JP7341487B2 - Interference power estimation device, interference power estimation program, and information collection station - Google Patents

Description

The present invention relates to an interference power estimation device, an interference power estimation program, and an information collection station for optimizing the spectrum spreading factor in an environment where a plurality of standards use the same frequency band.

In recent years, LPWA (Low Power Wide Area) has attracted attention as being suitable for utilization of IoT (Internet of Things) (Non-Patent Document 1). LPWA is a communication standard that enables long-distance communication with low power consumption. Even in environments where it is difficult to secure a power source, the battery can be used for a long time, so it is often used to meet the needs of IoT.

However, since the wireless communication systems classified as LPWA have been proposed independently (for example, LoRa, SIGFOX, Wi-SUN), interference (CCI: Co-Channel Interference) occurs from other wireless communication systems that use the same frequency. ) may be received. LoRa, which is a typical example of LPWA, uses chirp modulation, which is a type of spread spectrum technology (Patent Document 1). In this method, resistance to interference power can be improved by increasing the signal spreading factor. For example, when the frequency band of interference from other wireless communication systems is relatively narrow, there is a method of removing interference waves using a frequency mask (Patent Document 2).

JP 2019-193049 Publication Japanese Patent Application Publication No. 2009-58308

Ministry of Internal Affairs and Communications, “Trends in wireless systems related to LPWA”, March 7, 2018 (https://www.soumu.go.jp/main_content/000543715.pdf)

However, in order to increase resistance to interference, it is necessary to further widen the spectrum, and in the case of chirp modulation, broadening the spectrum also leads to a reduction in the transmission rate. Therefore, it is necessary to accurately estimate the interference power and design an appropriate spreading factor to match the interference.

An interference power estimating device according to an aspect of the present disclosure receives radio waves including a signal from an arbitrary wireless terminal in which transmission data is chirp-modulated with a predetermined spreading factor for each symbol belonging to a predetermined set. and an interference power estimating device that estimates the power of an interfering radio wave from the radio wave, wherein each of the symbols has a configuration of 2 bits or more , and a unit that converts a signal received from the radio wave into a received spectrum signal; A unit that prepares a chirp pattern for each symbol, and performs maximum likelihood estimation of the symbols included in the transmission data by comparing correlation values between the reception spectrum signal and each chirp pattern, and a unit that estimates symbols other than the maximum likelihood estimation of the symbols included in the transmission data. The apparatus further includes a unit that estimates the intensity of the interference radio wave by averaging the correlation values related to the chirp patterns or by selecting the maximum value of the correlation values related to the chirp patterns other than the symbol estimated by the maximum likelihood.

The interference power estimating device may further include a unit that estimates the strength of a signal radio wave from a correlation value related to a chirp pattern of symbols estimated by maximum likelihood.

In the interference power estimating device, the transmission data may be composed of a plurality of packets each including a predetermined number of the symbols.

In the interference power estimating device, the calculation process of the correlation value may be performed in parallel for each of the chirp patterns.

The interference power estimating device may further include a unit that enables estimation of the intensity of interference radio waves when the signal-to-interference power ratio is equal to or greater than a predetermined threshold.

An interference power estimation program according to an aspect of the present disclosure receives radio waves including a signal from an arbitrary wireless terminal in which transmission data is chirp-modulated at a predetermined spreading factor for each symbol belonging to a predetermined set. and an interference power estimation program for estimating the power of an interfering radio wave from the radio wave, each of the symbols having a configuration of 2 bits or more , and converting the signal received from the radio wave into a received spectrum signal. a step of converting, a chirp pattern is prepared for each symbol, and a step of estimating the symbols included in the transmission data by maximum likelihood by comparing correlation values between the received spectrum signal and each chirp pattern; The intensity of the interfering radio waves is determined by averaging the correlation values related to chirp patterns other than the most likely estimated symbols, or by selecting the maximum value of the correlation values related to chirp patterns other than the most likely estimated symbols. Execute the estimation step.

The interference power estimating program may further include a step of estimating the strength of a signal radio wave from a correlation value related to a chirp pattern of symbols estimated by maximum likelihood.

In the interference power estimation program, the transmission data may be composed of a plurality of packets each including a predetermined number of the symbols.

In the interference power estimation program, the calculation process of the correlation value may be executed within the transmission time of one symbol for each chirp pattern.

The interference power estimation program may further include a step of validating the estimation of the intensity of the interference radio waves when the signal-to-interference power ratio is equal to or greater than a predetermined threshold.

An information collection station according to an aspect of the present disclosure includes the interference power estimation device or the interference power estimation program.

The information collection station may transmit an instruction to increase the spreading factor to the wireless terminal when the intensity of the interference radio waves exceeds a predetermined value.

According to one aspect of the present disclosure, by using the properties of chirp modulation, it is possible to accurately grasp the influence of interference, and as a result, the necessary minimum spreading factor can be set, and efficient LPWA communication can be realized.

FIG. 1 is a block diagram of an embodiment (this embodiment) according to one aspect of the present disclosure. It is a block diagram of a power estimation unit in this embodiment. FIG. 3 is an explanatory diagram showing an example of chirp modulation. FIG. 7 is an explanatory diagram showing another example of chirp modulation. It is a flowchart showing the operation of this embodiment. FIG. 2 is a conceptual diagram showing an interference wave model for a chirp modulated signal. FIG. 3 is a conceptual diagram of a chirp modulation signal with an increased spreading factor. It is a graph showing simulation results of an example of the present disclosure. It is a graph showing simulation results of an example of the present disclosure.

Hereinafter, an interference power estimating device in an embodiment (hereinafter referred to as the present embodiment) according to one aspect of the present disclosure will be described in detail with reference to the drawings. In this embodiment, it is assumed that an information collection station equipped with an interference power estimation device receives radio waves transmitted from an arbitrary wireless terminal (not shown). Furthermore, assume that radio wave interference is being received from a wireless terminal other than the wireless terminal. FIG. 1 shows a block diagram of an interference power estimating device according to this embodiment, and FIG. 2 shows a block diagram of main parts. FIG. 5 also shows a flowchart of the operation of this embodiment.

1 and 2, which are block diagrams of the device, will be used in the description of this embodiment. However, as long as the flowchart shown in FIG. It does not have to be limited to what is made up of. For example, part or all of the device may be constructed from program modules resident on a microprocessor. Further, the program may be installed in the device in advance, or may be downloaded from an external server or the like.

In FIG. 1, reference numeral 1 denotes an antenna, which simultaneously receives signals from the wireless terminal (not shown) and interference radio waves from other terminals. The output of the antenna 1 is amplified to a predetermined amplitude by a head amplifier 2, and then sent to an FFT unit 3 as a received signal. The subsequent operations are also shown in the flowchart (S101 to S105) in FIG. The FFT unit 3 sequentially performs Fourier transform on this received signal. As a result, the frequency spectrum of the received signal is detected in real time and output as a received spectrum signal f(t) whose frequency components change over time (S101).

Here, it is assumed that the wireless terminal performs chirp modulation on each symbol of transmission data at a predetermined spreading factor, and sequentially transmits the data by radio. It is assumed that all symbols in the transmitted data belong to a predetermined set (s ₁ , s ₂ , . . . , s _k ). FIG. 3 shows an example of a frequency spectrum signal of a chirp modulated signal when k=2. All symbols belong to a set (s ₁ , s ₂ ) represented by 1 bit, and here it is assumed that symbol s ₁ is composed of data 0 and symbol s ₂ is composed of data 1. By assigning corresponding chirp patterns C(s ₁ ) (up-chirp) and C(s ₂ ) (down-chirp), transmission data (for example, 1100) in an arbitrary packet can be transmitted in frequency steps as shown in the figure. It can be sent as an up/down signal. In this embodiment, the number of up-chirp and down-chirp stages (spreading factor) is set to 4, but this spreading factor can be set arbitrarily. The higher the spreading factor, the better the signal-to-noise power ratio (SNR) and the signal-to-interference power ratio (SIR), but the communication speed decreases in inverse proportion to the spreading factor.

FIG. 4 shows an example of chirp modulation when k=4. In this example, each symbol belongs to a set of all patterns consisting of 2 bits (∈00, 01, 11, 10). That is, the 2-bit data 00, 01, 11, 10 are set as symbols s ₁ , s ₂ , s ₃ , s ₄ , and the corresponding chirp patterns C(s ₁ ), C(s ₂ ), C(s ₃ ), C(s ₄ ). Each chirp pattern is a pattern obtained by cyclically shifting the basic pattern C(s ₁ ) by one step. Note that the chirp patterns are not limited to the above two examples as long as they have no correlation with each other, and the chirp patterns do not have to be those in which the frequency is changed stepwise (for example, they may be hopping patterns).

In FIG. 1, memories 5a, 5b, and 5c are provided for the number (k) of symbols (∈s ₁ , s ₂ , . . . , s _k ) belonging to a predetermined set, and chirp patterns corresponding to each symbol are registered. has been done. For these chirp patterns (C(s ₁ ), C(s ₂ )...C(s _k )), correlators 4a, 4b, and 4c each perform a correlation calculation with the received spectrum signal f(t). are executed and output as correlation values (r ₁ , r ₂ , . . . , r _k ), respectively (S102). As a method of correlation calculation, for example, a method of calculating f(t)×C(s _i ) (i=1, 2, . . . k) on the frequency axis and integrating each calculated value on the time axis. It may be. At this time, the integration time only needs to be within one symbol period.

Note that the correlation calculation process is preferably executed in parallel for each chirp pattern (C(s ₁ ), C(s ₂ ), . . . C(s _k )). It is more preferable to execute the processes in parallel and simultaneously, but in the case of program processing, etc., time-sharing processing may be performed as long as the intervals are negligible. In any case, it is sufficient if it is executed within the transmission time of one symbol.

Assuming that there is no interference from sources other than the wireless terminal, if any chirp pattern matches the received spectrum signal f(t), the corresponding correlator will output the square of the received spectrum signal f(t) as the correlation value. In other words, a value having the dimension of signal power is output. On the other hand, zero (noise level value) is output from integrators whose chirp patterns do not match.

Therefore, the maximum likelihood estimation unit 6 mutually compares the outputs (r ₁ , r ₂ , . . . , r _k ) of each correlator, performs maximum likelihood estimation on the symbols included in the transmission data, and selects the most probable estimated symbol s _MLE. (S103). As a maximum likelihood estimation method, for example, a symbol related to a chirp pattern with a maximum correlation value is selected. The estimated symbols are sequentially output and separated into packets. The end of each packet is a parity, and the CRC unit 7 performs an error check.

The output of each correlator (r ₁ , r ₂ , . . . , r _k ) is also supplied to the power estimation unit 8 . In this embodiment, the power estimation unit 8 estimates the intensity of the interference radio wave from the correlation value related to the chirp pattern other than the maximum likelihood estimated symbol s _MLE (S104). Further, the intensity of the signal radio wave is estimated from the correlation value related to the chirp pattern of the symbol s _MLE estimated by the maximum likelihood (S105). A detailed block diagram of the power estimation unit 8 is shown in FIG.

In FIG. 2, the maximum likelihood estimated symbol s _MLE (m bits) is supplied to each input of the decoders 83a to 83c, and the positive logic output of each decoder is input to the control input of the gates 81a to 81c (k in total). The negative logic outputs are respectively supplied to control inputs of gates 82a to 82c (k in total). Each _decoder has a label corresponding to the symbols s ₁ , s ₂ , . (The negative logic output is designed to change from 1 to 0). For example, when the symbol s _MLE is supplied to the power estimation unit 8, only the gate 83b with the label corresponding to the symbol s _MLE reacts and opens the gate 81b (positive logic side). As a result, the correlation value (r _MSE =P _D ) is sent to the integrator 84 . This will be discussed later.

On the other hand, due to the negative logic output of the reacting decoder 83b, gates of systems other than the symbol s _MLE among the gates 82a to 82c (k in total) open, and the interference power estimation unit 85 receives these k-1 correlation values (r (other than _MSE ) is sent. The operation of the interference power estimation unit 85 will be explained below. First, the case where k=2 will be explained. The upper part of Fig. 6 shows that in the case of the received spectrum signal in Fig. 2, that is, (s ₁ , s ₂ ) = (0, 1), interference is superimposed on the up-chirp and down-chirp signals for each symbol. A received spectrum signal is schematically shown. If symbols s ₂ , s ₂ , s ₁ , s ₁ (1100 in data notation) are estimated from this received spectrum signal (second row in the figure), the interference power estimation unit 85 stores symbols other than those estimated. will be sent. In the case of K=2, there is only one type of "symbol other than the estimated one", that is, the inverted values of each symbol, s ₁ , s ₁ , s ₂ , s ₂ (0011 in data notation) are sent (same as (third row of figure).

Further, the interference power estimating unit 85 receives the results of correlation calculation between these inverted symbols and the corresponding chirp patterns (bottom row in FIG. 6). As mentioned above, this value is zero (or noise level) when there is no interference and there are no errors in the received data. However, when there is interference, since the interference component is not chirped, a signal component crosses the chirp, and therefore the correlation value does not become zero. It goes without saying that the correlation value at this time has a correlation with the intensity of the interference component.

If one symbol has a multi-bit structure (m (≧2) bits), K = 2 ^m > 2, and there are multiple (K-1) “symbols other than those estimated”, and the interference power estimation unit 85 A plurality of (K-1) correlation values are sent. Since these correlation values may differ for each chirp pattern, the interference power estimation unit 85 may average these values and output the averaged values. Alternatively, the correlation value having the maximum value may be selected and output. The output P _I of the interference power estimation unit 85 is accumulated for each packet by the integrator 86, and the estimation result is output as the estimated interference power P _I ^ (indicated by an overline in FIG. 2 and equation (1)).

The integrator 84 is supplied with only the correlation value r _MSE related to the estimated symbol s _MLE . This correlation value has a dimension of the square of the received signal when there is no interference or noise, and becomes a value (P _D ) indicating signal power. The integrator 84 integrates the signal power P _D for each packet and outputs it as the signal estimated power P _D ^ (indicated by an overline in FIG. 2 and equation (2)).

Here, N is the number of symbols making up the packet. Note that the description of the 1/N calculation means in the above equations (1) and (2) is omitted in FIG.

When the interference superimposed on the received signal becomes large to a certain extent, an error may occur in the maximum likelihood estimation of the symbol. The validity determination unit 87 validates the estimation of the intensity of the interference radio wave when the signal to interference power ratio (SIR) is equal to or higher than a predetermined threshold (S106). In other words, in other cases, the output of the estimated interference power PI^ is stopped. Alternatively, an identifiable code indicating that estimation is not possible may be output. The signal-to-interference power ratio (SIR) may be determined by calculating the ratio of the estimated signal power P _D ^ and the estimated interference power PI _^ . The value of the threshold value varies depending on the error target. It also changes depending on the packet length, so it is preferable to keep it in a state where it can be changed as appropriate. The relationship between the signal-to-interference power ratio and the error, and the threshold value will be explained again in a later example.

Note that the operation of the validity determination unit 87 needs to be completed before the estimated interference power P _I ^ is determined. Therefore, although not shown, some kind of time adjustment is required between the integrator 86 and the integrator 86. Further, the result of the CRC unit 7 can also be used to invalidate the estimation result. This will be explained again as a comparative example in the Examples described later.

The estimated interference power P _I ^ obtained as described above can be used to determine the chirp spreading factor of the next packet. For example, when the estimated interference power P _I ^ exceeds a predetermined value, the information collection station may transmit an instruction to the wireless terminal to increase the spreading factor. At this time, as shown in FIG. 7, the spreading factor may be expanded by increasing the number of chirp stages (from 4 to 8). This operation improves the SIR and reduces the occurrence of communication errors due to interference from other systems. However, as the spreading factor increases, the transfer speed decreases.

Note that when the number m of bits per symbol is large, k=2 ^m types of chirp patterns are required, and each chirp pattern becomes complicated. At this time, estimated values may differ for each chirp pattern. In such a case, the interference power may be estimated for each pattern and the spreading factor may be determined.

Examples of the present disclosure will be described below. In this example, the number of packets was 10,000, and the number of symbols (N) per packet was 1,024. Further, the spreading factor of chirp modulation was set to 1. The relationship between the estimation error (root mean square error (RMSE) between estimated interference power and true interference power) and signal-to-interference power ratio (SIR) was simulated and compared. As a comparative example, we also considered a case where only packets determined to be error-free by CRC were used.

Figure 8 shows the simulation results. In the figure, RMSE indicates the estimation error, and the smaller the value, the better the estimation accuracy has been achieved. The symbol “+” indicates that the SIR is estimated from all packets regardless of CRC, and the symbol “◇” indicates that estimation processing is performed only when CRC is successful (only when there is no error), and the symbol “・” indicates is the result when the transmitted bits are known on the receiver side, and is the lower bound for the current SIR estimation. From the figure, it was confirmed that when CRC is used, if the SIR is greater than 10 dB, estimation accuracy equivalent to the lower bound is achieved. However, when the SIR becomes 10 dB or less, errors occur in most packets, so the number of packets that can be used for estimation drastically decreases, making estimation difficult.

FIG. 9 shows the packet error rate results for each SIR. As shown in the figure, when the SIR is less than 10 dB, the PER is approximately 1.0 (all errors), and almost all packets cannot be used for SIR estimation. As a result, when CRC is used, estimation when the SIR is low becomes difficult. On the other hand, when CRC is not used, an RMSE of 0.1 or less is achieved if the SIR is 7 dB or more.

Therefore, in this embodiment, rather than using the CRC result, it is more preferable to remove estimated values with large RMSE (error) based on whether the estimated SIR is greater than or equal to a threshold (7 dB). Conceivable. Therefore, it can be seen that when estimation in a low SIR environment is required, a large estimation error can be avoided by comparing the estimated SIR with a predetermined threshold value to determine whether to adopt it as the SIR estimation value. In this embodiment, the threshold value is set to 7 dB, but since the threshold value differs depending on the target RMSE and also changes depending on the packet length, it is preferable to be able to change it as appropriate.

The present invention can be used not only for LPWA but also for communication systems after 5G, especially in fields related to IoT, such as connected cars, smart meters, wearable monitors, etc.

1 Antenna 2 Head amplifier 3 FFT units 4a to 4c Correlators 5a to 5c Memory 6 Maximum likelihood estimation unit
7 CRC unit 8 Power estimation unit 81a to 81c Gates 82a to 82c Gates 83a to 83c Decoder 85 Interference power estimation unit 87 Validity determination unit

Claims (12)

Interference involves receiving a radio wave containing a signal from an arbitrary wireless terminal in which transmission data is chirp-modulated with a predetermined spreading factor for each symbol belonging to a predetermined set, and estimating the power of the interfering radio wave from the radio wave. A power estimation device,
Each of the symbols is composed of 2 or more bits ,
a unit that converts the received radio wave signal into a reception spectrum signal;
A unit for preparing a chirp pattern for each symbol and estimating a symbol included in the transmission data by maximum likelihood by comparing a correlation value between the received spectrum signal and each chirp pattern;
The intensity of the interfering radio wave is determined by averaging the correlation values related to chirp patterns other than the maximum likelihood estimated symbols, or by selecting the maximum value of the correlation values related to chirp patterns other than the maximum likelihood estimated symbols. An interference power estimating device comprising a unit for estimating. The interference power estimating device according to claim 1, further comprising a unit that estimates the strength of a signal radio wave from a correlation value related to a chirp pattern of symbols estimated by maximum likelihood. 3. The interference power estimating device according to claim 1, wherein the transmission data is composed of a plurality of packets each including a predetermined number of the symbols. 4. The interference power estimating device according to claim 1, wherein the calculation processing of the correlation value is performed in parallel for each of the chirp patterns. The interference power estimating device according to any one of claims 1 to 4, further comprising a unit that enables estimation of the intensity of interference radio waves when a signal-to-interference power ratio is equal to or greater than a predetermined threshold. Receiving a radio wave containing a signal from an arbitrary wireless terminal in which transmission data is chirp-modulated with a predetermined spreading factor for each symbol belonging to a predetermined set, and estimating the power of the interfering radio wave from the radio wave. An interference power estimation program for performing
Each of the symbols is composed of 2 or more bits ,
converting the received radio wave signal into a reception spectrum signal;
A chirp pattern is prepared for each of the symbols, and a step of estimating the symbols included in the transmission data by maximum likelihood by comparing correlation values between the received spectrum signal and each chirp pattern;
The intensity of the interfering radio wave is determined by averaging the correlation values related to chirp patterns other than the maximum likelihood estimated symbols, or by selecting the maximum value of the correlation values related to chirp patterns other than the maximum likelihood estimated symbols. An interference power estimation program for executing the step of estimating the interference power. 7. The interference power estimating program according to claim 6, further causing the program to execute the step of estimating the strength of a signal radio wave from a correlation value related to a chirp pattern of a symbol estimated by maximum likelihood. 8. The interference power estimation program according to claim 6, wherein the transmission data is composed of a plurality of packets each including a predetermined number of the symbols. 9. The interference power estimating program according to claim 6, wherein the calculation process of the correlation value is executed within the transmission time of one symbol for each chirp pattern. The interference power estimation program according to any one of claims 6 to 9, for further executing the step of validating the estimation of the intensity of interference radio waves when the signal-to-interference power ratio is equal to or greater than a predetermined threshold. . An information gathering station comprising the interference power estimation device according to claim 1 to claim 5 or the interference power estimation program according to claim 6 to claim 10. The information gathering station according to claim 11, wherein when the intensity of the interference radio waves exceeds a predetermined value, an instruction to increase the spreading factor is transmitted to the wireless terminal.