阅读说明：本技术 一种多路径路由优化方法 (Multipath routing optimization method ) 是由 陈寿瑜 邹钟璐 李为 黄志才 黄贺平 全源 朱辉青 瞿振 袁志坚 翟柱新 陈皓宁 于 2019-09-29 设计创作，主要内容包括：本发明公开了一种多路径路由优化方法,包括构建类脑的知识库,获取包括用户状态和网络状态的状态信息,并发送至所述知识库中进行处理,其整体思路是通过类脑机制,在传统的底层路由算法的基础上,建立类脑的知识库。在通信过程中,通过类脑的知识库映射的方式,快速规划光前传网路由最优路径,而无需使用底层的长时间路由选择算法。且在路由选择策略上,同时规划当前服务路由选择策略和服务下一状态路由选择策略。从而有效地减少平均延迟,降低通信中断概率,在高移动性场景中提高用户的通信质量。(The invention discloses a multi-path routing optimization method, which comprises the steps of constructing a brain-like knowledge base, acquiring state information comprising a user state and a network state, and sending the state information to the knowledge base for processing. In the communication process, the optimal route of the optical fronthaul network is rapidly planned in a brain-like knowledge base mapping mode without using a bottom-layer long-time routing algorithm. And simultaneously planning a current service routing strategy and a service next state routing strategy on the routing strategy. Therefore, the average delay is effectively reduced, the communication interruption probability is reduced, and the communication quality of the user is improved in a high-mobility scene.)

1. A method for multipath routing optimization, comprising the steps of:

establishing a brain-like knowledge base, acquiring state information including a user state and a network state, and sending the state information to the knowledge base for processing:

when the knowledge base stores a routing strategy corresponding to the state information, the knowledge base outputs the routing strategy corresponding to the state information, and the routing strategy comprises a current service routing strategy and a next service routing strategy;

when the knowledge base does not have the routing strategy corresponding to the state information, the state information is calculated by a bottom routing algorithm and then a routing strategy is output, and new knowledge is established, wherein the new knowledge comprises the state information, the routing strategy and the mapping relation between the state information and the routing strategy;

obtaining communication quality feedback corresponding to the output routing strategy:

when the communication quality feedback does not reach the standard, the operation is carried out again until the communication quality feedback reaches the standard;

and when the communication quality feedback reaches the standard, updating the new knowledge into the knowledge base.

2. The multi-path routing optimization method of claim 1, wherein the knowledge bases include a state knowledge base storing the state information, a routing policy knowledge base storing the routing policies, and a mapping relationship knowledge base storing the mapping relationships.

3. The multi-path routing optimization method according to claim 1, wherein the state knowledge base matches the acquired state information with the stored state information, and when matching is successful, sends the acquired state information to the policy mapping relation knowledge base; otherwise, the obtained state information is calculated by a bottom layer routing algorithm and then the routing strategy is output, and the new knowledge is generated.

4. The multi-path routing optimization method according to claim 1, wherein when the mapping relation repository stores the mapping relation corresponding to the acquired state information, the routing policy repository is controlled to output a corresponding routing policy; and when the mapping relation knowledge base does not have the mapping relation corresponding to the acquired state information, calculating the state information through a bottom routing algorithm, outputting the routing strategy, and generating the new knowledge.

5. A multi-path routing optimisation method as claimed in claim 2 wherein the state knowledge base includes the current user state and the network state K_iIdentified by the following equation:

K_i＝(s₁，s₂，s₃，s₄，s₅)；

UB＝{K₁，K₂，…，K_i}；

wherein s1 to s5 represent the relative direction, relative position, relative speed, network bandwidth utilization, and link delay of the user, respectively; determining attribute value x according to q attribute of user and network state₀Finding membership function A corresponding to fuzzy set_qp(x₀) Determining the value x by using the principle of maximum membership degree₀Determining a fuzzy set, and converting the membership function into N fuzzy sets:

x is then₀∈A_qpAnd (1. ltoreq. p. ltoreq.N).

6. A multi-path routing optimization method as claimed in claim 2, wherein the mapping relation knowledge base is configured to construct q attribute closeness functions σ of two demand states_qJudging the user state K_mAnd user status K_nWhether or not they belong to the same pattern, the proximity function σ_q：

In the formula A_qmIndicating user and network status K_mMaximum membership fuzzy set of q-th attribute, A_qnIndicating user and network status K_nThe maximum membership fuzzy set of the ith attribute.

7. A multi-path routing optimization method as claimed in claim 6, wherein the mapping relation knowledge base uses similarity concept to represent the user state K_mAnd user status K_nThe state of the maximum similarity is selected as the best similar matching state:

wherein ω is_qIs the weight of the qth state attribute for selecting the state K from the user and network state knowledge bases_mMost similar state K₀。

8. A multi-path routing optimization method as claimed in claim 1, wherein the newly learned new knowledge is represented by similar stored new knowledge, and the similarity between the newly learned new knowledge and the stored new knowledge is expressed as:

wherein, c_simIs the similarity threshold.

9. A multi-path route optimisation method as claimed in claim 2, characterised in that the routing policy expression is r ═ b₁，b₂，…，b_iIn which b is_i1 denotes forwarding data to router b_i，b_iWhen 0, it means not to router P_iForwarding the data;

the routing policy knowledge base is the obtained routing policy set:

CR＝{R₁，R₂，…，R_i}；

R_i＝(r₁+r₂+…+r_x+a_x)；

wherein r is_xIs the current service routing path, a_xIs the next service predicted routing path.

Technical Field

The invention relates to the technical field of optical communication, in particular to a multipath routing optimization method.

Background

With the rapid development of technologies such as inspection robots, unmanned planes, intelligent wearable terminals and car networking, emerging services with high mobility have emerged as main features. These emerging services require faster access, and lower transmission delays, which are important guarantees for high quality communications for high mobility users. The high requirement of the communication on the transmission delay is guaranteed, and a new challenge is provided for the existing communication network.

An access network with convergence of optical communication and wireless communication is receiving more and more attention as an important support for development of a mobile communication network. The optical and wireless integrated access network has stronger flexibility and dynamic performance, and can provide larger bandwidth, thereby meeting the actual requirements of high-quality network service and user experience. The optical fronthaul network is an important part of the bearer service of the optical and wireless converged access network.

Disclosure of Invention

Therefore, the invention provides a multi-path route optimization method to solve the problems of too high communication interruption probability and too poor communication service quality caused by the unsolved important routing problem of the optical forwarding network with support in a high mobility scene in the prior art.

In order to achieve the above object, the embodiment of the present invention discloses the following technical solutions:

a method of multipath routing optimization, comprising the steps of:

establishing a brain-like knowledge base, acquiring state information including a user state and a network state, and sending the state information to the knowledge base for processing:

when the knowledge base stores a routing strategy corresponding to the state information, the knowledge base outputs the routing strategy corresponding to the state information, and the routing strategy comprises a current service routing strategy and a next service routing strategy;

when the knowledge base does not have the routing strategy corresponding to the state information, the state information is calculated by a bottom routing algorithm and then a routing strategy is output, and new knowledge is established, wherein the new knowledge comprises the state information, the routing strategy and the mapping relation between the state information and the routing strategy;

obtaining communication quality feedback corresponding to the output routing strategy:

when the communication quality feedback does not reach the standard, the operation is carried out again until the communication quality feedback reaches the standard;

and when the communication quality feedback reaches the standard, updating the new knowledge into the knowledge base.

Further, the knowledge base comprises a state knowledge base storing the state information, a routing strategy knowledge base storing the routing strategy, and a mapping relation knowledge base storing the mapping relation.

Further, the state knowledge base matches the acquired state information with the stored state information, and when the matching is successful, the acquired state information is sent to the strategy mapping relation knowledge base; otherwise, the obtained state information is calculated by a bottom layer routing algorithm and then the routing strategy is output, and the new knowledge is generated.

Further, when the mapping relation corresponding to the acquired state information is stored in the mapping relation knowledge base, controlling the routing strategy knowledge base to output a corresponding routing strategy; and when the mapping relation knowledge base does not have the mapping relation corresponding to the acquired state information, calculating the state information through a bottom routing algorithm, outputting the routing strategy, and generating the new knowledge.

Further, in the state knowledge base, the current user state and the network state K_iIdentified by the following equation:

K_i＝(s₁，s₂，s₃，s₄，s₅)；

UB＝{K₁，K₂，…，K_i}。

wherein s1 to s5 represent the relative direction, relative position, relative speed, network bandwidth utilization, and link delay of the user, respectively; determining attribute value x according to q attribute of user and network state₀Finding membership function A corresponding to fuzzy set_qp(x₀) Determining the value x by using the principle of maximum membership degree₀Determining a fuzzy set, and converting the membership function into N fuzzy sets:

x is then₀∈A_qpAnd (1. ltoreq. p. ltoreq.N).

Further, in the mapping relation knowledge base, a q-th attribute closeness function sigma of two demand states is constructed_qJudging the user state K_mAnd user status K_nWhether or not they belong to the same pattern, the proximity function σ_q：

In the formula A_qmIndicating user and network status K_mMaximum membership fuzzy set of q-th attribute, A_qnIndicating user and network status K_nThe maximum membership fuzzy set of the ith attribute.

Further, in the mapping relation knowledge base, similarity concepts are adopted to represent the user state K_mAnd user status K_nThe state of the maximum similarity is selected as the best similarity matching stateState:

wherein ω is_qIs the weight of the qth state attribute for selecting the state K from the user and network state knowledge bases_mMost similar state K₀。

Further, the newly learned new knowledge is represented by similar stored new knowledge, and the similarity expression of the newly learned new knowledge and the stored new knowledge is as follows:

wherein, c_simIs the similarity threshold.

Further, the routing policy expression is r ═ { b ═ b₁，b₂，…，b_iIn which b is_i1 denotes forwarding data to router b_i，b_iWhen 0, it means not to router P_iForwarding the data;

the routing policy knowledge base is the obtained routing policy set:

CR＝{R₁，R₂，…，R_i}；

R_i＝(r₁+r₂+…+r_x+a_x)；

wherein r is_xIs the current service routing path, a_xIs the next service predicted routing path.

The invention has the following advantages:

(1) on the basis of the traditional bottom layer routing algorithm, a state knowledge base and a mapping knowledge base mapped by the user state, the network state and the routing strategy are established. In the communication process, the optimal route of the optical fronthaul network route is rapidly planned through the state information-strategy mapping relation without using a bottom long-time routing algorithm;

(2) and simultaneously planning a current service routing strategy and a service next state routing strategy on the routing strategy. Therefore, the average delay is effectively reduced, the communication interruption probability is reduced, and the communication quality of the user is improved in a high-mobility scene.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It should be apparent that the drawings in the following description are merely exemplary, and that other embodiments can be derived from the drawings provided by those of ordinary skill in the art without inventive effort.

The structures, ratios, sizes, and the like shown in the present specification are only used for matching with the contents disclosed in the specification, so as to be understood and read by those skilled in the art, and are not used to limit the conditions that the present invention can be implemented, so that the present invention has no technical significance, and any structural modifications, changes in the ratio relationship, or adjustments of the sizes, without affecting the effects and the achievable by the present invention, should still fall within the range that the technical contents disclosed in the present invention can cover.

FIG. 1 is a flow chart of a route optimization strategy according to an embodiment of the present invention;

FIG. 2 is a diagram of a knowledge base building structure of a brain-like system in an embodiment of the present invention.

Detailed Description

The present invention is described in terms of particular embodiments, other advantages and features of the invention will become apparent to those skilled in the art from the following disclosure, and it is to be understood that the described embodiments are merely exemplary of the invention and that it is not intended to limit the invention to the particular embodiments disclosed. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The overall idea of the invention is to establish a state knowledge base and a mapping knowledge base mapped by user states, network states and routing strategies on the basis of the traditional bottom routing algorithm through a brain-like mechanism. In the communication process, the optimal route of the optical fronthaul network route is rapidly planned through the state information-strategy mapping relation without using a bottom long-time routing algorithm. And simultaneously planning a current service routing strategy and a service next state routing strategy on the routing strategy. Therefore, the average delay is effectively reduced, the communication interruption probability is reduced, and the communication quality of the user is improved in a high-mobility scene.

As shown in fig. 1 and fig. 2, the present invention discloses a multipath routing optimization method, which includes:

establishing a brain-like knowledge base, acquiring state information including a user state and a network state, and sending the state information to the knowledge base for processing:

when the knowledge base stores a routing strategy corresponding to the state information, the knowledge base outputs the routing strategy corresponding to the state information, and the routing strategy comprises a current service routing strategy and a next service routing strategy;

when the knowledge base does not have the routing strategy corresponding to the state information, the state information is calculated by a bottom routing algorithm and then a routing strategy is output, and new knowledge is established, wherein the new knowledge comprises the state information, the routing strategy and the mapping relation between the state information and the routing strategy;

obtaining communication quality feedback corresponding to the output routing strategy to select an optimal routing strategy, specifically:

when the communication quality feedback does not reach the standard, the operation is carried out again until the communication quality feedback reaches the standard;

and when the communication quality feedback reaches the standard, updating the new knowledge into the knowledge base.

Further, the user and network state knowledge base comprises two parts: user location and motion related information and network status information.

Further, the knowledge base comprises a state knowledge base storing the state information, a routing strategy knowledge base storing the routing strategy, and a mapping relation knowledge base storing the mapping relation.

Further, the state knowledge base matches the acquired state information with the stored state information, and when the matching is successful, the acquired state information is sent to the strategy mapping relation knowledge base; otherwise, the obtained state information is calculated by a bottom layer routing algorithm and then the routing strategy is output, and the new knowledge is generated.

Further, when the mapping relation corresponding to the acquired state information is stored in the mapping relation knowledge base, controlling the routing strategy knowledge base to output a corresponding routing strategy; and when the mapping relation knowledge base does not have the mapping relation corresponding to the acquired state information, calculating the state information through a bottom routing algorithm, outputting the routing strategy, and generating the new knowledge.

Further, in the state knowledge base, the current user state and the network state K_iIdentified by the following equation:

K_i＝(s₁，s₂，s₃，s₄，s₅)；

UB＝{K₁，K₂，…，K_i}。

wherein s1 to s5 represent the relative direction, relative position, relative speed, network bandwidth utilization, and link delay of the user, respectively. Determining attribute value x for qth attribute of user and network state₀The membership function A corresponding to the fuzzy set can be found_qp(x₀) Determining the value x by using the principle of maximum membership degree₀Determining a fuzzy set, converting the membership function into N fuzzy sets, and recording:

x is then₀∈A_qpAnd (1. ltoreq. p. ltoreq.N).

Further, during the process of accessing the network by a large number of users at different times, each user will haveThe different position movement states are different, and the network states at the current moment are different. In order to classify the user states, the q-th attribute closeness function sigma of two demand states is constructed in the mapping relation knowledge base_qJudging the user state K_mAnd user status K_nWhether or not they belong to the same pattern, the proximity function σ_q：

In the formula A_qmIndicating user and network status K_mMaximum membership fuzzy set of q-th attribute, A_qnIndicating user and network status K_nThe maximum membership fuzzy set of the ith attribute.

Further, in the mapping relation knowledge base, a similarity concept is introduced to represent the user state K_mAnd user status K_nTo select the state of maximum similarity as the best similarity matching state:

wherein ω is_qIs the weight of the qth state attribute, in order to select the state K from the user and network state knowledge bases_mMost similar state K₀。

And comparing all the user states and the network states in the user and network state knowledge base, and selecting the state with the maximum similarity as the best similar matching state so as to avoid the occurrence of data redundancy in the brain-like knowledge base.

Specifically, the function f is judged by using knowledge base increment_addTo determine whether the newly learned user and network status should be added to the knowledge base:

wherein, c_simIs the similarity threshold.

Further, in the optical forwarding network routing process, the router P0 needs to perform an operation of selecting and forwarding data to other routers, and therefore, the routing policy expression is r ═ { b ═ b%₁，b₂，…，b_iIn which b is_i1 denotes forwarding data to router b_i，b_iWhen 0, it means not to router P_iThe data is forwarded.

The routing policy repository is the obtained set of routing policies:

CR＝{R₁，R₂，…，R_i}；

R_i＝(r₁+r₂+…+r_x+a_x)；

wherein r is_xIs the current service routing path, a_xIs the next service predicted routing path. And in the process of continuously accessing different users, continuously carrying out routing selection to obtain a routing strategy, adding the obtained optimal path into a routing strategy knowledge base, and continuously enriching the knowledge base to develop the user. The development of the routing strategy is a process of continuously increasing knowledge accumulation of the routing strategy knowledge base.

And mapping the demand state-routing strategy according to the calculation of a bottom routing algorithm at the beginning of the network, and after finishing the access and communication of a large number of users, accumulating a large number of mapping relations to obtain a knowledge base of the demand state-routing strategy mapping relation. That is, any user and network state can find similar state patterns in the knowledge base, and a routing strategy matched with the user and network state can be obtained without passing through a bottom routing algorithm. But the routing strategy at this time is not necessarily optimal for the demand state. The mapping relation of brain-like development is an incremental learning mechanism, which not only directly outputs a routing strategy according to a demand state mode, but also dynamically improves the demand state-routing strategy mapping matching result on the basis of the increment of the demand relation mode, namely, when the service transmission quality is poor, the routing is selected again through a bottom algorithm after the transmission is finished, and the new mapping relation result is updated to a knowledge base.

Although the invention has been described in detail above with reference to a general description and specific examples, it will be apparent to one skilled in the art that modifications or improvements may be made thereto based on the invention. Accordingly, such modifications and improvements are intended to be within the scope of the invention as claimed.