path detection means for detecting a path having a timing at which the reception level measurement value exceeds the threshold as an effective path including the desired wave signal.
  Therefore, since an effective path is detected using a threshold that is not affected by the path including the desired wave signal, even if the reception level of the path that is the desired wave signal fluctuates, the number of incoming multipaths also changes. However, since the effective path estimation accuracy does not deteriorate, path detection is performed appropriately.
  In addition, since the effective path is detected using a threshold value based on the cumulative probability distribution of thermal noise, even if the thermal noise level fluctuates, the effective path estimation accuracy does not deteriorate. Done.
  The received signal in the code division multiple access communication system is the direct spreading signal described above, but is not necessarily limited to the code division multiple access system, and may be a direct spreading received signal in a one-to-one communication system, for example.
  Further, the threshold setting means described above estimates an average thermal noise power level based on a reception level having a predetermined cumulative probability value in an area not including the desired wave signal, and sets a predetermined value to the estimated average thermal noise power level. By performing weighting such as multiplication, a threshold value for determining a region including the desired wave signal and a region not including the desired wave signal may be set.
[0021]
ThereforeSince the threshold is set based on a plurality of predetermined cumulative probabilities, the influence of uncertainties such as variations in calculated cumulative probabilities can be reduced, and an appropriate threshold can be set. Accuracy is improved.
[0022]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, a case where the present invention is applied to MC / DS-CDMA will be described with reference to the drawings. However, a configuration that does not necessarily use a multicarrier can also be applied to DS-CDMA.
FIG. 1 is a block diagram showing an embodiment in which the present invention is applied to the MC / DS-CDMA system. The configuration on the transmission side is the same as in FIG. Each subcarrier channel is transmitted in the frame format shown in FIG. That is, it is assumed that the same number of pilot symbols are transmitted at the same timing in each subcarrier channel.
[0023]
In the figure, the same parts as those in FIGS. 4 and 7 used for explaining the prior art are denoted by the same reference numerals, and description thereof is omitted. 2₀, 2₁, ..., 2_Nc-1Are matched filters provided for subcarrier channels, 3₀, 3₁, ..., 3_Nc-1Is a phase compensation and Rake combiner provided for the subcarrier channel, 4 is a path detector, 6 is a power meter, 8 is a threshold setter, 9₀, ..., 9_Nc-1Are in-phase adders provided for the subcarrier channels, 10₀, ......, 10_Nc-1Is a power calculator provided for the subcarrier channel, 11 is an averager, 12 is a data storage, 13 is a cumulative probability calculator, 14 is an average thermal noise power detector, and 15 is a weighting unit.
[0024]
The received I signal and the received Q signal obtained by the orthogonal demodulation are input to the FFT unit 1 and converted into a received I signal and a received Q signal for each subcarrier channel, and the phase compensation and Rake combiner 3₀, 3₁, ..., 3_Nc-1And each in-phase adder 9 in the power measuring device 6₀, ..., 9_Nc-1Is output.
Phase compensation and Rake combiner 3₀, 3₁, ..., 3_Nc-1In, phase compensation and Rake synthesis of the selected path are performed, data determination (demapping) corresponding to primary modulation is performed, and the result is output to the P / S converter 5. The P / S converter 5 performs parallel-serial conversion on the received data determined in each subcarrier channel, and outputs a received data sequence corresponding to the transmission data sequence.
[0025]
The first feature of this embodiment is that one power delay profile common to all subcarrier channels is obtained from the power delay profiles of a plurality of subcarrier channels. As a result, since the number of samples of the power delay profile can be increased by the number of subcarrier channels, the reliability of cumulative probability calculation described later is increased. Since the threshold value is determined from the cumulative probability distribution, the accuracy of the threshold value is improved, and as a result, the reception quality is improved.
Further, since a large number of samples of the power delay profile can be obtained, as shown in FIG. 5, when one frame is composed of a plurality of slots, for example, the power delay profile is set only for the pilot symbol of the first slot of one frame. Even if the calculation is performed, a sufficient number of samples can be obtained to calculate the cumulative probability, so that the power delay profile does not have to be calculated over the subsequent slots in one frame only in terms of the number of samples.
[0026]
In the power meter 6, for each subcarrier channel, as described in the prior art, the in-phase adder 9₀, 9₁, ..., 9_Nc-1In each slot, a plurality of consecutive pilot symbols are subjected to in-phase addition over a plurality of consecutive symbols at each measurement timing, and each power of I-phase component and Q-phase component output after in-phase addition is calculated for each power. Vessel 10₀......, 10_Nc-1The power delay profile is calculated by squaring the I-phase component, squaring the Q-phase component, and adding both.
[0027]
An averager 8 averages all subcarrier channels to obtain (spreading factor) × (oversampling number) sample values per slot as one power delay profile common to all subcarrier channels. Output. Furthermore, power averaging in the time direction may be performed using a power delay profile of pilot signals of a plurality of slots (for example, one frame).
In the above description, the power delay profiles of all the subcarrier channels are used. However, the cumulative probability of using only the power delay profiles of a plurality of subcarrier channels is higher than that using only one subcarrier channel. Increased value reliability.
[0028]
A second feature of this embodiment is that a cumulative probability is calculated from the calculated power delay profile, a reception level that has a certain cumulative probability is obtained, and an average thermal noise power level Pave is estimated based on this reception level. is there.
The power delay profile output from the power measuring device 6 is also output to the threshold setting device 8, and the sample value of the reception level at each measurement timing is stored in the data storage 12.
The cumulative probability calculator 13 stores a sample value of the reception level at each measurement timing over a predetermined measurement period, for example, one period of the pilot signal in one slot. Among them, the number of appearances is cumulatively counted every time the reception level value increases in order from the lowest reception level.
[0029]
The total number of samples to be stored is predetermined. For example, the total number of samples stored over one period of the pilot signal in one slot is spreading factor (spread code chip number) × oversampling number × multicarrier number.
The cumulative probability is obtained by dividing the cumulative count value described above by the total number of samples. Therefore, the cumulative count described above is substantially the same as the cumulative probability calculation. Hereinafter, the cumulative probability will be described, but even if the cumulative count value itself is evaluated, it is substantially the same as the evaluation of the cumulative probability.
[0030]
FIG. 2 is a first graph showing a cumulative probability distribution of received power levels.
In the figure, the horizontal axis represents the reception level, which is represented by, for example, a relative reception level [dB] in which the average thermal noise power level is 0 [dB]. The vertical axis represents the cumulative probability distribution [%, but logarithmic scale%].
The thermal noise distribution is theoretically known, for example, a Rayleigh distribution. If it is assumed that there is no sample including a signal of an effective path, a cumulative probability distribution as shown by a solid line in the region on the left side of the alternate long and short dash line and a broken line in the region on the right side in the figure is shown. In this case, the cumulative probability of giving the average thermal noise power level Pave, which is the average value level of the thermal noise probability distribution, is 63%. A table of cumulative probability distribution may be set based on simulation or actual measurement values.
[0031]
However, actually, a sample including the signal component of the effective path exists in the region on the higher reception level side. Therefore, when the reception level increases to some extent, the cumulative probability increases gradually (solid line in the region on the right side of the dashed line in the figure).
If the number of chips of the spreading code is about 1000 and the number of multipaths is about 4, the number of samples of only the thermal noise signal is sufficiently larger than the number of samples of the effective path including the desired wave signal.
Therefore, when the cumulative probability calculator 13 shown in FIG. 1 counts the cumulative probability value 63%, the average thermal noise power detector 14 detects the reception level at that time and regards it as the average thermal noise power level Pave. Since the average thermal noise power level Pave does not include a signal component of an effective path, the average thermal noise power level Pave can be accurately calculated.
Further, as will be described later, if a value obtained by weighting this value is used as a threshold value, it is possible to avoid Rake synthesis of samples with only thermal noise power, and Rake synthesis can be performed only on samples of effective paths.
[0032]
The weighting unit 15 multiplies the average thermal noise power level Pave by a preset weighting factor ΔP (multiplying in the case of a linear scale, 10 log in decibel conversion)._TenThe sum of (ΔP) is output to the path selector 7 as a threshold for path selection. This threshold value is for discriminating between a region including the desired wave signal and a region not including the desired wave signal.
Therefore, the path selector 7 selects, as an effective path, a path having a power that is larger than the value obtained by multiplying the average thermal noise power level Pave by ΔP.
That is, the matched filter 2 corresponding to each subcarrier channel only at the measurement timing when the reception level output from the power meter 6 exceeds the threshold within one spreading code period.₀, 2₁, ..., 2_Nc-1The received I signal and Q signal (pilot symbol and data symbol) output from the phase compensation and Rake combiner 3₀, 3₁, ..., 3_Nc-1Output to Rake.
Since the accurately calculated average thermal noise power level Pave is used, the estimation accuracy of the effective path is improved, and the reception quality by Rake combining is improved.
[0033]
The value of the predetermined value ΔP described above may be experimentally determined or theoretically determined using a standard deviation of the cumulative distribution.
Also, the reception level itself corresponding to a predetermined cumulative probability value can directly be a threshold value.
However, if the reception level serving as the threshold is set so that, for example, the cumulative probability is about 90%, there is a possibility that an effective path including the desired wave is included in this region. Accordingly, when the cumulative probability value corresponding to the reception level of such a threshold is directly detected, an inappropriate threshold that is affected by an effective path may be given.
Therefore, once the cumulative probability corresponding to the reception level that is likely to contain no signal of the desired wave is detected, weighting is performed by multiplying the reception level at that time by a predetermined value (linear calculation) or the like. Thus, an accurate threshold can be obtained by setting the threshold.
[0034]
In the above description, the average thermal noise power level Pave is obtained by cumulative probability calculation, and the threshold for path selection is finally determined by multiplying by a preset weight coefficient ΔP.
When the number of multipaths is sufficiently small, the average thermal noise power level Pave can be obtained by calculating the cumulative probability value 63%. However, a signal having a level corresponding to the cumulative probability value 63% does not always include the desired wave component. Therefore, a certain reception level other than the average thermal noise power level Pave (Δ more than Pave₁(dB lower level) is preset and the corresponding cumulative probability value (CDF)₁%) May be detected.
That is, the cumulative probability calculator 13 calculates the cumulative probability value (CDF₁%), The average thermal noise power detector 14 adds a predetermined value (Δ corresponding to the difference from the average thermal noise power level Pave to the measured value of the reception level at this time.₁The average thermal noise power level Pave is output by adding (dB).
The weighter 15 multiplies this by ΔP (linear calculation), thereby outputting the same threshold as the threshold described above.
[0035]
In the above description, the threshold is set after obtaining the average thermal noise power level Pave. However, a certain reception level having a predetermined cumulative probability value in an area not including the desired signal is detected, and this predetermined level is detected. A threshold value for discriminating between a region including the desired wave signal and a region not including the desired wave signal may be immediately set by weighting the “certain reception level”.
For example, the cumulative probability is a predetermined value (CDF₁%), “A certain reception level” (Δ more than Pave₁dB lower)₁By adding +10 log (ΔP) (logarithmic calculation), a threshold for immediately discriminating between a region including the desired wave signal and a region not including it is set.
[0036]
For example, the “certain reception level” is 2 dB lower than the average thermal noise power level Pave (cumulative probability is about 50%), 10 dB lower than Pave (cumulative probability is about 10%), Pave Or 20dB lower (cumulative probability is about 1%). However, it should be noted that even if the reception level is lowered too much, the reception level does not exist, or when calculating in order from the lowest level, the number of samples is insufficient and statistical reliability cannot be obtained. There is a need to.
In the cumulative probability distribution described above, when the number of subcarriers used for power calculation increases or decreases, the total number of samples only increases or decreases, so the distribution curve does not change. On the other hand, if the in-phase addition number Np of pilot symbols is increased or decreased, the reception level increases by 10 log (Np) [dB] due to the addition, and the distribution curve shape does not change but shifts to the right.
[0037]
FIG. 3 is a second graph showing a cumulative probability distribution of received power levels. With reference to this figure, another method for obtaining the average thermal noise power level Pave will be described.
Similar to FIG. 2, the horizontal axis represents the reception level [dB], and the vertical axis represents the cumulative probability [% of logarithmic scale].
The average thermal noise power detector 14 shown in FIG. 1 detects a plurality of cumulative probability values output by the cumulative probability calculator 13, and calculates the average thermal noise power level Pave from the value of “a certain reception level” corresponding to each of the cumulative probability values. Is estimated.
As shown in Fig. 3, the cumulative probability value is CDF.₁, CDF₂, ……, CDF_N(N is an integer greater than or equal to 2) is set in advance, and the reception level corresponding to each is Pave or ΔΔ₁, Δ₂, ……, Δ_NSuppose that it is a small value.
[0038]
In this way, a plurality of small points with a cumulative probability distribution of 63% or less or less than 63% are selected as much as possible, and the average thermal noise power level Pave is selected based on the reception level corresponding to each cumulative probability. Alternatively, if the threshold value is estimated, the cumulative probability that the estimation is uncertain with only one point can be set with higher accuracy than the average thermal noise power level Pave, and thus the threshold value. This is effective when the number of chips of the spread code is small.
Although various forms can be considered as a specific estimation method, only an example will be described here. The number of selected points is 4.
[0039]
Cumulative probability CDF₁Reception level is P₁Is the estimated average thermal noise power level Pave₁The value of is as follows.
Pave₁= P₁+ Δ₁
Other cumulative probability CDF₂~ CDF_FourSimilarly, is as follows.
Pave₂= P₂+ Δ₂
Pave_Three= P_Three+ Δ_Three
Pave_Four= P_Four+ Δ_Four
Pave mentioned above₁~ Pave_FourShould theoretically be equal, but do not necessarily match due to variations in measured values. Accordingly, these are regarded as temporary average thermal noise power levels, and the following average thermal noise power level Pave is calculated.
Pave₁~ Pave_FourIf an error is within a predetermined value, for example, within 1%, the two are considered to be reliable estimates, and the average of the two values is selected. Estimate the average thermal noise power level Pave.
Also Pave₁~ Pave_FourTwo average values in which any two of the errors are minimum may be estimated as the average thermal noise power level Pave.
[0040]
Even in this method of calculating the average thermal noise power level Pave, without obtaining the average thermal noise power level Pave, a plurality of “certain reception levels” having a plurality of predetermined cumulative probability values in an area not including the desired wave signal are obtained. A threshold value for immediately discriminating between a region including the desired wave signal and a region not including the desired wave signal may be set by detecting and weighting each of the plurality of predetermined “certain reception levels”.
For example, cumulative probability CDF₁~ CDF_FourMultiple reception levels P corresponding to₁~ P_FourBased on a plurality of provisional thresholds P₁+ (Δ₁+ 10log (ΔP)), P₂+ (Δ₂+ 10log (ΔP)), P_Three+ (Δ_Three+ 10log (ΔP)), P_Four+ (Δ_Four+ 10log (ΔP)) (both are logarithmic calculations), any number of them are selected one after another, and if the error is within a predetermined value, for example, 1%, the two are reliable Assuming that it is an estimated value, the average of the two is used as a threshold value for discriminating between a region including the desired wave signal and a region not including the desired wave signal.
In addition, two average values in which two arbitrary errors are minimum may be used as a threshold value for discriminating between a region including the desired wave signal and a region not including the desired wave signal.
[0041]
In the above description, a correlator for despreading the effective path data and a correlator for performing power measurement for path detection are realized by a common matched filter. However, a correlator for despreading effective path data and a correlator for performing power measurement for path detection may be provided separately. In addition, although a matched filter is used as a correlator for despreading, a sliding correlator may be used instead. However, the sliding correlator requires one cycle of the spreading code to output the correlation value of one path. In particular, when used as a correlator for path selection, it is necessary to detect a plurality of paths by time division processing or parallel operation of a plurality of sliding correlators.
[0042]
In the above description, assuming the MC / DS-CDMA system, the power delay profile of all subcarrier channels is regarded as one power delay profile, the cumulative probability distribution is obtained therefrom, the threshold is determined, and all subcarriers are determined. The path is selected by comparing with the average value of the power delay profile of the channel.
However, even if not all subcarrier channels, one or a plurality of power delay profiles are regarded as one power delay profile, a cumulative probability distribution is determined therefrom, a cumulative probability distribution is determined, and a threshold value is determined. The average value of the power delay profile and the threshold value of one or a plurality of subcarrier channels may be compared.
In this case, the detected effective path may be limited to at least one subcarrier channel described above, but since the multipath is the same for all subcarrier channels, all the detected effective paths are Common to the subcarrier channels.
[0043]
Also, in the DS-CDMA path detection device, only one channel power delay profile can be obtained, but by detecting the path with the cumulative probability described above, the received signal of the effective path including the desired wave signal can be obtained. By separating the received signal from only the thermal noise by the threshold value, it is possible to perform Rake synthesis with good quality.
If there is a sufficient number of samples and variation in the cumulative probability distribution is seen, the pilot symbol delay profile may be measured and stored over multiple slots or multiple frames.
[0044]
【The invention's effect】
As apparent from the above description, according to the present invention, it is possible to accurately set a threshold for selecting an effective path, and there is an effect that the detection accuracy of the effective path is improved.
In addition, according to the present invention, when applied to the MC / CDMA system, multicarrier can be used effectively, and effective path detection accuracy is improved.
As a result, the reception quality of Rake combining is improved.
[Brief description of the drawings]
FIG. 1 is a block diagram showing an embodiment in which the present invention is applied to an MC / DS-CDMA system.
FIG. 2 is a first graph showing a cumulative probability distribution of received power levels.
FIG. 3 is a second graph showing a cumulative probability distribution of received power levels.
FIG. 4 is a block diagram schematically showing a conventional path detection apparatus.
FIG. 5 is a configuration diagram of a frame format of a transmission signal.
FIG. 6 is a schematic explanatory diagram of a power delay profile for explaining a conventional first path detection method.
FIG. 7 is a block diagram showing an outline of an MC / DS-CDMA transmitter and receiver.
[Explanation of symbols]
1 ... FFT section, 2₀~ 2_Nc-1... Matched filter, 3₀~ 3_Nc-1... Phase compensation and Rake combiner, 4 ... Path detector, 5 ... P / S converter, 6 ... Power measuring device, 7 ... Path selector, 8 ... Threshold setter, 9₀~ 9_Nc-1... In-phase adder, 10₀-10_Nc-1... Power calculator, 11 ... Averager, 12 ... Data storage, 13 ... Cumulative probability calculator, 14 ... Average thermal noise power detector, 15 ... Weighter

Claims (1)

A path detection device that detects an effective path including a desired wave signal according to a reception level measurement value of a power delay profile of a received direct spread signal,
A cumulative probability calculating means for calculating a cumulative probability of the reception level measurement value;
Among the calculated cumulative probabilities, first to Nth having predetermined cumulative probabilities from the first to the Nth (N is an integer of 2 or more) in a region not including the predetermined desired signal. Threshold setting for detecting a predetermined reception level and setting a threshold for discriminating between an area including the desired wave signal and an area not including the desired wave signal based on the first to Nth predetermined reception levels Means,
Path detection means for detecting a path at a timing when the reception level measurement value exceeds the threshold as an effective path including the desired wave signal;
A path detection apparatus comprising:

Images