JP2003198651A - Maximum doppler frequency estimating device and adaptively modulated radio communication equipment - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a maximum Doppler frequency estimating device that has a simple constitution and can estimate the maximum Doppler frequency with high accuracy. <P>SOLUTION: A data extracting section 23 for pilot symbol extracts a pilot symbol P inserted into each slot in one frame. A normalized moving distance calculating section 24 calculates 'normalized moving distances' which are the 'moving distances' of two averaged pilot symbols P separated from each other by a distance on a phase plane, and normalized in accordance with the distances (received signal intensities) from an origin 0 after averaging close pilot symbols. A sliding average output section 25 outputs the accumulated average value of the 'normalized moving distances' and an estimated fd value output section 26 outputs the estimated value of the maximum Doppler frequency fd. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a maximum Doppler frequency estimator in mobile radio communication performed in a fading environment, and an adaptive modulation radio communication device for improving transmission quality using the maximum Doppler frequency estimator. Is.

[0002]

2. Description of the Related Art In mobile radio communication, various fading measures are taken because the amplitude and phase of a received signal fluctuate due to fading. As a technique for compensating for fading fluctuation, "Pilot symbol insertion method" (P
(SAM), Seiichi Sampei, “16QAM Fading Distortion Compensation Method for Land Mobile Communication”, IEICE Transactions B
-II, Vol.J72-B-II, No.1, pp.7-15, 1989-1 etc. FIG. 15 is a block configuration diagram of a conventional wireless communication apparatus using the pilot symbol insertion method. Figure 15
15A is a block diagram of the transmitter side, and FIG. 15B is a block diagram of the receiver side. Fading distortion compensation is performed by inserting a known pilot symbol into the information symbol section and transmitting it, and estimating the fading fluctuation at the information symbol position by interpolation from the received signal of the pilot symbol on the receiving side.

In FIG. 15A, 1 is transmission data,
A serial-parallel converter (S / P) 2 converts the transmission data into parallel data every 4 bits. A baseband signal generator (BSG) 3 converts 4-bit parallel data into a baseband signal corresponding to one symbol of 16QAM modulation. A frame signal generation unit 4 periodically inserts pilot symbols at equal intervals in the information symbol section. 16 pilot symbols
Among the four symbols having the maximum amplitude in the IQ phase plane of QAM, the symbols are appropriately switched and used. A low-pass filter (LPF) 5 band-limits the baseband signal. 6 is a quadrature modulator, and 7 is a local oscillator.
The reference frequency signal and the quadrature reference frequency signal output from the local oscillator 7 are converted into band-limited baseband signals.
Modulate with QAM. Reference numeral 8 is an amplifier, and 9 is a transmission antenna, which amplifies and transmits the modulated signal.

In FIG. 15 (b), 11 is a receiving antenna, and 12 is a bandpass filter (BPF), which limits the band of the received signal in order to normally operate the AGC 13 and AFC 14 described later. An automatic gain control unit (AGC) 13 keeps the received signal level constant. An automatic frequency controller (AFC) 14 roughly adjusts the frequency offset between the transmitter side and the receiver side. 15 is a quadrature demodulator, 1
6 is a local oscillator, and 17 is a low pass filter (LPF). The received signal whose frequency offset is roughly adjusted is multiplied by the reference frequency signal and the quadrature reference frequency signal output from the local oscillator 16 to perform 16QAM quasi-synchronous detection, and the LPF
By limiting the band at 17, a reception signal (baseband signal of two channels of I phase and Q phase) is output. The frequency of the reference frequency signal is orthogonally demodulated in a state where it does not completely match the frequency of the reference frequency signal on the transmission side in FIG. Reference numeral 18 denotes a fading distortion estimation / compensation unit, which uses a pilot symbol to perform fading distortion estimation and fading distortion compensation, which will be described later, and also finely adjust the offset frequency. A symbol determination unit 19 outputs 4-bit output data 22 per symbol by determining a received signal in which fading distortion has been compensated for at a symbol timing.

FIG. 16 is an explanatory diagram showing a frame structure in the pilot symbol insertion method. The example shown is TD
MA (Time Division Multiple Access) / TDD (TimeDivis
Ion Duplex: A frame structure of a digital mobile communication system of time division duplex system. Since the same frequency band is used for transmission and reception, the first half of one frame is assigned to the downlink for transmitting from the base station to the mobile station, and the second half is assigned to the uplink for transmitting from the mobile station to the base station. The illustrated example shows an example of two-way multiplexing in which a base station transmits and receives two mobile stations. In the reception slots R1 and R2, for every (N-1) symbol, a known 1-symbol pilot symbol P
Are inserted periodically. On the mobile station side, the fading fluctuation is estimated based on the pilot symbol P, and the fading fluctuation is estimated for the information symbols between the pilot symbols by using the interpolation method. Receive slot R1,
R2 is a slot for a different mobile station, but a mobile station is not only the pilot symbol P of the reception slot for the mobile station from the base station, but also the pilot symbol of the reception slot assigned to another mobile station from the base station. P can also be used.

FIG. 17 is an explanatory diagram of the quadrant arrangement of pilot signals. The arrangement of pilot symbols is shown in the IQ phase plane showing the received signal points for the reference frequency signal of the complex baseband signal. As the pilot symbol, of the multi-valued symbols used, the maximum amplitude, here 1
Of the 6QAM symbols, one of the four symbols A, B, C and D is used. It should be noted that the mobile terminal side can also insert a pilot signal in its own transmission slot to allow the base station side to perform fading distortion estimation and compensation. In this case, the base station sequentially receives the received signals from different mobile terminals in the reception slot, and therefore the fading distortion estimation and compensation corresponding to each mobile terminal is performed based on the individual pilot signal received from each mobile terminal. Will be done.

The above-mentioned “pilot symbol insertion method”
Is a measure against fading on the receiving side. On the other hand, an adaptive modulation method is known as a measure against fading on the transmitting side. Depending on the propagation path characteristics at the time of reception, for example, the estimated values of the propagation path characteristics such as C / No (carrier power to noise power density ratio) and delay spread, the reference side BER (B
It is possible to improve the transmission quality by selecting the modulation multi-value number (modulation method) that satisfies the itError Rate) and maximizes the transmission speed and adaptively controlling the modulation multi-value number at the time of the next transmission. For example, Takashi Suzuki et al. 2 “Propagation path characteristic estimation method in adaptive modulation method”, IEICE Technical Report RCS94-65 (1994-09) pp.37-42, Nobuya Otsuki et al. Transmission Characteristics of Variable Adaptive Modulation System ”, IEICE Transactions B-II, Vol.J78-B-II, No, 6, (19
June 1995), pp.435-444, Shuichi Sasaoka, "Mobile Communications"
p.117-119, etc.

FIG. 18 is a block diagram for explaining a conventional adaptive modulation wireless communication device. An example in which the mobile terminal 41 and the base station 42 communicate with each other through the fading line 43 is conceptually shown. TDMA / TDD described with reference to FIG.
The method will be described as a specific example. Further, in any of the modulation methods, the same fading distortion estimation / compensation of the received signal is adopted by using the pilot symbol insertion method similar to that shown in FIG. 18, the illustration is omitted. In the mobile terminal 41 on the left side of the drawing, transmission data is modulated by the modulator 51 and transmitted from the transmission antenna 52 to the base station 42. At this time, the modulation multi-level number control unit 101 selects one modulation system from modulation systems having different transmission bit rates (the same symbol rate but different multi-level modulation numbers). On the other hand, the transmission signal from the base station 42 is received by the receiving antenna 54,
The received signal is demodulated in the demodulator 55 to output the received data, and the modulated multilevel number estimation unit 56, the instantaneous C /
No and input to the delay spread estimation unit 57. For example, the output of the LPF 17 of the orthogonally demodulated signal in FIG. 15 may be branched to each of the modulation multi-level number estimation unit 56, the instantaneous C / No and delay spread estimation unit 57.

In the modulation multi-value number estimation unit 56, the demodulation method is designated for the demodulator 55. The modulation multi-level number estimation unit 56 multiplies and integrates, for example, the modulation parameter estimation word MC included in the midamble of the received signal with a predetermined code sequence corresponding to the modulation multi-level number. By comparing the correlation degrees with each other, the multilevel modulation number of each frame performed on the base station side is estimated, and the demodulator 55 is instructed of the demodulation method.

FIG. 19 is a block diagram showing the inside of the instantaneous C / No and delay spread estimators 57 and 67 shown in FIG. In the figure, reference numeral 91 is a delay profile estimation unit of the reception slot, which estimates the complex delay profile at the time of reception using, for example, the channel estimation word CE included in the midamble described above. A modified M sequence is used as the channel estimation word CE, and the complex delay profile is measured by correlating the received channel estimation word CE with the modified M sequence. Assuming a two-wave Rayleigh model considering only delays within one symbol, the complex delay profile is represented by the following impulse response.

[Equation 1] Here, δ (t) is the Dirac δ function.

Reference numeral 92 denotes a transmission slot delay profile estimation unit, which is extrapolated assuming that the loci of h ₀ (t) and h ₁ (t) change smoothly to obtain the delay profile at the time of transmission. To be The delay profile at the time of transmitting the kth TDMA frame is as follows.

[Equation 2] Here, h ′ _i (k) is the delay profile measured in the k-th frame. Extrapolation includes 0th-order, 1st-order, and 2nd-order extrapolation. In the case of zero-order extrapolation, Q ₀ = 1, Q ₁ = 0, Q ₂ =
In the case of 0 and linear extrapolation, Q ₀ = 3/2, Q ₁ = -1 / 2, Q ₂ = 0,
In the case of quadratic extrapolation, Q ₀ = 15/8, Q ₁ = -10 / 8, Q ₂ = 3/8.

Reference numeral 93 is a noise power estimation unit, which convolves the delay profile of the received signal and the channel estimation word CE to generate a replica of the received signal, and subtracts the replica from the received signal to determine the noise power. Output. The noise power is equal in the receive slot and the transmit slot, and
It is assumed that the other station also has the same noise power. Reference numeral 94 denotes a received power estimation unit, which uses the delay profile at the time of transmission to
Estimate the received power at the partner station. 95 is a C / No calculator, which divides the estimated received power at the partner station by the noise power
Outputs the value of C / No (carrier power to noise power density ratio).

The modulation multi-value number control unit 101 of FIG.
From the C / No and delay spread estimation unit 57, the delay spread and C / No are input, the delay spread and the instantaneous C / No at the time of transmission are extrapolated, and according to the modulation parameter selection chart, among the modulation methods with different transmission bit rates. One of the modulators 5 is selected at least every frame.
Specify 1. Regarding the base station 42 side, the modulator 61 to the instantaneous C / No and delay spread estimation section 67 and the modulation multi-level number control section 102 respectively receive the instantaneous C / No and delay from the modulator 51 on the mobile terminal 41 side. Spread estimation unit 5
Up to 7 are the same as the modulation multi-level number control unit 101,
The description is omitted.

In the TDMA / TDD system digital mobile communication system shown in FIG. 16, the channel characteristic in the next transmission slot is estimated from the channel characteristic in the receiving slot, and the modulation multi-value number at that time is selected. However, the time difference is half the time interval of one frame. When one frame length of TDMA is comparatively long like the 400MHz band digital mobile communication system for business use, the fading state changes in one frame, and the estimation result and the actual propagation at the partner station at the time of transmission The error with the road characteristics becomes large. As a result, it becomes difficult to estimate the optimum modulation multilevel number.

On the other hand, in the FDD (Frequency Division Duplex) system, different frequency bands are used for the uplink and the downlink, so that the propagation path characteristics are different. Therefore, the channel characteristics in the uplink (transmission slot of the mobile station, the reception slot of the base station) are estimated, and the channel information estimated in the downlink (transmission slot next to the base station) is notified to the mobile terminal. Based on the notification, the mobile terminal applies a modulation multi-level number suitable for the propagation path characteristics expected at the time of uplink (mobile station next transmission slot, base station next reception slot) and transmits. To do.
It should be noted that the modulation multi-level number is similarly controlled for the transmission on the base station side. Therefore, when the frame length is relatively long, the fading state changes in the next frame, and it becomes difficult to estimate the optimum modulation multilevel number.

Therefore, it is necessary to perform adaptive modulation in consideration of fading fluctuation. The Doppler frequency generated according to the traveling of the mobile station is involved in the fading fluctuation.
Therefore, we would like to improve the reception quality of the adaptive modulation method by estimating the maximum Doppler frequency fd related to the speed of fading fluctuation. Maximum Doppler frequency (maximum Doppler shift) fd, which is the maximum Doppler frequency
Is expressed by the following equation. fd = V / λ where V is the moving speed of the mobile station and λ is the wavelength of the radio wave used. As a method of estimating the moving speed V, it can be measured by a vehicle speedometer if it is an in-vehicle wireless communication device. Also GPS
Enter the location information from (Global Positioning System),
The data processing unit obtains the moving distance L in the fixed time T,
It can also be measured. However, these methods cannot perform estimation by the wireless communication device alone.

Maximum Doppler frequency for wireless communication device alone
As a method of estimating fd, a method based on zero-cross point counting is used by Mitsuharu Kondo and 3 others.
Diversity characteristics and fd estimation of QAM ", 2001 IEICE General Conference, B-5-167, 2001/3.
This zero-cross point counting method uses pilot symbols inserted for the "Pilot symbol insertion method" (PSAM) described above.

The number of zero cross points of the I component or the Q component is counted, and the ratio of zero cross points per pilot symbol is obtained. Next, using the calculated ratio of the zero-cross points, fd is estimated from the relational expression between the previously determined zero-cross point ratio and fd. However, if there is a small phase variation near the zero cross point, there is a problem that extra counting is performed. Further, the ratio of the zero-cross points also changes due to the delay spread in the frequency selective fading environment, so it is necessary to estimate the maximum Doppler frequency fd in consideration of the estimated value of the delay spread.

On the other hand, as another method for estimating the maximum Doppler frequency fd by the wireless communication device alone, there is a method using a normalized inner product value, for example, Hidehiro Ando and 3 others, “Doppler frequency detection using pilot symbols”, 2000. It is known for the Annual Conference of the Society of Shinkai, B-5-59, 2000, etc. Figure 20
FIG. 6 is an IQ phase plan view for explaining the principle of conventional Doppler frequency estimation using a normalized inner product value. In the figure, the horizontal axis shows the I-phase component in phase with the carrier wave, and the vertical axis shows the Q-phase component orthogonal to the phase of the carrier wave. W-CDMA (Wideband Code Division
In Multiple Access), the normalized inner product value is calculated using the I / Q code multiple pilot signals that are present in succession in the first part of each slot of the uplink. The channel estimation value of each slot is calculated from the average value of the pilot symbols of each slot, and the normalized inner product value of the channel estimation values of adjacent slots or slots separated by a plurality of slots is calculated.

In this figure, the channel estimation value of each slot is represented as averaged received signal points 111 and 112. The normalized inner product value is the cosine value cos θ of the phase change amount θ from the averaged reception signal point 111 to the next averaged reception signal point 112 when viewed from the origin O. The vector of the averaged received signal points 101 and its components are
Let (x ₁ , y ₂ ) be the vector of the averaged received signal points 102 and its components be b (x ₂ , y ₂ ), which is expressed by the following equation. Normalized inner product value cos θ = (x ₁ x ₂ + y ₁ y ₂ ) / {(x ₁ ² + y ₁ ² ) (x ₂ ² + y ₂ ² )} ^1/2 (3) The calculated normalized inner product value is Since it is affected by noise and interference, the effect is reduced by averaging over multiple slots. Next, using the calculated normalized inner product value, the maximum Doppler frequency fd is estimated with reference to the relational expression between the normalized inner product value obtained in advance and fd. The normalized inner product method described above is a technique relating to multiple I / Q code multiple pilot signals that continuously exist in the first part of each slot in W-CDMA. Application to modulated signals has never been considered. Then, how to apply it becomes a problem. Further, when the maximum Doppler frequency fd is small, there is a possibility that the reception signal point may exist near the origin O for a long time. Since the normalized inner product value of the received signal points near the origin is greatly affected by noise, there remains a problem of giving a large error to the maximum Doppler frequency fd.

[0021]

SUMMARY OF THE INVENTION The present invention has been made in order to solve the above-mentioned problems, and increases the maximum Doppler frequency related to the speed of fading fluctuation with a simple structure by a wireless communication device alone. It is an object of the present invention to provide a maximum Doppler frequency estimator that can be accurately estimated. Another object of the present invention is to provide an adaptive modulation wireless communication device that improves reception quality by taking into account fading fluctuations by using this maximum Doppler frequency estimator.

[0022]

According to a first aspect of the present invention, the maximum Doppler frequency estimating apparatus orthogonally demodulates a received signal obtained by periodically inserting and modulating pilot symbols. The obtained complex baseband signal is input, and the phase of the complex baseband signal with respect to a predetermined first pilot symbol and a second pilot symbol separated from the first pilot symbol by a predetermined number of pilot symbols On a plane,
The received signal points of the plurality of adjacent pilot symbols including the first pilot symbol are averaged to output a first averaged received signal point, and the plurality of adjacent pilot symbols including the second pilot symbol are output. An averaged received signal point output means for averaging the received signal points to output a second averaged received signal point, and a first average depending on the reception strength of the first pilot symbol and / or the second pilot symbol. Normalized moving distance output means for normalizing the moving distance from the normalized received signal point to the second averaged received signal point and outputting the normalized moving distance; and the normalized moving distance for the first and second pilots. A normalized moving distance average output means for outputting a normalized moving distance average by sequentially updating the symbols and taking a time average. Flip and is intended to estimate the maximum Doppler frequency. Therefore, the maximum Doppler frequency is
It can be detected by a simple method of obtaining the moving distance of the pilot signal. Further, since the moving distance is normalized, the influence of long-term attenuation of reception intensity due to the distance difference from the transmitting device to the receiving device is reduced, so that the maximum Doppler frequency can be estimated regardless of the distance difference. The normalized moving distance average also changes depending on the frequency offset foff between the transmitter side and the receiver side. If this frequency offset foff cannot be ignored, it is stored in advance. The frequency offset-corrected maximum Doppler frequency fd can be estimated by subtracting the frequency offset foff from the estimated value of the maximum Doppler frequency fd obtained from the normalized moving distance average.

According to a second aspect of the present invention, in the maximum Doppler frequency estimating apparatus according to the first aspect, the normalized moving distance average output means successively outputs the first and second pilot symbols one pilot symbol at a time. It is updated and time averaged. Therefore, the positions of the received signal points of all pilot symbols can be used as samples, and since the calculation period is sequentially shifted by one pilot symbol, an averaged moving average that is less affected by noise or the like is obtained. be able to.

According to the third aspect of the present invention, in the maximum Doppler frequency estimator, a complex baseband signal obtained by quadrature demodulating a received signal in which symbols are periodically inserted and modulated is input, Regarding a predetermined first pilot symbol and a second pilot symbol that is separated from the first pilot symbol by a predetermined number of pilot symbols, the first pilot symbol is included on the phase plane of the complex baseband signal. While averaging the received signal points of the plurality of adjacent pilot symbols and outputting a first averaged received signal point,
Averaged received signal point output means for averaging received signal points of a plurality of adjacent pilot symbols including the pilot symbol and outputting a second averaged received signal point;
The moving distance average that outputs the moving distance average by sequentially updating the first and second pilot symbols to obtain the moving distance from the averaged received signal point to the second averaged received signal point An output means, a reception intensity average output means for outputting a reception intensity average by taking a time average of the reception intensity of the pilot signal, and a normalized movement distance by dividing the movement distance average by the reception intensity average. A normalized moving distance average output means for outputting an average is provided, and the maximum Doppler frequency is estimated according to the normalized moving distance average. Therefore, the maximum Doppler frequency can be detected by a simple method of obtaining the moving distance of the pilot signal. Further, since the moving distance is normalized, the influence of long-term attenuation of reception intensity due to the distance difference from the transmitting device to the receiving device is reduced, so that the maximum Doppler frequency can be estimated regardless of the distance difference. Since normalization is performed after averaging the moving distances,
Even if there is a period in which the received signal strength temporarily drops due to fading fluctuations and there is a large influence of noise, etc., the temporary moving distance itself during this period is not normalized, so the effect on the moving distance after averaging is not affected. Does not grow. Therefore, the estimation accuracy of the maximum Doppler frequency is improved.

According to a fourth aspect of the present invention, in the maximum Doppler frequency estimating apparatus according to the third aspect, the moving distance average output means sequentially updates the first and second pilot symbols by one pilot symbol at a time. Then, the time average is taken. Therefore, the positions of the received signal points of all pilot symbols can be used as samples, and since the calculation period is sequentially shifted by one pilot symbol, an averaged moving average that is less affected by noise or the like is obtained. be able to.

According to a fifth aspect of the present invention, there is provided an adaptive modulation wireless communication device for adaptively modulating transmission data by controlling the modulation multi-valued number according to the estimated value of the propagation path characteristic. According to the maximum Doppler frequency estimator according to any one of the above items and the output of the maximum Doppler frequency estimator, when the estimated value of the maximum Doppler frequency is high, the one with a small modulation multilevel number is selected. It has a modulation multi-value number control means for controlling as described above. Therefore, the estimated value of the maximum Doppler frequency can prevent the deterioration of the transmission quality in an environment where the transmission quality is deteriorated due to the rapid fading fluctuation. If the frequency offset foff cannot be ignored and the maximum Doppler frequency estimation device does not perform frequency offset correction, the frequency offset corrected maximum Doppler frequency fd may be used for control.

According to a sixth aspect of the present invention, in the adaptive modulation wireless communication apparatus according to the fifth aspect, the estimated value of the maximum Doppler frequency is classified into a high class and a small class according to the output of the Doppler frequency estimation apparatus. , The modulation multi-value number control means selects the smallest modulation multi-value number when a class with a high estimated value of the maximum Doppler frequency is set, and estimates the maximum Doppler frequency. When a class with a small value is set, adaptive modulation is performed. Therefore, it is possible to easily prevent deterioration of the transmission quality with a simple configuration in an environment where the transmission quality is deteriorated due to fading fluctuation.

According to a seventh aspect of the present invention, in the adaptive modulation wireless communication device according to the fifth aspect, the estimated value of the maximum Doppler frequency is small and high in class according to the output of the maximum Doppler frequency estimation device. Having a class setting means for setting a class and an intermediate class, the modulation multi-value number control means selects the smallest modulation multi-value number when a class with a high estimated value of the maximum Doppler frequency is set, When a class with a small estimated value of the maximum Doppler frequency is set, first adaptive modulation is performed, and when the intermediate class is set, second adaptive modulation having higher transmission quality than the first adaptive modulation is performed. It is something to do. Therefore, in an environment where the transmission quality deteriorates due to fading fluctuation, it is possible to easily prevent the deterioration of the transmission quality with a simple configuration, and gradually control the modulation multi-value number according to the magnitude of the normalized moving distance average. Because
The modulation multi-valued number can be finely adapted to the fading fluctuation.

The invention according to claim 8 is an adaptive modulation wireless communication apparatus for adaptively modulating transmission data by controlling a modulation multi-valued number according to an estimated value of a channel characteristic. And a maximum Doppler frequency estimator according to any one of the above items, and a delay profile at the time of transmission is extrapolated by extrapolating delay profiles of past plural times at the time of reception, and the output of the maximum Doppler frequency estimator is obtained. Accordingly, when the estimated value of the maximum Doppler frequency is high, the delay profile at the time of transmission is estimated by estimating the delay profile at the time of transmission by averaging the delay profiles of the past plural times at the time of reception instead of extrapolation. According to a propagation path characteristic estimating means for estimating the propagation path characteristic at the time of transmission by the delay profile at the time of transmission, and an estimated value of the propagation path characteristic at the time of transmission. And it has a modulation level controller means for selecting the modulation level. Therefore, even when the estimation accuracy by extrapolation deteriorates due to fading fluctuation,
Since the delay profile at the time of transmission can be estimated well, it is possible to easily prevent the deterioration of the transmission quality with a simple configuration in an environment where the transmission quality is deteriorated due to fading fluctuation.

According to a ninth aspect of the present invention, there is provided an adaptive modulation wireless communication device for adaptively modulating transmission data by controlling the modulation multi-level number according to the estimated value of the channel characteristic. And a maximum Doppler frequency estimator according to any one of the above items, and a delay profile at the time of transmission is extrapolated by extrapolating delay profiles of past plural times at the time of reception, and the output of the maximum Doppler frequency estimator is obtained. Accordingly, when the estimated value of the maximum Doppler frequency is high, the extrapolation order is reduced to estimate the delay profile at the time of transmission, thereby estimating the delay profile at the time of transmission, and the estimated delay profile at the time of transmission is estimated. A channel characteristic estimating means for estimating the channel characteristic at the time of transmission based on a delay profile, and a modulation multi-level number according to the estimated value of the channel characteristic at the time of transmission And it has a modulation level control means for selecting. Therefore, even when the estimation accuracy due to extrapolation deteriorates due to fading fluctuation, it is possible to perform good estimation by reducing the extrapolation order, so that the transmission quality deteriorates in an environment where the transmission quality deteriorates due to fading fluctuation. Can be easily prevented with a simple configuration.

According to a tenth aspect of the present invention, in the adaptive modulation wireless communication device according to any one of the fifth to ninth aspects, the estimated value of the propagation path characteristic is a carrier power to noise power density ratio. And an estimate of delay spread. Therefore, the propagation path characteristics can be well reflected.

[0032]

FIG. 1 is a block diagram for explaining an embodiment of a maximum Doppler frequency estimating device. On the transmitter side, the pilot symbol P is inserted in the data section as in the conventional configuration described with reference to FIG. FIG. 1A is a block diagram of the receiver side. The fd estimation unit 20 is added to the conventional configuration described with reference to FIG. 15B, and its internal configuration is shown in FIG. 1B. The fd estimation unit 20 can be realized using a hardware logic circuit, a DSP (Digital Signal Processor), or a general-purpose MPU (Micro Processing Unit).
The frequency offset foff is roughly adjusted in the AFC 14, but is still included in the LPF 17 output, and also includes the maximum Doppler frequency fd which is generally smaller than this. In FIG. 1B, 23 is a pilot symbol data extraction unit, 24 is a normalized moving distance calculation unit, 25 is a sliding average output unit, and 26 is fd.
It is an estimated value output unit.

FIG. 2 is an explanatory diagram of the pilot symbol extraction method and sliding average in the embodiment shown in FIG. A frame configuration diagram similar to that shown in FIG. 16 is used. FIG. 3 is an IQ phase plan view for explaining the principle of Doppler frequency estimation based on the normalized moving distance in the embodiment shown in FIG.

In FIG. 1B, a pilot symbol data extraction unit 23 inputs the symbol data that has been orthogonally demodulated and passed through the LPF 17, and is inserted into the first position of each of a plurality of downlink slots in one frame. In other words, the pilot symbols P inserted at intervals (slots) that are integer multiples of the data symbol length are extracted at kp pilot symbol intervals. FIG. 2 shows a specific example in which kp = 3. In the TDMA / FDD system, the mobile station can receive all pilot symbols P _{1 to} P ₁₂ of the downlink period sent from the base station, and using these, the maximum Doppler frequency fd can be estimated. To do.

The normalized moving distance calculating unit 24 is the "moving distance" on the phase plane of the two extracted pilot symbols P, and is normalized according to the distance from the origin O (received signal strength). The value "normalized movement distance" is output. That is, in FIG. 3, the distance from the reception signal point 31 of the first pilot symbol to the reception signal point 32 of the second pilot symbol, which is separated by a predetermined symbol interval, is defined as “moving distance”. It is estimated that the maximum Doppler frequency fd increases as the moving distance increases. However, when the received signal strength of the pilot symbol becomes small due to the attenuation of the received strength due to the distance difference from the transmitter to the receiver, the “moving distance” also becomes small and the maximum Doppler frequency fd is underestimated. Therefore, normalization is performed according to the received signal strength of pilot symbols. By normalizing, the maximum Doppler frequency fd can be estimated without worrying about the received power level.

An intersection of a unit circle 33 and a straight line connecting the origin O and the reception signal point 31 of the first pilot symbol is 34. A line parallel to the "moving distance" line is drawn from the intersection 34, and the origin O and the reception signal point 3 of the second pilot symbol are drawn.
The intersection with the straight line connecting 2 and 35 is 35. In the illustrated example, the “normalized movement distance” is the distance between the intersection 34 and the intersection 35 described above. That is, the received signal point 31 of the first pilot symbol 31 is retained while the value of the phase change angle θ is stored.
Is the “moving distance” when the distance from the origin of (1) (received signal strength) is normalized to 1. When expressed by a mathematical formula, it is as follows. Normalized moving distance = {(x ₁ −x ₂ ) ² + (y ₁ −y ₂ ) ² } / (x ₁ ² + y ₁ ² ) (4)

The "normalized movement distance" is replaced with the "movement" when the amplitude of the reception signal point 32 of the second pilot symbol is normalized to 1 while the value of the movement angle θ is retained. The distance may be defined as the following equation. Normalized moving distance = {(x ₁ −x ₂ ) ² + (y ₁ −y ₂ ) ² } / (x ₂ ² + y ₂ ² ) (5) Alternatively, received signal points of the first and second pilot symbols In consideration of the amplitudes of 31 and 32, it may be defined as the following expression. Normalized movement distance = {(x ₁ −x ₂ ) ² + (y ₁ −y ₂ ) ² } / {(x ₁ ² + y ₁ ² ) (x ₂ ² + y ₂ ² )} ^1/2 (6) Above Since all of the above equations use the ratio of the square of the distance, they cannot be said to be strictly the normalized distance unless the square root is taken. However, the distance referred to in the present invention is the same as the distance itself or the square of the distance.

In the above description, the received signal point 3 of the existing second pilot symbol 3 which is separated from the received signal point 31 of the existing first pilot symbol by a predetermined symbol interval.
We are looking for a travel distance of up to 2. This may be done, but in order to remove the effects of noise, etc., the adjacent pilot symbols are averaged for the received signal points, and then the “normal” between two averaged pilot symbols (kp pilot symbol interval) is used. It is also possible to calculate the “changed moving distance”. That is, with respect to the first pilot symbol, the reception signal points of a plurality of adjacent pilot symbols including the first pilot symbol are averaged, and the averaged first reception signal point is obtained. Similarly, for the second pilot symbol, the averaged second received signal point is similarly obtained. The method of determining the “proximity pilot symbol” may be arbitrary. The averaged reception signal points of the first and second pilot symbols are assigned to the reception signal points 31 and 32 shown in FIG. 3 to obtain the “moving distance” and the “normalized moving distance”.

The normalized moving distance calculating unit 24 calculates the above-mentioned "normalized moving distance" between the averaged received signal points by using the first and second pilot symbols, one pilot symbol at a time. Repeat by shifting (sliding) in the direction of travel. An example will be described with reference to FIG. In this figure, pilot symbols P are numbered. With kp = 3, the received signal points are averaged by a
It is assumed that v = 5 adjacent pilot symbols.
Initially, the “normalized moving distance” is calculated for the pilot symbols P ₁ and P ₄ (or P ₃ and P ₇ may be considered), but P _{1 to} P ₅ are close to each other for averaging. As the pilot symbol, the received signal points are averaged to obtain the received signal point a. Similarly, the received signal points are averaged by using P _{4 to} P ₈ as adjacent pilot symbols, and the received signal points b
Ask for. The “normalized moving distance” is obtained for these received signal points a and b. Next, by sliding one pilot symbol at a time, the received signal point a is obtained by using P _{2 to} P ₆ as the adjacent pilot symbols, the received signal point b is obtained by using P _{5 to} P ₉ as the adjacent pilot symbols, and the received signal point a is obtained. , B “normalized movement distance” is obtained. When the last pilot symbol P ₁₂ of the downlink is reached, the slide calculation may be stopped, or the number of averaging numbers av may be reduced and the slide calculation may be continued for a while.

The value of kp may be a value that extends over adjacent frames. If kp = 24, the moving distance is obtained at intervals of one frame length. kp = 24, av = 5
Then, for the first frame, the received signal points are averaged using P _{1 to} P ₅ as adjacent pilot symbols,
The reception signal point a is obtained. Similarly, for the second frame, P _{1 to} P ₅ are used as the adjacent pilot symbols, and the received signal points are averaged to obtain the received signal point b. A normalized moving distance is obtained for the received signal points a and b. Then 1
By sliding pilot symbols by one, P _{2 to} P ₆ of the first frame are used as adjacent pilot symbols to obtain a reception signal point a, and P _{2 to} P ₆ of the _second frame are used as adjacent pilot symbols to obtain a reception signal point b. , This received signal point a,
The “normalized movement distance” is obtained for b.

The sliding average output unit 25 reduces the influence of noise by calculating and outputting the cumulative average value of the "normalized moving distance" thus obtained over a long observation section. fd estimated value output unit 26
Is an estimation of the maximum Doppler frequency fd based on a preset relationship between the “normalized moving distance” and the maximum Doppler frequency fd, using the “normalized moving distance” cumulatively averaged by the sliding average output unit 25. Calculate the value. The relationship changes depending on the preset pilot symbol time interval T _pilot . The preset relationship between the "normalized moving distance" and the maximum Doppler frequency fd is stored in advance in a table format, and the "normalized moving distance" is cumulatively averaged.
Is used to refer to the table and output the corresponding maximum Doppler frequency fd. The maximum Doppler frequency fd
If the relationship between the "normalized moving distance" and the maximum Doppler frequency fd is known on the device side that uses, the "normalized moving distance" is used instead of the maximum Doppler frequency fd itself.
Alternatively, a value corresponding to this may be output.

This Doppler frequency estimation method using the "normalized moving distance" is based on the distance that a pilot symbol that has changed due to fading moves on the IQ phase plane per kp × (pilot symbol time interval T _pilot ). To estimate the maximum Doppler frequency fd. The normalized moving distance thus calculated converges to a certain range by performing sliding average in the time axis direction. Since this convergence range varies depending on the magnitude of the maximum Doppler frequency fd, the maximum Doppler frequency fd can be estimated. The time interval of the first and second pilot symbols used, that is,
It is necessary to set kp × T _pilot to an appropriate value depending on the magnitude of the estimated maximum Doppler frequency fd. Usually T
_pilot is determined by the frame format, so k
Set an appropriate value for p.

In the above description, in the calculation of the moving distance, the moving distance is first normalized and then the time average is obtained. Instead of this, the moving distance is simply calculated, the time average of the moving distances is obtained without normalization processing, and finally, the reception strength of the pilot symbols in all the observation intervals (in the case of the moving distance squared) May be divided by the average value of the received power obtained by squaring the received intensity) to obtain the normalized moving distance. The average received signal strength of the pilot symbols reflects the effect of attenuation as the distance between the transmitter and the receiver (base station to mobile terminal) increases. In addition to being able to reduce the number, it is possible to avoid the influence of noise at this time, since the temporary drop itself of the reception intensity due to the fading fluctuation is not normalized. As a result, the estimation accuracy of the maximum Doppler frequency is improved.

FIGS. 4 and 5 show simulation results when the maximum Doppler frequency fd is estimated using downlink pilot symbols in the TDMA / TDD2 multiplexing frame structure shown in FIG. 2 is a graph of 2. The horizontal axis represents the number of repetitions of the cumulative average calculation by the number of frames, and the vertical axis represents the “normalized moving distance average”. In FIG. 4, the pilot symbol interval kp for obtaining the normalized moving distance average is set to kp = 3, and in FIG. 5, kp = 24. The normalized moving distance uses the definition shown in FIG. 3 and Expression (4). Other simulation conditions will be described. The modulation method is 16QAM, and white noise (AWGN) is added to the frequency selective fading as the propagation path. Eb / N _O = 10 dB, delay spread σ = Ts / 3 (Ts is the time slot length of 1 symbol)
That is a poor propagation environment. The averaging process of the adjacent pilot symbols is av = 5 symbols. One pilot symbol is inserted every 15 data symbols. Ts is 62.5 μsec, pilot symbol interval Tpi
lot is 1 msec. The frequency offset is foff = 0, and a simulation is performed to evaluate the maximum Doppler frequency fd.

In FIG. 4, the pilot symbol interval
Since kp = 3, it is suitable for relatively fast estimation of the maximum Doppler frequency fd. Up to fd = 80Hz, 40Hz, 20Hz, fd can be estimated by "normalized moving distance average". However, fd = 10, 4, 1 Hz cannot be discriminated even if the number of repetitions reaches 40 frames. The reason that there are multiple broken lines for each fd is that several random numbers seed used in the simulation are shown. This random number
The value that converges varies depending on the seed. Therefore, when creating a table that refers to the maximum Doppler frequency fd from the "normalized moving distance average", for example, a plurality of thresholds are set and compared with the measured "normalized moving distance average" value. By doing so, the value of the maximum Doppler frequency fd that converges to the value of “normalized moving distance average” corresponding to the median between the threshold values is obtained from FIG. 4 and corresponds to the input “normalized moving distance average”. This is the estimated output of the maximum Doppler frequency fd.

On the other hand, in FIG. 5, since the pilot symbol interval kp = 24, it is suitable for estimating a relatively slow Doppler frequency. fd = 1Hz, 4Hz, 10Hz
Up to, fd can be estimated by "normalized moving distance". However, fd = 20Hz and 40Hz cannot be identified even if the number of repetitions reaches 40 frames. Therefore, when estimating the relatively high maximum Doppler frequency fd, the normalized moving distance may be calculated using a relatively small value of kp, for example, kp = 3. On the other hand, when estimating the relatively slow maximum Doppler frequency fd, the normalized moving distance is calculated using a relatively large value of kp, for example, kp = 24 (corresponding to one frame interval in FIG. 2). Just calculate. Estimate the maximum Doppler frequency fd range from 1 to 20Hz
Then, only kp = 24 should be used.

On the other hand, when the maximum Doppler frequency fd cannot be predicted, different values of kp may be used to perform a plurality of calculations simultaneously, and the more accurate maximum Doppler frequency fd may be adopted. That is, the maximum Doppler frequency is estimated when the fd is small and when the fd is large, the Doppler frequency is large by parallel calculation (for example, about 20 to 80 Hz) and small (for example, 1 to 20 Hz).

4 and 5, the frequency offset foff = 0. However, the normalized moving distance average is fd
A value corresponding to + foff will be output. Therefore,
Maximum Doppler frequency fd obtained from the results of Figs.
A value obtained by subtracting the frequency offset estimated value obtained in advance from the value of is used as the estimated value of the maximum Doppler frequency fd. For example, the device frequency offset foff
Or store the value (frequency) estimated by the maximum Doppler frequency estimation device as the estimated value of the frequency offset foff when it is clear that the mobile station is stationary. Good. While moving, a value obtained by subtracting the estimated value of the frequency offset foff from the estimated value of the maximum Doppler frequency fd output from the maximum Doppler frequency estimator is set as the fd estimated correction value. By performing the above measures, the maximum Doppler frequency estimation device can be realized even when the frequency offset foff cannot be ignored.

FIG. 6 is a block diagram showing an embodiment of the adaptive modulation wireless communication apparatus. 18, those parts which are the same as those corresponding parts in FIG. 18 are designated by the same reference numerals, and a description thereof will be omitted. This embodiment is intended to improve transmission quality in an adaptive modulation communication apparatus in consideration of fading fluctuation. The maximum Doppler frequency fd is classified into multiple classes, the modulation multi-level number is controlled according to each class into which fd is classified, and a delay profile estimation method for transmission is performed based on the delay profile at reception. To change.

In the first and second specific examples, the modulation multi-level number control unit 53 inputs the delay spread and C / No from the instantaneous C / No and delay spread estimation unit 57, and the fd estimation unit 58.
The maximum Doppler frequency fd is input to select one modulation method from modulation methods having different transmission bit rates and specify it to the modulator 51. The fd estimation unit 58 corresponds to the fd estimation unit 20 in the detailed block configuration diagram shown in FIG. 1, and switches the control mode of the modulation multilevel number control unit 53 according to this output. The modulation multi-value number control units 53 and 63 can be realized by using a hardware logic circuit, a DSP (Digital Signal Processor), or a general-purpose MPU (Micro Processing Unit).

FIG. 7 is an explanatory diagram of a concrete example of the modulation multi-level number control units 53 and 63 shown in FIG. 6 for explaining the first concrete example of the adaptive modulation wireless communication apparatus. FIG. 7A is a block configuration diagram thereof, in which reference numeral 71 is a modulation multi-level number selection unit, and 72 is a comparison unit. 7 (b) and 7 (c),
It is the 1st, 2nd modulation parameter selection charts 73 and 74. Horizontal axis is standardized delay spread (σ / Ts)
And the vertical axis is the instantaneous C / No value. The modulation method of the area to which the intersection determined by both values belongs is used for transmission. When the standardized delay spread is small and C / No is large, the 64QAM modulation system is adopted, and when the standardized delay spread is large or C / No is small, the QAM modulation system is adopted. It uses 16QAM.

The estimated value of the maximum Doppler frequency fd estimated by the fd estimation unit 58 (fd estimation unit 20 in FIG. 1) is
It is input to the comparison unit 72, is classified into classes, and controls the number of modulation levels in accordance with the value of the maximum Doppler frequency fd.
When the value is equal to or less than the predetermined threshold value Th ₀ set in advance, the adaptive modulation method is adopted according to the first modulation parameter selection chart 73. When the estimated value of the maximum Doppler frequency fd is larger than the predetermined threshold value Th ₀ , transmission is performed according to the second modulation parameter selection chart 74, that is, using the QPSK modulation scheme with the smallest modulation multilevel number fixedly. As a result, when the fading fluctuation is fast, the modulation method with the smallest modulation multi-level number is adopted, and the deterioration of BER quality due to the selection error due to the estimation error of the modulation multi-level number can be suppressed.

It is not always necessary to input the value (Hz) of the maximum Doppler frequency fd, but any value corresponding to the maximum Doppler frequency fd (Hz) may be used. Comparison unit 7
If the relationship with the maximum Doppler frequency fd in 2 can be identified, the normalized moving distance average may be input as being substantially the maximum Doppler frequency fd. Also,
The modulation multi-level number selection unit 71 does not necessarily have to have the first and second modulation parameter selection charts 73 and 74.
A calculation formula or a logical formula may be used as long as it produces the same result. Maximum input Doppler frequency fd
If the value of is not the frequency offset foff correction, it can be corrected on the modulation multi-value number control unit 53 side. When the value of the maximum Doppler frequency fd itself is input, the estimated value of the frequency offset foff may be subtracted from this. However, when the normalized moving distance average or the like is input, it is once converted into the maximum Doppler frequency fd, the estimated value of the frequency offset foff is subtracted from this, and again converted into the normalized moving distance average. Alternatively, the above-mentioned threshold Th ₀ may be corrected according to the estimated value of the frequency offset foff. By performing the above measures, even if the frequency offset foff cannot be ignored,
An adaptive modulation wireless communication device with maximum Doppler frequency fd estimation can be realized. It is also possible to perform adaptive modulation control according to the frequency offset off, but in the embodiment of the present invention, it is intended to analyze the influence of the maximum Doppler frequency fd on the adaptive modulation and control the adaptive modulation. is there.

The same applies to the base station 42 side. The modulation multi-level number control unit 63 is the same as the modulation multi-level number control unit 53 on the side of the mobile terminal 41, and therefore description thereof will be omitted. fd estimation unit 68
Also, the same can be done as the fd estimation unit 68 (fd estimation unit 20). However, since the number of uplink slots received from the individual mobile terminal 41 is one slot in one frame, the number of pilot symbols P is small. Therefore, regarding the fd estimation value performed on the mobile terminal 41 side, the mobile terminal 41
6 receives a notification from the side using the uplink, and notifies this to the comparison unit 72 in the modulation multi-level number control unit 63 on the base station 42 side in FIG.
(FIG. 7) may be input. That is, the fd estimation unit 68 does not necessarily have to be provided on the modulation multi-level number control unit 63 side. Alternatively, regarding the received signal point data of the pilot symbol P received on the mobile terminal 41 side, the mobile terminal 4
The fd estimation value may be output on the base station 42 side upon receiving the notification from the first side. That is, the fd estimation unit 68 may exist across the mobile terminal 41 and the base station 42. Note that the normalized moving distance average data or fd estimated value notification method can be applied to both the mobile terminal and the base station in the case of FDD.

FIG. 8 is an explanatory diagram of a concrete example of the modulation multi-level number control units 53 and 63 shown in FIG. 6 for explaining the second concrete example of the adaptive modulation wireless communication apparatus. FIG. 8A is a block configuration diagram, in which 81 is a modulation multilevel number selection unit,
Reference numeral 82 is a comparison unit. 8 (b), 8 (c), 8
(D) is the first, third, and second modulation parameter selection charts 73, 83, 74. fd estimation unit 58 (fd in FIG. 1
The estimated value of the maximum Doppler frequency fd estimated by the estimation unit 20) is input to the comparison unit 82 and classified into three classes. When it is less than or equal to a predetermined threshold value Th ₁ set in advance, the adaptive modulation method is adopted according to the first modulation parameter selection chart 83.

When the estimated value of the maximum Doppler frequency fd exceeds the predetermined threshold value Th ₁ and is equal to or less than the predetermined threshold value Th ₂ , the adaptive modulation method according to the third modulation parameter selection chart 83 is adopted. This third modulation parameter selection chart 8
3 has a higher transmission quality than the first modulation parameter selection chart 73, and the number of modulation levels is 1
The stage (rank) has been lowered. That is, the 64QAM area and the 16QAM area in the first modulation parameter selection chart 73 are changed to the 16QAM and QPSK areas, respectively. However, the lowest QPSK region in the first modulation parameter selection chart 73 is left as it is because no more modulation multi-value number is provided. Alternatively, a BPSK region having a smaller modulation level than QPSK may be provided.
When the estimated value of the maximum Doppler frequency fd exceeds the predetermined threshold value Th ₂ , the QPSK modulation method with the smallest modulation multi-level number is used for transmission. The modulation multi-value number selection unit 81 does not necessarily have to include the first, second, and third modulation parameter selection charts 7.
It is not necessary to have 3,74,83. A calculation formula or a logical formula may be used as long as it produces the same result.
Since the base station 42 side is the same as the first specific example shown in FIG. 7, description thereof will be omitted.

Next, the third and fourth specific examples will be described. The fd estimation units 58 and 68 are modulation multi-level number control units 53 and 6
Instead of controlling 3 as shown by the dashed line in FIG.
It controls the instantaneous C / No and delay spread estimation units 57 and 67. These are hardware logic circuits, DSP (Digital
Signal Processor) or general-purpose MPU (Micro Proce
ssing Unit). FIG. 9 is an instantaneous C / No and delay spread estimation unit 57 shown in FIG. 6 for explaining the third and fourth specific examples of the adaptive modulation wireless communication device.
It is a block block diagram of 67. In the description of the prior art, the instantaneous C / No and delay spread estimation unit 57 described with reference to FIG. 19 is modified, and description of the same parts as in FIG. 19 will be omitted.

In the third specific example, the extrapolation method is controlled by the output of the fd estimation unit 58 in consideration of fading fluctuation.
That is, the estimated value of the maximum Doppler frequency fd is compared with the comparison unit 9
Type in 6 to classify. Maximum Doppler frequency
If the estimated value of fd is equal to or smaller than the predetermined threshold Th ₃ , the extrapolation of the second order is performed according to the equation (2) already described using the estimated value of the delay profile in the reception slot. Predetermined threshold Th ₃
If it exceeds, the delay profile of the transmission slot is estimated using the arithmetic mean of the estimated values of the delay profile in each of the past reception slots instead of extrapolation. This arithmetic average is as follows.

[Equation 3] As a fourth specific example, if the estimated value of the maximum Doppler frequency fd exceeds a predetermined threshold value Th ₃ , the extrapolation order is reduced in the equation (2) already described. For example, what was normally quadratic extrapolation is changed to zero-order extrapolation. In the third and fourth specific examples described above, when the fading fluctuation is large, the estimation accuracy is high because of the time delay from the channel estimation word (CE) reception timing in the reception slot to the transmission data transmission timing of the transmission slot. It focuses on the deterioration caused by extrapolation.

FIG. 10 is a first graph showing the adaptive modulation BER characteristic with Doppler frequency estimation by the normalized moving distance average. The horizontal axis is C / No and the vertical axis is the bit error rate (Bit Er
rorRate). FIG. 11 is a second graph showing the adaptive modulation BER characteristic with Doppler frequency estimation by the normalized moving distance average. The horizontal axis represents the maximum Doppler frequency fd, and the vertical axis represents the bit error rate. 12
FIG. 6 is a graph showing an adaptive modulation average bit rate characteristic with Doppler frequency estimation by a normalized moving distance average.
The horizontal axis represents the maximum Doppler frequency fd, and the vertical axis represents the average bit rate. These graphs show simulation results corresponding to the first and second specific examples of the wireless communication device that performs adaptive modulation in consideration of fading fluctuation by performing maximum Doppler frequency fd estimation by the normalized moving distance average.

A mode (mode) for switching between adaptive modulation and QPSK using one threshold value for the fd estimation value shown in FIG.
2) and adaptive modulation using the two threshold values shown in FIG.
Three types of simulations were performed: an intermediate mode (decrease the number of modulation levels by one step) → a mode (mode3) that switches like QPSK, and a conventional adaptive modulation that does not classify because fd estimation is not performed. The simulation conditions are
It is as follows. The TDMA / TDD2 multiplex system frame structure shown in FIG. 2 was adopted. Select the modulation method from 64QAM, 16QAM, and QPSK. The pilot insertion method (PSAM) was used for fading distortion compensation. The propagation path includes flat Rayleigh fading (delay spread σ = 0), white noise (AW
GN) was added to the environment. The extrapolation was performed by first-order extrapolation. The av = 5 symbols are used for the averaging process of the neighboring pilot symbols when the normalized moving average distance is calculated, and the moving distance is kp = 24.
(1 frame interval). As the normalized moving average distance, the definition shown in FIG. 3 is used. The frequency offset is foff = 0. Maximum Doppler frequency fd
The threshold for the normalized moving distance corresponding to is set to 0.3 in mode2 and 0.1 and 0.3 in mode3. In terms of maximum Doppler frequency conversion, the threshold 0.3 is about 9 Hz and the threshold 0.1 is about 5 Hz. The number of pilot symbols used to find one fd estimate was 984 pilots. At the transmission symbol rate Rs = 16 ksps (kilosymbol per second), the pilot interval is 1 msec, so that the measurement time of 984 pilots is about 1 second.

In FIG. 10, BE increases as C / No increases.
Although the characteristic that R decreases is shown, at the maximum Doppler frequency fd = 1Hz, 20Hz, mode2 and mode3 are the same B
Has ER characteristics. On the other hand, when the maximum Doppler frequency fd is 8 Hz, the BER is worse in the 2mode method than in the 3mode method. 8Hz in the 3mode system is in the intermediate mode region, and in the mode3 system, when switching from adaptive modulation to QPSK, it is not suddenly switched, but by providing an intermediate region first and gradually lowering the number of modulation levels. This is because the quality is secured. Note that in the case of conventional adaptive modulation, at maximum Doppler frequency fd = 20 Hz, BE
R is much worse than mode2 and mode3. BER becomes worse than mode2 even at maximum Doppler frequency fd = 8Hz. At the maximum Doppler frequency fd = 1 Hz, there is almost no difference in BER.

In FIG. 11, if the conventional adaptive modulation (without fd est.) Without fd estimation and no classification is performed,
The BER has a characteristic of rapidly deteriorating as the maximum Doppler frequency fd increases. On the other hand, by adding the maximum Doppler frequency fd estimation function, mode2 or mode3
By so doing, the BER quality can be kept constant by selecting the QPSK modulation method even when the maximum Doppler frequency fd becomes high. In the Doppler frequency estimation of mode2, the BER is temporarily deteriorated near the threshold of 0.3 (about 9 Hz in fd conversion), especially when C / No is high.
The Doppler frequency estimation of can maintain almost constant BER quality for any C / No, and the maximum Doppler frequency f
Less susceptible to d.

In FIG. 12, the modulation method is such that when the Doppler frequency exceeds 10 Hz regardless of the magnitude of C / No,
Fixed to QPSK. Also, conventional adaptive modulation, mode2, mod
The transition of the average bit rate until it is fixed to QPSK differs depending on e3. In the case of conventional adaptive modulation,
The reason that the average bit rate slightly increases as fd increases is that modulation schemes other than QPSK may be temporarily estimated. In FIG. 11 above, C / No is 60
When dB is low, between conventional adaptive modulation, mode2 and mode3
There was not much difference in BER quality. However, as shown in FIG. 12, there is a difference in the average bit rates, and the transmission efficiency is worse in mode3 than in conventional adaptive modulation, mode2. In addition, in FIG. 11, when C / No is 70 dB, the BER deteriorates with the increase of fd in the conventional adaptive modulation.
In 3 improvement is observed, especially mode3 in at BER 10 ^{- 3} is ensured. Furthermore, when C / No is as high as 80 dB, mode2
Also, it can be seen that unless mode ^{3 is used} , the BER quality of 10 ⁻³ cannot be maintained below 10 Hz of fd. From the above simulation results, the adaptive modulation with maximum Doppler frequency fd estimation performs normal adaptive modulation when the maximum Doppler frequency fd is low to perform efficient communication, and transmission quality when the maximum Doppler frequency fd is high. By selecting a modulation multi-level number that raises, the reduction of the transmission bit rate is minimized and the BER quality is adaptively acted. Therefore, it can be said that the first embodiment of the present invention can extend the applicable range of the conventional adaptive modulation to the environment where the fading fluctuation is large.

FIG. 13 is a graph showing the adaptive modulation BER characteristic with Doppler frequency estimation by the normalized moving distance average. The horizontal axis is C / No and the vertical axis is the bit error rate (BitError Rat
e). This graph shows a simulation result corresponding to the third specific example of the wireless communication device that performs adaptive modulation in consideration of fading fluctuation. The simulation conditions are as follows. The TDMA / TDD2 multiplex system frame structure shown in FIG. 2 was adopted. Modulation method is 64QAM, 16QAM, QP
Select from SK. The pilot insertion method (PSAM) was used for fading distortion compensation. The propagation path was an environment in which white noise (AWGN) was added to frequency selective fading (delay spread σ / Ts = 1/16). The averaging process of the adjacent pilot symbols when obtaining the normalized moving average distance is av
= 5 symbols and the moving distance is kp = 24 (one frame interval). As the normalized moving average distance, the definition shown in FIG. 3 is used. When extrapolating the delay profile at the time of reception and estimating the delay profile at the time of transmission, the case of performing normal second-order extrapolation and the case of using the arithmetic mean are compared. In the arithmetic mean, the weighting factor for extrapolation is 1/3 for each of 3 points. The adaptive modulation itself is the same for both. When the maximum Doppler frequency fd = 1 Hz and 8 Hz, the BER quality is better in the second-order extrapolation. However, when the maximum Doppler frequency fd = 20 Hz, it can be seen that the BER quality is better in the result of arithmetic averaging than in the result of second-order extrapolation.

FIG. 14 is a fifth graph showing the adaptive modulation BER characteristic with Doppler frequency estimation by the normalized moving distance average. The horizontal axis is C / No and the vertical axis is the bit error rate (BitErr
or Rate). This graph shows a simulation result corresponding to the fourth specific example of the wireless communication device that performs adaptive modulation in consideration of fading fluctuation. Since the simulation conditions are the same as those in the case of FIG. 13, the description is omitted. Comparison is made between cases where different extrapolation orders are used when extrapolating the delay profile at the time of reception to estimate the delay profile at the time of transmission. The adaptive modulation itself is the same for both. At fd = 1Hz and 8Hz, extrapolation order is 2
BER quality is higher in the order of 1st, 1st and 0th order, and fd = 20Hz, conversely,
The BER is good in the order of extrapolation order 0, 1 and 2. Therefore, with respect to the estimation result of the maximum Doppler frequency fd, the threshold of fd is set to 10 Hz, the second order is selected as the extrapolation order when the fd estimation value is 10 Hz or less, and the zero order is selected when the fd estimation value is 10 Hz or more. Can be improved.

The adaptive modulation wireless communication apparatus with the maximum Doppler frequency fd estimation described above estimates the maximum Doppler frequency fd by using the normalized moving distance average, but the maximum Doppler frequency fd is estimated by another method. Good. For example, the method of counting zero cross points as described in the related art can be used. It is also possible to perform the maximum Doppler frequency fd estimation by applying the method using the normalized inner product value as described in the related art to a received signal in which pilot symbols are periodically inserted and modulated. Specifically, in FIG. 3, for the received signal points 31 and 32 of the first and second pilot symbols averaged by the adjacent pilot symbols, the normalized inner product value may be calculated and the sliding average may be performed. .
The frequency offset correction may be performed by subtracting from the maximum Doppler frequency estimated by the normalized inner product value in the same manner as the normalized moving distance method described above.

[0067]

As is apparent from the above description, according to the present invention, the maximum Doppler frequency related to the speed of fading fluctuation can be estimated by the wireless communication device alone without using the vehicle speed or the like. . Furthermore, by estimating the maximum Doppler frequency and controlling the modulation parameter of adaptive modulation according to the fading fluctuation, there is an effect that the deterioration of the transmission quality can be prevented.

[Brief description of drawings]

FIG. 1 is a block configuration diagram for explaining an embodiment of a maximum Doppler frequency estimation device.

FIG. 2 is an explanatory diagram of a pilot symbol extraction method and a sliding average in the embodiment shown in FIG.

FIG. 3 is an IQ phase plan view for explaining the principle of Doppler frequency estimation based on the normalized moving distance in the embodiment shown in FIG.

FIG. 4 is a first graph showing a simulation result when estimating the maximum Doppler frequency fd using downlink pilot symbols in the TDMA / TDD2 multiplex system frame configuration shown in FIG. 2;

5 is a second graph showing a simulation result when the maximum Doppler frequency fd is estimated using downlink pilot symbols in the TDMA / TDD2 multiplexing frame structure shown in FIG.

FIG. 6 is a block configuration diagram showing an embodiment of an adaptive modulation wireless communication device.

FIG. 7 is an explanatory diagram of a specific example of the modulation multi-level number control units 53 and 63 shown in FIG. 6 for explaining the first specific example of the adaptive modulation wireless communication device.

8 is an explanatory diagram of a specific example of the modulation multi-level number control units 53 and 63 shown in FIG. 6 for explaining a second specific example of the adaptive modulation wireless communication device.

9 is a block configuration diagram of instantaneous C / No and delay spread estimation units 57 and 67 shown in FIG. 6 for explaining third and fourth specific examples of the adaptive modulation wireless communication device.

FIG. 10 is an adaptive modulation BER with Doppler frequency estimation based on a normalized moving distance average corresponding to the first and second specific examples.
It is the 1st graph which shows a characteristic.

FIG. 11 is an adaptive modulation BER with Doppler frequency estimation by a normalized moving distance average corresponding to the first and second specific examples.
It is a 2nd graph which shows a characteristic.

FIG. 12 is a graph showing an average bit rate characteristic with Doppler frequency estimation based on a normalized moving distance average, corresponding to the first and second specific examples.

FIG. 13 is a graph showing the adaptive modulation BER characteristic with Doppler frequency fd estimation by the normalized moving distance average corresponding to the third specific example.

FIG. 14 is a graph showing an adaptive modulation BER characteristic with Doppler frequency fd estimation by a normalized moving distance average, corresponding to the fourth specific example.

FIG. 15 is a block configuration diagram of a conventional wireless communication device using a pilot symbol insertion method.

FIG. 16 is an explanatory diagram showing a frame structure in the pilot symbol insertion method.

FIG. 17 is an explanatory diagram of a quadrant arrangement of pilot signals.

FIG. 18 is a block diagram for explaining a conventional adaptive modulation wireless communication device.

19 is a block configuration diagram showing the inside of the instantaneous C / No and delay spread estimation unit 57 shown in FIG.

FIG. 20 is an IQ phase plan view for explaining the principle of conventional Doppler frequency estimation using a normalized inner product value.

[Explanation of symbols]

20, 58, 68 ... Maximum Doppler frequency fd estimation unit, 2
3 ... Pilot symbol data extraction unit, 24 ... Normalized moving distance calculation unit, 25 ... Sliding average output unit, 2
6 ... fd estimated value output unit, 31 ... Received signal point of first pilot symbol, 32 ... Received signal point of second pilot symbol, 33 ... Unit circle, 34, 35 ... Intersection point, 53, 63
... modulation multi-value number control unit, 56, 66 ... modulation multi-value number estimation unit,
57, 67 ... Instantaneous C / No and delay spread estimation unit, 7
1, 81 ... Modulation multi-level number selection unit, 72, 82, 96 ... Comparison unit, 73, 74, 83 ... First to third modulation parameter selection charts, 91 ... Reception slot delay profile estimation unit, 92 ... Transmission Slot delay profile estimator,
93 ... Noise power estimating unit, 94 ... Received power estimating unit, 95 ...
C / No calculator

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Fumihiro Sunada             Sanri, 299 Yamazaki, Kamakura City, Kanagawa Prefecture             Suville Mitsubishi Space Software Co., Ltd.             Inside the company F-term (reference) 5K004 AA08 JE00 JG00                 5K067 AA02 AA13 EE02 EE10 EE71                       GG01 GG11 HH21

Claims (10)

[Claims] 1. A complex baseband signal obtained by quadrature demodulating a received signal in which pilot symbols are periodically inserted and modulated is input, and a predetermined first pilot symbol and the first pilot are input. With respect to a second pilot symbol that is apart from the symbol by a predetermined number of pilot symbols, on the phase plane of the complex baseband signal, the reception signal points of the plurality of adjacent pilot symbols including the first pilot symbol are averaged. And receiving a first averaged reception signal point, and averaging the reception signal points of a plurality of adjacent pilot symbols including the second pilot symbol to output a second averaged reception signal point. Depending on the signal point output means and the reception strength of the first pilot symbol and / or the second pilot symbol, Normalized moving distance output means for normalizing the moving distance from the first averaged received signal point to the second averaged received signal point and outputting the normalized moving distance; Normalizing moving distance average output means for outputting a normalized moving distance average by sequentially updating the second pilot symbol and taking a time average, and determining a maximum Doppler frequency according to the normalized moving distance average. Maximum Doppler frequency estimator characterized by estimating. 2. The maximum Doppler according to claim 1, wherein the normalized moving distance average output means sequentially updates the first and second pilot symbols one pilot symbol at a time and takes a time average. Frequency estimator. 3. A complex baseband signal obtained by quadrature demodulating a received signal in which pilot symbols are periodically inserted and modulated is input, and a predetermined first pilot symbol and the first pilot are input. With respect to a second pilot symbol that is apart from the symbol by a predetermined number of pilot symbols, on the phase plane of the complex baseband signal, the reception signal points of the plurality of adjacent pilot symbols including the first pilot symbol are averaged. And receiving a first averaged reception signal point, and averaging the reception signal points of a plurality of adjacent pilot symbols including the second pilot symbol to output a second averaged reception signal point. The signal point output means and the moving distance from the first averaged reception signal point to the second averaged reception signal point are calculated as the first and second A moving distance average output means for outputting a moving distance average by sequentially updating the ilot symbol and taking a time average, and a receiving strength average for outputting a receiving strength average by taking a time average of the receiving strength of the pilot signal. Output means, and a normalized moving distance average output means for outputting a normalized moving distance average by dividing the moving distance average by the reception intensity average, and the maximum Doppler according to the normalized moving distance average. A maximum Doppler frequency estimator characterized by estimating a frequency. 4. The maximum Doppler frequency according to claim 3, wherein the moving distance average output means sequentially updates the first and second pilot symbols one pilot symbol at a time and takes a time average. Estimator. 5. An adaptive modulation wireless communication device for adaptively modulating transmission data by controlling a modulation multi-value number according to an estimated value of a channel characteristic, wherein the adaptive modulation wireless communication device according to any one of claims 1 to 4. Of maximum Doppler frequency estimation device, according to the output of the maximum Doppler frequency estimation device, when the estimated value of the maximum Doppler frequency is high An adaptive modulation wireless communication device comprising: a control unit. 6. A class setting means for setting a class having a high estimated value of the maximum Doppler frequency and a class having a small estimated value of the maximum Doppler frequency according to the output of the maximum Doppler frequency estimating device, wherein the modulation multi-level number control means comprises: When the class with a high estimated value of the maximum Doppler frequency is set, the smallest modulation multi-level number is selected, and when the class with a small estimated value of the maximum Doppler frequency is set, adaptive modulation is performed. The adaptive modulation wireless communication device according to claim 5. 7. The modulation multi-level number has class setting means for setting a class having a high estimated value of the maximum Doppler frequency, a class having a small estimated value of the maximum Doppler frequency, and an intermediate class according to an output of the maximum Doppler frequency estimating device. The control means selects the smallest modulation multilevel number when a class with a high estimated value of the maximum Doppler frequency is set, and performs the first adaptive modulation when a class with a small estimated value of the maximum Doppler frequency is set. The adaptive modulation wireless communication apparatus according to claim 5, wherein when the intermediate class is set, the second adaptive modulation having a higher transmission quality than the first adaptive modulation is performed. 8. An adaptive modulation wireless communication device for adaptively modulating transmission data by controlling a modulation multi-valued number according to an estimated value of a propagation path characteristic, wherein the adaptive modulation wireless communication device according to any one of claims 1 to 4. The maximum Doppler frequency estimator of and the extrapolation of the past several delay profiles at the time of reception to extrapolate the delay profile at the time of transmission,
According to the output of the maximum Doppler frequency estimator, when the estimated value of the maximum Doppler frequency is high, by arithmetic mean of the delay profile of the past multiple times at the time of reception instead of extrapolation, by the time of the transmission A propagation path characteristic estimating unit that estimates a delay profile and estimates the propagation path characteristic at the time of the transmission based on the estimated delay profile at the time of transmission, and a modulation multi-value according to an estimated value of the propagation path characteristic at the time of transmission. An adaptive modulation wireless communication device, comprising: a modulation multi-level number control means for selecting a number. 9. An adaptive modulation wireless communication device for adaptively modulating transmission data by controlling a modulation multi-valued number according to an estimated value of a channel characteristic, wherein the adaptive modulation wireless communication device according to any one of claims 1 to 4. The maximum Doppler frequency estimator of and the extrapolation of the past several delay profiles at the time of reception to extrapolate the delay profile at the time of transmission,
According to the output of the maximum Doppler frequency estimator, when the estimated value of the maximum Doppler frequency is high, the extrapolation order is reduced to estimate the delay profile during transmission, thereby estimating the delay profile during transmission. Then, a propagation path characteristic estimating means for estimating the propagation path characteristic at the time of transmission by the estimated delay profile at the time of transmission, and a modulation for selecting a modulation multi-value number according to the estimated value of the propagation path characteristic at the time of transmission An adaptive modulation wireless communication device, comprising: multi-value number control means. 10. The estimation value of the propagation path characteristic is an estimation value of a carrier power to noise power density ratio and an estimation value of a delay spread, according to any one of claims 5 to 9. An adaptive modulation wireless communication device as described.