WO2019010861A1 - 一种认知无线网络的频谱预测方法及装置 - Google Patents

一种认知无线网络的频谱预测方法及装置 Download PDF

Info

Publication number
WO2019010861A1
WO2019010861A1 PCT/CN2017/107314 CN2017107314W WO2019010861A1 WO 2019010861 A1 WO2019010861 A1 WO 2019010861A1 CN 2017107314 W CN2017107314 W CN 2017107314W WO 2019010861 A1 WO2019010861 A1 WO 2019010861A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
access
frequency band
target
primary user
Prior art date
Application number
PCT/CN2017/107314
Other languages
English (en)
French (fr)
Inventor
景晓军
杨威
黄海
Original Assignee
北京邮电大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京邮电大学 filed Critical 北京邮电大学
Priority to US16/619,207 priority Critical patent/US10674434B2/en
Publication of WO2019010861A1 publication Critical patent/WO2019010861A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W48/00—Access restriction; Network selection; Access point selection
    • H04W48/02—Access restriction performed under specific conditions
    • H04W48/06—Access restriction performed under specific conditions based on traffic conditions
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/08—Testing, supervising or monitoring using real traffic
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/391—Modelling the propagation channel
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/391—Modelling the propagation channel
    • H04B17/3913—Predictive models, e.g. based on neural network models
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/70—Admission control; Resource allocation
    • H04L47/83—Admission control; Resource allocation based on usage prediction
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/14—Spectrum sharing arrangements between different networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/18—Network planning tools
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W48/00—Access restriction; Network selection; Access point selection
    • H04W48/02—Access restriction performed under specific conditions
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/382—Monitoring; Testing of propagation channels for resource allocation, admission control or handover
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/0289—Congestion control
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W40/00—Communication routing or communication path finding
    • H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
    • H04W40/18—Communication route or path selection, e.g. power-based or shortest path routing based on predicted events
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W48/00—Access restriction; Network selection; Access point selection
    • H04W48/16—Discovering, processing access restriction or access information

Definitions

  • the present application relates to the field of cognitive wireless network technologies, and in particular, to a spectrum prediction method and apparatus for a cognitive wireless network.
  • Spectrum sensing is a key technology in current cognitive radio and a key technology in the future of smart antenna systems, military communications and interference countermeasures.
  • cognitive radio mainly solves the following problems: on the one hand, the radio spectrum resources are relatively scarce, and the spectrum resources available for allocation are less and less; on the other hand, according to the US Communications Commission As a result of the survey, the spectrum utilization rate currently authorized is very low, and in most cases only about 10%, resulting in a huge waste of spectrum resources.
  • Cognitive radio means that a wireless communication device with spectrum sensing capability can temporarily borrow an idle authorized channel for communication without affecting the normal communication of the primary user (authorized user), thereby improving spectrum utilization. Therefore, as an important part of the cognitive radio network, the prediction of the spectrum occupied by the primary user is an urgent problem to be solved.
  • the secondary user usually performs full-band scanning. This results in untargeted spectrum access and low accuracy.
  • the application of the embodiments of the present application is to provide a spectrum prediction method and apparatus for a cognitive wireless network, so as to achieve targeted access to secondary users, thereby improving spectrum utilization efficiency of the wireless network;
  • a spectrum prediction method for a cognitive wireless network comprising:
  • the secondary user is connected to the target frequency band according to the prediction model of the primary user occupied frequency band.
  • the spectrum sensing method using subspace filtering acquires real-time occupation information of the frequency band, including:
  • the detection probability P d ( ⁇ , ⁇ ) of the frequency band and the false alarm probability P f ( ⁇ , ⁇ ) are obtained;
  • ⁇ 0 is the expectation for spectrum sensing
  • Is the variance used for spectrum sensing
  • Is the variance of the noise
  • Q(x) is the complementary cumulative distribution function of the standard normal distribution
  • t is the energy value of the signal received by the secondary user
  • N is the number of samples of the primary user
  • H 1 is the case where the primary signal s(t) exists
  • H 0 is the wireless frequency band is not occupied by the primary user
  • y(n) is The signal observed after sampling by the receiver
  • n is the number of samples at the signal receiving end
  • ⁇ is the ratio of the signal variance to the noise variance after subspace filtering
  • ⁇ is the preset energy threshold
  • ⁇ is the time of spectrum sensing.
  • f s is the sampling frequency of the secondary user
  • u 1 (n) is the residual noise after subspace filtering
  • the real-time occupancy information of the frequency band is obtained by using the detection probability of the frequency band and the false alarm probability.
  • the predictive model for the occupied frequency band of the primary user is established according to the updated spectrum resource database of the radio environment map REM, including:
  • the target primary user access frequency band After receiving the target primary user access frequency band, searching for all users close to the target primary user access time from the updated spectrum resource database of the radio environment map REM, and all users who are close to the target primary user access time Recorded as set A, wherein the spectrum resource database includes names, access times, access bands, geographical locations, and occupancy of all users;
  • the user who searches for the geographical difference between the target primary user and the target value is smaller than the preset value, and the user whose geographical difference with the target primary user is less than the preset value is recorded as the set C, and the target primary user and the set C are calculated.
  • a prediction model of the occupied frequency band of the primary user is established.
  • the sub-user accesses the target frequency band according to the prediction model of the occupied frequency band of the primary user, including:
  • the target access request includes an access time according to a prediction model of a frequency band occupied by the primary user Whether the target frequency band is occupied by the user;
  • the secondary user is connected to the target frequency band
  • a spectrum prediction apparatus for a cognitive wireless network comprising:
  • Requesting an access module receiving a target access request of the secondary user
  • the spectrum real-time occupancy information acquisition module is configured to acquire real-time occupancy information of the frequency band by using a spectrum sensing method of subspace filtering;
  • a resource library establishing module configured to update a spectrum resource database of the radio environment map REM according to real-time occupancy information of the frequency band
  • a model obtaining module configured to establish a prediction model of a frequency band occupied by the primary user according to the updated spectrum resource database of the radio environment map REM;
  • the access behavior module is configured to access the secondary user to the target frequency band according to the prediction model of the primary user occupied frequency band.
  • the spectrum real-time occupation information acquisition module includes:
  • the detection and false alarm probability acquisition submodule is configured to obtain the detection probability P d ( ⁇ , ⁇ ) of the frequency band and the false alarm probability P f ( ⁇ , ⁇ ) by using the spectrum sensing method of the subspace filtering;
  • ⁇ 0 is the expectation for spectrum sensing
  • Is the variance used for spectrum sensing
  • Is the variance of the noise
  • Q(x) is the complementary cumulative distribution function of the standard normal distribution
  • t is the energy value of the signal received by the secondary user
  • N is the number of samples of the primary user
  • H 1 is the case where the primary signal s(t) exists
  • H 0 is the wireless frequency band is not occupied by the primary user
  • y(n) is The signal observed after sampling by the receiver
  • n is the number of samples at the signal receiving end
  • ⁇ is the ratio of the signal variance to the noise variance after subspace filtering
  • ⁇ is the preset energy threshold
  • ⁇ is the time of spectrum sensing.
  • f s is the sampling frequency of the secondary user
  • u 1 (n) is the residual noise after subspace filtering
  • the real-time occupation information acquisition sub-module is configured to obtain the real-time occupation information of the frequency band by using the detection probability of the frequency band and the false alarm probability.
  • model obtaining module includes:
  • a user search sub-module configured to search, from the updated spectrum resource database of the radio environment map REM, all users that are close to the target primary user access time after receiving the target primary user access band, and the target primary user All users with similar access times are recorded as set A, wherein the spectrum resource database includes names, access times, access bands, access habits, geographical locations, and occupancy of all users;
  • a user confirmation sub-module configured to train all users in the set A, and confirm an access user that most closely matches the daily access behavior of the target primary user according to the temporal similarity, and the most suitable access user Recorded as collection B;
  • the first determining sub-module is configured to determine whether the access behavior of the user in the set B after the training meets the access behavior of the target primary user; if not, the trigger returns to the sub-module, and if yes, triggers the calculation sub-module;
  • a first returning submodule configured to return to perform, after receiving the target primary user access spectrum, searching for, from the updated spectrum resource database of the radio environment map REM, all users that are close to the target primary user access time;
  • a calculation sub-module configured to: if the user in the collection B searches for a geographical difference between the target primary user and the target user is less than a preset value, the user who is less than the preset value of the target primary user is recorded as the set C, and the calculation is performed. The time interval between the target primary user and the access band of the user in the set C;
  • the correction submodule is configured to correct the user in the set C according to the modification rule and the user that meets the target primary user access rule;
  • a prediction submodule configured to predict an access habit of the target primary user according to the user in the modified set C
  • the prediction model establishing sub-module is configured to establish a prediction model of the main user occupied frequency band according to the access habit of the predicted target primary user.
  • the access behavior module includes:
  • a second determining sub-module configured to determine, according to the prediction model of the occupied frequency band of the primary user, whether the target access request includes the target frequency band of the access time is occupied by the user; if not, triggering the secondary user access sub-module, if Yes, triggering the second return submodule;
  • a secondary user access sub-module for accessing the secondary user to the target frequency band
  • a second returning sub-module configured to return to perform a prediction model according to a frequency band occupied by the primary user, and determine, by the user, whether the target access request includes a frequency band of the access time, and is occupied by the user.
  • a computer readable storage medium stores instructions that, when run on a computer, cause the computer to perform one of any of the foregoing A method of spectrum prediction enhancement in cognitive wireless networks.
  • an embodiment of the present application further provides a computer program product comprising instructions, when executed on a computer, causing a computer to perform a cognitive wireless network according to any of the foregoing Spectrum prediction enhancement method.
  • the spectrum prediction method of the cognitive wireless network can obtain the real-time occupation information of the frequency band by using the spectrum sensing method of the sub-space filtering; and update the pre-determination according to the real-time occupation information of the frequency band.
  • a spectrum resource database of the radio environment map REM according to the updated spectrum resource database of the radio environment map REM, a prediction model for the occupied frequency band of the primary user is established; and the secondary user is accessed according to the prediction model of the primary user occupied frequency band Target band Inside.
  • the method can target the secondary users, reducing the spectrum sensing time and energy, and thus improving the spectrum utilization efficiency of the wireless network.
  • any of the products or methods of the present application necessarily does not necessarily require all of the advantages described above to be achieved at the same time.
  • FIG. 1 is a flowchart of a spectrum prediction method of a first cognitive wireless network according to an embodiment of the present application
  • FIG. 3 is a flowchart of spectrum access prediction of a primary user according to an embodiment of the present application
  • FIG. 4 is a flowchart of a second method for predicting a spectrum of a cognitive wireless network according to an embodiment of the present application
  • FIG. 5 is a schematic diagram of a spectrum prediction apparatus for a cognitive wireless network according to an embodiment of the present application.
  • FIG. 6 is a schematic diagram of an electronic device according to an embodiment of the present application.
  • FIG. 1 is a flowchart of a method for predicting a spectrum of a first cognitive wireless network according to an embodiment of the present application, where the method includes:
  • the primary user is a user authorized to use a certain frequency band in the cognitive wireless network
  • the secondary user is a user in the cognitive wireless network that is not authorized to use a certain frequency band.
  • the execution subject of the present application is a base station.
  • the real-time occupation information includes an access time, an access frequency band, and a geographical location;
  • the spectrum sensing method of the subspace filtering is used to obtain the real-time occupation information of the frequency band;
  • the cognitive user obtains spectrum usage information in the wireless network through various signal detection and processing methods.
  • spectrum sensing technology mainly involves the physical layer and the link layer.
  • the physical layer mainly focuses on various specific local detection algorithms, while the link layer mainly focuses on cooperation between users and local perception. Collaborative awareness and awareness mechanisms are optimized in three areas.
  • Subspace filtering is an active noise reduction method. The essence is that after constructing the linear estimator H, the received signal r(n) is decomposed into two orthogonal subspaces: signal noise subspace Hy(n) and noise.
  • Subspace (IH)r(n), y(n) is the signal (including signal and noise) observed after sampling by the receiver, I is the same identity matrix as the H dimension, signal noise subspace and noise sub
  • IH the noise subspace contains only noise
  • the signal noise subspace contains segmentation noise and full signal
  • n is the number of samples at the signal receiving end.
  • Optimal subspace filtering is obtained after removing all noise in the noise subspace and noise in the signal noise subspace Can be expressed as:
  • the optimal linear estimator H opt can minimize signal distortion and make the residual noise smaller than a preset threshold, ie
  • ⁇ s is the distortion signal
  • ⁇ th is a preset threshold
  • the covariance matrix R r of the received signal can be expressed as the sum of the signal covariance matrix R s and the noise covariance matrix Ru ,
  • R s and R u have the same eigenvector matrix V, so the eigen decomposition of the matrix of R r can be obtained.
  • R s signal covariance matrix and the noise covariance matrix R (3) in u can be expressed as
  • i is the number of samples
  • N is the total number of samples
  • i and N are both natural numbers.
  • the remaining part can be expressed as
  • H 1 represents the presence of the main signal s(t); H 0 indicates that the radio frequency band of interest is not occupied.
  • equations (11) and (12) Under the ideal conditions of complete removal of background noise based on subspace filtering, the effects of equations (11) and (12) are better. However, there is still some ambient noise in the residual signal after subspace filtering. Obviously, equations (11) and (12) leave a lot of errors in estimating the detection and false alarm probability.
  • the corrected detection probability P d ( ⁇ , ⁇ ) can be expressed as
  • ⁇ represents the time of spectrum sensing
  • t is the energy value of the signal received by the secondary user
  • f s is the sampling frequency of the secondary user
  • ⁇ is the variance of the signal and the subspace
  • the ratio of the filtered noise variance, ⁇ is a preset energy threshold, Is the noise variance after subspace filtering.
  • ⁇ 0 is the expectation for spectrum sensing
  • u 1 (n) is the residual noise after subspace filtering
  • the detection probability and the false alarm probability reflect the perceptual error that the system can tolerate from one aspect.
  • the accuracy of the system perception is also different.
  • the purpose of spectrum sensing is to maximize the detection probability for a specific false alarm probability, that is, to detect the frequency band occupied by the primary user as accurately as possible.
  • the process of spectrum sensing based on subspace filtering is: inputting an observation signal of a secondary user, and the observation signal is filtered by a band pass filter to obtain a band pass signal; and using a sampler, after subspace filtering Obtaining an enhanced signal; using a decision maker to determine whether the current wireless network band is occupied; if the frequency band is occupied, storing the REM database after storing it; if not, accessing the corresponding secondary user And continue to detect the next band.
  • Figure 2 depicts the processing flow of spectrum sensing based on subspace filtering.
  • the subspace filtering in the virtual box corresponds to the equations (1) to (7); the decision of the decision maker depends on the mathematical analysis basis of the equations (11) to (15). It is worth noting that when the decision maker believes that the current wireless network band is occupied, the system will start spectrum sensing for other frequency bands within a certain period of time.
  • the detection probability P d ( ⁇ , ⁇ ) of the frequency band and the false alarm probability P f ( ⁇ , ⁇ ) are obtained;
  • ⁇ 0 is the expectation for spectrum sensing
  • Is the variance used for spectrum sensing
  • Is the variance of the noise
  • Q(x) is the complementary cumulative distribution function of the standard normal distribution
  • t is the energy value of the signal received by the secondary user
  • N is the number of samples of the primary user
  • H 1 is the case where the primary signal s(t) exists
  • H 0 is the wireless frequency band is not occupied by the primary user
  • y(n) is The signal observed after sampling by the receiver
  • n is the number of samples at the signal receiving end
  • ⁇ is the ratio of the signal variance to the noise variance after subspace filtering
  • ⁇ is the preset energy threshold
  • ⁇ is the time of spectrum sensing.
  • f s is the sampling frequency of the secondary user
  • u 1 (n) is the residual noise after subspace filtering
  • the real-time occupancy information of the frequency band is obtained by using the detection probability of the frequency band and the false alarm probability.
  • the spectrum refers to the distribution of signal energy at various frequencies.
  • the frequency band is calculated by calculating the spectrum of the signal, and then it can be visually seen in which frequency range (band) the energy of the signal mainly falls, that is, The frequency band is the frequency range;
  • the spectrum resource database of the radio environment map REM includes: the name, access time, access frequency band, access habit, geographical location and occupancy of all users;
  • the spectrum resource database of the radio environment map REM is already established.
  • the database, that is, the spectrum resource database of REM is the default database.
  • the environmental awareness of the cognitive wireless network can be enhanced, and the cognitive wireless network can train self-learning and reasoning from past experience and monitoring results, which is helpful for cognitive wireless network identification. More specific scenarios to meet specific user and global needs.
  • the content of the wireless environment map needs to be updated in real time to accommodate changes in the wireless environment.
  • the secondary user can obtain the global wireless environment map of the base station through the network, and then know the location of the nearby signal tower, the terrain of the area, the prohibited frequency band, the approximate distribution of the receiver, and the available channels, etc., which can assist the secondary users to Excellent transmission power selects the best spectrum opportunity.
  • Location information and geographic environment information are an important part of the wireless environment map.
  • Location awareness refers to the cognitive wireless network determining the absolute geographic location (or relative position relative to the reference node) in which it is located and the accuracy of this location estimate.
  • a conceptual model of the location awareness engine needs to be established.
  • Context awareness enables cognitive wireless networks to understand the geographical environment within their area, thereby providing different application scenarios for cognitive wireless networks, such as target and environment recognition, line-of-sight non-line-of-sight recognition, and seamless positioning.
  • the context-awareness of cognitive wireless networks can also provide location-based services for active spectrum access by secondary users.
  • the primary user occupied frequency band refers to a range between the lowest and highest frequency points where the primary user signal is located
  • the searched user is corrected, thereby obtaining a target problem (the target problem is which primary user occupies a certain frequency band for a specific time) Similar solution;
  • the access line of the primary user can be understood as that the primary user accesses a certain spectrum for a certain period of time; the access rule of the primary user can be understood as a normal user's habit for a specific time period. Which frequency band is connected to.
  • the spectrum resource database of the radio environment map REM includes: the name, access time, access frequency band, access habit, geographical location, and occupancy of all users;
  • Sim as the ratio of the features matched by the two of the two user sets to all its features, Sim ⁇ [0,1]. The larger the value, the higher the similarity between the two, equal to "l”. The same user, equal to "0" is a different user.
  • V A ⁇ a 1 ,...,a j ,...,a L ⁇
  • V B ⁇ b 1 ,...,b j ,...,b L ⁇
  • the corresponding weight can be set, and the weight coefficient w j is introduced into the expression of behavior similarity.
  • ⁇ j indicates the number of primary users accessed on the day
  • L indicates the total number of primary users accessed on the day
  • ⁇ j indicates the primary user accessed on the day
  • ⁇ j indicates that a specific time has been stored in the REM.
  • the user can flexibly set the size of the weight w j according to the specific situation.
  • the user who searches for the geographical difference between the target primary user and the target value is smaller than the preset value, and the user whose geographical difference with the target primary user is less than the preset value is recorded as the set C, and the target primary user and the set C are calculated.
  • the time interval between the calculation target target user and the access frequency band of the user in the set C may be a time interval between the access frequency band of the target primary user and the access frequency band of the user in the set C.
  • correction rule may be understood as a user that determines that the time interval between the access band of the user in the set C and the target primary user is the smallest;
  • a prediction model of the occupied frequency band of the primary user is established.
  • the access habit of the above-mentioned primary user can be understood as considering the time similarity, that is, which primary user is occupying a certain frequency band.
  • the spectrum resource database includes a name, an access time, an access frequency band, a geographical location, and an occupation status of all users, a name corresponding to multiple primary users, an access frequency band, a geographical location, and a connection.
  • Training all users in the set A can be called autoregressive lifting prediction), collecting the access characteristics of the target primary user spectrum according to the temporal similarity, and recording the users close to the target primary user spectrum access characteristics as a set B;
  • the user who searches for the closest geographical location of the target primary user is searched from the set B, and the time interval of the access frequency band of the target primary user and the user closest to the geographical location is calculated;
  • a prediction model of the occupied frequency band of the primary user is established.
  • the secondary user is connected to the target frequency band
  • the spectrum prediction method of the first cognitive wireless network acquires real-time occupation information of the frequency band; and updates the preset radio environment according to the real-time occupation information of the frequency band.
  • the spectrum resource database of the map REM according to the updated spectrum resource database of the radio environment map REM, a prediction model of the occupied frequency band of the primary user is established; and the secondary user is accessed into the target frequency band according to the prediction model of the occupied frequency band of the primary user.
  • the secondary user can grasp the environmental information of the cognitive wireless network. Not only can it effectively reduce the interference that secondary users can cause to the primary user, but also avoid the influence of hidden nodes and exposed nodes, thereby improving the overall performance of the entire cognitive wireless network.
  • FIG. 4 is a flowchart of a method for predicting a spectrum of a second cognitive wireless network according to an embodiment of the present application, where the method includes:
  • ⁇ 0 is the expectation for spectrum sensing
  • Is the variance used for spectrum sensing
  • Is the variance of the noise
  • Q(x) is the complementary cumulative distribution function of the standard normal distribution
  • t is the energy value of the signal received by the secondary user
  • N is the number of samples of the primary user
  • H 1 is the case where the primary signal s(t) exists
  • H 0 is the wireless frequency band is not occupied by the primary user
  • y(n) is The signal observed after sampling by the receiver
  • n is the number of samples at the signal receiving end
  • ⁇ is the ratio of the signal variance to the noise variance after subspace filtering
  • ⁇ is the preset energy threshold
  • ⁇ is the time of spectrum sensing.
  • f s is the sampling frequency of the secondary user
  • u 1 (n) is the residual noise after subspace filtering
  • S203 Acquire real-time occupancy information of the frequency band by using a detection probability of the frequency band and a false alarm probability.
  • the spectrum prediction method of the second cognitive wireless network is to use the mutual support of the spectrum sensing and the radio environment map REM to target the secondary users to the frequency band, thereby improving the accuracy of the prediction.
  • the embodiment of the present application further provides a A spectrum prediction apparatus for a cognitive wireless network.
  • FIG. 5 is a schematic diagram of a spectrum prediction apparatus for a cognitive wireless network according to an embodiment of the present application, where the apparatus includes:
  • the requesting access module 301 is configured to receive a target access request of the secondary user.
  • the spectrum real-time occupancy information acquiring module 302 is configured to acquire real-time occupancy information of the frequency band by using a spectrum sensing method of subspace filtering;
  • a resource library establishing module 303 configured to update a spectrum resource database of the radio environment map REM according to real-time occupancy information of the frequency band;
  • the model obtaining module 304 is configured to establish a prediction model of the occupied frequency band of the primary user according to the updated spectrum resource database of the radio environment map REM;
  • the access behavior module 305 is configured to access the secondary user into the target frequency band according to the prediction model of the primary user occupied frequency band.
  • the spectrum real-time occupancy information obtaining module 302 includes:
  • the detection and false alarm probability acquisition submodule is configured to obtain the detection probability P d ( ⁇ , ⁇ ) of the frequency band and the false alarm probability P f ( ⁇ , ⁇ ) by using the spectrum sensing method of the subspace filtering;
  • ⁇ 0 is the expectation for spectrum sensing
  • Is the variance used for spectrum sensing
  • Is the variance of the noise
  • Q(x) is the complementary cumulative distribution function of the standard normal distribution
  • t is the energy value of the signal received by the secondary user
  • N is the number of samples of the primary user
  • H 1 is the case where the primary signal s(t) exists
  • H 0 is the wireless frequency band is not occupied by the primary user
  • y(n) is The signal observed after sampling by the receiver
  • n is the number of samples at the signal receiving end
  • ⁇ is the ratio of the signal variance to the noise variance after subspace filtering
  • ⁇ is the preset energy threshold
  • ⁇ is the time of spectrum sensing.
  • f s is the sampling frequency of the secondary user
  • u 1 (n) is the residual noise after subspace filtering
  • the real-time occupation information acquisition sub-module is configured to obtain the real-time occupation information of the frequency band by using the detection probability of the frequency band and the false alarm probability.
  • the model obtaining module 304 includes:
  • a user search sub-module configured to search, from the updated spectrum resource database of the radio environment map REM, all users that are close to the target primary user access time after receiving the target primary user access band, and the target primary user All users with similar access times are recorded as set A, wherein the spectrum resource database includes names, access times, access bands, access habits, geographical locations, and occupancy of all users;
  • a user confirmation sub-module configured to train all users in the set A, and confirm an access user that most closely matches the daily access behavior of the target primary user according to the temporal similarity, and the most suitable access user Recorded as collection B;
  • the first determining sub-module is configured to determine whether the access behavior of the user in the set B after the training meets the access behavior of the target primary user; if not, the trigger returns to the sub-module, and if yes, triggers the calculation sub-module;
  • a first returning submodule configured to return to perform, after receiving the target primary user access spectrum, searching for, from the updated spectrum resource database of the radio environment map REM, all users that are close to the target primary user access time;
  • the calculation sub-module is configured to search for a user who is less than the preset value from the set B by the user of the target main user, and record the user whose geographical difference with the target main user is less than the preset value as the set C, and calculate the target main user. Time interval between access bands with users in set C;
  • the correction submodule is configured to correct the user in the set C according to the modification rule and the user that meets the target primary user access rule;
  • a prediction submodule configured to predict an access habit of the target primary user according to the user in the modified set C
  • the prediction model establishing sub-module is configured to establish a prediction model of the main user occupied frequency band according to the access habit of the predicted target primary user.
  • the access behavior module 305 includes:
  • a second determining sub-module configured to determine, according to the prediction model of the occupied frequency band of the primary user, whether the target access request includes the target frequency band of the access time is occupied by the user; if not, triggering the secondary user access sub-module, if Yes, triggering the second return submodule;
  • a secondary user access sub-module for accessing the secondary user to the target frequency band
  • a second returning sub-module configured to return to perform a prediction model according to a frequency band occupied by the primary user, and determine, by the user, whether the target access request includes a frequency band of the access time, and is occupied by the user.
  • the embodiment of the present application further provides an electronic device, as shown in FIG. 6, including a processor 401, a communication interface 402, a memory 403, and a communication bus 404, wherein the processor 401, the communication interface 402, and the memory 403 pass through the communication bus 404.
  • a processor 401 a communication interface 402
  • a memory 403 a communication bus 404
  • the processor 401, the communication interface 402, and the memory 403 pass through the communication bus 404.
  • the processor 401 is configured to perform the following steps when executing the program stored on the memory 403:
  • the secondary user is connected to the target frequency band according to the prediction model of the primary user occupied frequency band.
  • the electronic device provided by the embodiment can utilize the spectrum sensing and the radio ring.
  • the mutual support of the border map REM will connect the secondary users to the frequency band in a targeted manner, thereby improving the accuracy of the prediction.
  • the spectrum sensing method using the subspace filtering may acquire the real-time occupation information of the frequency band according to the target access request of the received secondary user, and use the spectrum sensing method of the subspace filtering to obtain the real-time occupation information of the frequency band.
  • the implementation of the spectrum prediction method of the related cognitive wireless network is the same as that of the cognitive wireless network provided by the foregoing method embodiment, and is not described here.
  • the communication bus mentioned in the above electronic device may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus.
  • PCI Peripheral Component Interconnect
  • EISA Extended Industry Standard Architecture
  • the communication bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one thick line is shown in the figure, but it does not mean that there is only one bus or one type of bus.
  • the communication interface is used for communication between the above electronic device and other devices.
  • the memory may include a random access memory (RAM), and may also include a non-volatile memory (NVM), such as at least one disk storage.
  • RAM random access memory
  • NVM non-volatile memory
  • the memory may also be at least one storage device located away from the aforementioned processor.
  • the above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; or may be a digital signal processing (DSP), dedicated integration.
  • CPU central processing unit
  • NP network processor
  • DSP digital signal processing
  • ASIC Application Specific Integrated Circuit
  • FPGA Field-Programmable Gate Array
  • the embodiment of the present application further provides a computer readable storage medium, where the computer readable storage medium stores a computer program, and when the computer program is executed by the processor, the following steps are implemented:
  • the secondary user is connected to the target frequency band according to the prediction model of the primary user occupied frequency band.
  • the spectrum sensing method using the subspace filtering may acquire the real-time occupation information of the frequency band according to the target access request of the received secondary user, and use the spectrum sensing method of the subspace filtering to obtain the real-time occupation information of the frequency band.
  • the mutual assistance of the spectrum sensing and the radio environment map REM can be utilized to target the secondary user to the frequency band, thereby improving the prediction. Accuracy.
  • the implementation of the spectrum prediction method of the related cognitive wireless network is the same as that of the cognitive wireless network provided by the foregoing method embodiment, and is not described here.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Computer Security & Cryptography (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请实施例提供了一种认知无线网络的频谱预测方法及装置，该方法接收次用户的目标接入请求；利用子空间滤波的频谱感知方法，获取频段的实时占用信息；根据频段的实时占用信息，更新预设的无线电环境地图REM的频谱资源数据库；根据更新后的所述无线电环境地图REM的频谱资源数据库，建立主用户占用频段的预测模型；根据主用户占用频段的预测模型，将次用户接入到目标频段内。本申请实施例提供的方法和装置能够有目标地接入次用户，进而提高无线网络的频谱使用效率。

Description

一种认知无线网络的频谱预测方法及装置
本申请要求于2017年7月11日提交中国专利局、申请号为201710561832.4发明名称为“一种认知无线网络的频谱预测方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及认知无线网络技术领域，特别是涉及一种认知无线网络的频谱预测方法及装置。
背景技术
频谱感知是当前认知无线电的关键技术，也是未来智能天线系统、军事通信和干扰对抗等领域的关键技术。认知无线电作为一种提高频谱资源利用率的无线电技术，其主要解决的问题包括：一方面，无线频谱资源相对稀缺，可供分配的频谱资源越来越少；另一方面，根据美国通信委员会的调查结果，目前已授权的频谱利用率非常低，大多数情况下只有百分之十左右，造成了频谱资源极大浪费。
基于上述可知，认知无线电得到了发展和推广。认知无线电指的是，在不影响主用户(授权用户)正常通信的前提下，具有频谱感知功能的无线通信设备，可以临时借用空闲的授权信道进行通信，从而提高频谱利用率。因此，作为认知无线电网络中的重要一环，主用户所占用频谱的预测是亟待解决的问题。
目前，在主用户接入无线网络的情况下，次用户通常是进行全频段扫描。这样就造成无目的地频谱接入，而且准确率较低。
发明内容
本申请提供了本申请实施例的目的在于提供一种认知无线网络的频谱预测方法及装置，以实现有目标地接入次用户，进而提高无线网络的频谱使用效率；
具体技术方案如下：
一种认知无线网络的频谱预测方法，所述方法包括：
接收次用户的目标接入请求；
利用子空间滤波的频谱感知方法，获取频段的实时占用信息；
根据频段的实时占用信息，更新无线电环境地图REM的频谱资源数据库；
根据更新后的所述无线电环境地图REM的频谱资源数据库，建立主用户占用频段的预测模型；
根据主用户占用频段的预测模型，将次用户接入到目标频段内。
进一步地，所述利用子空间滤波的频谱感知方法，获取频段的实时占用信息，包括：
利用子空间滤波的频谱感知方法，获得频段的检测概率Pd(ε,τ)和虚警概率Pf(ε,τ)；
其中，
Figure PCTCN2017107314-appb-000001
Figure PCTCN2017107314-appb-000002
μ0是用于频谱感知时的期望,
Figure PCTCN2017107314-appb-000003
是用于频谱感知时的方差,
Figure PCTCN2017107314-appb-000004
是噪声的方差，Q(x)是标准正态分布的互补累计分布函数，
Figure PCTCN2017107314-appb-000005
t是次用户所接收到的信号的能量值，N是主用户的采样数量，H1是主信号s(t)存在的情况，H0是无线频段未被主用户占用，y(n)是经过接收机采样后所观察到的信号，n是信号接收端的采样数，γ是信号方差与经过子空间滤波后的噪声方差的比值，ε是预先设定的能量阈值， τ是频谱感知的时间,fs是次用户的采样频率,u1(n)是子空间滤波后的剩余噪声；
利用频段的检测概率和虚警概率，获取频段的实时占用信息。
进一步地，所述根据更新后的所述无线电环境地图REM的频谱资源数据库，建立主用户占用频段的预测模型，包括：
在接收目标主用户接入频段后，从更新后的所述无线电环境地图REM的频谱资源数据库中搜索与目标主用户接入时间相近的所有用户，将与目标主用户接入时间相近的所有用户记为集合A，其中，所述频谱资源数据库包括所有用户对应的名称、接入时间、接入频段、地理位置和是否占用情况；
对集合A中的所有用户进行训练，根据时间相似度，确认与所述目标主用户的日常接入行为最符合的接入用户，将所述最符合的接入用户记为集合B；
判断训练后的集合B中的用户的接入行为是否符合目标主用户的接入行为；
若不符合，返回执行在接收目标主用户接入频谱后，从更新后的所述无线电环境地图REM的频谱资源数据库中搜索与目标主用户接入时间相近的所有用户的步骤；
若符合，从集合B中搜索与目标主用户地理位置差小于预设值的用户，将所述与目标主用户地理位置差小于预设值的用户记为集合C，计算目标主用户与集合C中的用户的接入频段的时间间隔；
根据修正规则和符合目标主用户接入规律的用户，对集合C中的用户进行修正；
根据修正后的集合C中的用户，预测目标主用户的接入习惯；
根据预测目标主用户的接入习惯，建立主用户占用频段的预测模型。
进一步地，所述根据主用户占用频段的预测模型，将次用户接入到目标频段内，包括：
根据主用户占用频段的预测模型，判断所述目标接入请求包括接入时间 的目标频段是否被用户占用；
若为否，将次用户接入目标频段上；
若为是，返回执行根据主用户占用频段的预测模型，判断所述目标接入请求包括接入时间的频段是否被用户占用的步骤。
一种认知无线网络的频谱预测装置，所述装置包括：
请求接入模块；用于接收次用户的目标接入请求；
频谱实时占用信息获取模块，用于利用子空间滤波的频谱感知方法，获取频段的实时占用信息；
资源库建立模块，用于根据频段的实时占用信息，更新无线电环境地图REM的频谱资源数据库；
模型获得模块，用于根据更新后的所述无线电环境地图REM的频谱资源数据库，建立主用户占用频段的预测模型；
接入行为模块，用于根据主用户占用频段的预测模型，将次用户接入到目标频段内。
进一步地，所述频谱实时占用信息获取模块包括：
检测和虚警概率获取子模块，用于利用子空间滤波的频谱感知方法，获得频段的检测概率Pd(ε,τ)和虚警概率Pf(ε,τ)；
其中，
Figure PCTCN2017107314-appb-000006
Figure PCTCN2017107314-appb-000007
μ0是用于频谱感知时的期望,
Figure PCTCN2017107314-appb-000008
是用于频谱感知时的方差,
Figure PCTCN2017107314-appb-000009
是噪声的方差，Q(x)是标准正态分布的互补累计分布函数，
Figure PCTCN2017107314-appb-000010
t是次用户所接收到的信号的能量值,N是主用户的采样数量，H1是主信号s(t)存在的情况，H0是无线频段未被主用户占用，y(n)是经过接收机采样后所观察到的信号，n是信号接收端的采样数，γ是信号方差与经过子空间滤波后的噪声方差的比值，ε是预先设定的能量阈值，τ是频谱感知的时间,fs是次用户的采样频率,u1(n)是子空间滤波后的剩余噪声；
实时占用信息获取子模块，用于利用频段的检测概率和虚警概率，获取频段的实时占用信息。
进一步地，所述模型获得模块包括：
用户搜索子模块，用于在接收目标主用户接入频段后，从更新后的所述无线电环境地图REM的频谱资源数据库中搜索与目标主用户接入时间相近的所有用户，将与目标主用户接入时间相近的所有用户记为集合A，其中，所述频谱资源数据库包括所有用户对应的名称、接入时间、接入频段、接入习惯、地理位置和是否占用情况；
用户确认子模块，用于对集合A中的所有用户进行训练，根据时间相似度，确认与所述目标主用户的日常接入行为最符合的接入用户，将所述最符合的接入用户记为集合B；
第一判断子模块，用于判断训练后的集合B中的用户的接入行为是否符合目标主用户的接入行为；若不符合，触发返回子模块，若符合，触发计算子模块；
第一返回子模块，用于返回执行在接收目标主用户接入频谱后，从更新后的所述无线电环境地图REM的频谱资源数据库中搜索与目标主用户接入时间相近的所有用户的步骤；
计算子模块，用于若符合，从集合B中搜索与目标主用户地理位置差小于预设值的用户，将所述与目标主用户地理位置差小于预设值的用户记为集合C，计算目标主用户与集合C中的用户的接入频段的时间间隔；
修正子模块，用于根据修正规则和符合目标主用户接入规律的用户，对集合C中的用户进行修正；
预测子模块，用于根据修正后的集合C中的用户，预测目标主用户的接入习惯；
预测模型建立子模块，用于根据预测目标主用户的接入习惯，建立主用户占用频段的预测模型。
进一步地，所述接入行为模块包括：
第二判断子模块，用于根据主用户占用频段的预测模型，判断所述目标接入请求包括接入时间的目标频段是否被用户占用；若为否，触发次用户接入子模块，若为是，触发第二返回子模块；
次用户接入子模块，用于将次用户接入目标频段上；
第二返回子模块，用于返回执行根据主用户占用频段的预测模型，判断所述目标接入请求包括接入时间的频段是否被用户占用的步骤。
在本申请实施的又一方面，还提供了一种计算机可读存储介质，所述计算机可读存储介质中存储有指令，当其在计算机上运行时，使得计算机执行上述任一所述的一种认知无线网络中的频谱预测提升方法。
在本申请实施的又一方面，本申请实施例还提供了一种包含指令的计算机程序产品，当其在计算机上运行时，使得计算机执行上述任一所述的一种认知无线网络中的频谱预测提升方法。
由上述的技术方案可见，本申请实施例提供的一种认知无线网络的频谱预测方法，可以利用子空间滤波的频谱感知方法，获取频段的实时占用信息；根据频段的实时占用信息，更新预设的无线电环境地图REM的频谱资源数据库；根据更新后的所述无线电环境地图REM的频谱资源数据库，建立主用户占用频段的预测模型；根据主用户占用频段的预测模型，将次用户接入到目标频段 内。该方法能够有目标地接入次用户，减少了频谱感知时间和能量，进而也提高无线网络的频谱使用效率。当然，实施本申请的任一产品或方法必不一定需要同时达到以上所述的所有优点。
附图说明
为了更清楚地说明本申请实施例和现有技术的技术方案，下面对实施例和现有技术中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本申请文件实施例提供的第一种认知无线网络的频谱预测方法流程图；
图2为本申请文件实施例提供的一种基于子空间滤波的频谱感知流程图；
图3为本申请文件实施例提供的一种主用户的频谱接入预测流程图；
图4为本申请文件实施例提供的第二种认知无线网络的频谱预测方法流程图；
图5为本申请文件实施例提供的一种认知无线网络的频谱预测装置示意图；
图6为本申请文件实施例提供的一种电子设备示意图。
具体实施方式
为使本申请的目的、技术方案、及优点更加清楚明白，以下参照附图并举实施例，对本申请进一步详细说明。显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
图1为本申请文件实施例提供的第一种认知无线网络的频谱预测方法流程图，所述方法包括：
S101，接收次用户的目标接入请求；
其中，主用户是认知无线网络中被授权使用某一特定频段的用户，次用户是认知无线网络中未被授权使用某一特定频段的用户。
需要说明的是，本申请的执行主体是基站。
S102，利用子空间滤波的频谱感知方法，获取频段的实时占用信息；
其中，实时占用信息包括接入时间、接入频段和地理位置；
具体的，本步骤是根据接收的次用户的目标接入请求，利用子空间滤波的频谱感知方法，获取频段的实时占用信息；
认知用户通过各种信号检测和处理手段来获取无线网络中的频谱使用信息。从无线网络的功能分层角度看，频谱感知技术主要涉及物理层和链路层，其中物理层主要关注各种具体的本地检测算法，而链路层主要关注用户间的协作以及对本地感知、协作感知和感知机制优化3个方面。
为了更加清楚的描述，对子空间滤波的频谱感知方法进行如下简述：
子空间滤波是一种主动的降噪方法，其本质在于构造线性估计量H之后，将收到信号r(n)分解成两个正交的子空间：信号噪声子空间Hy(n)和噪声子空间(I-H)r(n)，y(n)为经过接收机采样后所观察到的信号(包括信号和噪声)，I为与H维数相同的单位矩阵，信号噪声子空间与噪声子空间一起构成单位矩阵，故有I-H，噪声子空间只包含噪声，而信号噪声子空间包含分段噪声和全信号，n为信号接收端的采样数。在去除噪声子空间中的全部噪声以及信号噪声子空间中的噪声后,获得最优子空间滤波
Figure PCTCN2017107314-appb-000011
可表示为：
Figure PCTCN2017107314-appb-000012
因此，最优子空间滤波的主要工作是如何获得最优线性估计量Hopt。最优线性估计量Hopt可以通过最小化信号失真，并使残留噪声小于预先设置的阈值，即
Figure PCTCN2017107314-appb-000013
其中，εs为失真信号，
Figure PCTCN2017107314-appb-000014
为失真信号的平均功率值，Δth为预先设置的阈值。
由于背景噪声是白色的,因此所接收到信号的协方差矩阵Rr可以表示为信号协方差矩阵Rs与噪声协方差矩阵Ru的和，有
Rr＝Rs+Ru          (3)
同时，Rs和Ru具有相同的特征向量矩阵V，所以对Rr进行矩阵的特征分解可以得到
Rr＝VΛrVT           (4)
其中，Λr为由特征值λ1,...,λN所构成的向量矩阵，Λr＝diag[λ1,...,λk,...,λN]包含矩阵Rr的所有特征值。如果特征值按降序排列,如λ1≥...≥λk≥...≥λN，将得到λk>σ2，则k＝1,2,...,K，λK+1＝...＝λN＝σ2。如果λk>σ2代表信号子空间，那么剩余的则表示噪声子空间。因此,式子(3)中的信号协方差矩阵Rs和噪声协方差矩阵Ru可分别表示为
Figure PCTCN2017107314-appb-000015
Figure PCTCN2017107314-appb-000016
所以，Hopt可近似等价表示为
Hopt=VΛr(Λr+μΛr)-1V-1        （7）
其中，μ为拉格朗日常数。
由于在实际应用中，上述协方差矩阵不可能获得，因此将其替换为样本的协方差矩阵Rr
Figure PCTCN2017107314-appb-000017
其中，i为采样个数，N为采样的总数量，i、N均为自然数。
理论上讲，通过子空间滤波的方法去除背景噪声后，剩余的部分可以表示为
Figure PCTCN2017107314-appb-000018
如果将信号s(n)的能量Es定义为
Es=E(|s(n)|2)≈E(|Hopts(n)|2)        （10）
那么相应的检测概率Pd和虚警概率Pf则可以表示成Pd=P(E(|Hoptr(n)|2)>0|H1)       （11）
Pf=P(E(|Hoptr(n)|2)>0|H0)        （12）
其中，频谱感知是一个二元假设问题，接收机的接收信号可以表示为：
H1:r(t)＝s(t)+u(t)
H0:r(t)＝u(t)
其中，H1表示主信号s(t)存在的情况；H0表示感兴趣的无线频段未被占用。
在接收端经过采样后后，上述两种情况所观察到的信号可以改写为：
H1:r(n)＝s(n)+u(n)
H0:r(n)＝u(n)
在基于子空间滤波的背景噪声完全去除的理想条件下，式子(11)和(12)的效果较好。然而，在子空间滤波后的残差信号仍存在一定的环境噪声。显然，式子(11)和(12)在估计检测和虚警概率方面留下了许多误差。
根据中心极限定理，我们假设子空间滤波后的残差仍服从独立同分布的高斯分布。由于经过子空间滤波处理的噪声，其分布与高斯分布的方差和平均值有关，因此修正后的检测概率Pd(ε,τ)可表示为
Figure PCTCN2017107314-appb-000019
Figure PCTCN2017107314-appb-000020
其中，τ代表频谱感知的时间,t是次用户所接收到的信号的能量值，fs是次用户的采样频率,N＝τfs是主用户的采样数量,γ是信号方差与经 过子空间滤波后的噪声方差的比值，ε是预先设定的能量阈值，
Figure PCTCN2017107314-appb-000021
是子空间滤波后的噪声方差。
相应地，虚警概率Pf(ε,τ)可表示为
Figure PCTCN2017107314-appb-000022
其中，μ0是用于频谱感知时的期望,
Figure PCTCN2017107314-appb-000023
是用于频谱感知时的方差,
Figure PCTCN2017107314-appb-000024
u1(n)是子空间滤波后的剩余噪声，
Figure PCTCN2017107314-appb-000025
是噪声的方差。
检测概率和虚警概率从一个侧面反映了系统所能容忍的感知误差，当设定不同的检测概率和虚警概率条件时，相应地，系统感知的准确率也不尽相同。频谱感知的目的是针对某一特定的虚警概率，最大程度地提升检测概率，也就是尽可能准确地检测到主用户正在占用的频段。
如图2所示，基于子空间滤波的频谱感知的流程为：输入次用户的观测信号，所述观测信号通过带通滤波器进行滤波，获得带通信号；利用采样器，经过子空间滤波后，获得增强后的信号；利用决策器判断当前的无线网络频段是否被占用，若所述频段被占用，则将其存储后对REM数据库进行更新；若未被占用，则接入相应的次用户，并继续下一频段的检测。
图2描述了基于子空间滤波的频谱感知的处理流程。其中，虚框内表示子空间过滤，对应于式子(1)～(7)；决策器的判决有赖于式子(11)～(15)的数学分析基础。值得注意的是，当决策器认为当前的无线网络频段被占用时，系统将会在一定时间内开始对其他频段进行频谱感知。
具体的，
利用子空间滤波的频谱感知方法，获得频段的检测概率Pd(ε,τ)和虚警概率Pf(ε,τ)；
其中，
Figure PCTCN2017107314-appb-000026
Figure PCTCN2017107314-appb-000027
μ0是用于频谱感知时的期望,
Figure PCTCN2017107314-appb-000028
是用于频谱感知时的方差,
Figure PCTCN2017107314-appb-000029
是噪声的方差，Q(x)是标准正态分布的互补累计分布函数，
Figure PCTCN2017107314-appb-000030
t是次用户所接收到的信号的能量值，N是主用户的采样数量，H1是主信号s(t)存在的情况，H0是无线频段未被主用户占用，y(n)是经过接收机采样后所观察到的信号，n是信号接收端的采样数，γ是信号方差与经过子空间滤波后的噪声方差的比值，ε是预先设定的能量阈值，τ是频谱感知的时间,fs是次用户的采样频率,u1(n)是子空间滤波后的剩余噪声；
利用频段的检测概率和虚警概率，获取频段的实时占用信息。
S103，根据频段的实时占用信息，更新无线电环境地图REM的频谱资源数据库；
其中，频谱是指信号能量在各个频率上的分布，频段是通过计算出信号的频谱,然后就能直观地看到,该信号的能量主要落在哪些频率范围(频段)上，也就是说，频段是频率范围；
无线电环境地图REM的频谱资源数据库包括：所有用户对应的名称、接入时间、接入频段、接入习惯、地理位置和是否占用情况；
需要说明的是，无线电环境地图REM的频谱资源数据库是已经建立好的 数据库，也就是说，REM的频谱资源数据库是预设的数据库。
另外，通过查询和访问无线环境地图可以增强认知无线网络的环境意识，并使认知无线网络能够从过去的经验和监测结果中训练自我学习和推理的能力，有利于帮助认知无线网络识别更多特定的场景，以满足特定用户和全局的需求。无线环境地图的内容需要进行实时更新以适应无线环境的变化。
次用户可以通过网络获得基站的全局无线环境地图，进而获知附近信号塔的位置、所在区域的地形、禁止使用的频段、接收机的大致分布以及可用信道等信息，这些信息能够辅助次用户以最优的传输功率选择最好的频谱机会。
位置信息和地理环境信息是无线环境地图的重要组成部分。位置感知是指认知无线网络确定其所在的绝对地理位置(或相对参考节点的相对位置)以及这个位置估计的精度。为了实现位置感知，需要建立位置感知引擎的概念模型。
环境感知使认知无线网络能够了解其所在区域内的地理环境情况，从而为认知无线网络提供不同的应用场景，如目标和环境识别、视距非视距识别、无缝定位等。此外，认知无线网络的环境感知能力还可以为次用户的主动频谱接入提供基于位置的服务。
S104，根据更新后的所述无线电环境地图REM的频谱资源数据库，建立主用户占用频段的预测模型；
其中，主用户占用频段是指主用户信号所在的最低和最高频点之间的范围；
如图3所述，为了更加详细的描述主用户的频谱接入行为，先对其详细的处理步骤进行描述：
首先，当主用户开始接入频谱时，在已存储的REM中进行搜索，寻找与该主用户接入时间相近的所有用户；
其次，利用上述与主用户接近的所有用户进行学习，根据时间相似度，确认与所述主用户的日常接入行为最符合的接入用户，通过时间相似度分析并确定上述哪一个最符合该主用户的日常接入习惯。如果搜索失败，表明用 户行为的覆盖面还不够，要继续学习，即自回归提升预测；
接着，通过最符合该主用户接入行为的用户，找出与最符合主用户的用户最邻近的下一个用户，并计算二者的时间间隔；
然后，根据符合主用户接入规律的用户和修正规则，对该搜索到的用户进行修正，从而得到目标问题(所述目标问题为针对某一特定时间，占用某一频段的是哪个主用户)的相似解；
最后，对该用户进行审查修订，并评估其保留的必要性。
其中，主用户的接入行可以理解为对于某一时间段，主用户接入到某一频谱；主用户的接入规律可以理解为通常情况下，对于某一具体的时间段，主用户习惯接入到哪一个频段内。
具体的，
在接收目标主用户接入频段后，从更新后的所述无线电环境地图REM的频谱资源数据库中搜索与目标主用户接入时间相近的所有用户，将与目标主用户接入时间相近的所有用户记为集合A；其中，无线电环境地图REM的频谱资源数据库包括：所有用户对应的名称、接入时间、接入频段、接入习惯、地理位置和是否占用情况；
对集合A中的所有用户进行训练，根据时间相似度，确认与所述目标主用户的日常接入行为最符合的接入用户，将所述最符合的接入用户记为集合B；
其中，上述日常接入行为可以用特征集Feature和关系集Relation构成用户User的结构集U:U＝{Feature,Relation}描述；结构集中还可以包含其他的属性(比如结合主用户频谱接入行为的规律性不同，使结构集含有权重系数，即U:U＝{Feature,Relation,W})，其中，W表示权重系数的集合。收集主用户频谱接入特征；
判断训练后的集合B中的用户的接入行为是否符合目标主用户的接入行为；
为了更加描述清楚，现举一示例对其进行描述；
示例：假如用户User1和用户User2的所有特征全部匹配，那么U1＝U2，这两个用户的集合相同。如果部分特征值不相同，那么这两个用户为部分相似， 可以用相似度Sim来表示二者之间的相似程度。
将Sim定义为两个用户集合中二者所匹配的特征与其所有特征的比值，Sim∈[0,1]，该数值越大，表示两者间的相似程度越高，等于“l”即为相同的用户，等于“0”则为不同的用户。
任意两个用户的集合可定义为VA＝{a1,…,aj,…,aL}，VB＝{b1,…,bj,…,bL}，它们之间的行为相似度能够表示为
Figure PCTCN2017107314-appb-000031
如果用户的频谱接入行为不是很稳定，那么可以设置对应的权重，将权重系数wj引入行为相似度的表达式中，有
Figure PCTCN2017107314-appb-000032
其中
Figure PCTCN2017107314-appb-000033
j表示当天所接入的主用户的个数，L表示当天所接入的主用户的总数量，αj表示当天所接入的主用户，βj表示REM中已存储的对应某一特定时间所接入的主用户。当aj＝bj时，表示当前接入的主用户与REM中所存储的主用户是相同的，此时，sin(aj,bj)＝1；反之，sin(aj,bj)＝0。同时，用户可以根据具体的实际情况灵活地设置权重wj的大小。
若不符合，返回执行在接收目标主用户接入频谱后，从更新后的所述无线电环境地图REM的频谱资源数据库中搜索与目标主用户接入时间相近的所有用户的步骤；
若符合，从集合B中搜索与目标主用户地理位置差小于预设值的用户，将所述与目标主用户地理位置差小于预设值的用户记为集合C，计算目标主用户与集合C中的用户的接入频段的时间间隔；
具体的，上述计算目标主用户与集合C中的用户的接入频段的时间间隔可以是计算目标主用户的接入频段与集合C中的用户的接入频段的时间间隔。
根据修正规则和符合目标主用户接入规律的用户，对集合C中的用户进行修正；
需要说明的是，修正规则可以理解为确定集合C中的用户与目标主用户的接入频段的时间间隔为最小的用户；
根据修正后的集合C中的用户，预测目标主用户的接入习惯；
根据预测目标主用户的接入习惯，建立主用户占用频段的预测模型。
其中，上述主用户的接入习惯可以理解为利用时间相似度进行考量，也就是说，占用某一频段的是哪一个主用户。
需要说明的是，如图3所示，上述步骤也可以描述为：
根据主用户的既往频谱接入特征，从更新后的所述无线电环境地图REM的频谱资源数据库中搜索与目标主用户接入时间相近的所有用户，将与目标主用户接入时间相近的所有用户记为集合A，其中，所述频谱资源数据库包括所有用户对应的名称、接入时间、接入频段、地理位置和是否占用情况，多个主用户对应的名称、接入频段、地理位置和接入习惯；
对集合A中的所有用户进行训练(可以称为自回归提升预测)，根据时间相似度，收集目标主用户频谱的接入特征，将与目标主用户频谱接入特征相近的用户，记为集合B；
根据目标主用户频谱的接入特征，判断是否存在与目标主用户“相同”的用户；
若为否，返回执行在接收目标主用户接入频谱后，从更新后的所述无线电环境地图REM的频谱资源数据库中搜索与目标主用户接入时间相近的所有用户的步骤；
若为是，从集合B中搜索与目标主用户地理位置最近的用户，计算目标主用户与所述地理位置最近的用户的接入频段的时间间隔；
根据修正规则和符合目标主用户接入特征的用户，对集合B中的用户进行修正；
根据修正后的集合B中的用户，预测目标主用户的接入习惯；
根据预测目标主用户的接入习惯，建立主用户占用频段的预测模型。
S105，根据主用户占用频段的预测模型，将次用户接入到目标频段内。
具体的，
根据主用户占用频段的预测模型，判断所述目标接入请求包括接入时间的目标频段是否被用户占用；
若为否，将次用户接入目标频段上；
若为是，返回执行根据主用户占用频段的预测模型，判断所述目标接入请求包括接入时间的频段是否被用户占用的步骤。
由上可见，本实施提供的第一种认知无线网络的频谱预测方法，即利用子空间滤波的频谱感知方法，获取频段的实时占用信息；根据频段的实时占用信息，更新预设的无线电环境地图REM的频谱资源数据库；根据更新后的所述无线电环境地图REM的频谱资源数据库，建立主用户占用频段的预测模型；根据主用户占用频段的预测模型，将次用户接入到目标频段内。能够有目标地接入次用户，进而提高预测的准确率。
另外，在无线环境地图的辅助下，次用户可以掌握所在认知无线网络的环境信息。不但能够有效地降低次用户对主用户可能造成的干扰，而且还可以避免隐藏节点和暴露节点的影响，从而提高整个认知无线网络的综合性能。
图4为本申请文件实施例提供的第二种认知无线网络的频谱预测方法流程图，所述方法包括：
S201，接收次用户的目标接入请求；
S202，利用子空间滤波的频谱感知方法，获得频段的检测概率Pd(ε,τ)和虚警概率Pf(ε,τ)；
其中，
Figure PCTCN2017107314-appb-000034
Figure PCTCN2017107314-appb-000035
μ0是用于频谱感知时的期望,
Figure PCTCN2017107314-appb-000036
是用于频谱感知时的方差,
Figure PCTCN2017107314-appb-000037
是噪声的方差，Q(x)是标准正态分布的互补累计分布函数，
Figure PCTCN2017107314-appb-000038
t是次用户所接收到的信号的能量值,N是主用户的采样数量，H1是主信号s(t)存在的情况，H0是无线频段未被主用户占用，y(n)是经过接收机采样后所观察到的信号，n是信号接收端的采样数，γ是信号方差与经过子空间滤波后的噪声方差的比值，ε是预先设定的能量阈值，τ是频谱感知的时间,fs是次用户的采样频率,u1(n)是子空间滤波后的剩余噪声；
S203，利用频段的检测概率和虚警概率，获取频段的实时占用信息。
S204，根据频段的实时占用信息，更新无线电环境地图REM的频谱资源数据库；
S205，根据更新后的所述无线电环境地图REM的频谱资源数据库，建立主用户占用频段的预测模型；
S206，根据主用户占用频段的预测模型，将次用户接入到目标频段内。
由此可见，本实施提供的第二种认知无线网络的频谱预测方法是利用频谱感知和无线电环境地图REM的相互支持，将次用户有目标地接入频段中，进而提高预测的准确率。
与上述认知无线网络的频谱预测方法相对应，本申请实施例还提供了一 种认知无线网络的频谱预测装置。
图5为本申请文件实施例提供的一种认知无线网络的频谱预测装置示意图，所述装置包括：
请求接入模块301；用于接收次用户的目标接入请求；
频谱实时占用信息获取模块302，用于利用子空间滤波的频谱感知方法，获取频段的实时占用信息；
资源库建立模块303，用于根据频段的实时占用信息，更新无线电环境地图REM的频谱资源数据库；
模型获得模块304，用于根据更新后的所述无线电环境地图REM的频谱资源数据库，建立主用户占用频段的预测模型；
接入行为模块305，用于根据主用户占用频段的预测模型，将次用户接入到目标频段内。
具体的，
所述频谱实时占用信息获取模块302包括：
检测和虚警概率获取子模块，用于利用子空间滤波的频谱感知方法，获得频段的检测概率Pd(ε,τ)和虚警概率Pf(ε,τ)；
其中，
Figure PCTCN2017107314-appb-000039
Figure PCTCN2017107314-appb-000040
μ0是用于频谱感知时的期望,
Figure PCTCN2017107314-appb-000041
是用于频谱感知时的方差,
Figure PCTCN2017107314-appb-000042
是噪声的方差，Q(x)是标准正态分布的互补累计分布函数，
Figure PCTCN2017107314-appb-000043
t是次用户所接收到的信号的能量值,N是主用户的采样数量，H1是主信号s(t)存在的情况，H0是无线频段未被主用户占用，y(n)是经过接收机采样后所观察到的信号，n是信号接收端的采样数，γ是信号方差与经过子空间滤波后的噪声方差的比值，ε是预先设定的能量阈值，τ是频谱感知的时间,fs是次用户的采样频率,u1(n)是子空间滤波后的剩余噪声；
实时占用信息获取子模块，用于利用频段的检测概率和虚警概率，获取频段的实时占用信息。
所述模型获得模块304包括：
用户搜索子模块，用于在接收目标主用户接入频段后，从更新后的所述无线电环境地图REM的频谱资源数据库中搜索与目标主用户接入时间相近的所有用户，将与目标主用户接入时间相近的所有用户记为集合A，其中，所述频谱资源数据库包括所有用户对应的名称、接入时间、接入频段、接入习惯、地理位置和是否占用情况；
用户确认子模块，用于对集合A中的所有用户进行训练，根据时间相似度，确认与所述目标主用户的日常接入行为最符合的接入用户，将所述最符合的接入用户记为集合B；
第一判断子模块，用于判断训练后的集合B中的用户的接入行为是否符合目标主用户的接入行为；若不符合，触发返回子模块，若符合，触发计算子模块；
第一返回子模块，用于返回执行在接收目标主用户接入频谱后，从更新后的所述无线电环境地图REM的频谱资源数据库中搜索与目标主用户接入时间相近的所有用户的步骤；
计算子模块，用于从集合B中搜索与目标主用户地理位置差小于预设值的用户，将所述与目标主用户地理位置差小于预设值的用户记为集合C，计算目标主用户与集合C中的用户的接入频段的时间间隔；
修正子模块，用于根据修正规则和符合目标主用户接入规律的用户，对集合C中的用户进行修正；
预测子模块，用于根据修正后的集合C中的用户，预测目标主用户的接入习惯；
预测模型建立子模块，用于根据预测目标主用户的接入习惯，建立主用户占用频段的预测模型。
所述接入行为模块305包括：
第二判断子模块，用于根据主用户占用频段的预测模型，判断所述目标接入请求包括接入时间的目标频段是否被用户占用；若为否，触发次用户接入子模块，若为是，触发第二返回子模块；
次用户接入子模块，用于将次用户接入目标频段上；
第二返回子模块，用于返回执行根据主用户占用频段的预测模型，判断所述目标接入请求包括接入时间的频段是否被用户占用的步骤。
本申请实施例还提供了一种电子设备，如图6所示，包括处理器401、通信接口402、存储器403和通信总线404，其中，处理器401，通信接口402，存储器403通过通信总线404完成相互间的通信，
存储器403，用于存放计算机程序；
处理器401，用于执行存储器403上所存放的程序时，实现如下步骤：
接收次用户的目标接入请求；
利用子空间滤波的频谱感知方法，获取频段的实时占用信息；
根据频段的实时占用信息，更新无线电环境地图REM的频谱资源数据库；
根据更新后的所述无线电环境地图REM的频谱资源数据库，建立主用户占用频段的预测模型；
根据主用户占用频段的预测模型，将次用户接入到目标频段内。
由此可见，执行本实施例提供的电子设备能够利用频谱感知和无线电环 境地图REM的相互支持，将次用户有目标地接入频段中，进而提高预测的准确率。
上述利用子空间滤波的频谱感知方法，获取频段的实时占用信息可以是根据接收的次用户的目标接入请求，利用子空间滤波的频谱感知方法，获取频段的实时占用信息。
上述的相关认知无线网络的频谱预测方法的实施方式与前述方法实施例部分提供的认知无线网络的频谱预测提升方式相同，这里不再赘述。
上述电子设备提到的通信总线可以是外设部件互连标准(Peripheral Component Interconnect，PCI)总线或扩展工业标准结构(Extended Industry Standard Architecture，EISA)总线等。该通信总线可以分为地址总线、数据总线、控制总线等。为便于表示，图中仅用一条粗线表示，但并不表示仅有一根总线或一种类型的总线。
通信接口用于上述电子设备与其他设备之间的通信。
存储器可以包括随机存取存储器(Random Access Memory，RAM)，也可以包括非易失性存储器(Non-Volatile Memory，NVM)，例如至少一个磁盘存储器。可选的，存储器还可以是至少一个位于远离前述处理器的存储装置。
上述的处理器可以是通用处理器，包括中央处理器(Central Processing Unit，CPU)、网络处理器(Network Processor，NP)等；还可以是数字信号处理器(Digital Signal Processing，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现场可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。
本申请实施例还提供了一种计算机可读存储介质，所述计算机可读存储介质内存储有计算机程序，所述计算机程序被处理器执行时实现如下步骤：
接收次用户的目标接入请求；
利用子空间滤波的频谱感知方法，获取频段的实时占用信息；
根据频段的实时占用信息，更新无线电环境地图REM的频谱资源数据库；
根据更新后的所述无线电环境地图REM的频谱资源数据库，建立主用户占用频段的预测模型；
根据主用户占用频段的预测模型，将次用户接入到目标频段内。
上述利用子空间滤波的频谱感知方法，获取频段的实时占用信息可以是根据接收的次用户的目标接入请求，利用子空间滤波的频谱感知方法，获取频段的实时占用信息。
由此可见，执行本实施例提供的计算机可读存储介质中存储的应用程序时，能够利用频谱感知和无线电环境地图REM的相互支持，将次用户有目标地接入频段中，进而提高预测的准确率。
上述的相关认知无线网络的频谱预测方法的实施方式与前述方法实施例部分提供的认知无线网络的频谱预测提升方式相同，这里不再赘述。
需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
本说明书中的各个实施例均采用相关的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其他实施例的不同之处。尤其，对于装置、电子设备或存储介质实施例而言，由于其基本相似于方法实施例，所以描述的比较简单，相关之处参见方法实施例的部分说明即可。
以上所述仅为本申请的较佳实施例而已，并非用于限定本申请的保护范围。凡在本申请的精神和原则之内所作的任何修改、等同替换、改进等，均包含在本申请的保护范围内。

Claims (10)

  1. 一种认知无线网络的频谱预测方法，其特征在于，所述方法包括：
    接收次用户的目标接入请求；
    利用子空间滤波的频谱感知方法，获取频段的实时占用信息；
    根据频段的实时占用信息，更新无线电环境地图REM的频谱资源数据库；
    根据更新后的所述无线电环境地图REM的频谱资源数据库，建立主用户占用频段的预测模型；
    根据主用户占用频段的预测模型，将次用户接入到目标频段内。
  2. 如权利要求1所述的方法，其特征在于，所述利用子空间滤波的频谱感知方法，获取频段的实时占用信息，包括：
    利用子空间滤波的频谱感知方法，获得频段的检测概率Pd(ε,τ)和虚警概率Pf(ε,τ)；
    其中，
    Figure PCTCN2017107314-appb-100001
    Figure PCTCN2017107314-appb-100002
    μ0是用于频谱感知时的期望,
    Figure PCTCN2017107314-appb-100003
    是用于频谱感知时的方差,
    Figure PCTCN2017107314-appb-100004
    是噪声的方差，Q(x)是标准正态分布的互补累计分布函数，
    Figure PCTCN2017107314-appb-100005
    t是次用户所接收到的信号的能量值,N是主用户的采样数量，H1是主信号s(t)存在的情况，H0是无线频段未被主用户占用，y(n)是经过接收机采样后所观察到的信号，n是信号接收端的采样数，γ是信号方 差与经过子空间滤波后的噪声方差的比值，ε是预先设定的能量阈值，τ是频谱感知的时间,fs是次用户的采样频率,u1(n)是子空间滤波后的剩余噪声；
    利用频段的检测概率和虚警概率，获取频段的实时占用信息。
  3. 如权利要求1所述的方法，其特征在于，所述根据更新后的所述无线电环境地图REM的频谱资源数据库，建立主用户占用频段的预测模型，包括：
    在接收目标主用户接入频段后，从更新后的所述无线电环境地图REM的频谱资源数据库中搜索与目标主用户接入时间相近的所有用户，将与目标主用户接入时间相近的所有用户记为集合A，其中，所述频谱资源数据库包括所有用户对应的名称、接入时间、接入频段、地理位置、接入习惯和是否占用情况；
    对集合A中的所有用户进行训练，根据时间相似度，确认与所述目标主用户的日常接入行为最符合的接入用户，将所述最符合的接入用户记为集合B；
    判断训练后的集合B中的用户的接入行为是否符合目标主用户的接入行为；
    若不符合，返回执行在接收目标主用户接入频段后，从更新后的所述无线电环境地图REM的频谱资源数据库中搜索与目标主用户接入时间相近的所有用户的步骤；
    若符合，从集合B中搜索与目标主用户地理位置差小于预设值的用户，将所述与目标主用户地理位置差小于预设值的用户记为集合C，计算目标主用户与集合C中的用户的接入频段的时间间隔；
    根据修正规则和符合目标主用户接入规律的用户，对集合C中的用户进行修正；
    根据修正后的集合C中的用户，预测目标主用户的接入习惯；
    根据预测目标主用户的接入习惯，建立主用户占用频段的预测模型。
  4. 如权利要求1所述的方法，其特征在于，所述根据主用户占用频段的预测模型，将次用户接入到目标频段内，包括：
    根据主用户占用频段的预测模型，判断所述目标接入请求包括接入时间的目标频段是否被用户占用；
    若为否，将次用户接入目标频段上；
    若为是，返回执行根据主用户占用频段的预测模型，判断所述目标接入请求包括接入时间的频段是否被用户占用的步骤。
  5. 一种认知无线网络的频谱预测装置，其特征在于，所述装置包括：
    请求接入模块；用于接收次用户的目标接入请求；
    频谱实时占用信息获取模块，用于利用子空间滤波的频谱感知方法，获取频段的实时占用信息；
    资源库建立模块，用于根据频段的实时占用信息，更新无线电环境地图REM的频谱资源数据库；
    模型获得模块，用于根据更新后的所述无线电环境地图REM的频谱资源数据库，建立主用户占用频段的预测模型；
    接入行为模块，用于根据主用户占用频段的预测模型，将次用户接入到目标频段内。
  6. 如权利要求5所述的装置，其特征在于，所述频谱实时占用信息获取模块包括：
    检测和虚警概率获取子模块，用于利用子空间滤波的频谱感知方法，获得频段的检测概率Pd(ε,τ)和虚警概率Pf(ε,τ)；
    其中，
    Figure PCTCN2017107314-appb-100006
    Figure PCTCN2017107314-appb-100007
    μ0是用于频谱感知时的期望,
    Figure PCTCN2017107314-appb-100008
    是用于频谱感知时的方差,
    Figure PCTCN2017107314-appb-100009
    是噪声的方差，Q(x)是标准正态分布的互补累计分布函数，
    Figure PCTCN2017107314-appb-100010
    t是次用户所接收到的信号的能量值,N是主用户的采样数量，H1是主信号s(t)存在的情况，H0是无线频段未被主用户占用，y(n)是经过接收机采样后所观察到的信号，n是信号接收端的采样数，γ是信号方差与经过子空间滤波后的噪声方差的比值，ε是预先设定的能量阈值，τ是频谱感知的时间,fs是次用户的采样频率,u1(n)是子空间滤波后的剩余噪声；
    实时占用信息获取子模块，用于利用频段的检测概率和虚警概率，获取频段的实时占用信息。
  7. 如权利要求5所述的装置，其特征在于，所述模型获得模块包括：
    用户搜索子模块，用于在接收目标主用户接入频段后，从更新后的所述无线电环境地图REM的频谱资源数据库中搜索与目标主用户接入时间相近的所有用户，将与目标主用户接入时间相近的所有用户记为集合A，其中，所述频谱资源数据库包括所有用户对应的名称、接入时间、接入频段、接入习惯、地理位置和是否占用情况；
    用户确认子模块，用于对集合A中的所有用户进行训练，根据时间相似度，确认与所述目标主用户的日常接入行为最符合的接入用户，将所述最符合的接入用户记为集合B；
    第一判断子模块，用于判断训练后的集合B中的用户的接入行为是否符合目标主用户的接入行为；若不符合，触发返回子模块，若符合，触发计算子模块；
    第一返回子模块，用于返回执行在接收目标主用户接入频谱后，从更新后的所述无线电环境地图REM的频谱资源数据库中搜索与目标主用户接入时间相近的所有用户的步骤；
    计算子模块，用于从集合B中搜索与目标主用户地理位置差小于预设值的用户，将所述与目标主用户地理位置差小于预设值的用户记为集合C，计算目标主用户与集合C中的用户的接入频段的时间间隔；
    修正子模块，用于根据修正规则和符合目标主用户接入规律的用户，对集合C中的用户进行修正；
    预测子模块，用于根据修正后的集合C中的用户，预测目标主用户的接入习惯；
    预测模型建立子模块，用于根据预测目标主用户的接入习惯，建立主用户占用频段的预测模型。
  8. 如权利要求5所述的装置，其特征在于，所述接入行为模块包括：
    第二判断子模块，用于根据主用户占用频段的预测模型，判断所述目标接入请求包括接入时间的目标频段是否被用户占用；若为否，触发次用户接入子模块，若为是，触发第二返回子模块；
    次用户接入子模块，用于将次用户接入目标频段上；
    第二返回子模块，用于返回执行根据主用户占用频段的预测模型，判断所述目标接入请求包括接入时间的频段是否被用户占用的步骤。
  9. 一种电子设备，其特征在于，包括处理器、通信接口、存储器和通信总线，其中，处理器，通信接口，存储器通过通信总线完成相互间的通信；
    存储器，用于存放计算机程序；
    处理器，用于执行存储器上所存放的程序时，实现权利要求1-4任一所述的方法步骤。
  10. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质内存储有计算机程序，所述计算机程序被处理器执行时实现权利要求1-4任一所述的方法步骤。
PCT/CN2017/107314 2017-07-11 2017-10-23 一种认知无线网络的频谱预测方法及装置 WO2019010861A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US16/619,207 US10674434B2 (en) 2017-07-11 2017-10-23 Frequency spectrum prediction method and apparatus for cognitive wireless network

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710561832.4 2017-07-11
CN201710561832.4A CN107359948B (zh) 2017-07-11 2017-07-11 一种认知无线网络的频谱预测方法、装置及计算机可读存储介质

Publications (1)

Publication Number Publication Date
WO2019010861A1 true WO2019010861A1 (zh) 2019-01-17

Family

ID=60291923

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/107314 WO2019010861A1 (zh) 2017-07-11 2017-10-23 一种认知无线网络的频谱预测方法及装置

Country Status (3)

Country Link
US (1) US10674434B2 (zh)
CN (1) CN107359948B (zh)
WO (1) WO2019010861A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111565081A (zh) * 2020-04-28 2020-08-21 深圳中科国威信息系统技术有限公司 认知无线电中白频谱频段的扫描方法、介质、终端和装置
CN112968740A (zh) * 2021-02-01 2021-06-15 南京邮电大学 一种基于机器学习的卫星频谱感知方法

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10812512B1 (en) * 2018-04-10 2020-10-20 The United States Of America As Represented By The Secretary Of The Air Force Spectrum sensing falsification detection in dense cognitive radio networks
CN109150341B (zh) * 2018-09-11 2020-11-20 北京邮电大学 频谱感知方法和设备以及存储介质
CN109787697B (zh) * 2019-03-11 2021-03-16 重庆邮电大学 一种基于无线指纹数据库的频谱感知方法
CN110493873B (zh) * 2019-07-02 2023-06-30 全球能源互联网研究院有限公司 一种适应电力业务的无线专网频谱分配优化方法及装置
JP7313323B2 (ja) * 2020-08-28 2023-07-24 株式会社Ｋｄｄｉ総合研究所 時系列予測システム、時系列予測方法及びコンピュータプログラム
CN112260781B (zh) * 2020-09-11 2021-10-19 南京邮电大学 一种基于反向散射与空中计算的协作频谱检测方法
CN113852432B (zh) * 2021-01-07 2023-08-25 上海应用技术大学 基于rcs-gru模型的频谱预测感知方法
CN115209508B (zh) * 2021-04-13 2023-09-19 中国移动通信集团设计院有限公司 频谱接入方法、装置、设备及存储介质
CN113271611B (zh) * 2021-04-23 2022-10-25 中国空间技术研究院 一种认知卫星网络系统频谱切换优化方法
CN113595664B (zh) * 2021-08-13 2023-02-03 电子科技大学长三角研究院(湖州) 一种基于深度学习的多用户宽带通信系统的频谱感知方法
CN114900234B (zh) * 2022-07-14 2022-10-21 四川太赫兹通信有限公司 一种太赫兹频谱环境地图构建方法及设备
CN115549823B (zh) * 2022-11-23 2023-04-07 中国人民解放军战略支援部队航天工程大学 一种无线电环境地图预测方法
CN116056090B (zh) * 2023-03-29 2023-06-02 塔盾信息技术(上海)有限公司 一种基于人工智能的认知自组网通信方法
CN116318445B (zh) * 2023-05-22 2023-09-19 北京星河亮点技术股份有限公司 频谱感知方法、装置、电子设备和存储介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2753036A1 (en) * 2013-01-08 2014-07-09 Freescale Semiconductor, Inc. Channel estimation in wireless communication
US20150124972A1 (en) * 2011-12-23 2015-05-07 Nokia Corporation Audio processing for mono signals
CN106507370A (zh) * 2016-11-25 2017-03-15 重庆邮电大学 主动式频谱感知切换的方法

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160127058A1 (en) * 2013-06-07 2016-05-05 Hewlett-Packard Development Company, L.P. Target channel identification for a wireless communication
US9753131B2 (en) * 2013-10-09 2017-09-05 Massachusetts Institute Of Technology Motion tracking via body radio reflections
US9277413B2 (en) * 2013-12-20 2016-03-01 King Fahd University Of Petroleum And Minerals Cooperative cognitive radio spectrum sensing using a hybrid data-decision method
IL240281B (en) * 2015-08-02 2020-05-31 Fiereizen Moshe System and method for locating the source of a single
CN114980221A (zh) * 2016-12-30 2022-08-30 英特尔公司 用于无线电通信的方法和设备
US10444322B2 (en) * 2017-08-09 2019-10-15 Swfl, Inc. Systems and methods for coherence based positioning
US10405219B2 (en) * 2017-11-21 2019-09-03 At&T Intellectual Property I, L.P. Network reconfiguration using genetic algorithm-based predictive models
US10708274B2 (en) * 2018-02-20 2020-07-07 Accenture Global Solutions Limited System for controlling access to a plurality of target systems and applications

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150124972A1 (en) * 2011-12-23 2015-05-07 Nokia Corporation Audio processing for mono signals
EP2753036A1 (en) * 2013-01-08 2014-07-09 Freescale Semiconductor, Inc. Channel estimation in wireless communication
CN106507370A (zh) * 2016-11-25 2017-03-15 重庆邮电大学 主动式频谱感知切换的方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
SONG, GUOLIN ET AL.: "A Kind of Wireless Communication Signals Ultra Narrow-band Filtering Method Improvement", SCIENCE TECHNOLOGY AND ENGINEERING, vol. 16, no. 30, 31 October 2016 (2016-10-31), ISSN: 1671-1815 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111565081A (zh) * 2020-04-28 2020-08-21 深圳中科国威信息系统技术有限公司 认知无线电中白频谱频段的扫描方法、介质、终端和装置
CN112968740A (zh) * 2021-02-01 2021-06-15 南京邮电大学 一种基于机器学习的卫星频谱感知方法
CN112968740B (zh) * 2021-02-01 2022-07-29 南京邮电大学 一种基于机器学习的卫星频谱感知方法

Also Published As

Publication number Publication date
CN107359948B (zh) 2019-06-14
CN107359948A (zh) 2017-11-17
US10674434B2 (en) 2020-06-02
US20200128473A1 (en) 2020-04-23

Similar Documents

Publication Publication Date Title
WO2019010861A1 (zh) 一种认知无线网络的频谱预测方法及装置
US11910199B2 (en) System, method, and apparatus for providing dynamic, prioritized spectrum management and utilization
US11943628B1 (en) System, method, and apparatus for providing dynamic, prioritized spectrum management and utilization
US11800369B2 (en) System, method, and apparatus for providing dynamic, prioritized spectrum management and utilization
US12015927B2 (en) System, method, and apparatus for providing dynamic, prioritized spectrum management and utilization
US11963013B1 (en) System, method, and apparatus for providing dynamic, prioritized spectrum management and utilization
US11974135B2 (en) System, method, and apparatus for providing dynamic, prioritized spectrum management and utilization
Tidjani et al. Low complexity multichannel spectrum prediction algorithm based on optimized neural network for spectrum allocation in cognitive radio internet of things
US20240224056A1 (en) System, method, and apparatus for providing dynamic, prioritized spectrum management and utilization
US12028719B2 (en) System, method, and apparatus for providing dynamic, prioritized spectrum management and utilization
US20240224053A1 (en) System, method, and apparatus for providing dynamic, prioritized spectrum management and utilization
US20240224054A1 (en) System, method, and apparatus for providing dynamic, prioritized spectrum management and utilization

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17917420

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17917420

Country of ref document: EP

Kind code of ref document: A1