Wireless communication system, wireless communication device, wireless communication method, and wireless communication program

Abstract

To provide a wireless communication system which is suitable for realizing high-speed transmission beyond the limit of the symbol interval based on the Nyquist reference, and can efficiently achieve high communication quality and suppression of power consumption by appropriately switching between equalization assuming whitening of noise and equalization considering coloring of noise to be applied to FTN transmission.SOLUTION: A transmitting station determines a compression ratio τ on the basis of the required transmission capacity (step 100) and transmits wireless signals through FTN transmission at the compression ratio τ. The merits and demerits of white equalization processing assuming a white property of noise and color equalization processing considering a coloring property of noise under the compression ratio τ are determined (step 102). A method according to the determination result is selected (steps 104, 106) and the result is notified to a receiving station. The receiving station applies equalization processing to the received signals using the notified method.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present disclosure relates to a wireless communication system, a wireless communication device, a wireless communication method, and a wireless communication program, and in particular to a wireless communication system, a wireless communication device, a wireless communication method, and a wireless communication program that are suitable for achieving high-speed transmission that exceeds the symbol interval limit according to the Nyquist criterion. [Background technology]

[0002] Non-Patent Document 1 below discloses a technique related to FTN (Faster-Than-Nyquist) that enables high-speed transmission that exceeds the symbol interval limit according to the Nyquist criterion. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] T, Ishihara, S. Sugiura, and L. Hanzo, “The evolution of faster-than-Nyquist signaling,” IEEE Access, vol. 9, pp. 86535-86564, June 2021.

[0004] Figure 1 shows how symbols are transmitted using Nyquist transmission. In Nyquist transmission, data is transmitted using symbols that are orthogonal to each other in the time domain. As shown in the figure, in this case, the symbol interval T is 1 / 2 the period of the transmission wave.

[0005] Figure 2 shows how symbols are transmitted by FTN transmission. Here, the compression rate τ is 0.82. The symbol interval is the symbol interval T in Nyquist transmission multiplied by the compression rate τ. In FTN transmission, the relationship between symbols in the time domain is non-orthogonal, but the number of symbols that can be transmitted per unit time can be increased compared to Nyquist transmission. For this reason, FTN transmission can achieve a higher transmission rate than Nyquist transmission. Summary of the Invention [Problem to be solved by the invention]

[0006] However, signals transmitted and received via wireless communication may be affected by the characteristics of the transmission path. In such cases, the receiving station performs equalization processing to eliminate such effects and properly restore the signal.

[0007] When wireless communication is performed using Nyquist transmission, the received signal is superimposed with so-called white noise, that is, noise that has the same intensity across the entire power spectrum. In this case, highly accurate equalization can be achieved with relatively simple processing.

[0008] On the other hand, when wireless communication is performed using FTN transmission, the noise power spectrum density becomes non-flat, i.e., the noise becomes colored. In this case, more complex calculations are required to achieve advanced equalization compared to when the noise is white.

[0009] Due to the above circumstances, when wireless communication is performed using FTN transmission, if equalization processing is performed assuming noise whitening, the calculation load can be reduced, but the accuracy of equalization may deteriorate. Reducing the calculation load leads to a reduction in the power consumption required for calculation. On the other hand, deterioration of equalization accuracy leads to an increase in the required CNR (Carrier-to-Noise Ratio) to obtain a desired communication quality. For this reason, if equalization processing assuming noise whitening is performed in FTN transmission, the baseband power consumption can be reduced, but a large transmission power is required to obtain a large required CNR.

[0010] On the other hand, if wireless communication using FTN transmission is combined with equalization that takes into account noise coloration, the calculation load increases, but the accuracy of the equalization can be improved. Therefore, in this case, while the baseband power consumption increases, the required CNR is small, so the transmission power can be kept small.

[0011] As described above, when wireless communication is performed using FTN transmission, a trade-off occurs in terms of power consumption between the case where equalization processing is performed assuming noise whitening and the case where equalization processing is performed considering noise coloring. Therefore, regardless of which equalization processing is incorporated into wireless communication, it is difficult to always obtain the highest communication quality with the lowest power consumption under various conditions.

[0012] In order to solve the above-mentioned problems, the first objective of this disclosure is to provide a wireless communication system that efficiently achieves both high communication quality and reduced power consumption by appropriately switching between equalization that assumes noise whitening and equalization that takes into account noise coloration and applying this to FTN transmission.

[0013] In addition, a second objective of this disclosure is to provide a wireless communication device that efficiently achieves both high communication quality and reduced power consumption by appropriately switching between equalization that assumes noise whitening and equalization that takes into account noise coloration and applying this to FTN transmission.

[0014] In addition, a third objective of this disclosure is to provide a wireless communication method that efficiently achieves both high communication quality and reduced power consumption by appropriately switching between equalization that assumes noise whitening and equalization that takes into account noise coloration and applying this to FTN transmission.

[0015] In addition, a fourth objective of this disclosure is to provide a wireless communication program that efficiently achieves both high communication quality and reduced power consumption by appropriately switching between equalization that assumes noise whitening and equalization that takes into account noise coloration and applying this to FTN transmission. [Means for solving the problem]

[0016] In order to achieve the above object, a first aspect is a wireless communication system including a transmitting station and a receiving station, The transmitting station, A process of determining a compression ratio τ to be multiplied by a reference symbol interval T according to the Nyquist criterion based on a required transmission capacity of the transmission data; a process of transmitting a radio signal to the receiving station by FTN transmission at an FTN symbol interval τT obtained by multiplying the reference symbol interval T by the compression rate τ; a process of determining which of a whitening equalization process assuming the whiteness of noise superimposed on the wireless signal is superior to a colored equalization process taking into account the coloredness of the noise in terms of power saving at the compression ratio τ; and selecting a method having superior power saving capability from the white equalization processing and the chromatic equalization processing and notifying the receiving station of the selection. The receiving station, receiving the wireless signal; It is preferable that the wireless communication device is configured to execute a process of performing equalization processing on the received wireless signal by a method indicated in the notification, out of the white equalization processing and the chromatic equalization processing.

[0017] A second aspect is a wireless communication device having a function of transmitting a wireless signal to a receiving station, A process of determining a compression ratio τ to be multiplied by a reference symbol interval T according to the Nyquist criterion based on a required transmission capacity of the transmission data; a process of transmitting a radio signal to the receiving station by FTN transmission at an FTN symbol interval τT obtained by multiplying the reference symbol interval T by the compression rate τ; A process of determining which is superior in terms of power saving at the compression ratio τ between a whitening equalization process assuming that the noise superimposed on the wireless signal is white and a colored equalization process taking into account the colored nature of the noise; It is desirable that the equalization unit is configured to select one of the white equalization processing and the chromatic equalization processing, which is more power-efficient, and notify the receiving station of this selection.

[0018] A third aspect is a wireless communication method for performing wireless communication between a transmitting station and a receiving station, comprising: The transmitting station determines a compression ratio τ to be multiplied by a reference symbol interval T according to the Nyquist criterion based on a required transmission capacity of the transmission data; The transmitting station transmits a radio signal to the receiving station by FTN transmission at an FTN symbol interval τ T obtained by multiplying the reference symbol interval T by the compression rate τ; The transmitting station determines which of a whitening equalization process assuming the whiteness of noise superimposed on the wireless signal is superior to a colored equalization process taking into account the coloredness of the noise in terms of power saving under the compression ratio τ; the transmitting station selects one of the white equalization processing and the chromatic equalization processing, which is more power-saving, and notifies the receiving station of the selection; the receiving station receiving the radio signal; the receiving station performs equalization processing on the received wireless signal using a method indicated by the notification, out of the white equalization processing and the chromatic equalization processing; It is preferable that the present invention includes the following:

[0019] A fourth aspect is a wireless communication program for causing a wireless communication device to transmit a wireless signal to a receiving station, The wireless communication device includes: A process of determining a compression ratio τ to be multiplied by a reference symbol interval T according to the Nyquist criterion based on a required transmission capacity of the transmission data; a process of transmitting a radio signal to the receiving station by FTN transmission at an FTN symbol interval τT obtained by multiplying the reference symbol interval T by the compression rate τ; A process of determining which is superior in terms of power saving at the compression ratio τ between a whitening equalization process assuming that the noise superimposed on the wireless signal is white and a colored equalization process taking into account the colored nature of the noise; a process of selecting a method having superior power saving performance from the white equalization process and the chromatic equalization process and notifying the receiving station of the selection; It is preferable that the computer readable program be capable of executing the above-mentioned method. Effect of the Invention

[0020] According to the first to fourth aspects, by appropriately switching between equalization that assumes noise whitening and equalization that takes into account noise coloration and applying this to FTN transmission, it is possible to efficiently achieve both high communication quality and reduced power consumption. [Brief description of the drawings]

[0021] [Figure 1] FIG. 1 is a diagram showing a state in which symbols are transmitted by Nyquist transmission. [Diagram 2] FIG. 1 is a diagram showing how symbols are transmitted by FTN transmission. [Diagram 3] FIG. 2 is a block diagram for explaining basic functions of a transmitting station according to the first embodiment of the present disclosure. [Figure 4] FIG. 2 is a block diagram for explaining basic functions of a receiving station according to the first embodiment of the present disclosure. [Diagram 5] 1 is a flowchart for explaining a flow of characteristic processes executed in the first embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0022] Embodiment 1 [Configuration of the first embodiment] Fig. 3 is a block diagram for explaining basic functions of the transmitting station 10 included in the wireless communication system according to the first embodiment of the present disclosure. Specifically, the configuration shown in Fig. 3 is configured with dedicated hardware and a general computer system. The computer system includes an input / output interface, a central processing unit (CPU), and a memory device including a ROM, a RAM, a hard disk, and the like. The memory device stores a program executed by the CPU. The functions of the transmitting station 10 are realized by the CPU executing the program.

[0023] The transmitting station 10 supports spatial multiplexing transmission using MIMO (Multiple-Input and Multiple-Output), and includes a plurality of antennas 18. Fig. 3 illustrates a configuration in which the number of antennas 18 is Nt.

[0024] The transmitting station 10 includes a modulator 12 for each antenna 18. The modulator 12 is a block that modulates a transmission signal in a desired manner. The modulator 12 can change the modulation method and the symbol interval as necessary. More specifically, the modulator 12 can appropriately switch and set parameters related to modulation, including a modulation coding rate and a compression rate τ used in FTN transmission, as necessary.

[0025] The output of the modulator 12 is provided to a guard interval inserter 14. The guard interval inserter 14 sets a guard interval taking into account inter-symbol interference. The output of the guard interval inserter 14 is provided to a transmission filter 16. The transmission filter 16 performs precoding on the signal to be transmitted using various filter processes. The resulting transmission signals s1(t) to sNt(t) are transmitted from antennas 18 #1 to #Nt, respectively.

[0026] Fig. 4 is a block diagram for explaining basic functions of the receiving station 20 included in the first embodiment of the present disclosure. The configuration shown in Fig. 4 is configured with dedicated hardware and a computer system, similar to the case of the transmitting station 10. The functions of the receiving station 20 are realized by the CPU of the receiving station 20 executing a program stored in a memory device of the receiving station 20.

[0027] The receiving station 20 supports spatial multiplexing transmission by MIMO, like the transmitting station 10. Fig. 4 illustrates a configuration in which the receiving station 20 includes Nr antennas 22, numbered #1 to #Nr. The receiving station 20 includes a receiving filter 24 for each antenna 22. The receiving filter 24 provides signals y1[n] to yNr[n] obtained by filtering a signal acquired at each sampling time t to a guard interval remover 26.

[0028] As shown in Fig. 4, the sampling time t is expressed by multiplying the symbol interval T according to the Nyquist criterion by an integer n and a compression rate τ used in communication. Also, y1[n] to yNr[n] shown in Fig. 4 represent signals generated by the #i (i = 1 to Nr)-th receiving filter 24 based on the signal at the sampling time t corresponding to the integer n.

[0029] The guard interval remover 26 removes the guard intervals from the received signals y1[n] to yNr[n] and provides the signals to a serial-to-parallel converter 28. The serial-to-parallel converter 28 provides a number of parallel signals corresponding to a preset number of FFT (Fast Fourier Transform) points NF to an FFT processor 30. The FFT processor 30 performs a fast Fourier transform on the received parallel signals to convert them into parallel signals in the frequency domain.

[0030] NF FDE (Frequency Domain Equalization) processors 32 are provided at the rear of the FFT processor 30. Each of the FDE processors 32 performs equalization processing using one of weight matrices Wj (j=0 to NF-1).

[0031] For example, the FDE processor 32 using the weight matrix W0 is provided with the first signal from each of the NF FFT processors 30 provided in the preceding stage. Then, the FDE processor 32 performs equalization processing based on the weight matrix W0 on the received NF signals to generate NF processed signals. Similarly, the FDE processor 32 using the weight matrix Wj performs equalization processing on NF signals corresponding to the (j+1)th signal to generate NF processed signals.

[0032] In this embodiment, the FDE processor 32 can switch between different weighting matrices W as necessary. Specifically, the FDE processor 32 can selectively switch between a weighting matrix W that assumes the whiteness of noise and a weighting matrix W that takes into account the coloredness of noise. Hereinafter, equalization processing using the former is referred to as a "white equalization processing," and equalization processing using the latter is referred to as a "colored equalization processing."

[0033] NF Inverse Fast Fourier Transform (IFFT) processors 34 are provided in the subsequent stage of the FDE processor 32. As shown in Fig. 4, each of the IFFT processors 34 is provided with one signal from each of the NF FDE processors 32 located in the preceding stage. The IFFT processors 34 perform an inverse fast Fourier transform on the received NF parallel signals to convert the frequency domain parallel signals into NF signals in the time domain.

[0034] The IFFT processor 34 provides the generated parallel signals to a parallel / serial converter 36. The parallel / serial converter 36 converts the received parallel signals into a serial signal and provides it to a demodulator 38. The demodulator 38 performs demodulation processing on the received signal and provides the demodulated signal to a downstream communication device.

[0035] [Weight matrix W for whitening equalization] In this embodiment, the relationship between the transmission signal vector a sent from the antenna 18 of the transmitting station 10 and the reception signal vector y arriving at the antenna 22 of the receiving station 20 can be expressed by the following (Equation 1).

[0036]

number

[0037] As described with reference to Fig. 4, the received signal vector y is subjected to a fast Fourier transform by the FFT processor 30, and then an equalization process using a weight matrix W is performed by the FDE processor 32. The reproduced signal vector a^ generated as a result can be expressed by the following (Equation 2).

[0038]

number

[0039] HΓ included in the above (Equation 1) is a block circulant matrix, and can be expressed using a block diagonalization matrix diag as shown in the following (Equation 3).

[0040]

number

[0041] When the relationship in (Equation 3) above is applied to y in (Equation 2) above, that is, to HΓ in (Equation 1) above, the reproduced signal vector a^ shown in (Equation 2) above can be expressed as shown in (Equation 4) below.

[0042]

number

[0043] Assuming that linear precoding using, for example, Minimum Mean Square Error (MMSE) is used as an equalization criterion for generating a reproduced signal vector a^ from a received signal vector y, the weight matrix W used by the FDE processor 32 can be expressed as shown below (Equation 5).

[0044]

number

[0045] When the transmitting station 10 transmits the transmission signal vector a by Nyquist transmission, the noise η superimposed on the reception signal vector y can be regarded as white noise. In this case, the noise-related term in the above (Equation 5) can be expressed as the following (Equation 6).

[0046]

number

[0047] Using the relationship in (Equation 6) above, the weight matrix W shown in (Equation 5) above can be expressed in the form of a block diagonalization matrix diag as follows:

[0048]

number

[0049] In addition, in the block diagonalization matrix shown in (Equation 7) above, the weight matrix Wk (k=0 to NF) of subchannel k can be expressed as shown in (Equation 8) below. In (Equation 8) below, the subscript "white" is added to the weight matrix Wk of subchannel k to indicate that the noise η is regarded as white noise.

[0050]

number

[0051] When the weight matrix W is block diagonalized as described above, the computational load of the equalization process is relatively light. Therefore, when the reproduced signal vector a^ is generated from the received signal vector y by the whitening equalization process, the power consumption in the receiving station 20 can be kept low.

[0052] [Weight matrix W taking into account noise coloration] When the transmitting station 10 compresses the symbol interval and transmits the signal using FTN transmission, noise becomes colored due to the processing by the receiving filter 24 and the influence of the high-speed sampling specific to FTN transmission. If the white noise at time t is nq(t), the colored noise ηq(t) is expressed by the function g * and the compression rate τ, it can be expressed as follows (Equation 9).

[0053]

number

[0054] The weight matrix W obtained by substituting the noise η shown in the above (Equation 9) into the above (Equation 5) is not a block diagonalization matrix. Therefore, the calculation load of the receiving station 20 associated with the colored equalization process is extremely large. For example, when the number of antennas Nt on the transmitting side and the number of antennas Nr on the receiving side are both 4, and the number of FFT points NF is 2048, the receiving station 20 needs to calculate the inverse matrix of an 8192×8192 matrix for the colored equalization process.

[0055] Even in an environment where noise η becomes colored, if the elements shown in the following (Equation 10) in the calculation formula of the weight matrix W shown in the above (Equation 5) are approximated by a block diagonalization matrix, the weight matrix W used for equalization by the FDE processor 32 becomes the block diagonalization matrix diag shown in the above (Equation 7). In this case, it is possible to advance the colored equalization process by calculation for each subchannel k, and the power consumption of the receiving station 20 involved in the colored equalization process can be reduced compared to the case where the above approximation is not performed.

[0056]

number

[0057] However, when the above approximation is performed, the weight matrix W of subchannel k is not the one shown in the above (Equation 8) but is the one shown in the following (Equation 11). Note that in the following (Equation 11), the subscript "colored" is added to the weight matrix Wk of subchannel k to indicate that the coloring of the noise η is taken into consideration.

[0058]

number

[0059] In the above (Equation 11), the term shown in the following (Equation 12) is a correlation matrix of noise in the frequency domain.

[0060]

number

[0061] Moreover, the (q, q') element of φk shown in the above equation (12) is expressed by the following equation (13).

[0062]

number

[0063] As shown in the above (Equation 13), in order to determine the (q, q') element of φk, it is necessary to solve a complicated equation. Therefore, even if the weight matrix W is block diagonalized by the above approximation, the computation load associated with the chromatic equalization process is still larger than the computation load associated with the white equalization process. Therefore, when the chromatic equalization process is performed, the power consumption of the receiving station 20 is larger than when the white equalization process is performed.

[0064] [Data transmission using Nyquist transmission] In this embodiment, the transmitting station 10 has a function of appropriately setting the compression rate τ of data according to the required transmission capacity required for data transmission. This function can be implemented in the modulator 12 shown in Fig. 3. Alternatively, a controller may be provided in the transmitting station 10 separately from the components shown in Fig. 3, and the parameter setting of the compression rate τ and the like may be performed by the controller.

[0065] If the required transmission capacity can be satisfied by Nyquist transmission, the compression rate τ is set to "1.0". In this case, since white noise is superimposed on the received signal vector y, the receiving station 20 performs whitening equalization. Since the calculation load associated with the whitening equalization is light, the power consumption of the receiving station 20 can be kept low.

[0066] In addition, in Nyquist transmission, since the noise superimposed on the received signal vector y is actually white noise, highly accurate equalization can be obtained by performing white equalization processing. If the reproduced signal vector a^ can be generated accurately by equalization processing, the required CNR for obtaining a desired communication quality can be reduced. If the required CNR is small, the transmission power can also be reduced. Therefore, when Nyquist transmission is used, the power consumption can be reduced to a sufficiently small value in both the transmitting station 10 and the receiving station 20.

[0067] [Data transmission via FTN transmission] In the present embodiment, when the required transmission capacity cannot be obtained by Nyquist transmission, the transmitting station 10 switches the data transmission method to FTN transmission. In other words, in this case, the transmitting station 10 sets the compression ratio τ to less than 1.0 and shortens the symbol interval τT to expand the transmission capacity.

[0068] In FTN transmission, as mentioned above, noise coloration occurs. In this case, when white equalization is performed, the power consumption associated with the equalization can be reduced, but the required CNR increases as the equalization accuracy deteriorates, and the transmission power required to obtain the desired communication quality increases. On the other hand, when colored equalization is performed in combination with FTN transmission, the power consumption associated with the equalization increases, but the required CNR decreases as the equalization accuracy improves, and as a result, the transmission power can be reduced.

[0069] However, the noise coloration caused by FTN transmission does not always occur at the same level, but becomes more noticeable as the compression rate τ becomes smaller and the degree of data compression increases. Therefore, the disadvantage of implementing whitening equalization processing, that is, the increase in power consumption due to the increase in the required CNR, does not need to be considered a big problem when the compression rate τ is close to "1.0", but becomes a big problem as the compression rate τ becomes smaller than "1.0".

[0070] The advantage of the whitening equalization process, that is, the effect of suppressing power consumption due to a light computation load, is always the same regardless of the compression rate τ. Therefore, the whitening equalization process combined with FTN transmission has the characteristic that it always maintains a certain amount of advantage, but generates a large disadvantage as the compression rate τ becomes smaller.

[0071] The opposite characteristic is observed for chromatic equalization. That is, the advantage of chromatic equalization is the effect of suppressing transmission power due to high equalization accuracy. In the region where the compression rate τ is close to "1.0", high equalization accuracy can be obtained even with white equalization, so this advantage is not very significant. On the other hand, as the compression rate τ becomes smaller, the equalization accuracy of white equalization deteriorates, so this advantage becomes significant.

[0072] The disadvantages associated with colored equalization, that is, the increase in power consumption due to a high computation load, always appear in the same way regardless of the compression ratio τ. Therefore, the colored equalization process combined with FTN transmission has the characteristic that it always has certain disadvantages, but generates large advantages as the compression ratio τ becomes smaller.

[0073] As described above, the white equalization process is likely to be of significant benefit in the region where the compression rate τ is close to "1.0". In addition, the chromatic equalization process is likely to be of significant benefit as the compression rate τ is smaller than "1.0". For this reason, in the equalization process method combined with FTN transmission, the overall superiority of the white equalization process and the chromatic equalization process may be reversed depending on the compression rate τ. Therefore, in this embodiment, when FTN transmission is used as the data transmission method, the overall superiority of the white equalization process and the chromatic equalization process is judged, and the more superior method is selected.

[0074] [Operation of the first embodiment] 5 is a flowchart for explaining the flow of processing executed by the transmitting station 10 to realize the above-mentioned functions. This processing may be executed by the modulator 12 shown in FIG. 3, or may be executed by a controller (not shown) as described above.

[0075] In the routine shown in FIG. 5, first, the transmitting station 10 determines the modulation coding rate and the compression rate τ based on the required transmission capacity for the transmission data (step 100).

[0076] Next, it is determined whether or not it is more advantageous to assume the whiteness of the noise than to consider the coloredness of the noise under the adopted compression rate τ (step 102). In other words, it is determined whether or not it is more advantageous to adopt white equalization as an equalization method than to adopt colored equalization.

[0077] In constructing the wireless communication system of this embodiment, the following values ​​are predicted by simulation for each compression ratio τ that is planned to be adopted. Baseband power consumption when white equalization is used: BB_white Required CNR_white when whitening equalization is used Baseband power consumption when colored equalization is used: BB_colored Required CNR_colored when colored equalization is used

[0078] Specifically, in step 102, first, the following value is calculated for the adopted compression ratio τ based on the above simulation results: ·Transmission power Pt_white to achieve the required CNR_white · Transmitting electrodes Pt_colored to achieve the required CNR_colored Difference between the transmission power Pt_white and the transmission electrode Pt_colored ΔPt Difference between baseband power consumption BB_colored and baseband power consumption BB_white ΔBB

[0079] In step 102, it is then determined whether ΔBB>ΔPt holds. ΔBB is the extra power consumed by using the chromatic equalization process compared to using the white equalization process. Also, ΔPt is the extra power consumed by using the white equalization process compared to using the chromatic equalization process.

[0080] Therefore, if it is determined that ΔBB>ΔPt holds, it can be determined that the white equalization process is more advantageous than the chromatic equalization process in terms of reducing total power consumption. In this case, the determination in step 102 is affirmative, and a method that assumes whiteness is adopted as the equalization process method, that is, the white equalization process (step 104).

[0081] On the other hand, if it is determined in step 102 that ΔBB>ΔPt is not true, it can be determined that the white equalization process is not more advantageous than the chromatic equalization process in terms of reducing power consumption. In this case, a method that takes into account coloredness, that is, the chromatic equalization process, is adopted as the equalization process method (step 106).

[0082] Whether the transmitting station 10 has adopted the white equalization process or the colored equalization process is notified to the receiving station 20 in a state stored in the transmission data. In accordance with the notification, the receiving station 20 determines whether the weighting matrix W to be used by the FDE processor 32 is for the white equalization process or for the colored equalization process. This process may be executed by the FDE processor 32, or may be executed by a controller separately provided inside the receiving station 20.

[0083] In the routine shown in FIG. 5, following the processing of step 104 or 106, it is determined whether the following relationship is satisfied (step 108). "Average CNR" > "Required CNR" + margin α

[0084] The above "average CNR" is the "average CNR" of the transmission data achieved by the currently set transmission power Pt. The above "required CNR" + margin α is a value obtained by adding a predefined margin α to the "required CNR" required by the current transmission data. The average CNR value can be estimated based on the results of a simulation carried out in advance. Alternatively, the average CNR value may be actually measured at the receiving station 20 and fed back from the receiving station 20 to the transmitting station 10.

[0085] If the above relationship holds, it can be determined that the current transmission power is excessive for the desired communication quality. In this case, the transmission power is suppressed in steps smaller than the margin α (step 110).

[0086] Thereafter, the process of step 110 is repeatedly executed until the determination in step 108 is negative. Then, when the determination in step 108 is negative, the routine shown in Fig. 5 is terminated. This allows the transmission power Pt to converge to a value that is neither excessive nor insufficient for obtaining the required CNR. As a result, wireless communication having the desired communication quality is realized with the minimum necessary power consumption.

[0087] As described above, according to the wireless communication system of the present embodiment, it is possible to appropriately switch between Nyquist transmission and FTN transmission as a wireless communication method. In addition, in the FTN transmission, it is possible to set an appropriate compression rate τ according to the required transmission capacity. Furthermore, when FTN transmission is implemented, it is possible to adopt either the white equalization process or the chromatic equalization process, whichever is more advantageous in terms of reducing total power consumption, according to the compression rate τ to be adopted. Therefore, according to the wireless communication system of the present embodiment, it is possible to efficiently achieve both high communication quality and reduced power consumption under various circumstances.

[0088] [Modification of the first embodiment] In the above-mentioned first embodiment, in order to judge the superiority of the white equalization processing and the color equalization processing, it is determined whether or not ΔBB>ΔPt is satisfied. However, the present disclosure is not limited to this. For example, when the superiority of both can be determined by directly comparing the difference in required CNR and the difference in power consumption due to the different equalization methods, the superiority of both may be determined by the comparison. Alternatively, the superiority of two equalization processes may be determined based on the magnitude relationship between the ratio of the required CNR for each equalization method and the ratio of the power consumption for each equalization method. Furthermore, the superiority of both may be determined by comparing the sum of the transmission power for each equalization method to achieve the required CNR and the power consumption required for equalization for each equalization method. [Explanation of symbols]

[0089] 10 Transmitting Station 12 Modulator 16 Transmission Filter 18, 22 Antenna 20 Receiving Stations 32 FDE processor

Claims

1. A wireless communication system including a transmitting station and a receiving station, The transmitting station, A process of determining a compression ratio τ to be multiplied by a reference symbol interval T according to the Nyquist criterion based on a required transmission capacity of the transmission data; a process of transmitting a radio signal to the receiving station by FTN transmission at an FTN symbol interval τT obtained by multiplying the reference symbol interval T by the compression rate τ; a process of determining which of a whitening equalization process assuming the whiteness of noise superimposed on the wireless signal is superior to a colored equalization process taking into account the coloredness of the noise in terms of power saving at the compression ratio τ; and selecting one of the white equalization processing and the chromatic equalization processing, which is more power-saving, and notifying the receiving station of the selection. The receiving station, receiving the wireless signal; and performing an equalization process on the received wireless signal by a method indicated in the notification, out of the white equalization process and the colored equalization process. Wireless communication system.

2. The superiority / inferiority determination process includes: A process of calculating power consumption associated with the execution of the white equalization process and the chromatic equalization process, respectively; A process of calculating a required CNR for obtaining a desired communication quality under the compression rate τ for each of the white equalization process and the chromatic equalization process; a determination process for determining the superiority or inferiority based on the power consumption and the required CNR for the white equalization process and the power consumption and the required CNR for the chromatic equalization process; 2. The wireless communication system of claim 1, comprising:

3. The determination process includes: A process of calculating a difference ΔBB between the power consumption associated with the execution of the chromatic equalization process and the power consumption associated with the white equalization process; A process of calculating a difference ΔCNR between the required CNR for the white equalization process and the required CNR for the chromatic equalization process; A process of determining the superiority or inferiority based on a comparison between the ΔBB and the ΔCNR; 3. The wireless communication system of claim 2, comprising:

4. The determination process includes: A process of calculating a difference ΔBB between the power consumption associated with the execution of the chromatic equalization process and the power consumption associated with the white equalization process; A process of calculating a difference ΔPt between a transmission power for satisfying the required CNR for the white equalization process and a transmission power for satisfying the required CNR for the chromatic equalization process; A process of determining the superiority or inferiority based on a comparison between the ΔBB and the ΔPt; 3. The wireless communication system of claim 2, comprising:

5. The transmitting station, A process of calculating a required CNR for obtaining a desired communication quality under the compression rate τ for each of the white equalization process and the chromatic equalization process; obtaining an average CNR for the wireless signal being transmitted; a process of repeatedly reducing transmission power within a range in which the average CNR does not fall below the required CNR for the selected one of the white equalization process and the chromatic equalization process; The wireless communication system of claim 1 further configured to perform:

6. A wireless communication device having a function of transmitting a wireless signal to a receiving station, A process of determining a compression ratio τ to be multiplied by a reference symbol interval T according to the Nyquist criterion based on a required transmission capacity of the transmission data; a process of transmitting a radio signal to the receiving station by FTN transmission at an FTN symbol interval τT obtained by multiplying the reference symbol interval T by the compression rate τ; A process of determining which is superior in terms of power saving at the compression ratio τ between a whitening equalization process assuming that the noise superimposed on the wireless signal is white and a colored equalization process taking into account the colored nature of the noise; and selecting one of the white equalization processing and the chromatic equalization processing which is more power-efficient, and notifying the receiving station of the selection.

7. A wireless communication method for performing wireless communication between a transmitting station and a receiving station, comprising: The transmitting station determines a compression ratio τ to be multiplied by a reference symbol interval T according to the Nyquist criterion based on a required transmission capacity of the transmission data; The transmitting station transmits a radio signal to the receiving station by FTN transmission at an FTN symbol interval τ T obtained by multiplying the reference symbol interval T by the compression rate τ; The transmitting station determines which of a white equalization process assuming whiteness of noise superimposed on the wireless signal is superior to a colored equalization process taking into account coloredness of the noise in terms of power saving under the compression ratio τ; the transmitting station selects one of the white equalization processing and the chromatic equalization processing, which is more power-saving, and notifies the receiving station of the selection; the receiving station receiving the radio signal; the receiving station performs equalization processing on the received wireless signal using a method indicated by the notification, out of the white equalization processing and the chromatic equalization processing; A wireless communication method comprising:

8. A wireless communication program for causing a wireless communication device to transmit a wireless signal to a receiving station, The wireless communication device includes: A process of determining a compression ratio τ to be multiplied by a reference symbol interval T according to the Nyquist criterion based on a required transmission capacity of the transmission data; a process of transmitting a radio signal to the receiving station by FTN transmission at an FTN symbol interval τT obtained by multiplying the reference symbol interval T by the compression rate τ; A process of determining which is superior in terms of power saving at the compression ratio τ between a whitening equalization process assuming that the noise superimposed on the wireless signal is white and a colored equalization process taking into account the colored nature of the noise; a process of selecting a method having superior power saving performance from the white equalization process and the chromatic equalization process and notifying the receiving station of the selection; A program for wireless communication comprising a computer readable program for causing the computer to execute the above steps.

Images