CN106792739B

CN106792739B - Network slicing method, device and equipment

Info

Publication number: CN106792739B
Application number: CN201611011225.2A
Authority: CN
Inventors: 彭木根; 项弘禹; 唐丽雅; 黄晓霞
Original assignee: Beijing University of Posts and Telecommunications
Current assignee: Beijing University of Posts and Telecommunications
Priority date: 2016-11-17
Filing date: 2016-11-17
Publication date: 2020-01-21
Anticipated expiration: 2036-11-17
Also published as: CN106792739A

Abstract

The embodiment of the invention provides a network slicing method, a device and equipment, wherein the method comprises the following steps: monitoring a network under a current slicing scheme, and acquiring user information, terminal information, service information, a wireless access network state and network performance information; evaluating the slice running state corresponding to the current slice scheme according to the user information, the terminal information, the service information, the wireless access network state and the network performance information to obtain an evaluation result; rearranging the slices and reconfiguring resources according to the evaluation result to obtain a modified slice scheme, and informing each slice to operate according to the modified slice scheme; the above process is repeated until the difference value between the network performance corresponding to the changed slicing scheme and the network performance corresponding to the last slicing scheme is less than or equal to the preset threshold value, so that the network reconfiguration based on information perception is realized, and the purposes of adapting to various service applications and flexibly, conveniently and quickly networking with high performance are achieved.

Description

Network slicing method, device and equipment

Technical Field

The embodiment of the invention relates to a wireless communication technology, in particular to a network slicing method, a network slicing device and network slicing equipment.

Background

With the rapid growth of data services for mobile terminals, the consumer market for mobile broadband services, such as north america, europe, east asia, etc., is gradually saturated. Meanwhile, the number of machine type communication terminals and vertical market applications is increasing at an accelerated rate, and new and diversified service requirements are being placed on mobile networks. However, the conventional mobile network design is mainly considered to meet the requirements of mobile broadband consumption at the beginning, and only limited parameter adjustment is performed for some specific services. However, such a modification is difficult to meet the requirements of different services in the future, for example, for a dedicated network of a high-speed railway, only network coverage support along the railway needs to be provided, and transmission of control management in a high-speed mobile scene is met, and for high-traffic density areas such as an office building, the coverage network should be capable of providing network service support with a large data rate, a high peak rate and a high traffic density. Therefore, incorporating the differentiated service requirements of the emerging vertical industry into network design and operation and maintenance presents a huge challenge for mobile operators.

Fortunately, the Network Function Virtualization (NFV) technology is proposed, so that the Network component can realize software and hardware decoupling, and the Network deployment cost is reduced. According to the report of the European Telecommunications Standards Institute (ETSI), NFV and software defined networking technologies can reduce the program development period, accelerate market deployment and business speed, and promote technical field innovation. However, The types of products and services provided by current operators are still limited, so people put forward a concept of network slice (network slice), and logically divide network functions and resources in The same network according to specific technologies and commercial requirements, so as to provide flexible customized services for vertical industries, third party (Over The Top, OTT) manufacturers, and The like.

The network slice can provide special customized network service according to different requirements of mobile services, so that the operation cost of the special network is effectively reduced, and great attention is attracted to the industrial and academic fields. Both 5GPP (5rd generation information Public-Private Partnership) and third generation Partnership Project (3rd generation Partnership Project, 3GPP) have launched related projects in tandem, and the respective vendors have also issued related white papers on network slices.

However, the existing network slicing methods are all from top to bottom, and slicing the core network and the transmission network from the perspective of service requirements has the problems of network rigidity, performance to be improved and the like.

Disclosure of Invention

The embodiment of the invention provides a network slicing method, a network slicing device and network slicing equipment, which are used for solving the problems that the existing network slicing method is from top to bottom, a core network and a transmission network are sliced from the aspect of service requirements, and the network stiffness and the performance are to be improved.

The first aspect of the present invention provides a network slicing method, including:

a, monitoring a network under a current slicing scheme, and acquiring user information, terminal information, service information, a wireless access network state and network performance information;

b, evaluating the slice running state corresponding to the current slice scheme according to the user information, the terminal information, the service information, the wireless access network state and the network performance information to obtain an evaluation result;

c, rearranging the slices and reconfiguring resources according to the evaluation result to obtain a modified slice scheme, and informing each slice to operate according to the modified slice scheme;

and d, repeating the steps a to c until the difference value between the network performance corresponding to the changed slicing scheme and the network performance corresponding to the last slicing scheme is less than or equal to a preset threshold value.

Optionally, before monitoring the network under the current slicing scheme and acquiring the user information, the terminal information, the service information, the status of the wireless access network, and the network performance information, the method further includes:

slicing the wireless access network according to the application scene, the wireless access network capability and the configuration information, and generating a slicing scheme by taking a typical application scene as a slicing blueprint;

slicing the network according to the slicing scheme; wherein, when the slicing scheme is used for operation, different slices are in hard isolation.

Optionally, the typical application scenario includes any one scenario among wide-area seamless coverage, hot spot high capacity, large-connection internet of things, and low-latency high-reliability communication.

Optionally, the method further includes:

when the periodic report information or new trigger time information is detected, a slicing scheme is regenerated for the wireless access network according to the application scene, the wireless access network capacity and the configuration information, and slicing processing is carried out on the network according to the slicing scheme;

and repeating the steps a to c until the difference value between the network performance corresponding to the changed slicing scheme and the network performance corresponding to the last slicing scheme is less than or equal to the preset threshold value.

A second aspect of the present invention provides a network slicing apparatus, including:

the acquisition module is used for executing the step a, monitoring the network under the current slicing scheme and acquiring user information, terminal information, service information, wireless access network state and network performance information;

a processing module to perform:

Optionally, the processing module is further configured to:

Optionally, the typical application scenarios used by the processing module include any scenario of wide-area seamless coverage, high capacity of hot spots, large-connection internet of things, and low-latency high-reliability communication.

Optionally, the processing module is further configured to:

A third aspect of the present invention provides an access network slice organizer, comprising: a memory and a processor for storing a program of execution instructions;

the processor is configured to perform:

A fourth aspect of the present invention provides a core network device, including: the third aspect provides an access network slice orchestrator.

The network slicing method, the device and the equipment jointly consider the wireless network slicing method of the wireless access network state and the channel time-varying characteristic, the access network slice organizer evaluates and rearranges the operation state of the slices according to the information of the periodic report or the event trigger mode obtained by monitoring, and informs the result to each slice, and each slice carries out corresponding adjustment according to the notification of the access network slice organizer; and repeating the two steps until the network reaches a required state. The network reconfiguration based on information perception can be realized, and the purposes of adapting to various service applications, being flexible, convenient and high-performance networking can be achieved.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.

FIG. 1 is a flow chart of a first embodiment of a network slicing method according to the present invention;

FIG. 2 is a flow chart of one embodiment of a network slicing method of the present invention;

FIG. 3 is a schematic diagram of an access network slice orchestrator performing network slicing on an F-RAN to generate four exemplary slices;

fig. 3a is a schematic diagram of an embodiment of a wide area seamless coverage network slicing method based on an F-RAN according to the present invention;

fig. 3b is a schematic diagram of an embodiment of a hot-spot high-capacity network slicing method based on F-RAN according to the present invention;

FIG. 3c is a diagram illustrating an embodiment of a F-RAN-based large-link Internet of things slicing method according to the present invention;

fig. 3d is a schematic diagram of an embodiment of a low-latency high-reliability network slicing method based on F-RAN according to the present invention;

fig. 4 is a unified frame structure used by the F-RAN based slices of the present invention;

fig. 5 is a flowchart of the operation steps of the access network slice orchestrator modifying each slice scheme in step S102 to achieve hard isolation;

FIG. 6a is a schematic diagram of an access network slice orchestrator allocating cache resources to slices according to the present invention;

FIG. 6b is a schematic diagram of the access network slice organizer implementing hard isolation of wireless resource allocation to each slice according to the present invention

FIG. 7 is a schematic structural diagram of a network slicing apparatus according to a first embodiment of the present invention;

FIG. 8 is a schematic structural diagram of an access network slice organizer according to a first embodiment of the present invention;

fig. 9 is a schematic structural diagram of a core network device according to a first embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. 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.

Fig. 1 is a flowchart of a first embodiment of a network slicing method of the present invention, and as shown in fig. 1, an execution main body of the scheme is a core network device or an access network slice orchestrator in the core network device, and the network slicing method includes the specific implementation steps of:

step a: and monitoring the network under the current slicing scheme, and acquiring user information, terminal information, service information, wireless access network state and network performance information.

Before the step, slicing the wireless access network according to the application scene, the wireless access network capability and the configuration information, and generating a slicing scheme by taking a typical application scene as a slicing blueprint; slicing the network according to the slicing scheme; wherein, when the slicing scheme is used for operation, different slices are in hard isolation.

The meaning is as follows: in an initial state, an access network slice orchestrator or core network equipment performs slicing of a wireless access network according to various mobile applications and application scenarios of operation and maintenance, wireless access network capabilities and configuration, and measurement feedback of a user, and generates a corresponding slicing scheme by taking slice blueprints corresponding to four typical application scenarios as templates. The access network slice orchestrator not only has the capability of monitoring and analyzing the report and feedback data, but also can establish mapping between the mobile application and the access network slices, select a proper slice scheme for the application, and optimize the slice scheme by combining the wireless access network state and the channel time-varying characteristics, so that the slices and the network operate well. The slice blueprint is used as a configuration file generated by the slice and used for explaining the required network functions in the corresponding slice and arranging and connecting the network function components in series, and different slices can be generated by the same slice blueprint due to the removal of part of processes and functions as required. And then according to the sensed user, terminal, service and access network state information, the access network slice organizer modifies the wireless network slice scheme, optimizes the slice configuration and ensures the hard isolation during slice operation, wherein the hard isolation means that when slices are deployed and operated according to the slice scheme, different slices do not influence each other, and the normal work of other slices is not influenced if the operation condition of one slice is good or not.

After the initialization configuration process is completed, the step a is executed in the network operation process, specifically, according to the information such as the user information, the terminal information, the service information, the wireless access network status, the network performance information and the like obtained by monitoring.

Step b: and evaluating the slice running state corresponding to the current slice scheme according to the user information, the terminal information, the service information, the wireless access network state and the network performance information to obtain an evaluation result.

Step c: and rearranging the slices and reconfiguring resources according to the evaluation result to obtain a changed slice scheme, and informing each slice to operate according to the changed slice scheme.

In the above steps b and c, in this step, the access network slice orchestrator or the core network device evaluates the slice operation status according to the obtained information, and according to the evaluation result, the access network slice orchestrator negotiates the corresponding slice, performs slice rearrangement and resource reconfiguration, and informs the result to each slice that needs to be adjusted.

Step d: and repeating the steps a to c until the difference value between the network performance corresponding to the changed slicing scheme and the network performance corresponding to the last slicing scheme is less than or equal to the preset threshold value.

In this step, each slice performs adjustment such as corresponding cache working mode, wireless resource allocation and the like according to the notification of the access network slice orchestrator, and then performs normal operation. Then the core network equipment or the access network slice organizer repeats the steps a to c until the whole network reaches the required ideal state, and each slice can normally run and ensure the good support of the service; the network ideal state means that the overall performance of the network is good, and the difference value between the network performance such as the energy efficiency and the spectrum efficiency after reconfiguration and the network performance after the last reconfiguration is not more than a threshold (namely less than a preset threshold); the normal operation of each slice refers to independent operation among slices, each slice can provide required service support for corresponding application, and the dynamic optimization and adjustment in the slices cannot influence the normal operation of other slices.

Further, when the core network device or the access network slice composer detects the periodic report information or the new trigger time information, a slice scheme is regenerated for the wireless access network according to the application scene, the wireless access network capability and the configuration information, and the network is sliced again according to the slice scheme;

and then repeating the steps a to c until the difference value between the network performance corresponding to the changed slicing scheme and the network performance corresponding to the last slicing scheme is less than or equal to the preset threshold value.

The network slicing method provided by this embodiment is a method of wireless network slicing in which the wireless access network state and the channel time-varying characteristics are jointly considered, according to the information of the periodic report or the event trigger mode obtained by monitoring, the access network slice organizer evaluates and rearranges the slice operation state, and informs the result to each slice, and each slice performs corresponding adjustment according to the notification of the access network slice organizer; and repeating the two steps until the network reaches a required state. The network reconfiguration based on information perception can be realized, and the purposes of adapting to various service applications, being flexible, convenient and high-performance networking can be achieved.

Fig. 2 is a flowchart of an embodiment of a Network slicing method of the present invention, and as shown in fig. 2, based on the first embodiment and based on a Fog Radio Access Network (F-RAN), for example, the specific operation steps of the F-RAN based wireless Network slicing method of the present invention include:

s101: and according to the received global information and the slice blueprint, the access network slice organizer slices the wireless access network to generate a slice scheme.

In this step, according to various mobile applications and application scenarios of operation and maintenance, wireless access network capability and configuration, and measurement feedback of users, the access network slice composer slices the wireless access network, and generates a corresponding slice scheme using slice blueprints corresponding to four typical application scenarios as templates. The access network slice orchestrator has the capability of monitoring and analyzing reporting and feedback data, establishes mapping between mobile applications and access network slices, selects a proper slice scheme for the applications, and optimizes the slice scheme by combining the wireless access network state and channel time-varying characteristics, so that the slices and the network operate well. The slice blueprint is used as a configuration file generated by the slice and used for explaining the required network functions in the corresponding slice and arranging and connecting the network function components in series, and different slices can be generated by the same slice blueprint due to the removal of part of processes and functions as required.

FIG. 3 is a schematic diagram of an access network slice orchestrator performing network slicing on an F-RAN to generate four exemplary slices; fig. 3a is a schematic diagram of an embodiment of a wide area seamless coverage network slicing method based on an F-RAN according to the present invention; fig. 3b is a schematic diagram of an embodiment of a hot-spot high-capacity network slicing method based on F-RAN according to the present invention; FIG. 3c is a diagram illustrating an embodiment of a F-RAN-based large-link Internet of things slicing method according to the present invention; fig. 3d is a schematic diagram of an embodiment of a low-latency high-reliability network slicing method based on F-RAN according to the present invention; fig. 4 is a unified frame structure used by the F-RAN based slices of the present invention.

As shown in fig. 3a to 3d, the four typical application scenarios specifically include: wide area seamless coverage, high capacity of hot spots, large connection internet of things, low time delay and high reliability of communication; the wide-area seamless coverage performance target is to pursue connection without time and user experience rate of hundred megabits per second, the hot-spot high-capacity performance target is to pursue peak transmission rate of up to ten gigabits per second, the large-connection internet-of-things pursues connection number exceeding million per square kilometer, and the low-delay high-reliability communication pursues end-to-end delay of millisecond level and transmission reliability rate close to hundred percent.

In step S101, more specific operation contents include:

(1) for the wide area seamless coverage application scenario, as shown in fig. 3a, in order to pursue seamless coverage and service communication of non-hotspot area mobile users, in the whole range where a slice of the wide area seamless coverage radio access network needs to be deployed, according to the network plan and the traffic volume and coverage of each area, a High Power Node (HPN) is deployed, and the configured HPN completes the service data and control information processing function with the help of its own configured radio frequency transmitting device and baseband processing device, and a Local Controller, also called Local network Controller (Local Controller), and provides service data and/or control signaling. The HPN can be a traditional macro base station or a micro base station, an enhanced macro base station configured with a large-scale multi-antenna or distributed antenna configuration, or a cloud base station, and is used for realizing downward compatibility of the existing cellular mobile communication system and ensuring seamless coverage of a wireless network.

The HPN has complete physical layer, data link layer, network layer and part of the function of the core network sunk according to the requirement; referring to fig. 3a, in order to achieve a user experience rate of 100Mbps, the HPN employs a low frequency band below 6GHz with good channel propagation characteristics, improves spectrum efficiency by means of a large-scale antenna technology, and improves data transmission rate and reduces call drop rate by optimizing transmission power and switching parameters; in order to support a larger bandwidth, the data transmission rate is increased by using the larger bandwidth, as shown in fig. 4, the physical layer uses a larger subcarrier interval, such as an integral multiple of 15kHz and 30kHz, and shortens the frame length, such as 5ms which can be divided by 1ms, so as to be compatible with a Long Term Evolution (LTE) frame structure; the physical layer can use Orthogonal Frequency Division Multiplexing (OFDM) and Orthogonal Frequency Division Multiple Access (OFDMA) technologies, or (Filtered OFDM, F-OFDM) and other novel multi-carrier technologies, Sparse Code Multiple Access (SCMA) and other novel Multiple Access technologies, and Low Density Parity Check (Low Density Parity Check, LDPC) codes, polarization codes, super nyquist modulation and other novel modulation and coding technologies, so as to improve user connectivity and spectral efficiency; the data link layer and the network layer follow default configuration, including Hybrid Automatic Repeat reQuest (HARQ) with optimized coverage.

In addition, the HPN has a heterogeneous network fusion function, and can realize heterogeneous cooperative fusion with other heterogeneous wireless access nodes, such as wireless access nodes of a wireless local area network or a wireless metropolitan area network, or millimeter wave visible light access nodes; meanwhile, the HPN can communicate with other slices to provide service data and/or control signaling for other slices and nodes, for example, a dynamic real-time backhaul link interface is provided between the HPN and a Baseband processing Unit (BBU) pool in the cloud computing server, so as to implement cooperative signal processing, radio resource management and mobility management with the BBU pool in the cloud computing server.

Or, (2) for a hotspot high-capacity application scenario, as shown in fig. 3b, in order to increase the wireless transmission rate, a large-scale antenna, a centralized cloud wireless access, or a heterogeneous aggregation technology is adopted; in order to meet the demand of flow density, the method can be realized by increasing the deployment density of the access nodes, improving the frequency spectrum efficiency, increasing the bandwidth and the like. A data/control separation technology is adopted, and a control plane is supported by an HPN (high-performance network) provided with a large-scale antenna, so that the reliable transmission of signaling is ensured; the data plane adopts centralized cloud wireless access, Remote Radio Heads (RRHs) are densely deployed in areas where hot spot high-capacity wireless access network slices need to be deployed according to network planning and the traffic and coverage of each area, near field communication is achieved through a large-bandwidth high frequency and terminal, the configured RRHs finish receiving and sending of service data and simple processing through front-end wireless Radio frequency signals and simple symbol processing modules configured by the RRHs, the service data are transmitted back to a BBU pool through fronthaul lines, and service support of the data plane is achieved through cooperative signal processing and centralized resource scheduling of the BBU pool.

More specifically, a plurality of F-APs are deployed in an area where the traffic is greater than the designated traffic or the access density is greater than the designated density, and the traffic with high popularity and high repeated request rate is preferentially pushed to the F-APs and other base stations with caches, wherein the F-APs are provided with a fog calculation baseband signal processing module and can absorb the hot spot capacity.

The F-AP not only has a front-end wireless radio frequency signal and a simple symbol processing module of the traditional RRH, but also has a baseband signal physical processing and wireless resource management control module, is used for realizing distributed cooperative communication with other adjacent F-APs with ideal backhaul link conditions, and can also carry out direct communication and resource management with a communication terminal accessed to the F-AP. Each F-AP is directly connected with a BBU pool in the cloud computing server by using a forward link, and has service and control interfaces with surrounding F-APs, so that cooperative communication among the F-APs can be realized under the management of the BBU pool in the cloud computing server, interconnection and intercommunication of a plurality of F-APs in a local range can be realized through a wireless Mesh network Mesh, a tree topology, a linear topology or a star topology, service sharing is realized, and meanwhile, the interference among the F-APs under the condition of intensive deployment is reduced.

The BBU pool comprises a physical layer, a data link layer, a network layer and an upper layer function, the RRH only has a physical layer simple radio frequency function and a symbol processing function, and the rest base band processing functions of the physical layer are converged into the BBU pool; the functional design scheme of each node protocol stack is basically consistent with the wide area coverage slice, and as shown in fig. 4, specific parameters of a frame structure can be optimized correspondingly according to the characteristics of a hotspot high-capacity scene channel and a service; a longer value is set for a Transmission Time Interval (TTI) to carry more data streams, and flexible duplex or full duplex can be introduced when an interference environment is simpler; the modulation coding can adopt a higher-order modulation mode and a higher code rate. Preferably, the data plane communication can be performed by using a high frequency band because the data plane communication is close to the terminal, so that the requirement of high capacity of a hotspot is met. The BBU pool has a heterogeneous network slice fusion function, can perform resource management calculation on other slices, and provides service data and/or control signaling for other slices through communication among the slices, as same as the HPN; in consideration of the characteristics of ultra-large bandwidth, small coverage and strong signal directivity of a high frequency band, the frame structure should be correspondingly optimized, the subcarrier interval can be increased to 150kHz, the frame length can be further shortened, the waveform can use a single carrier besides OFDM used by a low frequency band, novel waveform F-OFDM and the like, and the power utilization rate is improved; a Time Division Duplex (TDD) mode and a large-scale antenna are adopted, and the large-scale antenna compensates the influence caused by high path loss through adaptive beam forming and tracking.

Or, (13) for a large-connection internet-of-things application scenario, as shown in fig. 3c, in order to reduce control signaling overhead caused by dynamic scheduling of user resources, a clustering-based scheduling-free competitive access mode is adopted, an intelligent terminal local ad hoc network located in the coverage range of the same node F-AP or the same high-power node HPN is adopted, a terminal direct connection technology and a relay multi-hop technology are adopted for a terminal in a cluster, and information of each node in the cluster can be processed by a node of a cache service or is collected to a cluster head and is accessed to a network through the cluster head; the cluster head is selected according to the caching capacity and the connectable number of the nodes, the stronger the caching capacity of a certain node in the cluster is, the higher the service processing capacity is, and the proper position is at the same time, so that more intelligent terminals can be simultaneously accessed into the node, and meanwhile, the better the connection condition between the node and the network access node is, the higher the probability that the node is selected as the cluster head is; each cluster head and a non-intelligent terminal without networking capability access to the node in a scheduling-free competitive access mode, so that signaling overhead is reduced, and a large number of terminals can be ensured to access to a network simultaneously.

For nodes with caching capacity in the cluster, the sinking of the control plane function is more inclined, so that the cluster formed by the nodes has distributed business service capacity; a plurality of terminal nodes in the cluster are connected with one another through interfaces to form a Mesh topology, a tree topology, a linear topology or a star topology structure of the wireless Mesh network; in the same F-AP coverage range, terminals are connected with each other, a wireless mesh network topological structure is established under the management of the F-AP, when a certain node in the network fails or cannot work, the F-AP can adjust and allocate resources and transmit information in a mode of constructing a new route; or the terminal nodes establish a wireless tree topology network architecture under the management of the F-AP, and the nodes in the wireless tree topology structure can be expanded by connecting new nodes, so that the management of the terminal nodes is facilitated.

Considering that the large-connection Internet of things application scene is lower in channel bandwidth, an access node F-AP or HPN responsible for a coverage area can adopt scattered low-frequency band, fragment spectrum or partial OFDM subcarrier; by adopting a narrow-band system design, the slice coverage capability is improved, the number of access devices is increased, and the power consumption and cost of the terminal are reduced; in order to reduce the Power consumption of the terminal, a greatly Enhanced energy-Saving mechanism is adopted, the connection state is optimized by enhancing technologies such as Enhanced Discontinuous Reception (eDRX), energy Saving Mode (Power Saving Mode, PSM) and the like, the starting time is reduced, the paging interval is increased, and the data acquisition and reporting frequency is optimized, so that the terminal is in a low-Power-consumption dormant state for a longer time.

The HPN and the F-AP are provided with a complete physical layer, a data link layer, a network layer and an upper layer consisting of partial function sinks; the physical layer mainly focuses on how to efficiently support massive connection, and can adopt a long frame with narrow bandwidth, so that the F-AP and the HPN can perform multi-user data inclusion and scheduling by virtue of the cluster head; the waveform can adopt a new waveform technology based on high-efficiency filtering, such as F-OFDM, GFDM and the like, so as to reduce out-of-band interference, and utilize scattered spectrum and fragment spectrum to effectively realize decoupling of the technical scheme between sub-bands, so that modulation codes and the like of different sub-bands can be independently configured; the multiple access mode adopts a novel multiple access technology such as a non-orthogonal access mechanism MUSA, can realize that single connection which is more than 3 times is multiplexed on single time-frequency resources, has an overload rate of more than 300 percent, and has lower complexity.

Compared with wide area coverage and high capacity of hot spots, the large-connection Internet of things service has the characteristics of small data packets, massive connection and strong burstiness, so that the protocol stacks of a data link layer and a network layer cannot be completely reused, and the corresponding protocol stacks need to be recombined and combined to reduce the overhead of the protocol stacks; preferably, in the attachment process, a scheduling-free competition access mode is adopted, so that the signaling flow is simplified, and the user access delay and the system signaling overhead are reduced; the header compression and encryption of a data link layer are omitted, and the data processing amount is reduced; considering that the total amount of data is large, part of error data can be omitted, and a Radio Link Control (RLC) entity is configured to UM accordingly, so that the retransmission times are reduced; the network layer optimizes signaling flows of simplified mobility management, connection management, roaming and the like, and reduces system overhead.

Or, (4) for a low-latency high-reliability communication application scenario, as shown in fig. 3d, in order to reduce forwarding nodes and network forwarding latency, a local cache fog access node F-AP or a terminal may be used for direct communication, where the F-AP has a front-end wireless radio frequency signal and symbol processing module, and also has a baseband signal physical processing and wireless resource management control module, and can perform direct communication and resource management with a communication terminal accessing the F-AP; a dynamic self-organizing network structure is carried out between the F-APs, a dynamic Mesh communication link is established, single-hop and multi-hop communication between equipment and a terminal is supported, distributed storage of resources and interaction of information are realized, the number of routing forwarding nodes is reduced, and end-to-end time delay is reduced; in order to improve link reliability, when a backhaul link is in a good condition, distributed cooperative communication can be performed by using the F-AP and other adjacent F-APs, joint transmission between the F-APs is realized under the management of a BBU pool in a cloud computing server, and when the backhaul link is in a less than ideal condition, dual connection is adopted, the F-AP is accessed simultaneously, and a HPN (Multi-input Multi-output, MIMO) technology is configured.

The F-AP and the HPN have extremely simplified physical layers and data link layers to ensure low delay of service transmission, so that a complete protocol stack function is not required, and cross-layer design of a protocol stack is required; as shown in fig. 4, the physical layer frame structure adopts a short frame structure of short TTI, which ensures short and fast scheduling period, and at the same time, in order to ensure a certain data load, a large bandwidth is used. Due to the fact that Cyclic Prefix (CP) overhead is too large due to short TTI design, a Filter Bank Multi-Carrier (FBMC) without CP can be considered to be adopted as a waveform, and filtering based on subcarriers can be achieved to adapt to channel change more flexibly; the multiple access technology adopts novel multiple access technologies such as SCMA (sparse code multiple access) and the like, so that a resource allocation process is avoided, and zero waiting time of uplink data packet scheduling is realized; the modulation coding adopts advanced coding, space-time frequency diversity and other technologies to improve the reliability of a single link.

The data link layer can simplify or omit header compression and encryption, and reduce processing time delay; the self-adaptive HARQ is adopted, under the condition of a high signal-to-noise ratio channel, the HARQ is omitted, the time delay is reduced, under the condition of a low signal-to-noise ratio channel, the enhanced optimization HARQ is adopted, the retransmission performance is improved, the link reliability is increased, and the retransmission times are reduced; and scheduling-free competitive access with high priority is adopted to ensure the rapid access of the terminal. Preferably, in some scenarios, the F-AP and the HPN need network layer functions, and at this time, the connection state needs to be optimized, and the RRC state switching time of the terminal needs to be reduced, and some functions need to be sunk to an Upper layer, such as local paging and localization of mobile content.

If the application scenario does not belong to the above four typical application scenarios, the access network slice composer may determine that the application is a mixed mode between the four typical applications according to the application characteristics and the wireless network operating state, and the slice corresponding to the application may also be composed of a mixture of slices of the above typical applications.

S102: and according to the sensed various state information, the access network slice organizer modifies the wireless network slice scheme to ensure hard isolation of the slices.

In the step, according to the sensed user, terminal, service and access network state information, the access network slice orchestrator modifies the wireless network slice scheme, optimizes slice configuration and ensures hard isolation during slice operation; the hard isolation means that when the slices are deployed and operated according to the slice scheme, different slices do not influence each other, and whether the operation condition of one slice is good or not does not influence the normal work of other slices. Fig. 5 is a flowchart of operation steps of the access network slice orchestrator modifying each slice scheme in step S102 to implement hard isolation, and referring to fig. 5, in step S102, more specific operation contents include;

s201, based on the determined slice blueprint, the access network slice organizer corrects the slice scheme by combining the obtained global perception information.

The access network slice organizer takes the determined slice blueprint as a template, combines the sensed user, terminal, service and access network state information, modifies and generates a wireless network slice scheme, performs corresponding slice hard isolation, and provides cache allocation and cache working modes, wireless resources and allocation and transmission modes under different slices; the slice blueprint comprises a slice blueprint pre-stored in a network, and also comprises a slice blueprint generated by mixing a plurality of slice blueprints according to service characteristics, and the same network slice blueprint can remove part of processes and functions according to actual requirements and network environments to generate different slice network deployment schemes;

s202, the slice orchestrator determines each slice caching resource and caching working mode.

The access network slice orchestrator jointly considers and determines cache resources and cache working mode modification of each slice according to the network state and the actual requirements of each slice; in order to reduce the resource overhead of frontaul and backhaul, or reduce the transmission delay of frontaul and backhaul, some services are pushed and issued to an access node with a buffer in advance; the allocation priority of different slice cache resources is in the following order: low time delay, high reliability communication, large connection Internet of things, high hotspot capacity and wide area seamless coverage; for low delay and high reliability, when the delay requirement is extremely low, the service is cached in a cache of a nearby device close to the requesting user, such as an F-AP or an adjacent intelligent terminal, and the content is updated in a short period; when the service requirement time delay is low, the service is pushed to other access nodes in the area where the requesting user is located, such as a two-hop or multi-hop reachable F-AP or an intelligent terminal, and is updated in a short period; when the service is time delay tolerable, the service is cached in the BBU pool, and the content is updated in a longer period; for large-connection Internet of things, when the number of connections is large, the service is preferentially pushed to the intra-cluster backbone nodes to perform distributed data processing, so that the total amount of cluster head summary information is reduced; for high capacity of the hot spot, the hot service is pushed to the hot spot area according to the popularity;

FIG. 6a is a schematic diagram of an access network slice orchestrator allocating cache resources to slices according to the present invention; fig. 6b is a schematic diagram of the access network slice orchestrator implementing hard isolation for wireless resource allocation of each slice according to the present invention. As shown in fig. 6a, the UE1 requests traffic requiring very low transmission delay, so UE1 requests traffic to be buffered in a nearby intelligent terminal (or F-AP) close to UE1 and updates the content in a short period; the UE2 requests a service with a low delay, and the requested service is pushed to other access nodes in the area where the UE2 is located, such as a two-hop or multi-hop reachable F-AP (or an intelligent terminal), and is updated in a short period; the UE3 requests that the traffic is delay tolerant, and its traffic is buffered in BBU pool (or other nodes with abundant buffer resources, or even core network) to update the content in longer period; for a hot spot area, part of high-popularity service is cached in the F-AP, and when the UE4 requests the service to be in the local F-AP, the UE4 accesses the F-AP to realize hot spot capacity absorption; for a large-connection Internet of things area, a backbone node such as a cluster head utilizes a cache device of the cluster head to cache part of services, service support is provided for part of users in the cluster, such as UE5 and UE7, and the request services of other users such as UE6 cannot be processed locally, are summarized by the cluster head and are transferred to other parts in a network to be completed.

S203, the slice composer determines the radio resource and allocation of each slice.

The access network slice orchestrator allocates a proprietary wireless resource for different slices according to the coverage area and the geographic position of each slice and the sensed user, terminal, service and access network state information, and reserves a part of shared wireless resource; the dedicated radio resources allocated for different slices may be orthogonal or shared; if different slices have overlapping in space, then allocate the orthogonal radio resource; aiming at multi-slice operation under the resource sharing situation, the maximum allowable transmitting power of each slice is determined by an access network slice orchestrator through open-loop power control, and the interference among different slices is ensured to be smaller than a preset threshold;

referring to fig. 6b, when the radio resources are abundant, due to the long distance between the geographical locations, the large-link internet-of-things slice and the low-delay high-reliability slice multiplex the same dedicated radio resources 1; the wide area seamless coverage slice shares the geographical position with the large-connection Internet of things slice and the low-delay high-reliability slice, so that the wide area seamless coverage slice is allocated to the special wireless resource 2 and is orthogonal to the special wireless resources 1 of the other two slices;

s204, according to the distributed wireless resource and the buffer resource, the slice composer determines each slice transmission mode.

The adjustment of the transmission mode of each slice comprises that the access network slice composer determines the transmission mode of each slice according to the allocated wireless resources and cache resources; the transmission mode comprises each node accessed to the F-RAN, such as HPN, F-AP, even an intelligent terminal with a direct module, and other heterogeneous wireless access nodes, such as wireless access nodes of a wireless local area network or a wireless metropolitan area network, or millimeter wave visible light access nodes; the concrete contents comprise the following steps:

and aiming at the low-delay high-reliability slice, if the moving speed of the user terminal is less than or equal to a preset moving speed threshold value M1, the distance between the user terminals to be communicated is less than a preset distance threshold value D1, and the user terminals to be communicated are provided with a direct connection module and are all accessed to the same F-AP, the communication is carried out in a terminal direct connection mode. In this case, the F-AP controls and manages the accessed terminal direct connection pair according to the channel packet loss rate, throughput, end-to-end network transmission delay, and according to the channel usage condition. Otherwise, adopting an F-AP mode, and carrying out management control on the non-terminal-direct-connection user by the F-AP according to the service volume, the packet loss rate, the throughput and the end-to-end network transmission delay;

aiming at a large-connection Internet of things slice, in order to reduce control signaling overhead brought by user resource dynamic scheduling, a clustering-based scheduling-free competitive access mode is adopted, intelligent terminal local ad hoc networks located in the coverage range of the same node F-AP or the same high-power node HPN are selected out to communicate with external access nodes, and each cluster head and a non-intelligent terminal access the nodes in a scheduling-free competitive access mode, so that the signaling overhead is reduced; when various network structures are formed by dynamic networking among nodes, the nodes of the cache service are preferentially selected to become cluster heads, and the nodes of the cache service can be ensured to be accessed into more nodes at the same time; the node without the cache takes the cache node as a cluster head under the management of the F-AP and the node with the cache to establish a wireless tree topology structure, so that the management node and a new connecting node can be conveniently expanded;

aiming at the high-capacity slices with high heat points, the slices are divided into two types according to terminal communication objects, when a user requests that the communication objects are other users, if the distance between user terminals to be communicated is greater than a preset threshold D2, the global BBU pool mode is adopted for communication, and when the distance between the user terminals is less than D2, the local F-AP mode is adopted for communication; when a user requests that a communication object is an access node and requests data, if the cached data is close to the F-AP and the distance between the user terminal and the F-AP is less than a preset threshold D3, a local F-AP mode is adopted for communication; and when the cached data is not in the adjacent F-AP or the distance between the user terminal and the F-AP is greater than a preset threshold D3, the communication is carried out by adopting a global BBU pool mode.

S205, the slice composer informs each slice of the slice scheme, and each slice is specifically configured according to the obtained scheme, so that the operation and the hard isolation of the slice are realized.

S103: and detecting to obtain information of periodic report or event trigger, and rearranging the slices by the access network organizer to optimize the network.

The user, terminal, service, wireless access network state, user and network performance information period or event trigger mode are reported to the access network slice orchestrator, the access network slice orchestrator carries out corresponding slice negotiation according to the slice to be adjusted, and the main key performance index of each slice in the slice set, reconfigures the wireless resources and cache resources allocated to different slices, and informs the slice to be adjusted of the reconfiguration result, which is specifically realized as follows:

the periodic report information obtained by the access network slice orchestrator comprises the periodic report information from the user, such as the user position, the moving speed and the like, monitored when the slice runs according to the slice scheme of the access network orchestrator; periodic report information from the terminal, such as terminal cache and residual capacity; periodic reporting information from the service, such as space-time distribution, delay requirements; radio access network status information, such as available resources, interference relationships; network performance information, such as energy efficiency, spectral efficiency, etc.; the reporting information of the event trigger mode obtained by the access network slice orchestrator comprises new service request information, slice operation request cancellation information, slice interference conflict information and the like.

According to the periodic report information or the report information of the event trigger mode, the access network slice orchestrator judges and checks the operation condition of the slice, and the access network slice orchestrator acquires that the main key performance index of a certain slice is lower than a preset threshold value and needs to perform resource reconfiguration or service shunting. For wide area seamless coverage, the main key performance indexes are coverage rate, call drop rate and handover failure rate; for high capacity of hot spots, the main key performance indexes are user experience rate, flow density and hot spot rate; for a large connection internet of things, the main key performance indexes are large connection and low power consumption; for low-delay and high-reliability communication, the main key performance indexes are air interface delay, end-to-end delay and reliability.

Comparing with other slices in the network according to the type of the slice to be adjusted, and confirming the influence of wireless resources and buffer resources on the slice; the cache configuration and the influence of the file push rule in the cache on different slices are in the following order: low time delay, high reliability communication, large connection Internet of things, high hotspot capacity and wide area seamless coverage; the high and low impact of radio resources and allocation on different slices is: low-delay and high-reliability communication, hot spot and high-capacity application, wide-area seamless coverage application and large-connection Internet of things application.

According to the scheme of adjusting the slices, selecting a proper main key performance optimization method, including reconfiguration of wireless resources and cache resources and the like, so as to reduce the influence on other slices; when the access network slice organizer has resource reservation or other slice special resources are sufficient, preferentially selecting to allocate special resources to the slice needing performance enhancement, wherein the allocated special resources come from the access network slice organizer to reserve or reduce the available special resources of other slices and are orthogonal to the resources used by other slices; if no available orthogonal dedicated resources exist, preferentially selecting dedicated resources with small influence on other slices by multiplexing according to the influence on other slices; if no special resource is available, multiplexing the shared resource, and preferentially selecting the shared resource with small influence on other slices; and if no shared resource is available, carrying out service distribution, and distributing the service to slices capable of providing similar services according to the service performance requirement.

S104: and each slice is correspondingly adjusted according to the notice of the access network slice organizer.

And each slice adjusts the corresponding cache working mode, wireless resources and allocation according to the notice of the access network slice organizer. In order to ensure that the main key performance index of the slice is greater than the threshold value, firstly, the cache working mode is adjusted, and then, the wireless resource allocation is adjusted, the specific steps are as follows:

firstly, adjusting a cache working mode: according to the reconfiguration result, the required cache file is pushed into the slice and is determined by the internal resource management function of the slice, the cache file needs to be cached to the position in the slice, the size of the cache file of each node in each position, and the updating and updating period of the content; when the slice is served by a low-delay high-reliability terminal, the required service is preferentially pushed to an F-AP or an intelligent terminal close to the terminal; when the slice serves a large-connection Internet of things, the service is preferentially pushed to an F-AP or a cluster node with a large connection number; when the slice serves the hot spot with high capacity, caching the files with high popularity and multiple request times in a proper F-AP in advance;

then, adjusting a wireless resource allocation mode: according to the cache adjusting position, optimizing and adjusting wireless resources in the same geographic area; preferably, for an F-AP or a terminal caching a file, and an application scene has high requirement on time delay or large repeated request quantity, wireless resources are preferentially allocated so that the terminal can be accessed with a higher probability, and the service performance of the terminal can be ensured to meet the requirement and the utilization of the cached resources;

and finally, adjusting a transmission mode: according to the reconfiguration or modification of the cache resources and the wireless resources, the networking mode and the transmission mode also need to be correspondingly adjusted and optimized; when the slice is used for serving low delay and high reliability, the access node is one or more of HPN, F-AP, a terminal with cache and other access nodes; when the slice serves a large-connection Internet of things, the access node is one or more of HPN, F-AP, a terminal with cache and other access nodes; when the slice is to serve a hotspot high capacity, the access node is one or more of a F-AP, a RRH, and other access nodes.

Waiting for the information of the next periodic report or event trigger mode, and repeating the two steps until the whole network reaches the required ideal state, and each slice can normally run and ensure the good support of the service; the network ideal state means that the overall performance of the network is good, and the difference between the network performance such as the energy efficiency and the spectrum efficiency after reconfiguration and the network performance before reconfiguration is not larger than a threshold value; the normal operation of each slice refers to independent operation among slices, each slice can provide required service support for corresponding application, and the dynamic optimization and adjustment in the slices cannot influence the normal operation of other slices.

Fig. 7 is a schematic structural diagram of a network slicing apparatus according to a first embodiment of the present invention, where the network slicing apparatus 10 includes:

an obtaining module 11, configured to execute step a, monitor a network in a current slicing scheme, and obtain user information, terminal information, service information, a wireless access network state, and network performance information;

a processing module 12 for performing:

Optionally, the processing module 12 is further configured to:

Optionally, the typical application scenarios used by the processing module 12 include any scenario of wide-area seamless coverage, high capacity of hot spots, large-connection internet of things, and low-latency high-reliability communication.

Optionally, the processing module 12 is further configured to:

The network slicing apparatus provided in this embodiment is configured to execute the network slicing method, and the implementation principle and the technical effect thereof are similar, and are not described herein again.

Fig. 8 is a schematic structural diagram of a first embodiment of the access network slice organizer according to the present invention, and as shown in fig. 8, the access network slice organizer 20 includes: a memory 21 and a processor 22 for storing programs of execution instructions;

the processor 22 is configured to perform:

The access network slice organizer provided in this embodiment is configured to execute the foregoing network slice method, and its implementation principle and technical effect are similar, which are not described herein again.

Fig. 9 is a schematic structural diagram of a first core network device according to the present invention, and as shown in fig. 9, the core network device 30 includes the access network slice orchestrator 20 shown in fig. 8.

The core network device provided in this embodiment is configured to execute the network slicing method, and the implementation principle and the technical effect of the core network device are similar, which are not described herein again.

In the above embodiments of the access network slice organizer and the core network device, it should be understood that the Processor may be a Central Processing Unit (CPU), or may also be other general purpose processors, Digital Signal Processors (DSP), Application Specific Integrated Circuits (ASIC), etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the embodiments of the present invention may be directly implemented by a hardware processor, or may be implemented by a combination of hardware and software modules in the processor.

Those of ordinary skill in the art will understand that: all or a portion of the steps of implementing the above-described method embodiments may be performed by hardware associated with program instructions. The aforementioned program may be stored in a computer-readable memory. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned memory (storage medium) includes: read-only memory (ROM), RAM, flash memory, hard disk, solid state disk, magnetic tape (magnetic tape), floppy disk (flexible disk), optical disk (optical disk), and any combination thereof.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method of network slicing, comprising:

c, rearranging the slices and reconfiguring resources according to the evaluation result to obtain a modified slice scheme, and informing each slice to operate according to the modified slice scheme; the rearrangement comprises the steps of determining the cache resources and cache working modes, wireless resources and allocation and transmission modes of each slice;

step d, repeating the steps a to c until the difference value between the network performance corresponding to the changed slicing scheme and the network performance corresponding to the last slicing scheme is less than or equal to a preset threshold value;

before monitoring a network under a current slicing scheme and acquiring user information, terminal information, service information, a wireless access network state and network performance information, the method further comprises the following steps:

2. The method according to claim 1, wherein the typical application scenarios include any scenario among wide area seamless coverage, hot spot high capacity, large connection internet of things, and low latency high reliability communication.

3. The method of claim 1, further comprising:

4. A network slicing apparatus, comprising:

a processing module to perform:

wherein the processing module is further configured to:

5. The apparatus of claim 4, wherein the typical application scenarios used by the processing module include any scenario among wide area seamless coverage, hot spot high capacity, large connection internet of things, and low latency high reliability communication.

6. The apparatus of claim 4, wherein the processing module is further configured to:

7. An access network slice orchestrator, comprising: a memory and a processor for storing a program of execution instructions;

the processor is configured to perform:

before the network under the current slicing scheme is monitored and user information, terminal information, service information, wireless access network state and network performance information are acquired, the method is further used for executing:

8. A core network device, comprising: the access network slice orchestrator of claim 7.