CN101541010B - Method for realizing cognitive function of wireless communication network system with cognitive function - Google Patents
Method for realizing cognitive function of wireless communication network system with cognitive function Download PDFInfo
- Publication number
- CN101541010B CN101541010B CN2009100827348A CN200910082734A CN101541010B CN 101541010 B CN101541010 B CN 101541010B CN 2009100827348 A CN2009100827348 A CN 2009100827348A CN 200910082734 A CN200910082734 A CN 200910082734A CN 101541010 B CN101541010 B CN 101541010B
- Authority
- CN
- China
- Prior art keywords
- cognitive
- information
- network
- base station
- decision
- Prior art date
- Legal status (The legal status 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 status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a method for realizing a cognitive function of a wireless communication network system with the cognitive function, which is based on an invention patent application named as Wireless Communication Network System with Cognitive Function. The method substantially comprises the following steps of: obtaining cognitive flow; processing the cognitive flow and making a strategic decision; and executing the cognitive decision. When the method is applied to the prior communication network and the further communication network, an essential difference is the obtaining mode of cognitive information and the executing process of the cognitive decision. Based on the invention patent application named as Wireless Communication Network System with Cognitive Function, the invention can realize high cognitive competence for wireless environment, network environment and customer environment and autonomously decision control the wireless network with a set decision criteria based on the cognitive so as to lead the network to develop from a static working mode to a dynamic self-adapting working mode. The invention has a good application development prospect.
Description
Technical field
The present invention relates to a kind of implementation method, belong to the technical field of cordless communication network with wireless communication network system cognitive function of cognitive function.
Background technology
The broadband wireless business demand of current rapid growth has proposed requirements at the higher level to cordless communication network, but, owing to there is the static management problem of resource in the cordless communication network, make resource distribution very unbalanced, the phenomenon coexistence of shortage of resources and waste, and occupation mode can not dynamically be adjusted with the variation of environment according to demand, thereby it is low with the level of resources utilization to cause network modes to ossify.These problems have become the main bottleneck of restriction cordless communication network development, and are on the rise.In order to address these problems, make network develop into the mode of operation of dynamic self-adapting from static mode of operation, must make network possess height cognitive ability to wireless environment, network environment, user environment, and cordless communication network need carry out autonomous Decision Control with certain weighing criteria on the basis of cognition, and reaches the purpose that adapts to variation by the means of reconstruct.As can be seen, the solution of these problems too busy to get away cognitive, make decisions on one's own control and these three key elements of reconstruct, and the cordless communication network architecture of sealing at present, static state does not possess these key elements and corresponding tenability.
In order to address the above problem, the applicant succeeds in developing an application for a patent for invention " wireless communication network system with cognitive function ", and this application for a patent for invention has proposed the wireless communication network system that a kind of structure is unified, have cognitive function.
Referring to Fig. 1, the structure that briefly introduces when wherein the wireless communication network system of this cognitive function is applied to the existing communication network is formed.As seen from the figure, the wireless communication network system of cognitive function is provided with cognitive information storehouse and three platforms of the static state of being responsible for storage networking and semi-static information: business platform, controlling platform and cognitive platform.Wherein business platform and controlling platform are that application layer in the business platform of existing communication network and controlling platform has increased the cognitive interface with cognitive platform intercommunication respectively, and need expand the cognitive interface of specific application layer for particular network, wherein cognitive platform and cognitive information storehouse are that original communication network does not have.
Cognitive platform is not stratified and be provided with three kinds of different unit: cognitive information detection/transfer unit, cognitive interactive unit, cognitive Decision unit.Wherein, cognitive information detection/transfer unit detects the transmission module by one to be formed, and is responsible for the cognitive information of detection, filtration and delivery applications layer, realizes the interface function of the cognition stream of cognitive platform and business platform, controlling platform.Cognitive interactive unit be used between cognitive information detections/transfer unit and the cognitive Decision unit in the implementation platform and in the cognitive Decision unit cognitive information between each cognitive Decision module alternately, and be the different unified descriptive languages of cognitive information structure; Can realize striding cognitive information mutual of layer, for the combined optimization of striding layer information creates conditions.The cognitive Decision unit is made up of a plurality of cognitive Decision modules of finishing difference in functionality, for example to the module of frequency spectrum resource decision-making, to the module of route decision-making, to the module of chip resource decision, to the module of congestion information decision-making.Direct interactive information between these functional modules, their information interaction is finished by cognitive interactive unit, makes the simple in structure and easy realization of platform like this.These cognitive Decision modules are responsible for the different behaviors of network being analyzed, and being made final decision-making according to the corresponding cognitive stream that obtains from cognitive interactive unit.
Referring to Fig. 2, the structure the when Radio Network System that briefly introduces cognitive function is applied to following cordless communication network is formed.
As seen from Figure 2, it is basic identical to be used for the Radio Network System framework of cognitive function of following cordless communication network and the front system architecture that is used for existing wireless communications network shown in Figure 1, and its difference is: each layer difference of the business platform of future communications network, the physical layer of controlling platform, data link layer, network layer, transport layer and application layer all will expand cognitive interface.Cognitive information detections/transfer unit in the cognitive platform is made up of a plurality of different detections transmission modules, and these a plurality of disparate modules are used for detecting respectively, the cognitive information of the different layers of filtration and business transferring platform and controlling platform.The layering that so just can realize cognitive information detects with mutual.Can carry out information interaction with each layer of business platform, controlling platform though it should be noted that cognitive platform, cognitive platform is not stratified, and reason is that cognitive information has characteristics of overall importance, so be not suitable for the cognitive function layering.And do not adopt the cognitive information layering to detect be not: if cognitive platform and business platform, each layer of controlling platform are all set up cognitive interface respectively with mutual reason in the network system architecture that existing network adds cognitive function, to make bigger change to existing network, and be difficult to realize.
Summary of the invention
In view of this, the purpose of this invention is to provide a kind of implementation method with wireless communication network system cognitive function of cognitive function, thereby the resource that solves in the present wireless network is in static management, can not dynamically adjust with the variation of environment according to demand, cause that network modes ossifys, the low inferior problem of the level of resources utilization, make network possess cognitive function, can develop into the dynamic self-adapting mode of operation from static mode of operation.
In order to achieve the above object, the invention provides a kind of implementation method, it is characterized in that described method comprises following operating procedure with wireless communication network system cognitive function of cognitive function:
(1) obtain cognitive stream: base station or travelling carriage obtain cognitive information by the cognitive style of open loop and/or closed loop; The cognitive information of base station or travelling carriage being obtained via the business platform in the network system and/or controlling platform is sent to the cognitive information detection/transfer unit in the cognitive platform again; After cognitive information detection/transfer unit detects and handles this cognitive information, be passed to the cognitive interactive unit of cognitive platform;
(2) cognitive stream is analyzed and made a policy: cognitive interactive unit makes up unified descriptive language, after different cognitive information is unified to describe, is sent to the cognitive Decision unit of cognitive platform; The cognition stream that the cognitive Decision unit converges according to cognitive interactive unit adopts the detection estimation method that comprises based on probability distribution, based on the information characterizing method of pattern recognition, artificial intelligence and neural net and information processing method, based on centralized and distributed information processing method, based on the information processing method of classic optimisation theory and modern optimization theory, the behavior of phase-split network, and make a policy; And in decision process, cognitive platform and cognitive information storehouse communicate, and know the static state of storing in the cognitive information storehouse and the semi-static network information and knowledge, make a policy with auxiliary cognitive Decision unit;
(3) carry out cognitive Decision: the decision information that generates in the cognitive Decision unit, via cognitive interactive unit and cognitive information detection/transfer unit, pass to controlling platform, after controlling platform receives this cognition that comprises decision information stream, executive control operation, business platform is carried out the result of decision, thereby realizes the cognitive function of network system.
Described method comprise cognitive stream obtain with transmission, the visit of each module of cognitive function, each unit of cognitive platform between mutual, each step to whole operating process of the analyzing and processing of cognition stream and decision-making and execution cognitive Decision in; all to carry out safety protecting mechanism, prevent illegal the access and unauthorized access cognition stream.
Described cognitive information is the information with cognitive function, comprise frequency spectrum resource, carrier/interface ratio situation, channel capacity and other radio environment information, the user profile of service quality rating, network are fitted the various network information of sex change, network operation cost, network matching degree; Converging of the cognitive information that described cognitive stream is different Information Levels, cognitive being classified as follows of flowing:
Be divided into according to geographic range: the local cognitive stream that other network node of getting along well is mutual, and need the global knowledge stream mutual with other network node;
Be divided into according to time scale: with millisecond with second be the cognitive stream of short-term of time scale, with minute and hour be the cognitive stream in mid-term of time scale, and with sky, week, moon or year be that the long-term cognition of time scale is flowed;
Be divided into according to action scope: the public cognitive stream of the cognitive information that all-network all has, and the privately owned cognitive stream of the cognitive information that just has of particular network just.
Described open loop cognitive style be base station or travelling carriage by monitoring the cognitive information that comprises wireless environment, user profile and the network information of periphery, promptly under unidirectional perceptive mode, obtain cognitive information by measurement report; Described closed loop cognitive style is that cognitive entity sends solicited message to perceived person earlier, and perceived person sends to cognitive entity with the cognitive information of oneself again.
Described closed loop cognitive style comprises following three kinds of methods:
Utilize the cognitive information of the DPCH active broadcast of physical layer, obtain the cognitive stream of short-term;
Utilize to expand existing signaling and increase cognitive Query Information, make described signaling on link layer, support the data interaction of closed loop and obtain cognitive stream in mid-term;
By with the cognitive information storehouse in the static or semi-static information of storing mutual, obtain long-term cognitive stream.
The described closed loop cognitive approach of the DPCH realization of physical layer that utilizes is divided into following two kinds:
Downlink transfer: be in travelling carriage in first network base station will obtain the to coexist cognitive information of second network base station of this area, this travelling carriage just sends the request that cognitive information is provided to second network base station, and sends the relevant information of oneself; After second network base station received this request,, make travelling carriage obtain the cognitive information of second network base station to second cognitive information that network base station is collected of travelling carriage transmission;
Uplink: second network base station wishes to obtain to be in the perception information of the travelling carriage in first network base station of this area of coexisting, the travelling carriage of this second network base station in being in first network sends the request that cognitive information is provided, and the relevant information of broadcast transmission oneself; After being in travelling carriage in first network and receiving, to the cognitive information that second network base station transmission oneself collected, second network base station just obtained to be in the cognitive information of the travelling carriage in first network.
Described utilization is expanded signaling and realized that the method for closed loop cognition is: first network base station wishes to obtain the cognitive information of second network base station, just sends the request that cognitive information is provided to second network base station, and the relevant information of broadcast transmission oneself; The relevant information of first network is carried on and expands when being delivered to gateway in the signaling, gateway carries out the conversion of agreement, and after this signaling passed to second network base station, second cognitive information that network base station adopts above-mentioned the same manner oneself to collect to first network base station transmission again makes first network base station obtain the cognitive information of second network base station.
Static or the semi-static network information and the knowledge of storing in the described cognitive information storehouse, wherein the network information is the cognitive information of each network element entity in the network, and knowledge is in network operation process, based on reasoning and study and accumulate the decision-making of acquisition and the set of tactful experience;
Described cognitive information storehouse includes by the method for finishing knowledge accumulation with the information interaction of cognitive platform: existing decision strategy in the cognitive information storehouse is strengthened handling, to accelerate speed of decision, increase the weight of tactful weight; The existing new knowledge of instant use is done to replace to the out-of-date knowledge in the cognitive information storehouse and is upgraded; Still the new knowledge that does not have in real-time storage and this cognitive information storehouse of accumulation;
Described cognitive information storehouse mutual by with cognitive platform, the process of auxiliary cognitive platform decision-making is: cognitive platform is earlier by searching and mate the knowledge of storing in the cognitive information storehouse and the network information, obtain similar scene and relevant parameter, and judge the similar degree of this scene, adopt the matching algorithm aid decision of pattern recognition then, and the result of decision returned to the cognitive information storehouse, carry out accumulation of knowledge for the cognitive information storehouse.
In the described step (2), cognitive interactive unit makes up unified descriptive language, and when the Different Cognitive information in this cognition stream was described, it was mutual to adopt unified format to realize; And the operating procedure that cognitive information is made up unified descriptive language is as follows:
(21) the autonomous analysis: earlier to the cognitive information classification, to extract the essential characteristic of detected Different Cognitive information;
(22) autonomous modeling:, use unified descriptive language to set up the descriptive model of uniform format according to essential characteristic to the Different Cognitive information extraction;
(23) the autonomous description: the descriptive model of the consolidation form of code requirement is described the essential characteristic of Different Cognitive information.
The present invention is a kind of implementation method with wireless communication network system cognitive function of cognitive function, it is cognitive function, Decision Control function and the suitable implementation method that becomes function of the wireless communication network system that proposes on the basis of another application for a patent for invention " wireless communication network system with cognitive function " the applicant, mainly comprises three steps: obtain cognitive stream, to cognitive stream analyze and make a policy, the execution of cognitive Decision.Because the object that is suitable for: the existing communication network is different with the future communications network, both are slightly different in the implementation of approach that obtains cognitive information and cognitive Decision thereof.Utilize this method can be on the basis of related invention patent application " wireless communication network system " with cognitive function, realization is to the height cognitive function of wireless signal environment, network environment, user environment, and can on cognitive basis, carry out autonomous Decision Control to wireless network, thereby make network develop into the dynamic self-adapting mode of operation from static mode of operation with the weighing criteria of setting.
Operating procedure of the present invention simply, realize easily and apply, in the acquisition methods of wherein cognitive stream, the closed loop cognitive style that the present invention adopts has been saved the expense of obtaining cognitive stream, can more effectively obtain cognitive information.And in the decision-making technique of cognition stream, the detection estimation method that the present invention adopts based on probability distribution, based on the information characterizing method of pattern recognition, artificial intelligence and neural net and information processing method, based on centralized and distributed information processing method, based on the information processing method of classic optimisation theory and modern optimization theory, for example these methods comprise obfuscation sign, the modeling of parameter Markov, the identification of PCFG syntactic pattern and the distributed treatment and the heuristic optimization of information, the behavior of phase-split network, and make a policy.Above-mentioned various distinct methods can both solve the problem that data volume is huge, data mode is various and Data Dynamic changes of cognitive information well.Another application for a patent for invention of patent application of the present invention and applicant " wireless communication network system with cognitive function " brings out the best in each other, form the complete wireless communication network system and the implementation method of correlation function thereof, have good development prospect with cognitive function.
Description of drawings
Fig. 1 is wireless communication network system with cognitive function structural representation when being used for existing network.
Fig. 2 is wireless communication network system with cognitive function structural representation when being used for future network.
Fig. 3 is the implementation method operational flowchart that the present invention has the wireless communication network system cognitive function of cognitive function.
Fig. 4 (A), (B) utilize physical channel to realize the uplink and downlink cognitive approach schematic diagram of closed loop cognition.
Fig. 5 is the transmission method schematic diagram that utilizes cognitive information between two network base stations that expand the cognition of signaling realization closed loop.
Fig. 6 is the mutual situation schematic diagram of cognitive information storehouse and cognitive platform.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, the present invention is described in further detail below in conjunction with accompanying drawing.
Obviously, method of work and the difference of the method for work of conventional wireless network system of the present invention with wireless communication network system of cognitive function only is the operation that is associated with cognitive information.So essence of the present invention is exactly the implementation method of cognitive function.Specifically, the implementation method of cognitive function is divided three steps: the obtaining of cognitive stream, and cognitive stream is analyzed and made a policy, carry out cognitive Decision.
Introduce earlier the notion of cognitive stream: the converging of the cognitive information of different Information Levels.Cognitive information of the present invention is meant the information with cognitive function, for example, wireless environment (as frequency spectrum resource, carrier/interface ratio situation, channel capacity etc.), user profile (service quality QoS etc.), network information cognitive information such as (as the suitable sex change of network, network operation cost, network matching degrees) all is cognitive stream.And traditional cognitive information includes only radio environment information, can not promote user/network satisfaction degree well.Being classified as follows of cognitive stream:
Be divided into local cognitive stream and global knowledge stream according to geographic range: local cognitive other network node of stream discord is mutual; Global knowledge stream needs with other network node mutual.
Be divided into the cognitive stream of short-term according to time scale, cognition in mid-term stream, long-term cognitive stream.The cognitive stream of short-term is time scale with millisecond or second; Mid-term cognitive stream with minute or hour be time scale; Long-term cognitive stream is time scale with sky, week, the moon or year.
Be divided into public cognitive stream and privately owned cognitive stream according to action scope.The former is the cognitive information that all-network all has, and for example power disturbs etc.; The latter is the cognitive information that particular network just has, the subcarrier among the LTE for example, the chip resource among the CDMA etc.
Two kinds of systems according to the present invention respectively below constitute the implementation method of introducing its cognitive function respectively.
Referring to Fig. 3, introduce the following step of operating process of implementation method that has the wireless communication network system cognitive function of cognitive function at the present invention:
(1) obtain cognitive stream: base station or travelling carriage be by the cognitive style acquisition cognitive information of open loop and/or closed loop, and the cognitive information of base station or travelling carriage being obtained via the business platform in the network system and/or controlling platform is sent to the cognitive information detection/transfer unit in the cognitive platform again; Cognitive information detection/transfer unit is passed to cognitive interactive unit after this cognitive information is detected and handles.
When the inventive method is used for the existing communication network, adopt the open loop cognitive style to obtain cognitive information; And it obtains cognitive information by open loop cognitive style and/or closed loop cognitive style when being used for the future communications network.Wherein open loop cognitive style be base station or travelling carriage by monitoring the cognitive information that comprises wireless environment, user profile and the network information of periphery, promptly under unidirectional perceptive mode, obtain cognitive information by measurement report.The closed loop cognitive style is that cognitive entity sends solicited message to perceived person earlier, and perceived person sends to cognitive entity with the cognitive information of oneself again.
(2) cognitive stream is analyzed and made a policy: cognitive interactive unit makes up unified descriptive language, after different cognitive information is unified to describe, is sent to the cognitive Decision unit; The cognition stream that the cognitive Decision unit converges according to cognitive interactive unit adopts the detection estimation method that comprises based on probability distribution, based on the information characterizing method of pattern recognition, artificial intelligence and neural net and information processing method, based on centralized and distributed information processing method, based on the behavior of the information processing method phase-split network of classic optimisation theory and modern optimization theory, and make a policy; And in decision process, cognitive platform and cognitive information storehouse communicate, and know the static state of storing in the cognitive information storehouse and the semi-static network information and knowledge, make a policy with auxiliary cognitive Decision unit.Wherein, the network information is the cognitive information of each network element entity in the network, and knowledge is in network operation process, based on reasoning and study and accumulate the decision-making of acquisition and the set of tactful experience.
In this step, cognitive interactive unit makes up unified descriptive language, when describing the Different Cognitive information in this cognition stream, realize mutual with unified format.At this moment, three operating procedures to the unified descriptive language of cognitive information structure are as follows:
A, the autonomous analysis: earlier to the cognitive information classification, with to detected its essential characteristic of Different Cognitive information extraction;
B, autonomous modeling:, use unified descriptive language to set up the descriptive model of uniform format according to essential characteristic to the Different Cognitive information extraction;
C, the autonomous description: the descriptive model of the consolidation form of code requirement is described the essential characteristic of Different Cognitive information.
In this operating procedure, the method that is adopted during the phase-split network behavior can adopt the method for obfuscation sign, the modeling of parameter Markov, the identification of PCFG syntactic pattern and the distributed treatment and the heuristic optimization of information that cognitive stream is carried out analyzing and processing.Wherein:
The obfuscation of information characterizes: well-known, the data volume of cognitive information is huge, and presents diversity and dynamic characteristic, has brought unprecedented complexity so just for the traditional detection estimation theory.In order to address this problem, it is just very necessary to study the new application of instrument in the cognitive information decision-making.Obfuscation sign by information is set up fuzzy logic, and is used as a kind of artificial intelligence technology of analyzing and processing cognitive information of the present invention.Because fuzzy logic is being handled complexity, uncertainty and reached powerful advantages aspect the high control performance, it has replaced conventional art gradually in engineering system (especially control system).When fuzzy logic carries out system's control, need not set up system mathematic model accurately, but according to system in the past control experience summarize, the control strategy that formation is described with language and fuzzy mathematics, thereby simplify the control that complication system and process are carried out, especially to uncertain and system dynamic change, its advantage is more obvious.
Hidden parameter markoff process (HMM): HMM is a kind of statistical model, is used to describe a Markov process that contains implicit unknown parameter.Be characterized in from observable parameter, determining the implicit parameter of this process, utilize these parameters to do further to analyze then.HMM can regard a kind of dynamic bayesian network of simplification as, be widely applied to all multi-direction of artificial intelligence field, as speech recognition or optical character identification, machine translation, bioinformatics and genomics, and obtained fine effect, embodied good case study and disposal ability and the broad applicability of HMM at artificial intelligence field.Because the analyzing and processing of cognitive information and the close ties of artificial intelligence, so HMM is highly suitable for the analyzing and processing cognitive information.By setting up rational HMM, can analyze and make a strategic decision a large amount of, multifarious dynamic cognitive information.
PCFG probability context-free grammar: this PCFG method has obtained extensive use in pattern recognition and intellectual analysis field.The irrelevant up and down syntax of probability comprise factor and two essential parts of rule, and factor is the various symbols by the data message mapping, and rule is based on the statistics training of data in the actual scene and produces.The main purpose of the syntax is to infer next step situation by direct or indirect information.The dynamic of cognition wireless network has increased the complexity of network, but by long-term training process, can obtain some characteristic that the probability context-free grammar characterizes corresponding cognition wireless network.This PCFG method itself has the structural model of language, can realize the supposition of network behavior and the processing of all kinds of hierarchical informations of network better.In a word, the probability context-free grammar has good artificial intelligence, is applied in to have very big advantage in the cognitive platform.
Distributed treatment and heuristic optimization:, make the centralized processing of cognitive information increase difficulty because the data volume of cognitive information is very huge.Distributed processing mode becomes the first-selection of cognitive information decision-making mode with its flexibility.Also just because of the mass data of cognitive information, the variable of optimization problem is a lot, and optimization problem need carry out combined optimization with multiple target, and it is very high that the degree of freedom of optimized Algorithm also becomes.This has brought stern challenge all for traditional optimized Algorithm.Heuritic approach because of its in the convergence of some optimization problem and the advantage aspect the algorithm complex, certainly will become the strong instrument of realizing cognitive function.The application of distributed treatment and heuritic approach can make the performance of cognition network system be optimized, thereby realizes the rapid decision-making on the macroscopic view, and realizes optimization on performance.
(3) carry out cognitive Decision: the decision information that generates in the cognitive Decision unit, via cognitive interactive unit and cognitive information detection/transfer unit, pass to controlling platform, after controlling platform receives this cognition that comprises decision information stream, executive control operation, business platform is carried out the result of decision, thereby realizes the cognitive function of network system.
Therefore, summing up the implementation method of cognitive function that the present invention has the wireless communication network system of cognitive function is: the detection in the at first cognitive platform and transfer unit obtain cognitive stream, when this cognition wireless network system applies during in the existing communication network, cognitive stream obtains by open loop approach, cognitive stream is carried in Business Stream and the control flows, and after the application layer of business platform and controlling platform is collected, be transferred to cognitive platform with the form of application data.And this cognition wireless network system applies is when the future communications network, cognitive stream obtains by open loop approach and closed-loop fashion, and cognitive stream is carried in Business Stream and the control flows, and collects and be transferred to cognitive platform in physical layer, data link layer, network layer, transport layer and the application layer of business platform and controlling platform.In addition; the inventive method comprise cognitive stream obtain with transmission, the visit of each module of cognitive function, each unit of cognitive platform between mutual, each step to whole operating process of the analyzing and processing of cognition stream and decision-making and execution cognitive Decision in; all to carry out safety protecting mechanism, prevent illegal the access and unauthorized access cognition stream.
Introduce three kinds of included cognitive approach of closed loop cognitive style below:
The one, utilize the cognitive information of the DPCH active broadcast of physical layer, obtain the cognitive stream of short-term.
The 2nd, utilize to expand existing signaling and increase cognitive Query Information, make signaling on link layer, support the data interaction of closed loop and obtain cognitive stream in mid-term.
The 3rd, by with the cognitive information storehouse in the static or semi-static information of storing mutual, cognitive information detect with transfer unit in each module obtain long-term cognitive stream.
Wherein, utilize the closed loop cognitive approach of the DPCH realization of physical layer to be divided into following two kinds:
Referring to Fig. 4 (A), introduce the cognitive transmission method of descending closed loop earlier: be in travelling carriage in first network A base station wish to obtain the to coexist cognitive information of second network B base station of this area, for example, travelling carriage (multimode) originally belonged to the network A base station, the signal of the network A of travelling carriage discovery then worse and worse, so travelling carriage, sends cognitive information to the network B base station of this area that coexists request is provided, and send the relevant information of oneself; After this request is received in second network B base station,, make travelling carriage obtain the cognitive information of second network B base station to second cognitive information that collect the network B base station of travelling carriage transmission.
Referring to Fig. 4 (B), introduce the cognitive transmission method of up closed loop again: the perception information of the travelling carriage in first network A base station of this area of coexisting is wished to obtain to be in second network B base station, the travelling carriage of this second network B base station in being in first network A sends the request that cognitive information is provided, and the relevant information of oneself is broadcasted away; After being in travelling carriage in first network A and receiving, to the cognitive information that second network B base station transmits oneself collected, second network B base station just obtained to be in the cognitive information of the travelling carriage in first network A.
Referring to Fig. 5, introduce the present invention and utilize the method that expands the cognition of signaling realization closed loop: the cognitive information of second network B base station is wished to obtain in first network A base station, just send the request that cognitive information is provided, and the relevant information of oneself is broadcasted away to second network B base station; The relevant information of first network A is carried on and expands when being delivered to gateway in the signaling, gateway carries out the conversion of agreement, and after this signaling passed to second network B base station, second cognitive information that the network B base station adopts above-mentioned the same manner oneself to collect to first network A base station transmits again makes first network A base station obtain the cognitive information of second network B base station.
Referring to Fig. 6, introduce cognitive information of the present invention storehouse by finishing three kinds of methods of knowledge accumulation: existing decision strategy in the cognitive information storehouse is strengthened handling,, increase the weight of tactful weight to accelerate speed of decision with the information interaction of cognitive platform; In time using existing new knowledge that the out-of-date knowledge in the cognitive information storehouse is done to replace upgrades; Still the new knowledge that does not have in real-time storage and this cognitive information storehouse of accumulation.
Cognitive information of the present invention storehouse mutual by with cognitive platform, the process of auxiliary cognitive platform decision-making is: cognitive platform is earlier by searching and mate the knowledge of storing in the cognitive information storehouse and the network information, obtain similar scene and relevant parameter, and judge the similar degree of this scene, adopt the matching algorithm aid decision of pattern recognition then, and the result of decision returned to the cognitive information storehouse, carry out accumulation of knowledge for the cognitive information storehouse.
Claims (9)
1. the implementation method of the cognitive function of the wireless communication network system with cognitive function is characterized in that described method comprises following operating procedure:
(1) obtain cognitive stream: base station or travelling carriage obtain cognitive information by the cognitive style of open loop and/or closed loop; The cognitive information of base station or travelling carriage being obtained via the business platform in the network system and/or controlling platform is sent to the cognitive information detection/transfer unit in the cognitive platform again; After cognitive information detection/transfer unit detects and handles this cognitive information, be passed to the cognitive interactive unit of cognitive platform;
(2) cognitive stream is analyzed and made a policy: cognitive interactive unit makes up unified descriptive language, after different cognitive information is unified to describe, is sent to the cognitive Decision unit of cognitive platform; The cognition stream that the cognitive Decision unit converges according to cognitive interactive unit adopts the detection estimation method that comprises based on probability distribution, based on the information characterizing method of pattern recognition, artificial intelligence and neural net and information processing method, based on centralized and distributed information processing method, based on the information processing method of classic optimisation theory and modern optimization theory, the behavior of phase-split network, and make a policy; And in decision process, cognitive platform and cognitive information storehouse communicate, and know the static state of storing in the cognitive information storehouse and the semi-static network information and knowledge, make a policy with auxiliary cognitive Decision unit;
(3) carry out cognitive Decision: the decision information that generates in the cognitive Decision unit, via cognitive interactive unit and cognitive information detection/transfer unit, pass to controlling platform, controlling platform receives this cognition that comprises decision information stream back executive control operation, business platform is carried out the result of decision, thereby realizes the cognitive function of network system.
2. method according to claim 1; it is characterized in that: described method comprise cognitive stream obtain with transmission, the visit of each module of cognitive function, each unit of cognitive platform between mutual, each step to whole operating process of the analyzing and processing of cognition stream and decision-making and execution cognitive Decision in; all to carry out safety protecting mechanism, prevent illegal the access and unauthorized access cognition stream.
3. method according to claim 1, it is characterized in that: described cognitive information is the information with cognitive function, comprise frequency spectrum resource, carrier/interface ratio situation, channel capacity and other radio environment information, the user profile of service quality rating, network are fitted the various network information of sex change, network operation cost, network matching degree; Converging of the cognitive information that described cognitive stream is different Information Levels, cognitive being classified as follows of flowing:
Be divided into according to geographic range: the local cognitive stream that other network node of getting along well is mutual, and need the global knowledge stream mutual with other network node;
Be divided into according to time scale: with millisecond with second be the cognitive stream of short-term of time scale, with minute and hour be the cognitive stream in mid-term of time scale, and with sky, week, moon or year be that the long-term cognition of time scale is flowed;
Be divided into according to action scope: the public cognitive stream of the cognitive information that all-network all has, and the privately owned cognitive stream of the cognitive information that just has of particular network just.
4. method according to claim 1, it is characterized in that: described open loop cognitive style be base station or travelling carriage by monitoring the cognitive information that comprises wireless environment, user profile and the network information of periphery, promptly under unidirectional perceptive mode, obtain cognitive information by measurement report; Described closed loop cognitive style is that cognitive entity sends solicited message to perceived person earlier, and perceived person sends to cognitive entity with the cognitive information of oneself again.
5. method according to claim 4 is characterized in that: described closed loop cognitive style comprises following three kinds of methods:
Utilize the cognitive information of the DPCH active broadcast of physical layer, obtain the cognitive stream of short-term;
Utilize to expand existing signaling and increase cognitive Query Information, make described signaling on link layer, support the data interaction of closed loop and obtain cognitive stream in mid-term;
By with the cognitive information storehouse in the static or semi-static information of storing mutual, obtain long-term cognitive stream.
6. method according to claim 5 is characterized in that: the described closed loop cognitive approach of the DPCH realization of physical layer that utilizes is divided into following two kinds:
Downlink transfer: be in travelling carriage in first network base station will obtain the to coexist cognitive information of second network base station of this area, this travelling carriage just sends the request that cognitive information is provided to second network base station, and sends the relevant information of oneself; After second network base station received this request,, make travelling carriage obtain the cognitive information of second network base station to second cognitive information that network base station is collected of travelling carriage transmission;
Uplink: second network base station wishes to obtain to be in the perception information of the travelling carriage in first network base station of this area of coexisting, the travelling carriage of this second network base station in being in first network sends the request that cognitive information is provided, and the relevant information of broadcast transmission oneself; After being in travelling carriage in first network and receiving, to the cognitive information that second network base station transmission oneself collected, second network base station just obtained to be in the cognitive information of the travelling carriage in first network.
7. method according to claim 5, it is characterized in that: described utilization is expanded signaling and realized that the method for closed loop cognition is: first network base station wishes to obtain the cognitive information of second network base station, just send the request that cognitive information is provided to second network base station, and the relevant information of broadcast transmission oneself; The relevant information of first network is carried on and expands when being delivered to gateway in the signaling, gateway carries out the conversion of agreement, and after this information passed to second network base station, second cognitive information that network base station adopts above-mentioned the same manner oneself to collect to first network base station transmission again makes first network base station obtain the cognitive information of second network base station.
8. method according to claim 5, it is characterized in that: the static or semi-static network information and the knowledge of storing in the described cognitive information storehouse, wherein the network information is the cognitive information of each network element entity in the network, and knowledge is in network operation process, based on reasoning and study and accumulate the decision-making of acquisition and the set of tactful experience;
Described cognitive information storehouse includes by the method for finishing knowledge accumulation with the information interaction of cognitive platform: existing decision strategy in the cognitive information storehouse is strengthened handling, to accelerate speed of decision, increase the weight of tactful weight; The existing new knowledge of instant use is done to replace to the out-of-date knowledge in the cognitive information storehouse and is upgraded; Still the new knowledge that does not have in real-time storage and this cognitive information storehouse of accumulation;
Described cognitive information storehouse mutual by with cognitive platform, the process of auxiliary cognitive platform decision-making is: cognitive platform is earlier by searching and mate the knowledge of storing in the cognitive information storehouse and the network information, obtain similar scene and relevant parameter, and judge the similar degree of this scene, adopt the matching algorithm aid decision of pattern recognition then, and the result of decision returned to the cognitive information storehouse, carry out accumulation of knowledge for the cognitive information storehouse.
9. method according to claim 1 is characterized in that: in the described step (2), cognitive interactive unit makes up unified descriptive language, when different cognitive information is described, adopt unified lattice
Formula realizes mutual; And the operating procedure that cognitive information is made up unified descriptive language is as follows:
(21) the autonomous analysis: earlier to the cognitive information classification, to extract the essential characteristic of detected Different Cognitive information;
(22) autonomous modeling:, use unified descriptive language to set up the descriptive model of uniform format according to essential characteristic to the Different Cognitive information extraction;
(23) the autonomous description: the descriptive model of the consolidation form of code requirement is described the essential characteristic of Different Cognitive information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100827348A CN101541010B (en) | 2009-04-28 | 2009-04-28 | Method for realizing cognitive function of wireless communication network system with cognitive function |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100827348A CN101541010B (en) | 2009-04-28 | 2009-04-28 | Method for realizing cognitive function of wireless communication network system with cognitive function |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101541010A CN101541010A (en) | 2009-09-23 |
| CN101541010B true CN101541010B (en) | 2010-06-23 |
Family
ID=41123929
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009100827348A Expired - Fee Related CN101541010B (en) | 2009-04-28 | 2009-04-28 | Method for realizing cognitive function of wireless communication network system with cognitive function |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101541010B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102547717B (en) * | 2010-12-09 | 2016-06-15 | 中兴通讯股份有限公司 | Cognitive radio system operation mode conversion method and device |
| CN102158294B (en) * | 2011-02-25 | 2013-06-05 | 北京邮电大学 | Cognitive wireless network system and method for transmitting cognitive streams thereof |
| CN103188152B (en) * | 2011-12-31 | 2015-07-08 | 南京邮电大学 | Cognitive network quality of service (QoS) routing method based on service differentiation |
| CN103068058B (en) * | 2012-12-24 | 2015-08-26 | 中国人民解放军总参谋部第六十一研究所 | A kind of scheduling method for wireless resource based on double-layer loop model |
| WO2014121459A1 (en) * | 2013-02-06 | 2014-08-14 | Nokia Corporation | A method and apparatus for cooperative spectrum sensing |
| CN103458423B (en) * | 2013-09-17 | 2016-03-30 | 北京邮电大学 | The methods, devices and systems of transmitting cognitive stream between isomery cognition wireless network |
| CN105592024A (en) * | 2014-11-14 | 2016-05-18 | 江苏威盾网络科技有限公司 | Network protection system and method based on cognition network |
| CN105007191B (en) * | 2015-07-31 | 2018-09-11 | 上海红神信息技术有限公司 | Perception, decision and the execution separation method of mimicry route switching system |
| CN108229685B (en) * | 2016-12-14 | 2021-11-02 | 中国航空工业集团公司西安航空计算技术研究所 | Air-ground integrated unmanned intelligent decision-making method |
-
2009
- 2009-04-28 CN CN2009100827348A patent/CN101541010B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN101541010A (en) | 2009-09-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101541010B (en) | Method for realizing cognitive function of wireless communication network system with cognitive function | |
| Wang et al. | Online UAV-mounted edge server dispatching for mobile-to-mobile edge computing | |
| Lin et al. | Artificial intelligence-empowered resource management for future wireless communications: A survey | |
| US20200196155A1 (en) | Utilizing machine learning to provide closed-loop network management of a fifth generation (5g) network | |
| Huang et al. | Wireless big data: transforming heterogeneous networks to smart networks | |
| WO2022000365A1 (en) | Machine learning based downlink channel estimation and prediction | |
| WO2020152389A1 (en) | Machine learning for a communication network | |
| CN104144425A (en) | Network management device and method and device in wireless communication system | |
| CN111405575B (en) | Deployment method and deployment device of RSU | |
| Ali et al. | Reinforcement-learning-enabled massive internet of things for 6G wireless communications | |
| US20230262489A1 (en) | Apparatuses and methods for collaborative learning | |
| Hou et al. | Radio resource allocation and power control scheme in V2V communications network | |
| Christopoulou et al. | Artificial Intelligence and Machine Learning as key enablers for V2X communications: A comprehensive survey | |
| Aguilar-Garcia et al. | Location-aware self-organizing methods in femtocell networks | |
| Zhang et al. | Endogenous security-aware resource management for digital twin and 6G edge intelligence integrated smart park | |
| Montero et al. | Proactive radio-and QoS-aware UAV as BS deployment to improve cellular operations | |
| Koudouridis et al. | An architecture and performance evaluation framework for artificial intelligence solutions in beyond 5G radio access networks | |
| Jalilvand Aghdam Bonab et al. | QoS-aware resource allocation in mobile edge computing networks: Using intelligent offloading and caching strategy | |
| CN116744261B (en) | Millimeter wave communication network and edge calculation fusion method | |
| CN109769257A (en) | Heterogeneous network robust resource allocation and duration method for mixing and optimizing based on efficiency | |
| Das et al. | Cognitive radio-enabled Internet of Things (CR-IoT): An integrated approach towards smarter world | |
| Biswash | Device and network driven cellular networks architecture and mobility management technique for fog computing-based mobile communication system | |
| Santana et al. | LoRaWAN-based smart parking service: Deployment and performance evaluation | |
| Erman et al. | Modeling 5G wireless network service reliability prediction with bayesian network | |
| Meng et al. | Bayesian network prediction of mobile user throughput in 5g wireless networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100623 Termination date: 20120428 |