Frequency optimization method and device based on spectrum sensing

Abstract

The invention discloses a frequency optimization method and device based on spectrum sensing, and belongs to the field of wireless communication, and the method comprises the steps: firstly, starting a spectrum sensing function before the start of a current receiving time slot; frequency configuration is continuously carried out in the spectrum sensing process; after the frequency is configured, sampling frequency mixing and low-pass filtering are carried out on a radio frequency signal corresponding to the current frequency, and a baseband signal is output; calculating the average power of the baseband signal in the time domain, calculating the modulus and the retrieval peak value after fast Fourier transform in the frequency domain, and completing the spectrum sensing calculation of two dimensions of time and frequency; and finally, according to a time-frequency sensing calculation result, configuring a subsequent frequency, performing frequency optimization according to an optimization strategy, and outputting a final optimization frequency. The method does not occupy the system bandwidth, does not depend on inter-node prediction and cooperation, and can complete synchronization pulse spectrum sensing and frequency optimization only by using a low calculation amount by the receiving node so as to avoid frequency domain interference and improve the synchronization capability.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of wireless communication, more particularly, to a frequency optimization method and device based on spectrum sensing. BACKGROUND

[0002] Frequency hopping technology is a technology of transmitting signals on multiple predefined frequencies. Compared with fixed frequency technology, it has stronger anti-interference ability, better spectrum utilization and better anti-multipath fading ability, so it is widely used in the field of wireless communication.

[0003] In order to improve the anti-interference ability of the synchronization section and the synchronization probability of the receiving node, the traditional frequency hopping communication system generally configures multiple frequency hopping synchronization pulses when designing the frame structure. When the sending node transmits synchronization information through the predefined frequency, the traditional frequency hopping receiving node cannot adjust the receiving frequency according to the channel quality or environmental changes, but can only wait in order according to the frequency hopping pattern, which has obvious limitations. The spectrum sensing technology can detect abnormal spectrum activity and obtain the spectrum state by real-time detection of the current spectrum state, which can better adapt to the instantaneous changes of the wireless environment. When applied to frequency hopping synchronization pulse reception, spectrum sensing can help the receiving node to avoid interference frequencies and complete the reception signal synchronization on the optimized frequency, thereby improving the synchronization ability of the node. However, the architecture and processing logic of the existing spectrum sensing technology are often too complex when running in real time, and the system needs to have core capabilities such as detection, classification, prediction and cooperation. Real-time detection and classification require a large amount of computing resources, and prediction and cooperation require nodes to exchange sensing results, which increases communication overhead and processing delay. If a complex algorithm such as deep learning is used for spectrum sensing, a higher-performance processor or a dedicated chip needs to be provided to provide computing power, resulting in high hardware cost. SUMMARY

[0004] The present application aims to overcome the shortcomings of the prior art and provide a frequency optimization method and device based on spectrum sensing, which does not occupy system bandwidth and does not rely on node prediction and cooperation. The receiving node only needs to complete the synchronization pulse spectrum sensing and frequency optimization with a low amount of calculation to avoid frequency domain interference and improve the synchronization ability.

[0005] The purpose of the present application is achieved by the following scheme: A frequency optimization device based on spectrum sensing, comprising a time synchronization acquisition module, a frequency configuration module, a frequency conversion sampling module, a spectrum sensing module and a sensing decision module; The time synchronization acquisition module is used to acquire system time information and start the spectrum sensing function before the start of the current receiving time slot; The frequency configuration module is used to configure the synchronization pulse frequency during spectrum sensing; The variable frequency sampling module is configured to complete sampling and mixing of the radio frequency signal corresponding to the current frequency and low-pass filtering after frequency configuration, and output a baseband signal. The spectrum sensing module is configured to calculate average power in the time domain, calculate the modulus after fast Fourier transform in the frequency domain, and search for a peak value, to complete spectrum sensing calculation in two dimensions of time and frequency. The sensing decision module is configured to control frequency configuration of the frequency configuration module according to the sensing calculation result of time and frequency, perform frequency optimization according to an optimization strategy, and output a final optimized frequency.

[0006] A frequency optimization method based on spectrum sensing, based on the device described above, includes the following steps: S1, starting spectrum sensing function before the start of the current receiving time slot; S2, continuously performing frequency configuration during spectrum sensing; S3, sampling and mixing the radio frequency signal corresponding to the current frequency and low-pass filtering after frequency configuration, and outputting a baseband signal ; S4, calculating average power in the time domain, calculating the modulus after fast Fourier transform in the frequency domain, and searching for a peak value, to complete spectrum sensing calculation in two dimensions of time and frequency. S5, configuring subsequent frequency according to the sensing calculation result of time and frequency, performing frequency optimization according to an optimization strategy, and outputting a final optimized frequency.

[0007] Further, the continuously performing frequency configuration during spectrum sensing specifically includes the following sub-steps: Suppose that a burst frame contains N synchronization pulses, and spectrum sensing of the node starts at the end of the propagation protection segment of the last frame before the start of the current time slot, and the frequency of synchronization pulse 1 is configured.

[0008] Further, the sampling and mixing of the radio frequency signal corresponding to the current frequency and low-pass filtering, and outputting a baseband signal specifically includes the following sub-steps: Sampling at the current frequency, and mixing and filtering the sampled signal to obtain L I and Q baseband signals, and the complex form of the baseband signal is represented as , wherein i represents an imaginary unit, I l represents the lth I baseband signal, Q l represents the lth Q baseband signal, I represents in-phase, and Q represents quadrature.

[0009] Further, the calculating average power in the time domain, calculating the modulus after fast Fourier transform in the frequency domain, and searching for a peak value, to complete spectrum sensing calculation in two dimensions of time and frequency specifically includes the following sub-steps: ​​The average power of L complex baseband signals is calculated in time domain to obtain average power values The FFT of L I, Q baseband signals is calculated in frequency domain to obtain L-point complex signals in frequency domain , is a natural exponent, The modulus of the L frequency domain complex signals is continuously calculated to obtain The score retrieval is performed on the L frequency domain modulus value signals to find the maximum peak value .

[0010] Further, the sensing calculation result according to time and frequency is used to configure subsequent frequencies, and the preferred strategy is used for frequency preference, and the final preferred frequency is output, and the specific steps include: The average power values and the maximum peak values in frequency domain are compared with the set power threshold and the frequency domain peak threshold respectively, if and , it is considered that there is no interference on the current frequency, the frequency preference is ended, and the current frequency is output as the non-interference frequency, otherwise, it is considered that there is interference on the current frequency, the average power value and the maximum peak value in frequency domain of the current frequency are stored, and the frequency control is switched to the next frequency point, and steps S3, S4 and S5 are repeatedly executed.

[0011] Further, if all the N synchronization pulse frequencies are completed, and the non-interference frequency is still not found, the fusion parameters are obtained by fusing the average power values and the square of the frequency domain peak values of the remaining frequencies except the first frequency The fusion parameters are compared, and the frequency with the minimum fusion parameter is output, and the frequency preference is ended.

[0012] The beneficial effects of the present application include: (1) The present application does not occupy additional system bandwidth. Specifically, the present application uses the propagation protection end period of the last time slot in the frequency hopping system to control the first synchronization pulse frequency, and decides whether to control the subsequent frequency for spectrum sensing according to the sensing decision result each time after entering the current time slot. The silent period of the propagation protection period and the difference between the frequencies of the synchronization pulses are fully utilized, so that the spectrum environment without pollution from system nodes can be obtained without separately dividing the spectrum sensing period by the system, thereby saving the system bandwidth.

[0013] (2) This invention does not rely on inter-node prediction, cooperation, or upper-layer protocol assistance. Specifically, this invention only requires a single node to perform unidirectional sampling calculation at the receiving time slot signal processing level to complete the synchronization pulse spectrum sensing and frequency optimization. It does not require the transmission of sensing information between nodes, the coordination of other nodes, or the participation of upper-layer protocols, which can greatly simplify the system complexity.

[0014] (3) The present invention has low computational complexity, fast processing speed, and good real-time performance. Specifically, the present invention only needs to perform basic power calculation, FFT calculation and peak retrieval calculation on a small number of sampled signals in the time and frequency domains when sensing the spectrum. It does not rely on complex algorithms such as deep learning and has low computational requirements for the system. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 A diagram of the frame format for a frequency hopping system with multiple synchronization pulses; Figure 2 This is a structural block diagram of the device according to an embodiment of the present invention; Figure 3 This is a flowchart of the method according to an embodiment of the present invention; Figure 4 This is a timing diagram of frequency control in the method of this embodiment of the invention; In the diagram, there are: synchronization segment 101, data segment 102, propagation protection segment 103, time synchronization acquisition module 201, frequency configuration module 202, frequency conversion sampling module 203, spectrum sensing module 204, and sensing decision module 205. Detailed Implementation

[0017] All features disclosed in all embodiments of this specification, or steps in all methods or processes implied in the disclosure, may be combined and / or extended or replaced in any way, except for mutually exclusive features and / or steps.

[0018] As a first aspect of the present invention, a frequency optimization device based on spectrum sensing is provided, such as... Figure 2 As shown, it includes: a time synchronization acquisition module 201, which acquires system time information and starts the spectrum sensing function before the current receiving time slot begins; a frequency configuration module 202, which completes the configuration of the synchronization pulse frequency during the spectrum sensing process; and a frequency conversion sampling module 203, which, after the frequency is configured, completes the sampling, mixing, and low-pass filtering of the radio frequency signal corresponding to the current frequency, and outputs the baseband signal. a spectrum sensing module 204, for sensing the baseband signal The spectrum sensing is calculated in time domain and frequency domain, and the peak value is searched after FFT, and the spectrum sensing is calculated in time and frequency dimensions. The sensing decision module 205 controls the frequency configuration of the frequency configuration module 202 according to the sensing calculation results of time and frequency, and performs frequency optimization according to the preferred strategy, and outputs the final preferred frequency.

[0019] As a second aspect of the present application, a spectrum sensing-based frequency optimization method is provided, particularly a method for selecting frequency by using spectrum sensing in a frequency hopping communication system, which specifically includes the following steps: First, the spectrum sensing function is started before the start of the current receiving time slot; the frequency configuration is continuously performed during the spectrum sensing process; after the frequency configuration, the radio frequency signal corresponding to the current frequency is sampled, mixed and low-pass filtered, and the baseband signal is output ; the baseband signal The spectrum sensing is calculated in time domain and frequency domain, and the peak value is searched after FFT, and the spectrum sensing is calculated in time and frequency dimensions. Finally, the subsequent frequency is configured according to the sensing calculation results of time and frequency, and the frequency optimization is performed according to the preferred strategy, and the final preferred frequency is output.

[0020] In other embodiments, based on the above-mentioned embodiment method, a spectrum sensing-based frequency optimization method is provided, which includes the following contents: In Figure 1 In the burst frequency hopping system data frame format shown, in order to improve the anti-interference performance of the synchronization section 101, the burst frame contains N synchronization pulses, each of which has a different frequency, which is determined by the frequency hopping pattern in advance, and the receiving end completes signal acquisition, timing synchronization and frequency offset estimation by selecting one of the synchronization pulses, and then demodulates and decodes the signal in the data section 102 to recover the source data. 103 is a propagation protection section. Referring to Figure 3 and Figure 4 , the present embodiment specifically provides a spectrum sensing-based frequency optimization method, which is implemented by using the following steps: Step A, the spectrum sensing of the node starts at the end of the propagation protection of the last frame before the start of the current time slot, and the frequency of the synchronization pulse 1 is configured; Step B, after the configuration is completed, sampling is performed on the current frequency, and the sampling signal is mixed and filtered to obtain L I and Q baseband signals. The complex form of the baseband signal can be represented as , , represents the imaginary unit, , represents the lth I baseband signal, , represents the lth Q baseband signal, I represents in-phase, and Q represents quadrature; Step C, average power of L complex baseband signals is calculated in time domain to obtain average power value , L complex baseband signals are calculated in frequency domain to obtain L-point complex signals in frequency domain , in the above formula is natural index ; continue to calculate modulus of L complex signals in frequency domain to obtain , L modulus signals in frequency domain are subjected to score searching to find maximum peak value ; ; Step D, average power value , maximum peak value in frequency domain are compared with set power threshold , frequency domain peak threshold , if and , it is considered that there is no interference in current frequency, and frequency optimization is ended, and current frequency is output, otherwise, it is considered that there is interference in current frequency, and average power value and maximum peak value in frequency domain of current frequency are stored, meanwhile, frequency configuration module 202 switches next frequency point, and the above steps B, C and D are repeated; Step E, if until N synchronization pulse frequencies are all switched, and no interference-free frequency is found, then average power value of remaining frequencies except first frequency and square of frequency domain peak value are fused to obtain fusion parameter , all fusion parameters are subjected to comparison optimization, and the frequency with minimum fusion parameter is output, and frequency optimization is ended.

[0021] The units described in the embodiments of the present application can be implemented in software or hardware, and the units described can be arranged in a processor. In some cases, the names of the units do not constitute a limitation on the units themselves.

[0022] According to an aspect of an embodiment of the present application, a computer program product or computer program is provided, and the computer program product or computer program includes computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method provided in various optional implementation manners.

[0023] As another aspect, the embodiments of the present application also provide a computer readable medium, which can be included in the electronic device described in the above embodiments, or exist independently without being assembled into the electronic device. The computer readable medium carries one or more programs, which, when executed by the electronic device, enable the electronic device to implement the method described in the above embodiments.

Claims

1. A spectrum-sensing based frequency preference apparatus, characterized by, The device comprises a time synchronization acquisition module, a frequency configuration module, a frequency conversion sampling module, a spectrum sensing module and a sensing decision module. The time synchronization acquisition module is configured to acquire system time information and start the spectrum sensing function before the start of a current receiving time slot. The frequency configuration module is configured to complete the configuration of the frequency of the synchronization pulse during the spectrum sensing process. The frequency conversion sampling module is configured to complete the sampling and mixing of the radio frequency signal corresponding to the current frequency and low-pass filtering after the frequency configuration, and output a baseband signal. The spectrum sensing module is configured to calculate the average power in the time domain and calculate the modulus and retrieve the peak value after the fast Fourier transform in the frequency domain, and complete the spectrum sensing calculation in the time-frequency two dimensions. The sensing decision module is configured to control the frequency configuration of the frequency configuration module according to the sensing calculation result in the time-frequency two dimensions, perform frequency optimization according to an optimization strategy, and output the final optimized frequency.

2. A spectrum-sensing based frequency preference method, characterized in that, The device according to claim 1 comprises the following steps: S1, starting the spectrum sensing function before the start of a current receiving time slot; S2, continuously configuring the frequency during the spectrum sensing process; S3, when the frequency is configured, the radio frequency signal corresponding to the current frequency is sampled, mixed and low-pass filtered, and a baseband signal is output ; S4, to the baseband signal The average power is calculated in time domain, and the fast Fourier transform (FFT) is calculated in frequency domain, and the modulus and peak value are searched after the FFT, so as to complete the spectrum sensing calculation in time-frequency two dimensions. S5, configuring the subsequent frequency according to the sensing calculation result in the time-frequency two dimensions, performing frequency optimization according to an optimization strategy, and outputting the final optimized frequency.

3. The spectrum-sensing based frequency preference method of claim 2, wherein, The continuously configuring the frequency during the spectrum sensing process comprises the following sub-steps: Supposing that a burst frame contains N synchronization pulses, the spectrum sensing of the node starts at the end of the propagation protection segment of the last frame before the start of the current time slot, and the frequency of the synchronization pulse 1 is configured.

4. The spectrum-sensing based frequency preference method of claim 3, wherein, The sampling, mixing and low-pass filtering of the radio frequency signal corresponding to the current frequency to output a baseband signal comprises the following sub-steps: The L I, Q baseband signals are obtained by sampling at the current frequency and mixing the filtered sampled signals down, and the complex form of the baseband signals is represented as , j represents the imaginary unit, I l represents the lth I baseband signal, Q l represents the lth Q baseband signal, I represents in-phase, and Q represents quadrature.

5. The spectrum-sensing based frequency preference method of claim 4, wherein, The pair of baseband signals The average power is calculated in time domain, and the modulus and peak value are searched after fast Fourier transform (FFT) is calculated in frequency domain, specifically including the following sub-steps: The average power values are obtained by calculating the average power of the L complex baseband signals in the time domain The L-point complex signals in the frequency domain are obtained by calculating the FFT of the L I, Q baseband signals in the frequency domain , The natural exponent is The modulus of the L frequency domain complex signals is continued to be calculated to obtain The score search is performed on the L frequency domain modulus value signals to find the maximum peak value .

6. The spectrum-sensing based frequency preference method of claim 5, wherein, The configuring the subsequent frequency according to the sensing calculation result in the time-frequency two dimensions, performing frequency optimization according to an optimization strategy, and outputting the final optimized frequency comprises the following sub-steps: The average power value , the frequency domain maximum peak value are compared with the set power threshold , the frequency domain peak threshold respectively, if and , it is considered that the current frequency has no interference, the frequency selection is ended, and the current frequency is output as the non-interference frequency, otherwise it is considered that there is interference on the current frequency, the current frequency average power value and the frequency domain maximum peak value are stored, and the frequency control switches to the next frequency point, and steps S3, S4 and S5 are repeatedly executed.

7. The spectrum-sensing based frequency preference method of claim 6, wherein, If no interference-free frequency is found until all N synchronization pulse frequencies are cut, then the average power value of the remaining frequencies except the first frequency is fused with the square of the frequency domain peak value to obtain a fusion parameter Then, all fusion parameters are compared, and the frequency with the smallest fusion parameter is output, and the frequency selection is completed.

Images