JPS592220B2 - Interference elimination method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for eliminating interference, and particularly when transmitting digital information using a propagation path in which interference, fading, and impulsive noise exist, reception errors due to interference are eliminated. This paper relates to an interference cancellation method for

The causes of code errors in digital communications using radio circuits as propagation paths are fading, interference, and impulsive noise.

Among these, fading occurs for a large percentage of the time, but the magnitude of its interference is generally predictable, and its influence can be reduced by using error correction codes and the like. Furthermore, since impulsive noise exists only for a small period of time, errors can be removed by using an error correction code or the like. On the other hand, there are various causes of interference, but in this case, interference refers to interference that occurs when a receiver receives the signal and converts it to an intermediate frequency or audio frequency band, which ends up dropping to the same frequency as the signal frequency band. This includes spurious sensitivity such as image frequencies caused by the characteristics of the receiver, spurious transmission from the transmitter, and overlapping frequency assignments (for example, in shortwave communication, the transmission distance of radio waves is long, so This includes interference caused by interference from far away (overseas). This interference does not always exist, but when it does exist, it has a large effect, and the use of error correction codes alone cannot sufficiently eliminate the effect. For low-speed transmission, there are methods such as spectrum pattern modulation, but especially when performing relatively high-speed transmission, communication often becomes impossible due to the presence of interference, and there has been no effective solution. In view of the above points, the present invention aims to provide an interference cancellation method that can reliably eliminate reception errors due to interference.

Its unique feature is that it uses the FDM-PSK method, which divides the transmission band into multiple subchannels, phase modulates each subchannel with digital information, and transmits it. The purpose of this technology is to distinguish between subchannels where interference exists and those without interference, and to transmit information using only subchannels where interference does not exist. Embodiments of the interference removal method according to the present invention will be described below with reference to the drawings.

In FIG. 1, a transmitting device 1 divides the transmission band into a plurality of subchannels and transmits input information via a propagation path 2 using the FDM-DPSK method, which differentially phase modulates each subchannel with digital information. and transmits it to the receiving device 3.

In this case, error-detectable encoding or signal processing is performed for each subchannel. Therefore, encoding etc. is performed separately for each subchannel, such that one character of the code is placed on one subchannel. conduct. The receiving side device 3 is an FD
It has a receiving section 10 that receives and amplifies information transmitted by the M-DPSK method, and a demodulating section 11 provided corresponding to each subchannel, and the digital information demodulated by the demodulating section 11 is It is sent to the output memory 13 via the reception control circuit 12. The output memo 1/3 temporarily stores digital information and outputs the output information when all desired information is correctly stored. Also, an error detection signal due to interference detected at the same time as demodulation is applied to the reception control circuit 12, where it is determined whether each subchannel is good or bad. This determination result is transmitted by the transmitting unit 14 to the transmitting side device 1 via the propagation path 2. The transmitting device 1 transmits information using only the subchannels for which it is determined that no interference exists based on the half-separation results. In other words, initially it is transmitted from all subchannels, and the receiving side receives it and determines whether it is good or bad, but once the determination results are fed back from the receiving side to the transmitting side and controlled, it is used for transmitting information. This means that there will be subchannels that are currently in use, and subchannels that are not used and do not emit their own radio waves. Here, error detection due to interference in the demodulator 11 is performed based on the following principle.

In other words, when comparing fading, which is a cause of reception errors defined as code errors (when the transmission code and reception code are different) in digital information received and demodulated by the receiving side equipment, and interference, it is found that interference is caused by fading. It is known both theoretically and from measurement results of actual propagation paths that most errors occur when the electric field strength is significantly lower than the average signal electric field strength. On the other hand, errors due to interference, for example, in 4-phase PSK, the phase difference between the phase of the composite vector of the signal and interference and the original phase of the signal is determined by the decision threshold (±4
Occurs when the number exceeds 5th place). Therefore, the amplitude of the composite wave does not necessarily become small. Therefore, the cause of the error can be identified based on the amplitude when the error occurs.
When distinguishing between the two error-generating factors, when looking at individual errors, it is possible that the identification may be incorrect.

However, when averaged over time for each subchannel, subchannels with no interference will have a low average value, and subchannels with interference will have an average value that is equal to or higher than the signal level, allowing for more reliable identification. It will be done.
FIG. 2 shows the configuration of the demodulator 11 based on the above principle.

In this figure, the detection circuit DET separates each FDM subchannel from the input signal S using a filter or a circuit that functions equivalent to a filter (for example, an integrating/dumping circuit), and separates each subchannel component from the input signal S. , a sine wave with a frequency approximately equal to the carrier frequency of each input subchannel (preferably a frequency difference of approximately 1/10 or less of the modulation speed), and two sine waves with the same frequency but with a phase number less than 90. Two local signals are generated within the circuit, and two outputs are obtained by synchronously detecting the separated subchannel components using these two local signals. This process calculates the in-phase component and orthogonal component to the local signal for each subchannel component, using the phase of one local signal as a reference (0.). Say. Although these processes can be realized by ordinary analog circuits, since it is convenient for the processing after the converter CONVl, the input signal S is converted into a digital code by an analog-to-digital converter, and then the above processes are performed by digital calculation. method is often used. In this way, the detection circuit DET separates the received signal component of the corresponding subchannel from the received wave S, and as shown in FIG. The amplitude C of the component and the amplitude B of the orthogonal component are detected for each element constituting the digital information, and a circuit or the like that performs digital arithmetic processing is used. The detection circuit DE
The amplitude C of the in-phase component and the amplitude B of the quadrature component due to T are applied to the conversion circuit CONl. This conversion circuit CON
Vl is the amplitude C of the in-phase component and the amplitude B of the quadrature component.
From this, the phase and logarithmically expressed amplitude of the received signal S1 are determined, the details of which are shown in FIG. In other words, the conversion circuit CONVl receives the signal from the detection circuit DET.
This circuit uses two outputs (in-phase component and quadrature component) to calculate the amplitude of each subchannel component and the phase (phase difference with the local signal) based on the phase of the local signal generated in the above-mentioned detection circuit DET. is output in decibel (DB) display. When the amplitude is A, the phase is θ, the in-phase component is C, and the orthogonal component is B, the amplitude is expressed as follows, and when the logarithm of this is taken, it becomes.

In actual processing, in order to reduce the size of the converter's ROM table, 1B1 and 1C1 are compared in size,
10gIB1 (when B≧C) or 10g1C1 (when C>B)) and 10gV "'De Cabinet Seal'? ] (when B≧C) or 1
Calculate 0gVd = nkenwo] (when C>B) and get 10
Find gA.

The phase is expressed as θ=Tan-1 (B/C), but in order to make the ROM table of the converter small (to make the table in the range of 0≦θ≦45 devices), θ
-Tan-1 (1BI/1C1), (when 1C1≧1B1) or Tan-1 (1CI/IBI), (1B1>
When 1CI) is obtained from the table, and using the positive/negative of B and C and the magnitude relationship between (B and 1C1), θ≦θ is calculated as 360 (
Calculate the phase θ in the range of .

For example, B is positive, C is negative, and if 1C1>1B1,
From the ROM table, θ5-Tan-1 (IBl/ICI
) and calculate the phase θ using θ-1800-θ5. In Fig. 4, the amplitude C of the in-phase component and the amplitude B of the quadrature component
is applied to the comparator COMP2 and the switch SW.

The comparator COMP2 compares the absolute values of both amplitudes B and C, and sends the comparison result to the switch SW and the converter CON.
Add to V5. As a result, the switch SW applies the amplitude with the larger absolute value to the divider DIl as a division input, and also adds the smaller amplitude as the human power to be divided. Furthermore, the amplitude with the larger absolute value is supplied to the converter CON4. The divider DIVl calculates the amplitude ratio by dividing the small amplitude by the large amplitude, and if the received signal Si has a phase as shown in Fig. 3, it outputs the amplitude C/B to the converters CON2 and CON3. do. The converter CON2 is a circuit that obtains the aforementioned θ52Tan-1 (1BI/1CI) or Tan-1 (1C1/1B1).
A ROM table is used in which a code of y is obtained as an output when y=Tan-x is input manually using x as an address.

Then, the value corresponding to θ5 from the amplitude ratio C/B, that is, the received signal Si out of each phase of O, π/2, π, 3π/2
The absolute value of the phase seen from the closest position is determined and output to the converter CON5. The converter CON5 converts the CO
A signal Si(θ) representing the phase θ of the received signal Si is output using the comparison result of the positive/negative and magnitude relationship of both components B and C from MP2. On the other hand, the converter CONV3 is 10gV ``U5''? ) [or 10g V-cho "mouth?fu〒7)]", z
=Tan-1(1+x), and a ROM table is used in which when X is input as an address, a code of z is obtained as an output.

That is, upon receiving the output of the divider DIl, the logarithm of the value corresponding to C/B is added to the adder ADDl, and the converter CO
N4 converts the amplitude B having the larger absolute value into a logarithmic value and adds it to the adder ADDl. As a result, adder ADDl receives 101B1 (or 10gIC1) from converter CONV4.
This is a circuit that obtains 10 gA by adding 10 gvn required from the converter CON3 and 10 gvn required from the converter CON3. Outputs level signal Si (IOgA).

Now, the signal Sl(θ) representing the phase θ of the received signal Si obtained by the converter CONVl as described above is applied to the determination circuit DEC as shown in FIG.

This determination circuit DEC determines the phase difference between the front and rear elements, determines the phase difference, decodes it into a digital code, and applies the code to the error detection circuit CHK. This error detection circuit CHK is an error detection circuit that supports error-detectable encoding on the transmitting side, and detects errors in demodulated digital information. Various codes are known as error-detectable codes, but any block code can be applied. For example, Hamming code, BC
H code, cyclic code, etc. can be applied. Then, this error detection circuit CHK detects an error for each character, and this error detection signal is applied to an AND gate ANDl.
On the other hand, the reception level signal S1 (
10gA) is applied to the character level detection circuit AVE, where the value of the signal amplitude within the interval of one character, for example, the average value, is determined and outputted to the comparator COMPl. Comparator COM
Pl is the reception level for each character and the threshold level SH
When the received level is higher than the threshold level SH, a level comparison output is output to the AND gate AND1. This makes the AND gate AND
l is the reception level greater than the threshold level,
When an error exists, an interference detection signal CO is output. Note that information when there is no error is output from the error detection circuit CHK as digital information DO. Incidentally, in the above case, the threshold level SH for determining the level of each subchannel is usually calculated based on the instantaneous reception level (the average value of the signal levels measured for each signal element).

However, if the average value is calculated from the number of subchannels in use and the AGC characteristics of the receiving section 10, if strong interference exists, the AGC of the receiving section will work to suppress this interference to a certain level. Therefore, the level of the signal fluctuates and may not match the value determined as described above. Therefore, the average value of the reception level is calculated excluding the subchannel in which the error was detected. In this case, in addition to subchannels with interference, subchannels that have errors due to fading are also excluded, but since errors due to fading account for a small proportion of all subchannels, the average value of the reception level The calculation error is small;
It's not a problem. FIG. 5 shows a circuit that calculates the average value of the reception level as described above.

In this figure, the gate G has a character level detection circuit AV.
The output LL of E is applied to the AND gate AND2, and the no-error signal NE of the error detection circuit CHK and the channel use indication signal CU are applied to the AND gate AND2. Here, the error-free signal NE in FIG. 5 is an output determined to be error-free by the error detection circuit CHK in FIG. 2, and this output is manually switched in the order of subchannels. The channel use display signal CU is a signal stored in the reception control section 12 of FIG. 1 as the content of the judgment control, and is read out from the memory and input manually. In this case, these signals NE, CU
is applied at different times and switched for each subchannel. As a result, the counter CTR receiving the output of the AND gate AND2 counts the number of subchannels used for transmission in which no errors have been detected. At the same time, the received levels of the subchannels in use for the transmission and in which no errors have been detected are added via the gate G to the accumulative adder ADD2, where they are cumulatively added. After the above processing is performed for all subchannels, the divider DIV2 divides the cumulative sum of the reception levels by the counted number of subchannels to calculate a desired average reception level SAVL. Then, with this level as a reference, the threshold level SH is set to a value lower than this, for example, a value several DB lower. Now, when the interference detection signal CO is output from the demodulator 11, the reception control circuit 12 counts the number of times this signal CO appears for a certain period of time, and determines whether the subchannel is good or bad based on the magnitude of this counted value. Based on this result, the transmitting device 1 performs transmission using only the subchannels whose count values are below a certain value and are determined to be in good condition. As explained above, according to the above embodiment, FDM
- When transmitting using the DPSK method, the receiving device identifies the subchannel where interference exists, sends it back to the transmitting device, and transmits information using only subchannels without interference. Reception errors caused by interference can be reliably removed.

Therefore, relatively high-speed code transmission is possible even in the presence of interference. Additionally, since the number of subchannels used for transmission changes depending on the state of interference, information can be transmitted at high speed if the propagation path conditions are good, or at low speed if the conditions are bad, making it possible to adapt to a wide range of situations. . Note that although there is a time lag between the time when determining whether there is interference and the time when the results are used to control the transmitting device and actually transmit information, the radio waves causing interference may also contain some kind of information. This is a radio wave that is transmitting radio waves, and is thought to last for a certain amount of time. Therefore, for a short period of time, it can be considered that a state similar to the state at the time of discrimination continues. (Normal interference typically continues for several minutes or more) In addition, in order to respond to changes in the situation during transmission, the same discrimination operation is continued during transmission, and if the situation changes, the control signal is fed back again. Make it functional. In addition, it is possible to feed back the result of determining whether the information has been correctly decoded (error detection is possible because an error-detectable code is used) at the same time as feedback of the control signal, and to use processing such as retransmission. , information can be reliably conveyed to the other party. Various methods are known for feeding back error detection results and retransmitting information. In the above embodiment, the presence of interference is identified from the reception error and reception level of the subchannel used for transmission, but if the transmission time of the information to be transmitted is long and the interference situation changes, It is necessary to determine whether interference persists even on subchannels that are not used for transmission, and it is appropriate to determine the subchannel to be used for transmission not only once but also from time to time during transmission. It is. To determine whether there is interference on subchannels that are not used for transmission,
Since this is not possible using a method based on error detection, the determination must be made based only on the magnitude of the reception level. The threshold level at this time is set lower than when determining the subchannel used for transmission. Specifically, it depends on the number of modulation phases, but the signal at the limit where an error occurs,
A value corresponding to the level ratio with interference, for example, in the case of 4-phase DPSK, is set to be about 8 dB lower. In addition, when determining the level of each subchannel, the reception level may be detected by using the average value within a one-character interval of the instantaneous reception level value for the subchannel used for transmission, but for subchannels not used for transmission. For this, the level changes based on the modulation of the interference wave appear as they are, so there is a risk of erroneous detection using a simple average.

For example, if the interference wave is Al (AM telegraph), the received character will be erroneous even if the mark section is part of the character being transmitted, but the average level of the interference wave will be Depending on the ratio of time and space, the value will be considerably smaller than the peak level of the interference wave. In order to avoid this inconvenience, for subchannels that are not used for transmission, one character interval is divided into multiple parts, the average value within that interval is calculated for each divided interval, and the average value within that one character interval is calculated. The maximum value of the average values may be used as the reception level for the character section. Further, in the above embodiment, the case where differential phase modulation is applied to each subchannel is illustrated, but synchronous phase modulation may be adopted as necessary.

As described above, according to the present invention, there is provided an interference cancellation method that can reliably eliminate reception errors due to interference.

[Brief Description of the Drawings] Fig. 1 is a block diagram showing an embodiment of the interference cancellation method according to the present invention, Fig. 2 is a block diagram showing the configuration of the demodulation section in the embodiment, and Fig. 3 is a block diagram showing the operation of the embodiment. FIG. 4 is a block diagram showing the details of the converter that extracts the phase and logarithmically expressed amplitude of the received signal in the demodulation section.
FIG. 5 is a circuit diagram showing a circuit for calculating the average value of reception levels. 1... Transmitting side device, 2... Propagation path,
3. . . . Receiving side device, 10... Receiving unit, 11... Demodulating unit, 12... Reception control unit, 1
3...Output memory, 14...Transmission unit, ADD
I... Adder, ADD2... Cumulative adder,
ANDI, AND2...And gate, AVE
...Character level detection circuit, CONVI to CO
NV5...Converter, COMPI, COMP2...
... Comparator, CHK ... Error detection circuit, DEC
...Judgment circuit, DET...Detection circuit, D
IVIDIV2... Divider.

Claims (1)

[Claims] 1. In the case where a transmission band is divided into a plurality of subchannels and a transmitting device modulates each subchannel with digital information and transmits the modulated digital information to a receiving device, the receiving device A reception error is detected for each channel, the reception level at the time of the reception error is compared with a predetermined threshold level, and the number of times the reception error exists and the reception level is higher than the threshold level is counted. When this count value is above a certain value, it is determined that there is interference, thereby identifying the presence or absence of interference for each subchannel, and returning this identification result from the receiving device to the transmitting device, and confirming that there is no interference. An interference cancellation method characterized in that the digital information is transmitted using only subchannels. 2. The interference removal method according to claim 1, wherein the threshold level is set lower than the average value of the reception level for each character of the subchannel in which the reception error is not detected. 3. For subchannels in use for transmission, the first threshold level is used as the threshold level, and for subchannels not in use, the second threshold level is used, and the second threshold level is set to the second threshold level. 1. The interference removal method according to claim 1, wherein the interference removal method is set lower than the threshold level of 1. 4. Interference removal according to claim 3, in which the reception level and the first threshold level are compared for each character, and the reception level is the average value of instantaneous reception level values within one character interval. method. 5 The reception level of the unused subchannel and the second
The threshold level is compared with the threshold level for each character, and the interval of one character is divided into multiple parts, and the average value of the instantaneous reception level value is calculated for each divided interval, and among those average values, The interference removal method according to claim 3 or 4, in which the maximum value of is used as the reception level.