Sliding window sub-channel detection method of self-adaptive threshold

Abstract

The invention discloses a sliding window sub-channel detection method of a self-adaptive threshold, and belongs to the field of wireless communication signal processing. According to the invention, by introducing an adaptive threshold mechanism, the problem of low detection flexibility caused by limitation of threshold setting is solved while the advantage of accurate signal arrival point detection performance of a traditional method is retained; an adaptive threshold detection algorithm based on autocorrelation has the advantage that a threshold can be adaptively updated along with the change of a signal-to-noise ratio, but in the implementation process, in order to realize total probability receiving, channels generally adopt a 50% aliasing non-blind area division mode, so that certain energy leakage of sub-channels is caused, and a detection result is influenced. When the signal-to-noise ratio is high, the threshold value is relatively high at the moment, the leakage energy can cause the result to be judged as the channel is effective, the detection result is misjudged, and the subsequent sub-channel processing is influenced, so that the problem is well solved by the method provided by the invention.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of wireless communication signal processing, in particular to the detection technology of sub-channels after digital channelization processing. BACKGROUND

[0002] In the development history of communication technology, the emergence of frequency hopping technology has improved the situation that traditional fixed frequency communication signals are easily intercepted and interfered, and has enhanced the confidentiality and reliability of communication. According to the working principle of frequency hopping communication, the carrier frequency will be discretely transformed according to the pre-set sequence rule when the transmitting end and the receiving end transmit signals. With the continuous development of tracking interference technology, the anti-interference requirements of frequency hopping communication system are also constantly improving. The higher the frequency hopping rate, the lower the probability of capturing the frequency hopping signal frequency, which can effectively reduce the impact of tracking interference on the frequency hopping system. Therefore, high frequency hopping communication has become the focus of current research.

[0003] In the design of high-speed frequency hopping receiver, the main technical problem is how the dehopping unit performs real-time and continuous dehopping on the frequency hopping signal. For a frequency hopping signal with extremely short duration, it is necessary to synchronously monitor and analyze all channels within the entire sampling bandwidth, and complete the receiving and demodulation. If the search speed of the receiver used for search or monitoring is not fast enough, the frequency hopping signal may be missed or lost, resulting in missed detection and leading to the failure to achieve full probability signal interception. To effectively solve this problem, some researchers have proposed a scheme of using a digital channelization receiver.

[0004] The digital channelization scheme is used to complete the reception of high-speed frequency hopping signals, which has good advantages in hardware implementation and real-time performance. However, after channelization processing, the signals are dispersed in each sub-channel, and the dispersion channel occupation result can be obtained by calculating the channel detection technology. The common traditional time domain channel detection methods include amplitude detection method and energy detection method, among which the amplitude detection method has the advantages of simple implementation and less hardware resource consumption; the energy detection algorithm is a kind of semi-blind detection method, which can be attributed to the non-coherent detection in signal detection. This method has the advantages of low complexity and easy implementation, and is widely used in the detection field. In addition, the adaptive threshold detection algorithm based on time domain autocorrelation has been mentioned and used in many documents in recent years, which has better detection performance compared with the traditional time domain detection algorithm.

[0005] The existing various sub-channel detection methods of digital channelization receiver still have certain practical application limitations, which are specifically shown as follows:

[0006] (1) Frequency domain detection method converts to frequency domain by implementing FFT transformation on input signal, and then completes signal detection by comparing each frequency spectrum amplitude with threshold. FFT operation has signal-to-noise ratio gain characteristics, and the method has good detection performance, but due to high algorithm complexity, high hardware implementation cost, and large delay after hardware implementation, it is difficult to promote the use.

[0007] (2) The traditional amplitude detection method is a simple time domain detection algorithm, and its implementation process is simple, and the hardware resources consumed are also relatively less, but due to the weak anti-noise ability of the method, the detection performance is poor under low signal-to-noise ratio conditions, and in today's practical application, the amplitude detection method is rarely used to complete channel detection.

[0008] (3) The energy detection method is a classical signal detection method, which detects signals according to the energy difference between noise and signals, and the double sliding window energy detection method can accurately judge the signal arrival time by collecting the peak value, but due to the existence of cross-channel signals in channelized reception, the energy proportion of different signals in the sub-channel is different, and the detection peak value also changes greatly, which cannot guarantee the detection performance under the same fixed detection threshold.

[0009] (4) The adaptive threshold detection method based on time domain autocorrelation can better adapt to the noise environment and has better detection performance because the adaptive threshold depends on the noise signal, and has been widely cited, but when the signal-to-noise ratio is high, the aliasing between sub-channels may cause the method to misjudge the channel detection result. SUMMARY

[0010] The application proposes a sliding window sub-channel detection method with adaptive threshold, aiming to accurately express the effective signal arrival time point and accurately estimate the sub-channel occupation of the arriving signal.

[0011] The technical scheme adopted by the application is as follows: a sliding window sub-channel detection method with adaptive threshold, which comprises the following steps:

[0012] Step 1: Channelized calculation is performed on the input received signal to obtain the output signals of a plurality of sub-channels;

[0013] Step 2: Calculate the sliding window energy accumulation results of the output signals of each sub-channel, including the front window energy accumulation value the rear window energy accumulation value and the full window energy accumulation value The subscript k is the sub-channel index, and N is the set sliding window length;

[0014] Step 3: Perform signal detection processing based on the sliding window energy values of each sub-channel:

[0015] Calculate the cumulative energy values ​​before and after the window. The difference is denoted as the energy difference between the two sliding windows.

[0016] Search for the full-window energy accumulation value of all sub-channels. The minimum value E in max and maximum value E min ;

[0017] The ratio E between the maximum and minimum values max / E min Compare with the preset value η, if E max / E min If ≤η, then a valid signal has been determined to have arrived at this time, and the detection threshold U is... R Update; if E max / E min If η < , then it is assumed that there is only pure noise signal at this time, and the threshold remains unchanged;

[0018] Energy difference between the two sliding windows Compared with the current detection threshold U R In comparison, when Greater than or equal to the detection threshold U R When a valid signal has arrived, the flag V is updated. k [n] is 1; otherwise, set V. k [n] is 0.

[0019] Furthermore, in step 1, a digital channelization design method based on polyphase filter discrete Fourier transform (DFT) is used to perform channelization calculation on the input received signal.

[0020] Furthermore, in step 3, the detection threshold U is... R The update method is: U R =E max / 4.

[0021] Furthermore, in step 3, the cumulative energy value of the front window... Rear window energy accumulation value Specifically:

[0022]

[0023] Among them, e k [i] represents the energy sample value of the i-th window position in the sliding window, where i represents the symbol position and k is the sub-channel index.

[0024] Furthermore, the preset value η is set to 4.

[0025] Furthermore, the sliding window length N is set to 16.

[0026] The technical scheme provided by the application brings at least the following beneficial effects:

[0027] Compared with the traditional double sliding window energy detection method, the application introduces an adaptive threshold mechanism, retains the advantages of the traditional method in signal arrival point detection performance accuracy, and solves the problem of low detection flexibility caused by threshold setting limitations; the adaptive threshold detection algorithm based on autocorrelation has the advantage that the threshold can be updated adaptively with the signal-to-noise ratio, but in the implementation process, in order to realize full probability reception, the channel usually adopts a 50% mixed blind area division method, which may cause a certain energy leakage in the sub-channel, and the detection result is affected, when the signal-to-noise ratio is high, the threshold value is high at this time, and the leakage energy will cause the result to be judged as a valid channel, resulting in detection result misjudgment, thereby affecting the subsequent sub-channel processing, and the method of the application better solves the problem. BRIEF DESCRIPTION OF DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiment description, and obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0029] Figure 1 The adaptive threshold sliding window detection method block diagram;

[0030] Figure 2 The adaptive threshold sliding window detection method channelization signal processing flowchart;

[0031] Figure 3 The digital channelization structure based on polyphase filtering DFT;

[0032] Figure 4 The double sliding window energy detection value and the adaptive threshold comparison chart;

[0033] Figure 5 The autocorrelation detection value and the adaptive threshold comparison chart;

[0034] Figure 6 The detection result missed detection probability comparison chart;

[0035] Figure 7 The detection result false alarm probability comparison chart. DETAILED DESCRIPTION

[0036] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described in detail and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Generally, the components of the embodiments of the present application described and shown in the drawings can be arranged and designed using different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not only to limit the scope of the claimed present application, but only to represent selected embodiments of the present application.

[0037] The embodiment of the present application provides a sliding window sub-channel detection method with adaptive threshold, which firstly performs channelization calculation on an input signal, then performs sliding window energy accumulation on a sub-channel signal, obtains a channel detection signal by calculating the difference between the front and rear sliding windows, simultaneously judges the signal arrival condition according to the energy accumulation value and adaptively adjusts the decision threshold, has the advantages of accurately expressing the effective signal arrival time point, combines the advantages of real-time updating of the adaptive threshold, and can realize accurate estimation of the sub-channel occupation condition of the arrival signal.

[0038] In one embodiment, referring to Figure 1 and Figure 2 , the sliding window sub-channel detection method with adaptive threshold provided by the embodiment of the present application specifically comprises:

[0039] (1) Firstly, the received wideband digital signal x[n] is subjected to channelization processing, and in the embodiment of the present application, a digital channelization design method based on polyphase filtering DFT is adopted to complete this step, and the structural schematic diagram is as shown in Figure 3 After the channelization processing, the time domain expression of the kth sub-channel signal y k [m] is obtained.

[0040]

[0041] Wherein, k takes the value of 1, 2, …, D, D is the number of decimation, p represents the pth group of polyphase decimation decomposition, M represents the decimation multiple of time domain channelization, x' p [m] represents the decimated pth data and the convolution result of the polyphase decimation decomposition pth branch filter coefficient, DFT[] represents discrete Fourier transform, h P [m] is a polyphase decomposition form of a prototype low-pass filter, and x p (m) is the decimation result of the input signal.

[0042] (2) The sliding window energy accumulation result of each sub-channel output signal (y k [m]) is calculated, which includes the front and rear window energy accumulation results with a length of N and a full window energy accumulation result with a length of 2N, wherein Energy accumulated value for front window, Energy accumulated value for back window, Energy accumulated value for whole window:

[0043]

[0044] wherein, N represents the length of the set sliding window, e k [i] represents the energy sample value of the i-th window position in the sliding window, i represents the symbol position, and k is the subchannel index.

[0045] Similarly, this step can also be recursive to reduce the amount of calculation, for example, the front window energy:

[0046]

[0047] (3) The energy window signal generated by the above steps is processed in two ways, wherein the front sliding window energy is subtracted from the back sliding window energy as the sliding window difference channel detection data corresponding to the channel:

[0048]

[0049] (4) For the whole window accumulated energy value Let the accumulated value corresponding to each channel be:

[0050]

[0051] Traverse E all [n], search for the maximum value and the minimum value:

[0052] E max = max{E allk [n]}, k = 0, 1,..., K-1

[0053] E min = min{E allk [n]}, k = 0, 1,..., K-1

[0054] Find the maximum value and the minimum value ratio E max / E min , and compare it with a fixed value η (preset value). If E max / E min ≤ η, it is determined that the effective signal arrives at this time, and the threshold is updated; if E max / E min > η, it is considered that only pure noise signal exists at this time, and the threshold remains unchanged:

[0055]

[0056] In this embodiment, η is set to 4, and it can also be set to other values based on actual application scenarios.

[0057] The principle of the threshold updating rule in this step in the embodiment is as follows: because of the channel division overlap, when the signal-to-noise ratio condition is good, if the noise energy threshold is used for judgment, the existence of the overlap signal will affect the judgment result, and the double sliding window energy difference value has the advantage that the peak value of the signal arrival time detection is significant, so that the peak value point of the signal energy threshold can be accurately captured. Therefore, the threshold is updated based on the signal energy, and the detection threshold is also adaptively changed with the change of the signal energy, so that the signal arrival peak value can be effectively detected, and the misjudgment influence caused by the overlap signal can be eliminated. Compared with the traditional sliding window detection method, the adaptive threshold mechanism is introduced, the effective signal arrival time is roughly estimated by comparing the full window energy value, the threshold is updated after the arrival, the threshold is changed based on the channel effective signal energy, and the detection flexibility of the sliding window detection method is increased.

[0058] (5) The double sliding window energy difference value calculated is compared with the updated threshold value U R When the threshold is exceeded, it is determined that the effective signal arrives and the flag V k [n] is updated, that is:

[0059]

[0060] The obtained flag result will be transmitted to the subsequent peak value judgment module for further processing, so as to judge whether the current arrival signal is a single channel signal or a cross channel signal, and facilitate subsequent sub-channel signal processing.

[0061] The detection performance of the method provided in the embodiment of the application is further verified through experiments:

[0062] The sampling rate is set to 200 MHz, the baseband signal is 10 Mbit / s, the narrowband signal with a bandwidth of 12.5 MHz and a sampling rate of 200 MHz is generated through cosine filter and up-sampling filter, 10000 hops / s is selected as the experimental simulation frequency hopping rate, the sub-channel bandwidth is consistent with the baseband signal bandwidth, 62.5 MHz, 18.75 MHz, 81.25 MHz and 12.5 MHz are selected as the frequency hopping frequency points, and each occupies channel 6, channel 2&3, channel 7&8 and channel 2.

[0063] Under the noise environment with a signal-to-noise ratio SNR of 15 dB, the sliding window length N is 16, the adaptive threshold is changed to a rough estimation ratio η of 4, the initial value of the threshold is set according to the accumulated calculation value of the cross channel signal energy, and the detection method provided in the design is used for sub-channel detection of channel 2. The double sliding window energy detection value and the adaptive threshold change are as shown in Figure 4 .

[0064] It can be found that the threshold does not change in the time period only with noise, and the initial value of the threshold is large enough to effectively filter out the influence of noise; when the first hop single channel signal arrives, the threshold detection value changes adaptively with the energy accumulation value of the arriving signal. Since the threshold value is based on the change of effective signal energy at this time, its value is much larger than the aliasing signal energy, so the influence of channel aliasing signal can be effectively eliminated; since the detection threshold will be changed in real time according to the maximum value of the current channel energy, when the effective signal changes from single channel to cross channel, the arrival point of the current effective channel signal can also be successfully judged, thereby solving the missed detection problem caused by the fixed double sliding window threshold.

[0065] Under the same input signal condition, the detection value of the existing commonly used adaptive threshold detection method based on autocorrelation signal is compared with the threshold as shown in Figure 5 It can be found that in the detection value threshold comparison graph of channel 2, in the time period of 100 μs-200 μs, the detection value generated by the autocorrelation of the channel aliasing signal is obviously higher than the threshold value, which will bring a large influence to the channel decision result.

[0066] The performance of the commonly used adaptive threshold detection method based on time domain autocorrelation and the method proposed in the embodiment of the application is compared and analyzed, and the detection results under different signal-to-noise ratios are observed. If a point exceeding the threshold is found in the expected time period or subchannel, it is considered that a false alarm occurs; if no corresponding effective channel monitoring result is found in the expected time period, it is considered that a missed detection occurs. The input frequency hopping detection signal parameters and channelization parameters are the same as above, the SNR value is selected between -5 dB and 15 dB, and the step is 1 dB. The simulation results show that the performance comparison of the missed detection probability and the false alarm probability is as shown in Figure 6 and Figure 7 It can be found that the adaptive threshold sliding window detection method has better performance in both the missed detection probability under low signal-to-noise ratio and the false alarm probability under high signal-to-noise ratio than the time domain autocorrelation adaptive threshold detection algorithm, because in the low signal-to-noise ratio condition, the difference operation in the sliding window energy detection algorithm can make the detection signal have more obvious peak characteristics, and in the high signal-to-noise ratio condition, the sliding window can filter out the influence of aliasing signals by using the threshold based on signal energy, so the false alarm rate will also be better.

[0067] The present application is directed to the sub-channel detection problem after digital channelization receiving processing. The traditional amplitude detection method is usually to take the complex signal data of each sub-channel after the channelization analysis filter output as I and Q, and then calculate the result by complex modulus, and compare the result with a fixed threshold. The fixed threshold can be calculated according to the formula of noise variance and false alarm probability, but the defect is that the anti-noise ability is poor, and the detection performance is poor under low signal-to-noise ratio conditions. The sliding window sub-channel detection method with adaptive threshold proposed in the embodiment of the present application has the advantage of high detection accuracy of the arrival time of effective signal, but due to the limitation of the fixed threshold, the flexibility of sub-channel detection is not enough when the signal energy and noise environment change. In the embodiment of the present application, the detection threshold is updated by adding an adaptive threshold design based on the rough estimated arrival time of the effective signal, and the update value depends on the signal energy, which ensures the effectiveness of energy detection of the sliding window under different channel occupation conditions.

[0068] The adaptive threshold detection algorithm based on time domain autocorrelation estimates the noise energy adaptive threshold before the arrival of the signal through rough estimation of the signal arrival, and obtains the signal detection result by comparing the autocorrelation calculation result with the noise threshold. However, due to the certain inter-channel aliasing in the design of digital channelization, when the aliasing signal energy is higher than the noise threshold, the detection result is misjudged. The method proposed in the embodiment of the present application uses signal energy as the threshold reference value, which is much higher than the aliasing signal energy level, and can effectively eliminate the aliasing signal influencing factor.

[0069] In the method proposed in the embodiment of the present application, the energy of the sub-channel signal after channelization processing is calculated first, and then the energy accumulation difference of the front and rear windows is calculated, that is, the channel detection value can be obtained. The sum of the overall energy accumulation result is calculated, the maximum and minimum values are searched by traversing the whole sub-channel result, the adaptive threshold is updated by comparing the maximum and minimum value ratio with the fixed threshold, and the detection result can be obtained by comparing the sub-channel detection value with the adaptive threshold value. Compared with other methods, this method has better arrival time detection accuracy, flexibility and avoidance of aliasing signal influence.

[0070] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

[0071] The above merely describes some embodiments of the present application. For those skilled in the art, without departing from the concept of the present application, several modifications and improvements can be made, which are within the protection scope of the present application.

Claims

1. An adaptive threshold sliding window channel detection method, characterized in that, Includes the following steps: Step 1: Perform channelization calculation on the input received signal to obtain the output signals of multiple sub-channels; Step 2: Calculate the sliding window energy accumulation result for the output signal of each sub-channel, including the front window energy accumulation value. Rear window energy accumulation value And the total energy accumulation of the window In the subscript, k is the subchannel index, and N is the set sliding window length; Step 3: Perform signal detection processing based on the sliding window energy values ​​of each sub-channel. Calculate the cumulative energy values ​​before and after the window. The difference is denoted as the energy difference between the two sliding windows. Search for the full-window energy accumulation value of all sub-channels. The minimum value E in max and maximum value E min ; The ratio E between the maximum and minimum values max / E min Compare with the preset value η, if E max / E min If ≤η, then a valid signal has been determined to have arrived at this time, and the detection threshold U is... R Update; if E max / E min If η < , then it is assumed that there is only pure noise signal at this time, and the threshold remains unchanged; Energy difference between the two sliding windows Compared with the current detection threshold U R In comparison, when Greater than or equal to the detection threshold U R When a valid signal has arrived, the flag V is updated. k [n] is 1; otherwise, set V. k [n] is 0.

2. The method as described in claim 1, characterized in that, In step 1, a digital channelization design method based on polyphase filter discrete Fourier transform (DFT) is used to perform channelization calculation on the input received signal.

3. The method as described in claim 1, characterized in that, In step 3, the detection threshold U is set. R The update method is: U R =E max / 4.

4. The method as described in claim 1, characterized in that, In step 3, the cumulative energy value of the front window Rear window energy accumulation value Specifically: Among them, e k [i] represents the energy sample value of the i-th window position in the sliding window, where i represents the symbol position and k is the sub-channel index.

5. The method according to any one of claims 1 to 4, characterized in that, The default value η is set to 4.

6. The method according to any one of claims 1 to 4, characterized in that, The sliding window length N is set to 16.

Images