WO2013104281A1 - Method and system for spectrum sensing - Google Patents

Abstract

Disclosed are a method and system for spectrum sensing. The method comprises: a cognitive radio system receiving a signal during one or multiple discrete quiet periods, and processing the signal received to acquire a baseband signal; the cognitive radio system acquiring a detection statistic on the basis of the baseband signal, and determining the result of a spectrum sensing on the basis of the detection statistic. In embodiments of the present invention, the method for spectrum sensing allows for acquisition of a robust detection performance in an environment having certain interferences, for improved detection performance, for completion of the spectrum sensing within a shortened time frame, thus conserving the time overhead for detection, and for completion of spectrum sensing detection during the multiple discrete quiet periods, thus for increased flexibility in terms of applicability in various cognitive radio systems.

Description

 A spectrum sensing method and system The present application claims priority to Chinese Patent Application No. 201210008725.6, entitled "A Spectrum Sensing Method and System", filed on January 12, 2012, the entire contents of which is incorporated herein by reference. This is incorporated herein by reference. Technical field

 The present invention relates to the field of communications technologies, and in particular, to a spectrum sensing method and system. Background technique

 With the rapid development of mobile communication industry, the contradiction between the growing demand for broadband wireless communication and limited spectrum resources is becoming more and more obvious; and with the rapid development of mobile services in the future, telecom operators will face more severe shortage of spectrum resources; On the one hand, the spectrum usage of some wireless systems is almost idle in time and geographical area (such as the broadcast TV frequency band. With the development of broadcast transmission systems from analog transmission to digital transmission, digital transmission can greatly increase the transmission capacity, making many broadcast televisions The frequency band is idle for a long time. Based on this, in order to solve the problem of shortage of spectrum resources, CR (Cognitive Radio) technology has been widely concerned; it is an effective means to solve the shortage of spectrum resources, through cognitive radio technology. The radio system can adaptively adjust its operating parameters according to its working environment to meet its own needs; it can cooperate with software radio to implement flexible spectrum usage, realize spectrum sharing, and improve spectrum efficiency.

 In cognitive radio technology, it is generally assumed that the authorized system on a licensed frequency band works normally, and the cognitive radio system detects the spectrum usage of the authorized system; when the authorized system is found to be inoperative and the licensed frequency band is idle, the cognitive radio system uses This licensed band continues to detect the operating status of the licensed system in this band. Once the authorized system is found to resume operation, the cognitive radio system will exit the licensed band and transfer to the new licensed band to continue working.

In the cognitive process of cognitive radio systems, Spectrum Sensing discovers the idle spectrum by detecting wireless RF (radio frequency) signals; its Spectrum Analysis is measured by perceived results and wireless RF environment. Information analysis idle spectrum can be cognitive The line system provides communication capacity and the like; its Spectrum Decision determines the wireless parameters used by the cognitive radio system based on the perceived results and the analysis results. In summary, spectrum sensing is a prerequisite for the work of cognitive radio systems.

The spectrum sensing problem is actually a binary detection problem. Suppose there is an authorization system (such as a television system). The signal of the authorization system received at the spectrum sensing receiver is s( ⁿ ), the noise is v( ⁿ ), and the total received signal is ^x. ( ⁿ ) ; Assume that if the authorization system is working, it is HI; suppose that if the authorization system does not work, it is H0; based on this, the received signals can be:

 χ(η) = ν(η)

(η) = s(n) + v(n) where s( ⁿ ) is an I/Q (in-phase quadrature) signal whose signal energy is ^s ; v( ⁿ ) is an independent and identically distributed Gaussian white noise signal, And the I/Q component is independent, and the noise energy is °" ^v . In addition, for the above binary detection problem, it is assumed that the detection statistic is T, and if T is greater than a certain threshold, the authorization system is considered to be working, and the decision result is (logic True, or "1"); If T is less than a certain threshold, the authorization is considered to work, and the decision result is ^D. (logical false, or "0"), as shown below:

Further, for the above detection statistics, an energy detection method may be used, that is, directly counting the energy of a signal received at a certain frequency point, and determining whether the frequency point is occupied according to the intensity of the energy; the process of energy detection is as follows: A total of N DTMB (Digital Television Multimedia Broadcasting) samples are captured in the window, and the energy check statistics obtained directly in the detection time window are:

It should be noted that for a cognitive radio system operating in the broadcast television band, it is necessary to accurately detect whether there is a broadcast television signal at a certain frequency point, thereby judging the availability of the frequency point; for the DTMB system of the broadcast television frequency band, as shown in FIG. Shown as a structure diagram of a data frame. The basic unit of the data frame structure is a signal frame, and the signal frame is composed of a frame header and a frame body two-part time domain signal; the superframe is defined as a group of signal frames, the framing is defined as a set of superframes, and the top layer of the frame structure is a day frame. (Calendar Day Frame, CDF), the signal structure is periodic and synchronized with natural time; the baseband symbol rate of the frame header and frame body signals is the same (7.56Msymbols/s); the frame header portion is composed of PN (pseudo-noise) sequence, and the frame header length can be There are three options; the frame header signal is modulated by the same 4QAM (Quadature Amplitude Modulation) of I and Q; the frame body part contains 36 symbols of system information and 3744 symbols of data, a total of 3780 symbols; The frame length is 500 μs (3780 X 1/7.56 μs).

 As shown in Figure 2, in order to adapt to different applications, three optional frame header modes and corresponding signal frame structures are defined. The frame sequence used in frame header mode is defined as a cyclically extended 8-bit m-sequence, which can be implemented by an LFSR (Linear Feedback Shift Register), which is "0" to +1 value and "Γ to -1 value". The mapping is transformed into a non-returned binary symbol; the initial condition of the LFSR determines the phase of the generated PN sequence; there are 225 signal frames in a superframe; the frame header of each signal frame in each superframe uses a different phase The PN signal is used as the signal frame identifier. Based on the initial state of the LFSR, 255 different phase PN420 sequences can be generated, from sequence number 0 to sequence number 254; 225 PN420 sequences are selected in this standard, from sequence number 0 to sequence number 224; In order to minimize the correlation of adjacent serial numbers, the signal frame sequence sequence and the initial state of the LFSR are formed through computer optimization; the LFSR is reset to the initial phase of sequence number 0 at the beginning of each superframe.

 The frame header mode 2 is truncated by a 10th-order maximum-length pseudo-random binary sequence, and the length of the frame header signal is 595 symbols, which is the first 595 chips of the column of length 1023; the pseudo-random sequence generated by FIG. 2 The first 595 chips, the "0" to +1 value and the "1" to -1 value map are transformed into non-returned binary symbols; there are 216 signal frames in one superframe; each signal in each superframe The frame header of the frame uses the same PN sequence.

The PN sequence used in frame header mode 3 is defined as a cyclically extended 9P-m sequence, which can be implemented by an LFSR, which is transformed into a non-return-to-zero binary symbol by a "0" to +1 value and a mapping of "Γ to -1 value." The initial condition determines the phase of the generated PN sequence, a total of 200 signal frames in a superframe, and the frame header of each signal frame in each superframe uses a PN signal of a different phase as a signal frame identifier; based on the LFSR In the initial state, 511 PN945 sequences of different phases can be generated, from sequence number 0 to sequence number 510; 200 PN945 sequences are selected from the standard, from sequence number 0 to sequence number 199; in addition, in order to minimize the correlation of adjacent sequence numbers, The computer optimizes the selection, the sequence of signal frame numbers formed and the initial state of the LFSR. The LFSR is reset to the initial phase of sequence number 0 at the beginning of each superframe. The correlation of the PN sequence is frequency or time synchronized, and the principle is shown in FIG.

 In the prior art, the spectrum sensing problem for broadcast television signals in cognitive radio systems requires high detection sensitivity (for example, a cognitive radio system operating in the broadcast television band needs to perceive a signal of -114 dBm), according to receiver characteristics. The spectrum sensing algorithm is required to satisfy the SNR (Signal to Noise Ratio) greater than -22 dB, and the authorized system signal is correctly detected with a detection probability greater than 0.9, and the false alarm probability is required to be less than 0.1; in addition, the spectrum sensing of the cognitive radio system is recognized The interference of the radio system (ie, when detecting the current working frequency, whether there is an authorized system working, when the wireless RF signal receiving the cognitive radio system operating frequency is used for spectrum sensing, the RF of the cognitive radio system communication is also received. Signal), so the current spectrum sensing adopts the strategy of silent period sensing, that is, when the cognitive radio performs spectrum sensing, the cognitive radio system stops working and remains silent.

 In the process of implementing the present invention, the inventors have found that at least the following problems exist in the prior art: Since the cognitive radio system needs to remain silent in the spectrum sensing process, the cognitive radio system cannot perform normal communication during the spectrum sensing process, and wastes Radio resources; Further, for the energy detection algorithm, the detection performance is not high, and the sensitivity sensitivity requirement cannot be met, which requires a longer detection time overhead. On the other hand, energy detection is susceptible to interference from other wireless signals, such as adjacent channel leakage interference of TV or adjacent channel leakage interference of cognitive radio systems or interference from remote systems, etc., detection in such low SNR environments, energy The detection of the influence of noise uncertainty caused by factors such as temperature leads to the non-robustness of the detection. On the other hand, the detection performance of the DTMB system synchronization method is not high, and the time required to reach the sensitivity requirement is relatively large, which is not applicable to the spectrum sensing of the DTMB signal. Summary of the invention

 The embodiments of the present invention provide a spectrum sensing method and system to improve spectrum sensing performance and save time consumption of spectrum sensing.

 In order to achieve the above object, an embodiment of the present invention provides a spectrum sensing method, including: a cognitive radio system receiving a signal during one or more discrete silent periods, and processing the received signal to obtain a baseband signal;

The cognitive radio system obtains a detection statistic based on the baseband signal, and according to the check The measured statistic determines the spectrum sensing result.

 The embodiment of the invention provides a spectrum sensing system, including:

 a first acquiring module, configured to receive a signal during one or more discrete silent periods, and process the received signal to obtain a baseband signal;

 a second acquiring module, configured to obtain a detection statistic according to the baseband signal;

 And a determining module, configured to determine a spectrum sensing result according to the detection statistic.

 Compared with the prior art, the embodiment of the invention has at least the following advantages:

 In the embodiment of the present invention, the spectrum sensing method can obtain robust detection performance under a certain interference environment, and the detection performance is high; the spectrum sensing can be completed in a short time, thereby saving the detection time overhead; and being able to be discrete The spectrum sensing detection is completed in multiple silent periods, so that it is more flexible for various cognitive radio systems. DRAWINGS

 1 is a schematic structural diagram of a data frame of a DTMB system in the prior art;

 2 is a schematic diagram of three optional frame header modes and corresponding signal frame structures in the prior art; FIG. 3 is a schematic diagram of a principle of a DTMB system synchronization method in the prior art;

 4 is a schematic flowchart of a spectrum sensing method according to Embodiment 1 of the present invention;

 FIG. 5 is a schematic block diagram of detecting by using a frame header in the first embodiment of the present invention; FIG.

 6 is a schematic block diagram of detecting by using multiple frame headers in Embodiment 1 of the present invention;

 FIG. 7 is a schematic structural diagram of a spectrum sensing system according to Embodiment 3 of the present invention. detailed description

 The technical solutions in the present invention will be clearly and completely described in the following with reference to the accompanying drawings in the drawings. It is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

 Embodiment 1

Embodiment 1 of the present invention provides a spectrum sensing method for implementing in a cognitive radio system. Spectrum sensing, as shown in Figure 4, the method includes the following steps:

 Step 401: A cognitive radio system (such as a spectrum sensing module of a cognitive radio system) receives signals during one or more discrete silent periods, and processes the received signals to obtain a baseband signal.

 In the embodiment of the present invention, in the TD-LTE (TD-SCDMA Long Term Evolution) system, one or more discrete silent periods are specifically one or more GPs of the TD-LTE system. (protected time slot); or, is one or more discrete TD-LTE silent subframes.

 Step 402: The cognitive radio system obtains a detection statistic according to the baseband signal, and determines a spectrum sensing result according to the detection metric.

 Specifically, the cognitive radio system obtains detection statistics according to the baseband signal, including: the cognitive radio system filters the baseband signal through the filter, and performs modulo processing on the filtered result; the cognitive radio system selects the modulo processing The maximum value in the subsequent results is calculated as the average of the results after the modulo processing; the cognitive radio system obtains the detection statistic based on the maximum value and the average value.

 Further, the cognitive radio system filters the baseband signal through the filter, including: the cognitive radio system performs zero-padding processing on the baseband signal, and filters the baseband signal after the zero-padding process by the filter; wherein, the zero-padding The number is determined by the filter length.

 In the embodiment of the present invention, the signal is a signal having a frame header plus frame body (or a known sequence plus data sequence) structure, and may be a signal having multiple frame header (known sequence) modes, and the signal includes But not limited to digital TV multimedia broadcast DTMB signals. For convenience of description, the subsequent description uses the signal as a DTMB signal as an example, and will not be described again for other types of signals.

 For the DTMB signal, the baseband signal is the DTMB baseband signal, and the DTMB signal corresponds to the frame header mode PN420, the frame header mode PN595, and the frame header mode PN945. Assuming that the frame header length is N symbol and the entire signal frame length is M symbol, then the frame header is In mode PN420, the frame header length N is 420, and the signal frame length M is 4200. In the frame header mode PN595, the frame header length N is 595, the signal frame length M is 4375, and in the frame header mode PN945, the frame header length is N. At 945, the signal frame length M is 4725.

On the other hand, it is assumed that the initialization header header symbol sequence PN420 mode of SpN42o (n), n = 0 , ..., 419. Header header symbol sequence _N595 pattern p is ^s _PN595 ^{(n), n} = ^Q , '", ⁵⁹⁴ ; The frame header sequence of the frame header mode PN945 is ^s Pw ( ⁿ ), n = ^Q , , ⁹⁴⁴ . Further, it is assumed that the cognitive radio system can obtain a DTMB baseband signal ^x ( ⁿ ) whose signal is at a symbol rate of 7.56 MHz; and the DTMB baseband signal can pass the received signal (a signal received during one or more discrete silent periods) The sampling filtering is obtained. In the embodiment of the present invention, how to process the received signal to obtain the DTMB baseband signal is not described.

 Based on the above, the cognitive radio system (or its spectrum sensing module) may be one or more GPs of the TD-LTE system for one or more discrete silent periods (one or more discrete silent periods for the TD-LTE system) Or, the reception of the DTMB signal is completed in one or more discrete TD-LTE silence subframes, and the DTMB baseband signal is obtained.

 Sufficient DTMB signals are captured for detection during a discrete silent period. The detection process includes: The cognitive radio system (or its spectrum sensing module) processes the received signal to obtain the DTMB baseband signal; passes the DTMB baseband signal through 3 filters. The tap coefficients of the three filters are respectively a function of the entire frame header sequence or a partial frame header sequence; the three filter output results are subjected to modulo processing; for each filter output modulo result, the largest one is selected. Value, and calculate the average value; obtain 3 detection statistics by the maximum value than the average value; determine the three detection statistics separately (the decision threshold is a function of the target false alarm probability, DTMB frame length, filter length), The result of the logic false or the first identifier (such as 0) (the authorization system is not working at this time, no DTMB signal), or the logical result or the second identifier (such as 1). Work, there is DTMB signal); logically or merge the three judgment results to get the final judgment result.

During the K discrete silent periods, sufficient DTMB signals are captured for detection during each discrete silent period. The detection process includes: a process based on a discrete silent period, cognitive radio system (or its spectrum) Perceptual module) repeatedly performs processing in a discrete silent period during K discrete silent periods (ie, the cognitive radio system processes the received signal to obtain the DTMB baseband signal; passes the DTMB baseband signal through 3 filters, 3 filters The tap coefficients of the device are respectively a function of the entire frame header sequence or a partial frame header sequence; the three filter output results are subjected to modulo processing; for each filter output modulo result, the maximum value thereof is selected, and the average is calculated. The value is obtained by obtaining the three detection statistics by the maximum value; the three detection statistics are respectively determined, and the decision threshold is the target false alarm probability, the DTMB frame length, the filter length, and the number of discrete silent periods K. Function) to get 3K decisions (logical false or first identifier, or logically true or The second identifier); logically or merging the above 3K judgment results to obtain a final judgment result. During a discrete silent period, sufficient DTMB signals are captured for detection during each discrete silent period. The detection process includes: The cognitive radio system (or its spectrum sensing module) receives signals during a discrete silent period , processing the received signal to obtain a DTMB baseband signal;

The DTMB baseband signal is passed through a filter to obtain a filtering result of each segment of the signal, and the tap coefficient of the filter is a function of a frame header sequence of the frame header mode ΡΝ420 at initialization or a partial frame header sequence of the frame header mode ΡΝ420 at initialization; The filtering result is subjected to modulo processing; marking the segment filtering and taking the position of the maximum value of the modulo result, and combining the filtering results corresponding to the maximum value positions, and modulating the combined result to obtain a value a; Summing and modulo, obtaining the value b; obtaining the detection statistic by a/b; determining the detection statistic and the decision threshold (the decision threshold is the target false alarm probability, the DTMB frame length, the filter length, and the number of silent periods K) a function, obtaining a logical false or first identified decision result, or a logically true or second identified decision result; repeating the above steps twice, the filter tap coefficients are respectively initialized frame headers during two repeated executions a header sequence of the mode PN595 or a function of a partial header sequence of the frame header mode PN595 at initialization; and an initial PN945 sequence header when the header mode or the initialization mode function header part header of the PN945 sequence; the result of the judgment of the 3 obtained or logically combined to give the final verdict.

 Case 1, the block diagram shown in Figure 5, which uses a frame header for detection, that is, enough DTMB signals are captured for detection during a discrete silent period.

In this case, the cognitive radio system obtains the detection statistic according to the baseband signal further includes: Step 1. In a DTMB baseband signal ^x ( ⁿ ) corresponding to a discrete silent period, the cognitive radio system arbitrarily intercepts the DTMB of the M symbol length The baseband signal obtains the truncated signal χ(η), η = 0,...,Μ Step 2. The cognitive radio system fills the signal ^χ ( ^η ) to zero to obtain the signal u( ⁿ ), and the number of zeros is filtered according to The length of the device is determined.

Step 3. The cognitive radio system filters the signal ^u (n) through three filters, and the filtered output is:

y _P N42o ( ^m ) = ∑ ^u ( ^{m+ n} ) ^h PN42o (I^— n— l), m ₌ 0, l, 2~M— 1 _{ypN59 5 (m) = 7Σu (} m + n) h N595 (L 2 -nl), m = 0, l, 2 ... Ml

L _n= o

1 ^l3 — ¹

^{_{! y P N945 (m) =}} Σ U (m + n) h PN945 f 1 ] - Π = 012 M - 1 where, ^h () is a common fate operator *; tap coefficients of the filter for the entire sequence header Or a function of a partial frame header sequence, and the tap coefficients of the filter can be:

hpN420 ( L ¹ - 1) '― ^S PN420 ¹ (i), i =. ,· ••L ¹

 ·> Its L=164

hpN595 ( L ² - 1) 1― ^S PN595 ' (0,1 = 0,· ••L ²

 ·> Its L=594

hpN945 ( L ³ - 1) '― ^S PN945 ' (0,i = 0,· ••L ³

 ·> Its L=433

 Or,

hpN420 ( L ¹ - 1) ― ^S PN420 ' (i),i = 0,· ••L ¹

 ·> Its L=419

hpN595 ( L ² - 1) 1― ^S PN595 ' (0,1 = 0,· ••L ²

 ·> Its L=594

hpN945 ( L - 1) '― ^S PN9451 (0,i = 0,· ••L ³

 ·> Its L=944

Wherein, the frame header symbol sequence of the frame header mode PN420 is ^s PN «W, ^{n = ()} , , ⁴ⁱ⁹ ; the frame header mode

The frame header sequence of PN595 is 3⁄4^5 ( ^η ), ^{η = ()} , ···, ⁵⁹⁴ ; The frame header sequence of frame header mode ΡΝ945 is s _PN 94 ₅ (n), n = 0,... , 944 Step 4: The cognitive radio system performs the following modulo processing on the filtered output (ie, each sample result):

3⁄4420 (m) = abs ( y _P (m)) , m = 0, 1, 2 M _ 1

Zp 5 (m) = abs ( y _P (m)), m = 0, 1, 2 M _1

ZpNWi (m) = abs ( y _PNg45 (m)), m = 0, 1, 2 M -1 Step 5. The maximum value of the result after the cognitive radio system selects the modulo processing, and calculates the result after the modulo processing The average value, and the ratio of the maximum value to the average value, gives the following three detection statistics: max [zp _N420 (m), m = 012 M - 1}

Further, the cognitive radio system determines the spectrum sensing result according to the detection statistic, including: The electrical system performs the following judgment by using three detection statistic and a decision threshold:

In order to obtain the judgment result ¹³ , ^υ , ^υ , in the above judgment, ^υ . Represents logical false or first identification (0), at which point the authorization system is considered to be not working, no DTMB signal; ^D i" represents logical truth or second identification (1), at which point the authorized system is working, there is a DTMB signal; To determine the threshold, and the decision threshold is determined according to the target false alarm probability, M value, and L value.

Step 7. The cognitive radio system combines the decision result ^{D (} ", ^{D (} , ^ logically OR (or) to obtain the final decision result ¹ ^ ⁰¹ ^ ¹ ^), ^ ¹ , ² , ³ , and based on the decision result D determines the spectrum perception result.

Case 2: In this case, multiple frame headers are used for detection, and detection is performed by an OR combining method. In this case, the cognitive radio system obtains the detection statistic according to the baseband signal, and further includes: Step 1. The cognitive radio system obtains the same length of the K segment in the DTMB baseband signal ^x ( ⁿ ) corresponding to the plurality of discrete silent periods. the baseband signal DTMB ^{xW (n), k = (} , K a ^{1, η = 0, ···,} Μ a ^1, and a signal length of each segment Μ.

Step 2. The cognitive radio system processes each DTMB of the K-segment DTMB baseband signal according to the processing of the DTMB baseband signal ^x ( ⁿ ) corresponding to a discrete silent period (ie, step 1 - step 5 of the case). The baseband signals are processed separately to obtain corresponding 3*K detection statistics.

Further, the cognitive radio system determines the spectrum sensing result according to the detection statistic, and includes: Step 3: The cognitive radio system processes the statistic corresponding to the detection statistic corresponding to a discrete silent period (ie, step 6 of the case) ), judge 3*Κ detection statistic, To get 3*K decision results ^Dtw , ^D , ^D ', k = o, .„, K -i.

Step 4. The cognitive radio system logically logically 3*K decision results ^D(W , ^{D( 2'k} ), ^{D( 3'k} )

The OR merges to obtain the final decision result D ^ ¹ ^ ¹ ^)) '^ ¹ ' ² ' ³ '^^ - ¹ , and the spectrum sensing result is determined based on the decision result D. Case 3: The block diagram shown in Figure 6, which uses multiple frame headers for detection and is detected by the maximum value combination method.

In this case, the cognitive radio system obtains the detection statistic according to the baseband signal, and further includes: Step 1. The cognitive radio system obtains the same length of the K segment in the DTMB baseband signal ^x ( ⁿ ) corresponding to the plurality of discrete silent periods. the baseband signal DTMB ^{xW (n), k = (} , K a ^{1, η = 0, ···,} Μ a ^1, and a signal length of each segment Μ.

Step 2. The cognitive radio system fills the intercepted DTMB baseband signal x ^W ( ⁿ ) to obtain a signal.

(k) \

u ( ⁿ ) , the number of zeros is determined by the length of the filter.

Step 3. The cognitive radio system filters the signal ^u(k ) through three filters, and the filtered output is: ∑ ^u(k) ( ^{m + n} ) ^h py ^(k) PN595 H = ∑ u ^(k) (m+n)h; (L ² -nl),m= 0,1,2...M-1

L _n= oy ^(k) _PN945 (m) = -∑u ^(k) (m+n)h; _N945 (L ³ — n— l), m= 0,1,2~M— 1 where ^h4 ( * is a common operation; the tap coefficient of the above filter is a function of the entire frame header sequence or a partial frame header sequence, and the tap coefficients of the filter can be:

 hpN42. —1) 42. = 0,".L , its L=164

 hpN595 (L - 1) 595 (1),1 = 0,··丄 , its L=594

hpN945 (L -1) = ^S PN945 (l) J = 0,...L ^ or

 hpN42. — 1) 42. = 0,".L its L=419

hpN595 (L - 1) 595 (1),1 =◦,··丄其L=594 hpN945 (L -1) = ^S PN945 ( '^ = , its L ₌ 944

The frame header symbol sequence of the frame header mode PN420 is ^42. ( ^η ), ^{η = ()} , ···, ⁴¹⁹ ; frame header mode

The frame header sequence of ΡΝ595 is ⁸ ΡΝ^ (η), ^{η = ()} , ···, ⁵⁹⁴ ; The frame header symbol sequence of frame header mode PN945 is s _PN 94 ₅ ( ⁿ ), n = 0,... , 944. Step 4. The cognitive radio system performs the following processing on the filtered output: z ^(k) PN42o ( ^m ) = abs ( y ^(k) _PN420 (m)) , m = 0, 1, 2~ M _ 1 z ^(k) _PN595 (m) = abs(y« _PN595 (m)),m= 0,1, 2...M -1 z ^(k) _PN945 (m) = abs ( y ^(k) _PN945 (m)), m= 0, 1, _ 1 Step 5. The cognitive radio system marks m corresponding to the maximum value of the modulo result corresponding to each k by:

3⁄4 ^(k) = ir^max(z ^(k) _PN420 (m))

Step 6. The cognitive radio system combines the matched filtered output results corresponding to the m values of the above markers by the following manner, and performs modulo processing on the combined results:

Step 7. The cognitive radio system averages the matched filtering result of the k-segment signal by the following method, and performs modulo processing on the averaged result:

 ,

Mean _PN42 . = y ^(k) _PN42 . (m)

 ( K-l M-1

Mean _PN595 = abs _∑∑ y ^(k) _PN595 (m)

V k=0 m=0 Mean _p

Step 8. The cognitive radio system takes the maximum value of step 6 (that is, the result of modulo processing the combined result) and the average value of step 7 (that is, after modulo processing the result after averaging) The result is divided by the following test statistic: mean _P]

 w.

 Hey.

Mean _P] mean _p Further, the cognitive radio system determines the spectrum sensing result based on the detection statistic, including: Step 9. The cognitive radio system performs the detection statistic (ie, the three detection statistic in step 8) and the decision threshold. The following judgment:

In order to obtain the judgment result ¹⁵ , ^D , D , in the above judgment, ^D ^ represents logical false or first identification (0), at this time, it is considered that the authorization system is not working, no DTMB signal; ^D i" represents logical truth or second identification (1) At this time, it is considered that the authorization system is working, there is a DTMB signal; for the decision threshold, and the decision threshold is determined according to the target false alarm probability, K value, M value, and L value.

Step 10: The cognitive radio system logically ORs the decision results ^D(1) , ^D(2) , and ^D(3) to obtain a decision result D = ^QR ( ^D(1 ) ^{= 1} ' ² ' ³ , and based on the judgment Result D determines the spectrum perception result.

In the foregoing embodiment of the present invention, the foregoing spectrum sensing detection method may be completed in other discrete silent periods, or may be completed in a continuous silent period for a long time; in addition, it may be completed in the GP of TD-LTE, and the specific implementation is implemented. The method is that the spectrum sensing module keeps synchronization with the TD-LTE system and receives the DTMB signal in the GP of the TD-LTE. If a single GP is used for the detection, the usage is used. In the detection method of one, if multiple GPs are used for detection, the detection methods of cases two and three are used. The spectrum sensing of current cognitive radio faces the problem of detecting weak signals, requiring high detection performance of the algorithm, and requiring the sensing algorithm to quickly detect the authorized system signal, requiring less detection time overhead; in addition, spectrum sensing may be authorized The interference of the system and even the cognitive radio system requires the perceptual algorithm to have a relatively high robustness. In the prior art, the spectrum sensing algorithm is energy detection, but the energy detection is difficult to meet the requirements of detection performance and detection time overhead, and may be affected by other interferences, thereby causing a serious decline in detection performance; and the synchronous detection method of the broadcast television system Not fully suitable for spectrum sensing of cognitive radio systems.

 For the above findings, the embodiment of the present invention provides a spectrum sensing method for a DTMB signal, which effectively detects a DTMB signal by detecting a DTMB frame header signal; the method can obtain a higher relative energy detection algorithm. Detecting performance to achieve weak signal detection requirements; and the method can effectively combat the influence of wireless environment interference and system noise level changes on spectrum sensing, thereby greatly improving the robustness of the cognitive system and obtaining robustness under certain interference environments. The detection performance is high, and the detection performance is high; the method is applicable not only to the spectrum sensing during the continuous long-term silent period, but also to the sensing of multiple discrete short-time silent periods, and the spectrum sensing can be completed in a short time. Thereby, the time overhead of detection is saved; and the spectrum sensing detection can be completed in a plurality of discrete silent periods, thereby being more flexible and applicable to various cognitive radio systems.

 Embodiment 2

 Embodiment 2 of the present invention provides a spectrum sensing method, which is assumed that TD-LTE has a cognitive radio function, that is, a TD-LTE base station has a cognitive module to discover available DTMB broadcast system idle frequency bands,

The special subframe configuration of TD-LTE is configured as 0, and the spectrum sensing process includes:

 Step 1: The spectrum sensing module on the TD-LTE base station side maintains downlink synchronization with the TD-LTE base station; Step 2: The spectrum sensing module receives the DTMB signal of a certain frequency point in the GP of the TD-LTE; Step 3: Spectrum sensing module pair The received signal is filtered and sampled to obtain a symbol rate of

7.56 MSymbols/s DTMB baseband signal;

Step 4: intercepting the DTMB baseband signal to obtain a number x(n) having a length of 4725 symbols, n = 0, ..., 4724. Zeroing the ^X ( ⁿ ) signal to obtain ^U (n), and then Filtered separately through 3 filters After the output:

 1

ΥΡ 2 (m+n)h; (L ¹ -l),m= 0,l, 2...4119

 L

1 ¹

ΥΡ 595 ( ^m ) 7∑u(m+n)h _N595 (L ² -l),m = 0,l, 2...4374

L _n= o

1 ¹

YP 945 \ ^m ) ∑u(m+n)h; _N945 (l -l),m = 0,l, 2...4724 and the tap coefficients of the three filters are:

h VpNN4422O (LL - 1) == ss _PPNN44220 (1 _丄 ) J,,1丄== 0 U,,....丄.L 立立L 1^ ¹ -—419

h 1⁄2】Ν595 (L -1) = ^S PN595 = 0,...L Its L ² —594

1⁄2N945 (L --1l)) == ss _PPN945 (1),1 == 0 C, ....LL ³ , L ³ —=994444

Among them, ^S PN42 ² 0° 0W , ^SpN595 (') , ^SpN945 (') are the initial frame header sequences of the three frame header modes respectively; Step 5 ^ for each sample result of the filter output:

3⁄4420 (mi)) == ab. s ( _ yy _PpN420 (mmj)),, m= 0, 1, 2...4119

Zp 5 ( ^m )) == aabbss (( yy _PPNN559955 ((mm)))),,mm==00,, 11,, 22...44337744

Zp _N9 45 ( ^m ) = abs ( y ₅ (m)) , m = 0, 1, 2 4724 Step 6. Using the above modulo operation results, use the maximum ratio to average to obtain three detection statistics: max { Zp _N42 . (m) , m = 0, 1, 2...4119

 1

, m max { Zp _N595 (m) , m = 0, 1, 2...4374

 τ. , 1

m max { Zp _N945 (m),m= 0,l, 2...4724

 X

 m

M Step 7. Determine the three detection statistics and the decision threshold respectively, and obtain three judgment results··

D. D. 0

In the above judgment, ^D. Representing "0", ^D i stands for "Γ; for the decision threshold, the decision threshold is determined according to the target false alarm probability (0.1), Μ value, L value;

Step 8. Perform the above steps 2 - 7 in 5 consecutive TD-LTE GPs to obtain 15 decision results ¹ ^ ', ^ ¹ , ² , ³ ; ¹ ^ ¹ ,..., ⁵ ;

Step 9. Perform a logical OR operation on the above 15 decision results, that is, as long as any one of D, i = l, ² , ³ _; k = l, ⁵ is logic, then the decision result is logically true (ie, there is DTMB signal); Otherwise, the decision is logic "0", then the decision result is logic ^ (ie no DTMB signal).

 Embodiment 3

 Based on the same inventive concept as the above method, the embodiment of the present invention further provides a spectrum sensing system (which may specifically be a cognitive radio system, or a spectrum sensing module in a cognitive radio system), as shown in FIG. The system includes:

 a first acquisition block 11 for receiving a signal during one or more discrete silent periods, and processing the received signal to obtain a baseband signal;

 a second obtaining module 12, configured to obtain a detection statistic according to the baseband signal;

 The determining module 13 is configured to determine a spectrum sensing result according to the detection statistic.

 The second obtaining module 12 is specifically configured to filter the baseband signal by using a filter, and perform modulo processing on the filtered result; select a maximum value in the result of the modulo processing, and calculate and take An average of the results after the modulo processing; and obtaining a detection statistic based on the maximum value and the average value.

 The second obtaining module 12 is further configured to perform zero-padding processing on the baseband signal, and perform filtering processing on the baseband signal after zero-padding processing by using the filter; wherein, the number of zero-padding is performed by the filter The length is determined.

 The signal is a signal having a frame header plus frame structure, and the signal includes a digital television multimedia broadcast DTMB signal.

When the signal is a DTMB signal, the baseband signal is the DTMB baseband signal ^x ( ⁿ ), and DTMB The signal corresponds to a frame header mode PN420, a frame header mode PN595, and a frame header mode PN945. The second obtaining module 12 is further configured to arbitrarily intercept M in a DTMB baseband signal ^x ( ⁿ ) corresponding to a discrete silent period. The symbol length DTMB baseband signal is obtained by ^χ ( ^η ), ^{η = ()} , ···, ^Μ ; the intercept signal ^Χ (η) is zero-padded to obtain the signal u( ⁿ ), and the signal u( ⁿ ) is passed through three The filter is filtered and the filtered output is:

 1

y _P (m) = ₇ ∑u(m+n)h (L ¹ — n— l),m=0,l,2...M— 1

 L;

1 ¹

y _P 7∑ ^u ( ^m+n ) ^h PN595 (L ² - n - 1), m = 0, 12 M - 1

_{L n = 0 y PN945 (m} ) = K m + n) h 45 (L 3 - n - 1), m = 0, 12 M - 1 where, ^{h ()} is a co-doom operator *;

 Perform the following modulo processing on the filtered output:

3⁄4420 (m) = abs ( y _PN420 (m)) , m = 01, 2 M _ 1

3⁄4595 ( ^m ) = as ( y _{P 5} (m)), m=0, l, 2...Ml

3⁄4945 ( ^m ) = abs ( y _PN945 (m)) , m = 0, 1, 2 M _ 1 Select the maximum value of the result after modulo processing, calculate the average value of the result after the modulo processing: The ratio of the maximum value to the average value is obtained as follows:

Max{zp _N420 (m),m = 01,2—M -ll

 T

The determining module 13 is further configured to: determine, by using the detection statistic and the threshold:

J ίχ ^X PN420 _< , _γ PN420 D "0 ⁽¹⁾

 I γ

lT PN420 > z PN420 n "1 ⁽¹⁾ D. D. 0

The decision result ¹⁵ , ^D , D is obtained, where W represents a logical false or first identifier, represents a logical truth or a second identifier, is a decision threshold, and the decision threshold is determined according to a target false alarm probability, an M value, and an L value. ;

The decision results ^D(1 ), ^D ( ² ), ^D(3) are logically OR combined to obtain a decision result D = OR(D«) ' i = 1'2' ³ , and the spectrum is determined based on the decision result D Perceive the results. The second obtaining module 12 is further configured to obtain a DTMB baseband signal x( ^k ), 3⁄4 0, . 1 of the same length of the K segment in the DTMB baseband signal ^x ( ⁿ ) corresponding to the plurality of discrete silent periods. (, and each segment of the signal length is _M; according to the processing method of the DTMB baseband signal ^x ( ⁿ ) corresponding to a discrete silent period, each DTMB baseband signal in the K-segment DTMB baseband signal is processed separately to obtain a corresponding 3*K detection statistics;

The determining module 13 is further configured to: determine, according to a processing manner for determining a detection statistic corresponding to a discrete silent period, to determine the 3*Κ detection statistic to obtain 3*Κ deterministic results ^{D. (W} , ^D(2 ' ^k ), ^D(3 ' ^k ), k = 0,...,Kl.

The 3*K decision results ^D(W , ^{D( 2'k} ), ^{D( 3'k} ) are logically OR combined to obtain a decision result D = OH(D ⁽¹ ' ^k )), i ₌ l, 2,3,k ₌ 0_K-1, and the spectrum sensing result is determined based on the decision result _D. When the signal is a DTMB signal, the baseband signal is a DTMB baseband signal ^x ( ⁿ ), and the DTMB signal corresponds to a frame header mode PN420, a frame header Mode PN595, frame header mode PN945;

The second obtaining module 12 is further configured to obtain a DTMB baseband signal x( ^k ), 3⁄4 0, . 1 of the same length of the K segment in the DTMB baseband signal ^x ( ⁿ ) corresponding to the plurality of discrete silent periods. (, and each segment of the signal length is _M; the DTMB baseband signal ^x(k )( ⁿ ) is padded to obtain the signal ^u(k )( ⁿ ), and the signal is ¹ ^^

The three filters are filtered and the filtered output is:

y ^(k) PN4 ₂ o (m) = ^-∑u ^(k) (m+n)h; 诵(L ¹ -nl),m= 0,1,2...M -1 _V ( ^k ) PN595 I, m) = -^ u ^(k) (m+n) h; _N595 ( L ² - n - 1), m = 0, 1, 2~M - 1

L _n= oy( ^k ) PN945 I, m) = - ^k) (m+n) h; _N945 ( L ³ — n— 1), m = 0, 1, 2~M— 1

L _n =0 where , * is a common erection operation;

 Perform the following processing on the filtered output:

PN420 abs ( y ^(k) _PN420 (m)) , m = 0, 1, 2 to M - 1 z ^(k) PN595 (m) abs ( y ^(k) _PN595 (m)), m=0, l, 2...Ml z ^(k) PN945 (m) abs ( y( ^k) (m)), m=0, l, 2...Ml The maximum value of the modulo result corresponding to each k is marked as follows Corresponding m:

( ^k) =m|max(z ^(k) _PN595 (m))

 The matched filtered output results corresponding to the m values of the markers are combined as follows, and after the merge

 The matching filtering results of the k-segment signals are averaged as follows, and the averaging results are subjected to modulo processing:

 ί K-1M-1 Λ

Mean _PN420 = abs y ^(k) _PN42 . (m)

 k=0 m=0 )

 ί K-lM-1 Λ

Mean _PN595 = abs _∑∑ y ^(k) _PN595 (m)

 Vk=0 m=0 J

 ί K-lM-1 Λ

Mean _PN945 = abs ∑∑y° _N945 (m)

Vk=0 m=0 j The result of the modulo processing on the combined result is divided by the result of the modulo processing on the averaging result, and the following detection statistic is obtained: mean, mean, mean, the determination module 13, further Used to pass the detection statistic and the decision threshold as follows

In order to obtain the judgment result ^D '", D, ^Ό where, representing a logical false or first identifier, representing a logical truth or a second identifier, ^^ is a decision threshold, and the decision threshold is based on the target false alarm probability, K value, M The value and the L value are determined;

The decision result ^{D(1 D(} , " is logically OR combined to obtain a decision result D = OR(D ⁽¹⁾ ) ' i = 1'2' 3 , and the spectrum sensing result is determined based on the decision result D.

 The tap coefficients of the filter are a function of the entire frame header sequence or a partial frame header sequence, and the tap coefficients of the filter are specifically:

(0, 1 0. • L ¹

— ^T its L=164

(0, 1 0. • L ²

 Its L=594

(0, 1 0. • L ³

 Its L=433

 Or,

 (L ( '^ its L=419

( L-1) — ( '^ Its L=594 hpN945 (L -1) = ^S PN945 ( '^ = , its L ₌ 944

The frame header symbol sequence of the frame header mode PN420 is ^42. ( ^η ), ^{η = ()} , ···, ⁴¹⁹ ; frame header mode

The frame header sequence of ΡΝ595 is ⁸ ΡΝ^ (η), ^{η = ()} , ···, ⁵⁹⁴ ; The frame header symbol sequence of frame header mode PN945 is s _PN 94 ₅ ( ⁿ ), n = 0,... , 944. In a time division synchronous code division multiple access long term evolution TD-LTE system, the one or more discrete silent periods are specifically one or more guard slots GP of the TD-LTE system; or, for one or more discrete TD-LTE silent subframe.

 The modules of the device of the present invention may be integrated into one or may be deployed separately. The above modules can be combined into one module, or they can be further split into multiple sub-modules.

 Through the description of the above embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary general hardware platform, and of course, can also be through hardware, but in many cases, the former is a better implementation. the way. Based on such understanding, the technical solution of the present invention, which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium, including a plurality of instructions for making a A computer device (which may be a personal computer, server, or network device, etc.) performs the methods described in various embodiments of the present invention.

 A person skilled in the art can understand that the drawings are only a schematic diagram of a preferred embodiment, and the modules or processes in the drawings are not necessarily required to implement the invention.

 Those skilled in the art can understand that the modules in the apparatus in the embodiments may be distributed in the apparatus of the embodiment according to the embodiment, or may be correspondingly changed in one or more apparatuses different from the embodiment. The modules of the above embodiments may be combined into one module, or may be further split into a plurality of sub-modules.

 The serial numbers of the embodiments of the present invention are merely for the description, and do not represent the advantages and disadvantages of the embodiments.

 The above disclosure is only a few specific embodiments of the present invention, but the present invention is not limited thereto, and any changes that can be made by those skilled in the art should fall within the protection scope of the present invention.

Claims

Rights request

A spectrum sensing method, comprising:

 The cognitive radio system receives signals during one or more discrete silent periods and processes the received signals to obtain a baseband signal;

 The cognitive radio system obtains a detection statistic based on the baseband signal and determines a spectrum sensing result based on the detected statistic.

 2. The method of claim 1, wherein the cognitive radio system obtains a detection statistic based on the baseband signal, comprising:

 The cognitive radio system filters the baseband signal through a filter, and performs modulo processing on the filtered result;

 The cognitive radio system selects a maximum value in the result of the modulo processing, and calculates an average value of the result after the modulo processing;

 The cognitive radio system obtains a detection statistic based on the maximum value and the average value.

3. The method according to claim 2, wherein the cognitive radio system performs filtering processing on the baseband signal by a filter, including:

 The cognitive radio system performs zero-padding processing on the baseband signal, and performs filtering processing on the baseband signal after zero-padding processing through the filter;

 Wherein, the number of zeros is determined by the length of the filter.

 The method according to any one of claims 1 to 3, wherein the signal is a signal having a frame header plus frame structure, and the signal comprises a digital television multimedia broadcast DTMB signal.

5. The method according to claim 3, wherein when the signal is a DTMB signal, the baseband signal is a DTMB baseband signal ^x ( ⁿ ), and the DTMB signal corresponds to a frame header mode PN420, a frame header mode PN595, and a frame header. Mode PN945; the cognitive radio system obtains a detection statistic according to the baseband signal, and further includes:

The cognitive radio system arbitrarily intercepts the DTMB baseband signal of the M symbol length in a DTMB baseband signal ^x ( ⁿ ) corresponding to a discrete silent period to obtain ^χ ( ^η ), ^{η = ()} , ···, ^Μ ;

The cognitive radio system adds a signal ^Χ ( ^η ) to zero to obtain a signal u( ⁿ ), and divides the signal ^U (n) Do not filter through three filters, and the filtered output is:

y _P 7∑u (m+ n)h (L ¹ - n - 1), m = 0, 1, 2~M - 1

L _n= o

1 ¹

y _P N595 (m) = -∑u (m+ n) h; _N595 (L ² — n— 1), m = 0, 1, 2 M— 1

L _n= oy ∑u (m+ n)h ₅ (L ³ - n - 1), m = 0, 1, 2 M - 1

L _n= 0 where * is the common erg operation of ^ht ();

 The cognitive radio system performs the following modulo processing on the filtered output result:

ZPN420 ( ^m ) = abs ( y _PN420 (m)), m = 0, 1, ..M-1

ZPN595 ( ^m ) = abs ( y _PN595 (m)), m = 0, 1, 2...M-1

ZPN945 ( ^m ) = as ( y ₅ (m)), m= 0, 1, 2...M - 1 The cognitive radio system selects the maximum value in the result of the modulo processing, and calculates the modulo processing The average of the results, and by the ratio of the maximum to the average, the following test statistic is obtained:

 Max zpj, m) , m = 0,l,2—M -l}

 τ J595 ' :

^Α ΡΝ595

 1 M-1

 3⁄4595

Μ m Σ=0 τ ^Α ΡΝ945

The method according to claim 5, wherein the cognitive radio system determines the spectrum sensing result according to the detection statistic, including:

 The cognitive radio system makes the following decision by using the detection statistic and the decision threshold:

The decision result ¹⁵ , ^D , D is obtained, where W represents a logical false or first identifier, represents a logical truth or a second identifier, is a decision threshold, and the decision threshold is determined according to a target false alarm probability, an M value, and an L value. ;

The cognitive radio system will determine the decision result ^D( ", ^{D(2 0(} " logically OR, obtain a decision result D = ^QR ( ^D(1 ) ^{= 1} ' ² ' ^{3 )} , and determine based on the decision result D Spectrum sensing results.

 The method according to claim 6, wherein the cognitive radio system obtains a detection statistic according to the baseband signal, and further includes:

The cognitive radio system obtains a DTMB baseband signal ^xW ( ⁿ ) of the same length of K segments in a DTMB baseband signal ^x ( ⁿ ) corresponding to a plurality of discrete silent periods, ^{k = ()} , '", ^K - ^{1 ,} η ^{= 0,} ···, ^Μ a ^1, [mu] and a length of each segment of the signal;

The cognitive radio system processes each DTMB baseband signal in the DTMB baseband signal according to a DTMB baseband signal ^χ ( ^η ) corresponding to a discrete silent period to obtain a corresponding 3*K Detection statistics;

Determining, by the cognitive radio system, the spectrum sensing result according to the detection statistic, comprising: processing, by the cognitive radio system, a detection statistic corresponding to a discrete silent period, to the 3*Κ A test statistic is made to obtain 3*Κ decision knots n(l,k) _n (2,k) _n (3,k)

, k = 0,..., K -l . n(l,k) _n (2,k) _n (3,k) The cognitive radio system will treat the 3*K decision results u, ^υ , ^υ logical OR merge, get the decision result ^{D = QR} ( ^D(1 ' ^k ))' ^{i = 1} ' ² ' ³ ' ^{k = (} ' ^K — and determine the spectrum perception result based on the decision result D.

8. The method according to claim 3, wherein when the signal is a DTMB signal, the baseband signal is a DTMB baseband signal ^x ( ⁿ ), and the DTMB signal corresponds to a frame header mode PN420, a frame header mode PN595, and a frame header. Mode PN945; the cognitive radio system obtains a detection statistic according to the baseband signal, and further includes:

The DTMB baseband signal ^x ( ⁿ ) corresponding to the cognitive radio system in a plurality of discrete silent periods In the DTMB baseband signal ^x ( ⁿ ), ^{k = ()} , , ^{K 1} , ^{η = 0} , ···, ^{Μ 1} , and the length of each segment is Μ;

The cognitive radio system fills the DTMB baseband signal ^x(k ) ( ⁿ ) to obtain the signal ^u(k ) ( ^η ) and will

(k) \

The signal ^U ( ⁿ ) is filtered by three filters, and the filtered output is: y ^(k) PN42o (m) = -∑u ^(k) (m+n)h; _N42 (L ¹ — N- l),m = 0,1, 2...M-1

L _n= o

L ² -l

 , (k)

v PN595 \ ) = - L ^k) (m+ n) h (L ² - n - 1), m = 0, 1, 2 M - 1

L _n= o

L ³ —l

 , (k)

' PN945 ) ' = _ L u ^(k) (m+ n) h (L ³ - n - 1), m = 0, 1, 2 M - 1

The cognitive radio system performs the following processing on the filtered output: z ^(k) PN42o ( ^m ) = ^abs ( y ^(k) PN42o ( ^m )) ^ πι = 0, 1, 2~M - 1

z ^(k) PN5 ₉ 5 (m) = abs (y( ^k ) _PN595 (m)),m = 0,l, 2...M -1

z ^(k) PN945 ( ^m ) = abs ( y ^(k) p (m)) , m = 0, 1, 2...M - 1 The cognitive radio system marks each k corresponding corresponding by The maximum value of the modulo result corresponds to m:

3⁄4 ^{(k) = m} | ^m ^ ^x ( ^z(k) PN42o ( ^m )) q ₂ ^(k) = m| max (z ^(k) _PN595 (m)) _3⁄4 ^(k) = H max ( z ^{(k )} _P (m)) said matched filter output m cognitive values corresponding radio system by way of the tag are combined, and the combined result modulo process:

^W PN595 abs PN595 :q ₂ ^(k)

k=0

The cognitive radio system averages the matched filtering results of the k-segment signals by performing modulo processing on the averaged results as follows:

 K-1 M

Mean _PN , _n = abs ∑∑y' ₁ m

 k=0 m=0

 K-l M-1

Mean _PN595 = abs (k)

 ∑∑y' m

 k=0 m=0

 K-l M -l

 Mean 5 = abs (k)

 ∑∑y' PN945 m

 \k=0 The cognitive radio system divides the result of the modulo processing of the combined result with the result of the modulo processing after the averaged result, and obtains the following detection statistics:

T ^A PN420

Mean, τ ' ^ν ΡΝ595

 Mean

 w PN945

 Mea PN945

The method according to claim 8, wherein the cognitive radio system determines the spectrum sensing result according to the detection statistic, including:

The cognitive radio system makes the following decision by the detection statistic and the decision threshold: τ ^A PN420 < Υ ΡΝ 420

 τ > Υ ΡΝ 420

To obtain a decision result ^DD(3) , where represents a logical false or first identifier, ^D " represents a logical truth or a second identifier, ^^ is a decision threshold, and the decision threshold is based on a target false alarm probability, K value, M The value and the L value are determined;

The cognitive radio system merges the decision result ^D( ", ^{D(2 D(3} ) i row logical OR,

D = OR( D ⁽ⁱ⁾ ),i = 1, 2,3

Get the verdict and the verdict based on the results of spectrum sensing to determine D _t

The method according to claim 5 or 8, wherein the tap coefficient of the filter is a function of a whole frame header sequence or a partial frame header sequence, and the tap coefficients of the filter are specifically: hpN« . (L _1) ⁼ 42. (1), 1 = 0, „.L , L=164

hPN595 ― ¹ ) = ^S PN595 ( ,1 =〇"丄² , L=594

hPN945 - ¹ ) = ^S PN945 W' ¹ = ⁰ ,... , L=433

 Or,

hpN42. (L _1) ^{= S} PN42. (1), 1 = 0, „.L, its L=419

hPN595 - ¹ ) = ^S PN595 W,l =〇"丄^2, its L=594

hPN945 " = ^S PN945 W' ¹ = ⁰ ,... ¹ ^ Its L=944

Wherein, the frame header symbol sequence of the frame header mode PN420 is ^s PN^ ( ⁿ ), ^{n = Q} — ⁴¹⁹ ; the frame header mode

The frame header sequence of PN595 is ^ ₅₉₅ ( ^η ), ^{η = ()} , ···, ⁵⁹⁴ ; The frame header symbol sequence of frame header mode PN945 is s _PN945 (η) , n = 0, 944 .

The method according to any one of claims 1-3, wherein in the time division synchronous code division multiple access long term evolution TD-LTE system, the one or more discrete silent periods are specifically TD-LTE One or more guard slots GP of the system; or, one or more discrete TD-LTE silence subframes.

 12. A spectrum sensing system, comprising:

 a first acquiring module, configured to receive a signal during one or more discrete silent periods, and process the received signal to obtain a baseband signal;

 a second acquiring module, configured to obtain a detection statistic according to the baseband signal;

 And a determining module, configured to determine a spectrum sensing result according to the detection statistic.

 13. The system of claim 12, wherein:

 The second acquiring module is specifically configured to filter the baseband signal by using a filter, and perform modulo processing on the filtered result; select a maximum value in the result of the modulo processing, and calculate a modulo An average of the processed results; and a detection statistic is obtained based on the maximum value and the average value.

 14. The system of claim 13 wherein:

The second obtaining module is further configured to perform zero-padding processing on the baseband signal, and pass the The filter performs filtering processing on the baseband signal after the zero-padding process;

 Wherein, the number of zeros is determined by the length of the filter.

 The system according to any one of claims 12-14, wherein the signal is a signal having a frame header plus frame structure, and the signal comprises a digital television multimedia broadcast DTMB signal.

The system according to claim 14, wherein when the signal is a DTMB signal, the baseband signal is a DTMB baseband signal ^x ( ⁿ ), and the DTMB signal corresponds to a frame header mode PN420, a frame header mode PN595, and a frame header. Mode PN945;

The second obtaining module is further configured to arbitrarily intercept the DTMB baseband signal of the M symbol length in a DTMB baseband signal ^x ( ⁿ ) corresponding to a discrete silent period to obtain ^x ( ⁿ ), ⁿ = ^Q , , ^M ; The intercept signal ^x (n) is zero-padded to obtain the signal u( ⁿ ), and the signal u( ⁿ ) is filtered by three filters respectively, and the filtered output result is: y _P N42o ( ^m ) = ∑ ^u ( ^{m+ n} ) ^h PN42o ^— n— l), m = 0, 1, 2~M— 1

 n=0

1 L ² -l

L ² - ¹

y _P ∑u(m+n)h; (L ² m = 0,l,2...M

LL; _n= o

L ³ -l

y ∑u(m+n)h ₅ (L ³ m=0,l,2...M

L; where ^{h 0} is a total of bad luck:

 Perform the following processing on the filtered output:

Zp ( ^m ) = abs ( y _P (m)), m=0, l, 2...Ml

3⁄4595 ( ^m ) = abs ( y _PN595 (m)) , m = 0, 1, _ 1

ZPN945 ( ^m ) = abs ( y ₅ (m)) , m = 0, 1, _ 1 Select the maximum value in the result of the modulo processing, calculate the average value of the result after the modulo processing, and pass the maximum The ratio of the value to the average results in the following test statistic:

T ^Α ΡΝ420

Max|Zp _N595 (mm = 0,l,2 M -1

 TpN595 — 丄 ψ

-, 595 I m Max |zp _N945 (m),m= 0,1,2 M -l|

 TpN945— 1 -丄

—∑ ZpN94 ₅ (m)

i ^Vi m=0 _o

17. The system of claim 16 wherein:

 The determining module is further configured to perform the following judgment by using the detection statistic and the decision threshold

T _< , _Y D ⁽¹⁾

Τ _> ζ _Γ D ⁽¹⁾ ρΝ595 D. D.

To obtain the decision result ^{D (} ", ^{D (} , ° ⁽³⁾ , where represents a logical false or first identifier, W represents a logical truth or a second identifier, ^γ ^ is a decision threshold, and the decision threshold is based on the target false alarm Probability, devaluation, and L value are determined;

The decision result ^{D(1 D(} , " logically OR merges to obtain the decision result V ) ''', and the spectrum sensing result is determined based on the decision result D.

 18. The system of claim 17 wherein:

The second obtaining module is further configured to obtain a DTMB baseband signal x ^k (* 0 , · , 1, , of the same length of the K segment in the DTMB base signal ^x ( ⁿ ) corresponding to the plurality of discrete silent periods. And the length of each segment of the signal is M; according to the processing method of the DTMB baseband signal ^x ( ⁿ ) corresponding to a discrete silent period,

Each segment of the DTMB baseband signal in the K-band DTMB baseband signal is processed separately to obtain the corresponding

3*K detection statistics;

 The determining module is further configured to determine, according to a processing manner for determining a detection statistic corresponding to a discrete silent period, the 3* 检测 detection statistic to obtain 3* Κ

 J^(l,k) r» (2,k) pv(3,k)

The judgment results D, ^D , ^D , k = (U, Kl;

n(l,k) _n (2,k) p, (3,k)

The 3*K decision results ¹ ^ , ^U , ^U are logically OR combined to obtain a judgment The result is D^^D^)' ^{1 1} ' ² ' ³ ' ¹ "^''''^- ¹ , and the spectrum sensing result is determined based on the decision result D.

The system according to claim 14, wherein when the signal is a DTMB signal, the baseband signal is a DTMB baseband signal ^x ( ⁿ ), and the DTMB signal corresponds to a frame header mode PN420, a frame header mode PN595, and a frame header. Mode PN945;

The second obtaining module is further configured to obtain a DTMB baseband signal x( ^k ) of the same length of the K segment in the DTMB baseband signal ^x ( ⁿ ) corresponding to the plurality of discrete silent periods ( ^ * 0 , . , ^ 1 , , and the length of each segment of the signal is _M;

Zeroing the DTMB baseband signal ^xW (n) to obtain the signal u ^W ( ⁿ ), and the signal

The three filters are filtered and the filtered output is:

y ^(k) H = T∑u ^(k) (m+ η) _Ν42 (L ¹ _ n _ 1), m = 0, 1, 2...M _ 1

^{L y (m) = fΣu (} k) (m + n) h; N595 (L 2 -nl), m = 0,1,2 ... M -1

LL _n= o

L ³ -l

y ^(k) ₅ = ∑u ^(k) (m+n)h _N945 (L ³ — n— l),m= 0,1,2...M -1

Wherein L _n = 0, ^{h 0} * In co doom operator;

 Perform the following processing on the filtered output:

z ^(k) PN42o ( ^m ) = abs ( y ^(k) o (m)) , m = 0, 1, 2~M - 1 z ^(k) PN595 ( ^m ) = abs ( y ^(k) _PN595 (m )),m= 0,1,2...M -1 z ^(k) PN945 ( ^m ) = abs ( y ^(k) ₅ (m)) , m = 0,1, 2~M - 1 The mode marks the m corresponding to the maximum value of the modulo result corresponding to each k:

3⁄4 ^(k) =m|max(z ^(k) _PN420 (m)) q ₂ ^(k) =m|max(z ^(k) _PN595 (ml

The matched filtered output results corresponding to the m values of the markers are combined as follows, and the combined results are subjected to modulo processing:

The matching filtering result of the k-segment signal is averaged as follows, and the averaging result is subjected to modulo processing:

_{mean PN420 = PN420 (m) mean} PN595 = (m) mean PN945 = (m)

The result of the modulo processing of the combined result is divided by the result of the modulo processing after the averaging result, and the following detection statistic is obtained:

 T

 Mean.

20. The system of claim 19, wherein:

 The determining module is further configured to perform the following judgment by using the detection statistic and the decision threshold

Τ <, γ D ⁽¹⁾

T > z / y D ⁽¹ ) (2)

I Tp _N 595 < YP 595 D.

I IT ^A PN595 > y PN595 D( ² )

To obtain a decision result ^D( ", ^D '\ ^D ' ³⁾ , where represents a logical false or first identifier, A" represents a logical truth or a second identifier, which is a decision threshold, and the decision threshold is based on the target false alarm probability , K value, M value, L value determination; logically OR the decision results ^D(1) , ^D(2) , ^D(3) to obtain a decision result D = OR(D« ) ' i = 1'2' 3 , and based on the decision result D, the spectrum sensing result is determined.

The system according to claim 16 or 19, wherein the tap coefficient of the filter is a function of a whole frame header sequence or a partial frame header sequence, and the tap coefficients of the filter are specifically: i 丄

hpN42. (L -1) = s _PN42 . = 0,...L its L

h _PN5 95 (L - 1) = ^S PN595 = 0,...L ² with L=594

hpN945 (L -1) = s _PN945 = 0,...L Its L=433

= 0," stands

^PN595 (L-1) = ^S PN595 = 0,...L =594

hpN94 ₅ -1) = s _PN945 (1), 1 = 0, ... L Its L = 944 where the frame header symbol sequence of the frame header mode PN420 is ⁸ ΡΝ «. ( ^η ), ^{η = ()} , ···, ⁴¹⁹ ; frame header mode

The frame header sequence of the PN595 is ^s PN ⁵⁹ » ^{= Q} -, ⁵⁹⁴ ; The frame header sequence of the frame header mode PN945 is ^s _PN945 ( ⁿ ), ⁿ = 0, '", 9 4.

The system according to any one of claims 12 to 14, wherein in the time division synchronous code division multiple access long term evolution TD-LTE system, the one or more discrete silent periods are specifically TD-LTE One or more guard slots GP of the system; or, one or more discrete TD-LTE silence subframes.