Frequency spectrum occupancy rate calculation method based on full-band frequency spectrum data

Abstract

The invention relates to a spectrum occupancy rate calculation method based on full-band spectrum data, and belongs to the field of radio signal spectrum analysis. According to the invention, based on full-band spectrum data at the same sampling time, frequency band division is carried out, and a full-band signal is divided into a plurality of sub-bands; calculating level thresholds of all sub-bands based on the full-band spectrum data at the primary sampling moment; and estimating the occupancy rate based on the frequency band spectrum data at different sampling moments within the measurement time. According to the method, a plurality of interval thresholds in each sub-band are obtained by means of the histogram, the between-class variance and the like, the method can be used for an abnormal electromagnetic signal monitoring scene, the abnormal signal alarm false alarm rate is further reduced, and the electromagnetic space safety supervision capability is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of radio signal spectrum analysis, and particularly relates to a spectrum occupancy calculation method based on full-band spectrum data. BACKGROUND

[0002] There are various types and sources of electromagnetic signals in the electromagnetic environment, and the spectrum occupancy can be used as an important indicator for evaluating the complexity of the electromagnetic environment. Higher spectrum occupancy usually means that the electromagnetic environment is more complex, and the possibility of mutual interference between signals is larger, which has an important influence on the normal operation of electronic equipment, the reliability of wireless communication systems, and electromagnetic compatibility design. By long-term monitoring of the spectrum occupancy, the trend of the electromagnetic environment can be grasped, and a basis can be provided for the evaluation and management of the electromagnetic environment. First, in the aspect of abnormal signal monitoring, some illegal frequency modulation broadcasts interfere with the very high frequency communication system (118MHz-137MHz) of civil aviation, which poses a hidden danger to aviation safety. The existence of some "pseudo base stations" interferes with normal mobile communication. By performing spectrum analysis on the signals received at the receiving end, the illegal frequency components that do not belong to the normal communication range can be captured under the condition that the legal frequency information is known, and the purpose of reasonably distinguishing illegal signals can be achieved by comparing the spectrum of illegal signals with that of legal signals, thereby providing a theoretical basis for monitoring and positioning illegal signals.

[0003] At present, the spectrum monitoring method based on the energy monitoring principle, the value of the noise threshold has a very important influence on the evaluation of the spectrum occupancy. If the noise threshold is too low, the frequency band will be occupied all the time due to the existence of environmental noise. On the other hand, selecting a high noise decision threshold may lead to underestimation of the spectrum occupancy. Therefore, the value of the noise threshold is very important in spectrum monitoring. In the existing spectrum occupancy estimation method, the ITU recommendation mentions that only 20% of the samples with the lowest level are used to calculate the average root mean square level as the noise level threshold, however, this method may result in a too low noise level, which may lead to some channels that are not actually occupied being detected as "occupied", resulting in a high false alarm rate.

[0004] The application is used for the occupancy problem of the full-band spectrum data pushed by the electromagnetic monitor in real time. Firstly, the full-band spectrum is grouped, and the dynamic threshold statistics are performed according to the business frequency band, which can be used in the abnormal electromagnetic signal monitoring scene, further reduces the false alarm rate of abnormal signal alarm, and improves the electromagnetic space safety supervision capability. SUMMARY

[0005] (I) Technical problems to be solved

[0006] The technical problem to be solved by the application is to provide a spectrum occupancy calculation method based on full-band spectrum data, so as to solve the problem of high false alarm rate caused by the existing spectrum occupancy estimation method.

[0007] (II) Technical Solution

[0008] To solve the above technical problems, the present application provides a spectrum occupancy calculation method based on full-band spectrum data, which comprises the following steps:

[0009] S1, based on the full-band spectrum data at the same sampling time, the frequency band is divided, and the full-band signal is divided into several sub-frequency bands;

[0010] S2, based on the full-band spectrum data at the same sampling time, the level threshold of all sub-frequency bands is calculated

[0011] S21, generate all sub-frequency band spectrum data histogram

[0012] For the T n th acquisition of full-band data, all frequency points in the A n th sub-frequency band range are arranged in ascending order according to the power value and divided into R power intervals, and the number of frequency points falling in each power interval is counted;

[0013] S22, calculate the level threshold of all sub-frequency bands

[0014] Traverse each power interval of the sub-frequency band spectrum data histogram, select the median value of each power interval as a possible threshold value of the sub-frequency band, and each sub-frequency band has R possible threshold values. The frequency points in the sub-frequency band whose power value is less than or equal to the possible threshold value are noise, and the frequency points whose power value is greater than the possible threshold value are signals. Calculate the inter-class variance of the sub-frequency band spectrum data under each possible threshold value, so that the possible threshold value with the maximum inter-class variance is the level threshold value of the sub-frequency band. Traverse all sub-frequency bands of single sampling full-band spectrum data to obtain all sub-frequency band level threshold values;

[0015] S3, estimate the occupancy based on the frequency band spectrum data at different sampling times within the measurement time

[0016] According to the full-band spectrum data collected by the monitor at different times, the sub-frequency band level threshold values are calculated respectively, the power value of any frequency point is compared with the corresponding sub-frequency band level threshold value, and whether the frequency point exists signal is counted. When there is a signal, the number of times the signal exists is accumulated by 1, otherwise it is 0. Under the T n group full-band data, the ratio of the number of times the signal exists to T n is the time occupancy of the frequency point.

[0017] In the S1, the spectrum data in the full-band [9k-6G] Hz collected by the monitor at the T n th time is divided into A n sub-frequency bands, marked as [1, 2..., A n ].

[0018] Wherein, the S21, the sub-band spectrum data power interval calculation formula is as follows:

[0019]

[0020] Wherein N is the number of frequency points in the sub-band, R is the number of histogram intervals.

[0021] Wherein, the S21 includes: assuming that the sampling in the sub-band obtains N frequency points, each frequency point has a power value, the power value of the i-th frequency point is represented as P(i), wherein i=1, 2, …, i, … N; the power value of the sub-band with length N is arranged in ascending order to obtain a 1×N dimension vector P=[P(1), …, P(i), …, P(N)]; the power range of the sub-band is evenly divided into R parts, the number of frequency points falling into each power interval is counted, and then the sub-band histogram with R power intervals is obtained.

[0022] Wherein, N=65536, R∈[256, 512].

[0023] Wherein, in the S22, assuming that the T n th sampling of the A n th sub-band in the k power interval has a possible threshold M k , R power intervals have R possible thresholds M=[M1,…,M k ,…,M R ], k=[1,…,R];

[0024] After ascending arrangement in the sub-band, the position sequence number of M k in the sub-band is represented as j, all frequency points less than or equal to M k in the sub-band are noise, and their power values are [P(1), …, P(j)], all frequency points greater than M k are signals, and their power values are [P(j+1), …, P(N)];

[0025] The cumulative probability and mean value of the signal and noise in the sub-band are calculated respectively:

[0026] The cumulative probability w noise (M k ) and the mean value μ noise (M k ) of the noise are calculated, and the formula is as follows

[0027]

[0028] The cumulative probability w signal (M k ) and the mean value μ signal (Mk ), the formula is as follows

[0029]

[0030] Calculate the inter-class variance of the sub-band under the possible threshold The formula is as follows:

[0031]

[0032] Wherein μ G is the global mean of the spectrum data, M k is the possible threshold;

[0033] Calculate the inter-class variance of all sub-bands of the T n th full-band data under the R possible thresholds

[0034] In the S22, compare the inter-class variances of the sub-bands under different possible thresholds of all sub-bands The threshold M k that makes the inter-class variance maximum is the level threshold of the sub-band:

[0035]

[0036] In the S22, suppose that the number of all sub-bands of the T n th full-band spectrum data is A n , then traverse the inter-class variances under all possible thresholds of the sub-bands to obtain the level threshold of all sub-bands of the T n th full-band spectrum data:

[0037] In the S3, suppose that the spectrum occupancy degree statistical measurement duration is S seconds, and the electromagnetic monitoring device uploads full-band data once every s seconds, so that a total of T n = S / s sets of full-band data are uploaded in the statistical time; for the T n th set of full-band spectrum data, each set of full-band spectrum data is divided into A n sub-bands, and the threshold of all sub-bands is calculated as

[0038]

[0039] In the S3, for the T n th set of full-band data, the number of times that the signal of each frequency point exists is accumulated as L, and then the spectrum occupancy degree is

[0040]

[0041] (Three) beneficial effects

[0042] The application provides a spectrum occupation degree calculation method based on full-band spectrum data. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 The application provides a spectrum occupation degree calculation method based on full-band spectrum data.

[0044] Figure 2 The application provides a spectrum occupation degree calculation method based on full-band spectrum data.

[0045] Figure 3 The application provides a spectrum occupation degree calculation method based on full-band spectrum data. DETAILED DESCRIPTION

[0046] In order to make the objects, contents and advantages of the application clearer, the specific embodiments of the application are further described in detail below in combination with the drawings and examples.

[0047] The application discloses a spectrum occupation degree calculation method based on full-band spectrum data.

[0048] The application provides a spectrum occupation degree calculation method based on full-band spectrum data.

[0049] The application provides a spectrum occupation degree calculation method based on full-band spectrum data.

[0050] (1) based on the full-band spectrum data of the same sampling time, the frequency band is divided

[0051] The main types of indoor electromagnetic space signal distribution include broadcast television, trunked intercom, WIFI, Bluetooth, 4 / 5G mobile communication, RFID, etc. Based on the frequency band distribution range of mainstream wireless signals published by the Radio Regulatory Commission, the full frequency band signal is divided into several sub-frequency bands.

[0052] (2) Calculate the level thresholds of all sub-bands based on the full-band spectrum data at a single sampling time.

[0053] 1) Generate histograms of spectrum data for all sub-bands.

[0054] For the Tth n The full-band data collected in the second acquisition will be the Ath n All frequency points within a sub-band range are sorted in ascending order of power value and divided into R power intervals. The number of frequency points falling within each power interval is counted.

[0055] The number of power intervals is a key parameter in sub-band histogram generation. Too many power intervals can mask the true spectral distribution trend, such as bimodalities or skewness; too few power intervals can lead to the aggregation of key features. The formula for calculating the power interval of sub-band spectral data is as follows:

[0056]

[0057] Where N is the number of frequency points within the sub-band, and R is the number of histogram intervals.

[0058] 2) Calculate the level threshold for all sub-bands.

[0059] By traversing each power interval of the sub-band spectrum histogram, and considering that radio signal spectral power values ​​are typically affected by noise or signal strength fluctuations, the median value of each power interval is selected as a possible threshold for that sub-band. Thus, each sub-band has n possible thresholds. Frequency points within a sub-band whose power values ​​are less than or equal to a possible threshold are considered noise, and frequency points whose power values ​​are greater than a possible threshold are considered signals.

[0060] Calculate the inter-class variance of the sub-band spectral data for each possible threshold. The threshold with the largest inter-class variance is the level threshold for that sub-band. By traversing all sub-bands of the single-sample full-band spectrum data, the level thresholds of all sub-bands can be obtained.

[0061]

[0062] Where argmax() refers to taking the inter-class variance. Possible threshold M for maximizing the value k The value of .

[0063] (3) Based on the frequency spectrum data of different sampling time in the measurement time, the occupation degree is estimated

[0064] The level threshold of the full-band spectrum data of different sampling time changes dynamically, so the level threshold of the sub-band should be calculated respectively based on the full-band spectrum data collected by the monitor at different time.

[0065] Suppose the spectrum occupation degree statistical measurement time is S seconds, and the electromagnetic monitoring equipment uploads full data every s seconds, then the total number of uploads T n = S / s times full-band data. For T n times full-band spectrum data, each group of full-band spectrum data is divided into A n sub-bands, and the threshold M * of all sub-bands in the occupation degree statistical time is calculated.

[0066]

[0067] Among them, is the full-band data of the T n th sampling, and the level threshold of the A n th sub-band.

[0068] Compare the power value of any frequency point with the corresponding sub-band threshold, and count whether there is a signal at the frequency point. When there is a signal, the number of times the signal exists at the frequency point is accumulated by 1, otherwise it is 0. Under T groups of full-band data, the ratio of the number of times the signal exists to T n is the time occupation degree of the frequency point.

[0069] For T n times full-band data, the number of times the signal exists at each frequency point is accumulated to L, and the spectrum occupation degree is

[0070]

[0071] Example 1:

[0072] S1, based on the full-band spectrum data of the same sampling time, the frequency band is divided, and the full-band signal is divided into several sub-bands

[0073] The main types of indoor electromagnetic space signal distribution include broadcast television, group talk, WIFI, Bluetooth, 4 / 5G mobile communication, RFID, etc. According to the frequency band distribution range of the mainstream wireless signal published by the Radio Committee, the full-band signal is divided into several sub-bands.

[0074] In specific implementation, different signal types occupy different frequency ranges, such as computer video leakage signals mainly occupy the megahertz frequency range; ultra-high frequency broadcast television signals mainly occupy the 470-560 MHz frequency range; WIFI and Bluetooth signals are concentrated in the ISM frequency range; and mobile communication frequency bands mainly occupy the 2-5 GHz frequency range. In implementation, the spectrum data in the full frequency band [9k-6G] Hz collected by the monitor at the T n th time is divided into A n sub-frequency bands, marked as [1, 2..., A n ], as shown in Figure 2 .

[0075] S2, based on the full frequency band spectrum data at the sampling time, calculate the level threshold of all sub-frequency bands S21, generate the histogram of all sub-frequency band spectrum data.

[0076] For the full frequency band data collected at the T n th time, all frequency points in the A n th sub-frequency band range are arranged in ascending order according to the power value, and then divided into R power intervals, and the number of frequency points falling in each power interval is counted.

[0077] For the full frequency band data collected at the T n th time, all frequency points in the A n th sub-frequency band range are processed, and each sub-frequency band in the full frequency band data obtained at each sampling time is processed in this way.

[0078] The number of power intervals is a key parameter for generating a sub-frequency band histogram. If the histogram power interval is too many, the real distribution trend of the spectrum such as bimodal and skewness will be hidden; if the histogram power interval is too few, the key features will be aggregated. The power interval calculation formula of the sub-frequency band spectrum data is as follows:

[0079]

[0080] Where N is the number of frequency points in the sub-frequency band, and R is the number of histogram intervals.

[0081] In practice, assuming N frequency points are sampled within a sub-band, each with a power value, the power value of the i-th frequency point is represented as P(i), where i = 1, 2, ..., i, ... N. Arranging the sub-band spectral power values ​​of length N in ascending order yields a 1×N dimension vector P = [P(1), ..., P(i), ..., P(N)]. Assuming N = 65536, the number of power intervals in the sub-band histogram ranges from R ∈ [256, 512]. R can take any value within this range, such as R = 256. Dividing the sub-band power range into 256 equal parts and counting the number of frequency points falling into each power interval, a sub-band histogram with 256 power intervals is obtained.

[0082] S22. Calculate the level threshold for all sub-bands.

[0083] By traversing each power interval of the sub-band spectrum histogram, and considering that radio signal spectral power values ​​are typically affected by noise or signal strength fluctuations, the median value of each power interval is selected as a possible threshold for that sub-band. Thus, each sub-band has R possible thresholds. Frequency points within a sub-band with power values ​​less than or equal to the possible threshold are considered noise, and frequency points with power values ​​greater than the possible threshold are considered signals. The inter-class variance of the sub-band spectrum data under each possible threshold is calculated, and the possible threshold with the largest inter-class variance is the level threshold for that sub-band. By traversing all sub-bands of a single sample of the full-band spectrum data, the level thresholds for all sub-bands can be obtained.

[0084] In practical implementation, assume that the Tth n The Ath sampling n The possible threshold for the k-th power interval of each sub-band is M. k There are R possible thresholds M = [M1, ..., Mn] for R power intervals. k ,…,M R ], k = [1,…,R]. After sorting in ascending order within the sub-bands, M k The position index within this sub-band is denoted as j, and all sub-bands less than or equal to M are represented by this index. k The frequency point is noise, and its power value is [P(1),…,P(j)], all of which are greater than M. k The frequency point is the signal, and its power value is [P(j+1),…,P(N)]. Calculate the cumulative probability and mean of the signal and noise in each sub-band.

[0085] The cumulative probability w of noise noise (M k ) and mean μ noise (M k The formula is as follows:

[0086]

[0087] Cumulative probability w of signal signal (M k ) and mean μ signal (M k ), as follows

[0088]

[0089] Calculate the inter-class variance of the sub-band under the possible threshold As follows:

[0090]

[0091] Where μ G is the global mean of the spectrum data, M k is the possible threshold.

[0092] Iterate to calculate the inter-class variance of all sub-bands R of the T n th full-band data under the possible threshold

[0093] Compare the inter-class variances of the sub-band under different possible thresholds of all sub-bands The threshold M k that maximizes the inter-class variance is the level threshold of the sub-band.

[0094]

[0095] Assuming that the number of all sub-bands of the T n th full-band spectrum data is A n , iterating the inter-class variance under all possible thresholds of the sub-band can obtain the level threshold of all sub-bands of the T n th full-band spectrum data as

[0096] S3, based on the spectrum data of the frequency band at different sampling times within the measurement time, estimate the occupation degree

[0097] The level threshold of the full-band spectrum data at different sampling times dynamically changes, so the level threshold of the sub-band should be calculated according to the full-band spectrum data collected by the monitor at different times.

[0098] Assuming that the spectrum occupation degree statistical measurement time is S seconds, and the electromagnetic monitoring equipment uploads full data every s seconds, then the total number of full-band data uploaded within the statistical time is T n = S / s. For T n times of full-band spectrum data, each group of full-band spectrum data is divided into A n sub-bands, and the threshold of all sub-bands is calculated as

[0099]

[0100] Compare the power value of any frequency point with the corresponding sub-band level threshold, and count whether the signal exists in the frequency point. When the signal exists, the signal existing times of the frequency point is accumulated by 1, otherwise, it is 0. n The ratio of the number of times that the signal exists to T n is the time occupation degree of the frequency point.

[0101] For T n full-band data, the signal existing times of each frequency point is accumulated as L, and the spectrum occupation degree is

[0102]

[0103] Embodiment 2:

[0104] In the specific implementation, the system collects 1 hour of full-band spectrum data, and the spectrum data is uploaded every 1 second, a total of 3600 times, T n = 3600. Assuming that the frequency point is a 9 kHz spectrum located in the first sub-band of the full-band data, the position sequence number is 1, and the power value is P(1) = -40 dBm, for the first sampled full-band data threshold M 11 * = -30 dBm, then for this sampling, the frequency point has no signal existence; for the second sampled full-band data threshold M 21 * = -50 dBm, then for this sampling, the frequency point has signal existence, L = 1; in this way, finally L = 1800 is obtained, and the time occupation degree of the frequency point of 9 kHz is 50%.

[0105] The present application proposes a spectrum occupation degree calculation method based on full-band spectrum data. For the full-band spectrum data occupation degree problem of the real-time push of the electromagnetic monitor, first, the full-band is grouped, the inter-class variance of the sub-band spectrum data under each possible threshold is dynamically counted based on the business frequency band, and then the occupation degree is estimated based on the frequency band spectrum data at different sampling times within the measurement time. The present application obtains a plurality of interval thresholds in each sub-band through histogram, inter-class variance and other means, which can be used in abnormal electromagnetic signal monitoring scenarios, further reduces the false alarm rate of abnormal signal alarm, and improves the electromagnetic space safety supervision ability.

[0106] The above only describes the preferred embodiments of the present application, and it should be noted that for ordinary skilled persons in the technical field, without departing from the technical principles of the present application, a number of improvements and modifications can be made, and these improvements and modifications should be considered as the protection scope of the present application.

Claims

1. A method for calculating spectrum occupancy based on full-band spectrum data, characterized in that, The method includes the following steps: S1. Based on the full-band spectrum data at the same sampling time, divide the frequency band into several sub-bands; S2. Based on the full-band spectrum data at a single sampling time, calculate the level thresholds for all sub-bands. S21. Generate histograms of spectrum data for all sub-bands. For the Tth n The full-band data collected in the second acquisition will be the Ath n All frequency points within a sub-band range are sorted in ascending order of power value and divided into R power intervals. The number of frequency points falling within each power interval is counted. S22. Calculate the level threshold for all sub-bands. Traverse each power interval of the sub-band spectrum data histogram, and select the median value of each power interval as a possible threshold for that sub-band. Each sub-band has a total of R possible thresholds. Frequency points with power values ​​less than or equal to the possible thresholds within a sub-band are considered noise, and frequency points with power values ​​greater than the possible thresholds are considered signals. Calculate the inter-class variance of the sub-band spectrum data under each possible threshold. The possible threshold with the largest inter-class variance is the level threshold for that sub-band. Traverse all sub-bands of the single-sample full-band spectrum data to obtain the level thresholds for all sub-bands. S3. Estimate occupancy based on frequency band spectrum data at different sampling times within the measurement period. Based on the full-band spectrum data collected by the monitor at different times, the sub-band level thresholds are calculated respectively. The power value of any frequency point is compared with the corresponding sub-band level threshold to determine whether there is a signal at that frequency point. If a signal is present, the count of occurrences of that signal at that frequency is incremented by 1; otherwise, it is incremented by 0. n Under the full-band data set, the number of times the signal exists and T n The ratio of is the time occupancy of that frequency point.

2. The method for calculating spectrum occupancy based on full-band spectrum data as described in claim 1, characterized in that, In S1, the Tth monitoring unit... n The spectrum data collected in this acquisition, covering the entire frequency band [9kHz-6GHz], is divided into A... n Each sub-band is labeled [1,2...,A]. n ].

3. The method for calculating spectrum occupancy based on full-band spectrum data as described in claim 1, characterized in that, In step S21, the formula for calculating the power range of sub-band spectrum data is as follows: Where N is the number of frequency points within the sub-band, and R is the number of histogram intervals.

4. The method for calculating spectrum occupancy based on full-band spectrum data as described in claim 1, characterized in that, S21 includes: assuming that N frequency points are sampled within the sub-band, each frequency point has a power value, and the power value of the i-th frequency point is represented as P(i), where i = 1, 2, ..., i, ... N; after arranging the sub-band spectrum power values ​​of length N in ascending order, a 1×N dimension vector P = [P(1), ..., P(i), ..., P(N)] is obtained; the power range of the sub-band is divided into R equal parts, and the number of frequency points falling into each power interval is counted, thus obtaining a sub-band histogram with R power intervals.

5. The method for calculating spectrum occupancy based on full-band spectrum data as described in claim 4, characterized in that, N=65536, R∈[256,512].

6. The method for calculating spectrum occupancy based on full-band spectrum data as described in claim 4, characterized in that, In S22, it is assumed that the Tth n The Ath sampling n The possible threshold for the k-th power interval of each sub-band is M. k There are R possible thresholds M = [M1, ..., Mn] for R power intervals. k ,…,M R ], k = [1, ..., R]; After sorting in ascending order within the sub-bands, M k The position index within this sub-band is denoted as j, and all sub-bands less than or equal to M are represented by this index. k The frequency point is noise, and its power value is [P(1),…,P(j)], all of which are greater than M. k The frequency point is the signal, and its power value is [P(j+1),…,P(N)]; Calculate the cumulative probability and mean of signal and noise within each sub-band: The cumulative probability w of noise noise (M k ) and mean μ noise (M k The formula is as follows: The cumulative probability w of the signal signal (M k ) and mean μ signal (M k The formula is as follows: Calculate the inter-class variance of the sub-band at this possible threshold. The formula is as follows: Where μ G M is the global mean of the spectrum data. k It is a possible threshold; Traverse and calculate the Tth n Inter-class variance of all sub-bands of the sub-full-band data under R possible thresholds 7. The method for calculating spectrum occupancy based on full-band spectrum data as described in claim 6, characterized in that, In step S22, the inter-class variance of sub-frequency bands under different possible thresholds is compared. The threshold M that maximizes the inter-class variance k The level threshold for this sub-frequency band:

8. The method for calculating spectrum occupancy based on full-band spectrum data as described in claim 7, characterized in that, In S22, it is assumed that the Tth n The number of all sub-bands in the sub-full-band spectrum data is A. n Then, by iterating through all possible threshold values ​​for all sub-bands, we can obtain the Tth class variance. n The level thresholds for all sub-bands of the sub-full-band spectrum data are:

9. The method for calculating spectrum occupancy based on full-band spectrum data as described in claim 8, characterized in that, In S3, assuming the spectrum occupancy statistical measurement time is S seconds, and the electromagnetic monitoring device uploads full data every s seconds, then a total of T data will be uploaded within the statistical time. n = S / s full-band data; for T n Sub-full-band spectrum data, each group of full-band spectrum data is divided into A n For each sub-band, calculate the threshold for all sub-bands as follows:

10. The method for calculating spectrum occupancy based on full-band spectrum data as described in claim 9, characterized in that, In S3, for T n For sub-full-band data, the number of times the signal exists at each frequency point is accumulated to L, then the spectrum occupancy is...

Images