CN106358202A - Improved C-RAN network architecture and resource dispatching method - Google Patents

Abstract

The present invention provides an improved C-RAN network architecture and resource scheduling method, wherein the method includes the following steps: a remote wireless unit receives user request information, and sends the request information to a baseband processing unit; the baseband processing unit receives the request information, and send the request information to the controller; the controller calculates the load required by the request and compares it with the load threshold of the local baseband processing unit; if the required load is greater than the load threshold of the local baseband processing unit, the local baseband processing According to the scheduling mechanism, the unit maps user requests to other baseband processing units in the baseband processing pool for resource processing. The method can effectively solve the resource allocation problem among the baseband processing units in the communication field.

Description

An improved C-RAN network architecture and resource scheduling method

technical field

The present invention relates to the technical field of convergence access network, and more specifically, relates to an improved C-RAN network architecture and resource scheduling method.

Background technique

In the wireless communication cellular network system, the number of access users and the amount of business data that each cell needs to carry are random and unevenly distributed, especially problems such as traffic bursts and network congestion caused by hotspot cells, making The load in the whole system is in an unbalanced state, therefore, realizing the resource scheduling of each cell has become one of the problems to be solved urgently in the mobile communication system. At the same time, with the gradual rise of 5G wireless technology, in addition to traffic increases, traffic bursts or fluctuations, the access network also needs to meet other requirements, such as providing business users with higher security, network availability and quality of service (QoS). , QoS) guarantee, etc. It is also an urgent problem to improve the traditional C-RAN architecture to obtain a new network architecture that can meet "large capacity, wide coverage, few offices, and flattening".

In the traditional radio access network system, major operators are faced with the challenges of reducing costs, increasing profits, and reducing energy consumption and emissions. China Mobile and others took the lead in proposing a new radio access network architecture C-RAN for green evolution. As shown in Figure 1, C-RAN has a virtual baseband resource pool management structure, which mainly includes three parts: 1) a distributed wireless network composed of remote radio units (Remote Radio Unit, RRU) and antennas; 2) The remote wireless radio frequency unit is connected by a high-bandwidth and low-latency optical transmission network; 3) A centralized baseband processing pool composed of high-performance general-purpose processors and real-time virtualization technology.

Centralized baseband processing can ensure resource sharing, that is, the virtual base station can share information sent and received by all users, service data, and channel quality information in the baseband processing unit (BBU), thereby realizing joint processing and dynamic scheduling, and enhancing The reliability of the overall operation of the BBU. In a BBU pool, reasonable dynamic resource allocation between BBUs is of great significance. For example, for traffic distribution scenarios with "tidal effects", according to the periodic traffic changes in different regions, fewer static resources can be allocated for them to ensure basic coverage, and the remaining resources form a dynamic carrier pool. Through flexible virtual Real-time and dynamic allocation of remaining resources through an integrated system, and flexible scheduling according to the needs of the business load of each cell in different business hours.

In the prior art, there is no access network architecture that can realize the converged access network architecture that is compatible with the traditional C-RAN and the cellular network, and thus effectively cope with the problem of resource scheduling in the big data era of the rise of 5G wireless technology.

How to implement resource scheduling between BBUs has become a key concern of the technical community.

Contents of the invention

The present invention provides a method that overcomes the above-mentioned problems or at least partially solves the above-mentioned problems.

The present invention provides an improved C-RAN network architecture, which is characterized in that it includes a metropolitan area core network, an access network, and a plurality of fusion nodes; wherein the fusion nodes include user end nodes and return nodes, so The user end node integrates the remote wireless unit and the optical network unit, and the backhaul node is arranged at the edge of the metro core network, and integrates the baseband processing unit, the central station, the reconfigurable optical add-drop multiplexer and the optical line terminal.

The present invention also provides a resource scheduling method, in which a general controller is designed as the control center of all BBUs in a BBU pool, and the method includes the following steps:

The remote wireless unit receives the user request information, and sends the request information to the baseband processing unit;

The baseband processing unit receives the request information, and sends the request information to the controller;

The controller calculates the load required for the request and compares it with the load threshold of the local baseband processing unit; if the required load is greater than the load threshold of the local baseband processing unit, the local baseband processing unit maps the user request to the baseband processing unit according to the scheduling mechanism Other baseband processing units in the pool perform resource processing.

This application proposes an improved C-RAN network architecture and resource scheduling method, designs a new type of network architecture under the traditional C-RAN architecture, and studies the connection between the baseband processing units at the converging end on the basis of this architecture. Cooperative communication and the algorithm mechanism for resource scheduling can effectively improve network performance, meet the new demand for capacity and bandwidth of the access network, and at the same time realize dynamic resource allocation among baseband processing units and resource scheduling.

Description of drawings

FIG. 1 is a schematic diagram of a traditional radio access network C-RAN architecture according to an embodiment of the present invention.

FIG. 2 is a schematic diagram of an improved C-RAN network architecture suitable for long-distance and relatively sparse deployment scenarios according to an embodiment of the present invention.

FIG. 3 is a schematic diagram of an improved C-RAN network architecture suitable for short-distance denser deployment scenarios according to an embodiment of the present invention.

Fig. 4 is a schematic diagram of overall steps of a resource scheduling method according to an embodiment of the present invention.

Fig. 5 is a schematic flowchart of the step "the local baseband processing unit maps user requests to other baseband processing units in the baseband processing pool for resource processing according to the scheduling mechanism" in the resource scheduling method according to an embodiment of the present invention.

Fig. 6 is a schematic diagram of the step "selecting the destination baseband processing unit of the optimal route based on the improved Dijkstra algorithm" in the resource scheduling method according to an embodiment of the present invention.

detailed description

The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

First, some technical terms involved in specific embodiments or drawings are explained: RRU: Remote Radio Unit, BBU: Baseband Processing Unit, ODN: Optical Distribution Network, ONU: Optical Network Unit, OLT: Optical Line Terminal, WDM: Wavelength Division Multiplexing, RoF: Optical Carrier Wireless Communication, OTN: Optical Transport Network, CO: Central Office, Core Mesh: Core Network, OADM: Optical Add-Drop Multiplexer, ROADM: Reconfigurable Optical Add-Drop Multiplexing device.

As shown in Figure 1, in a specific embodiment of the present invention, in general, an improved C-RAN network architecture is characterized in that it includes a metropolitan area core network, an access network, and multiple converged nodes ; wherein the converged node includes a user end node and a backhaul node, the user end node integrates a remote wireless unit and an optical network unit, and the backhaul node is arranged at the edge of the metro core network and integrates a baseband processing unit, Central station, reconfigurable optical add-drop multiplexer and optical line terminal.

In a specific embodiment of the present invention, in general, an improved C-RAN network architecture is characterized in that the converged access network network architecture is divided into three parts:

Fronthaul: The analog RoF technology is used for simple and flexible deployment. The combination of RoF and PON improves spectrum resource utilization. The user end node adopts a fusion node, which combines RRU and ONU, effectively realizing the integration of optical network and wireless network.

Backhaul network: OTN sinking technology is used, and tree or ring topology is logically adopted according to different scenarios, so as to better realize the characteristics of low cost, low latency and virtualization, and meet the requirements of 5G.

Converged node-Advanced BBU: Convergence node (i.e., backhaul node) at the edge of the metropolitan area, integrating BBU and central office (central office, CO) functions. Here, it belongs to the OTN node at the metropolitan area network end, and belongs to the Optical line terminal (OLT) node, with ROADM and OTN technology, is a multi-functional convergent node that supports flexible uplink and downlink, and directly aggregates upward to the metropolitan core network to realize the integration of heterogeneous network resources Handle control.

The network architecture proposed by the present invention includes two application scenarios: a long-distance relatively sparse deployment scenario and a short-distance dense deployment scenario, fully solving the problems of long-distance, large-capacity, and traffic bursts in the next-generation mobile communication system.

In another specific embodiment of the present invention, the user end node is configured to receive user request information, and send the request to the fronthaul network, so as to realize the integration of the optical network and the wireless network.

In another specific embodiment of the present invention, the backhaul network in the network architecture adopts optical transport network sinking technology to receive the user request information sent by the user end node, and send the request information to the metropolitan area Core Network.

In another specific embodiment of the present invention, the backhaul node is configured to aggregate data information sent by the access network, and send the data information to the metro core network.

In another specific embodiment of the present invention, the backhaul network backhaul uses OTN sinking technology to meet the demand for large-capacity bandwidth and provide high-speed transmission of large-grained services. According to different scenarios, tree or ring topology is logically adopted, and more Good low-cost, low-latency and virtualization features to meet 5G requirements.

The distributed wireless network sets the corresponding topology according to the following scenarios:

In the long-distance and sparse deployment scenario, a ring topology is adopted, as shown in Figure 2;

In short-distance and dense deployment scenarios, a tree topology is used. As shown in Figure 3.

In another specific embodiment of the present invention, the fronthaul network adopts CoMP: Coordinated Multiple Points Transmission/Reception (CoMP: Coordinated Multiple Points Transmission/Reception), and through joint scheduling or joint processing, interference between different cells is avoided, while improving The signal-to-noise ratio received by the terminal.

In another specific embodiment of the present invention, in the tree-type topology distributed wireless network, the transmission network between the BBU and the ODN adopts the RoF-WDM-PON fusion technology, and simultaneously realizes wireless over light and passive optical network communication.

As shown in Fig. 4 and Fig. 5, the present invention also proposes a resource scheduling method for baseband processing units based on any of the aforementioned networks. In this method, all BBUs in a BBU pool are controlled by the central controller Controller. The controller decides whether a scheduling mechanism needs to be executed among the BBUs in the pool. Overall, the method includes the following steps:

Initialize user resource information;

The RRU receives the request information from the user, and sends the request information to the baseband processing unit;

BBU receives requests from users, establishes a user resource database, and dynamically updates service information;

The baseband processing unit receives the request information, and sends the request information to the controller;

The controller calculates the load required for the request and compares it with the load threshold of the local baseband processing unit; if the required load is greater than the load threshold of the local baseband processing unit, the local baseband processing unit maps the user request to the baseband processing unit according to the scheduling mechanism Other baseband processing units in the pool perform resource processing.

In another specific embodiment of the present invention, "the local baseband processing unit maps user requests to other baseband processing units in the baseband processing pool for resource processing according to the scheduling mechanism" includes:

Step 1, select the destination baseband processing unit of the best route, and determine its location;

In step 2, the controller performs load information transmission from the source baseband processing unit to the destination baseband processing unit according to the scheduling mechanism.

In another specific embodiment of the present invention, step 1 further includes: selecting the destination baseband processing unit of the optimal route based on the improved Dijkstra algorithm.

In another specific embodiment of the present invention, the improvement of the Dijkstra algorithm lies in: using the bandwidth utilization rate U(i, j) and the baseband processing unit business occupancy rate ω as weights, determine the shortest path between the node pairs and Establish a link; mark the link whose remaining bandwidth is less than the bandwidth required by the load as "busy", thereby obtaining the remaining network G', which is brought into the subsequent Dijkstra algorithm.

More specifically, as shown in Figure 6, the specific steps of the step 1 "in the BBU pool, determine the best path destination BBU of the local BBU":

a), detect the link request, if there is an establishment request r(s, d, b) arrives, execute step b);

b), ensure that the link that meets the required bandwidth in the entire network, marks the link with the remaining bandwidth of the link less than the required bandwidth as "busy", and obtains the remaining network G';

c), calculate the bandwidth utilization U(i,j) of each link in G';

d) Using the link bandwidth utilization rate and service occupancy rate as the weight of the link (i, j), use the improved Dijkstra algorithm to determine the shortest path (s, d) between the node pairs in the remaining network G'. If the path exists, enter step e), otherwise reject the LSP establishment request, return to step a), and get ready for the next LSP establishment request;

e), from s to d along the shortest path routing to establish an LSP whose bandwidth requirement is b, and update the remaining bandwidth on the link;

f), return to step a), and get ready for the next LSP request.

In another specific embodiment of the present invention, the scheduling mechanism in step 2 includes: a), determining the need for scheduling, the local baseband processing unit transmits the demand signal information to the controller to find the optimal destination baseband processing unit, Establishing a link; b), the local baseband processing unit transmits the load information to the destination baseband processing unit; c), deleting the link, and updating the remaining bandwidth.

In another specific embodiment of the present invention, in the BBU pool proposed by the present invention, each baseband processing unit adopts a resource reservation scheme of 10% bandwidth reservation for local users, so as to be suitable for local small-scale and small-scale services Burst, avoid frequently calling the scheduling mechanism, reduce routing overhead, and maximize resource utilization.

Finally, the method of the present application is only a preferred embodiment, and is not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (10)

1. a kind of improved c-ran network architecture is it is characterised in that include Metro core network, access network and multiple pattern of fusion section Point；Wherein said pattern of fusion node includes user endpoints, passback node, described user endpoints merge remote radio units and Optical network unit, described passback inserting knot, in described metro core network edge, merges baseband processing unit, central station, can weigh Structure optical add/drop multiplexer and optical line terminal.

2. network as claimed in claim 1, will it is characterised in that described user endpoints are used for receive user solicited message Described request sends to forward pass network, realizes the fusion of optical-fiber network and wireless network.

3. network as claimed in claim 1 is it is characterised in that the return network in the described network architecture adopts under optical transfer network Heavy technology, the user request information sending for receive user end node, described solicited message is sent to Metro core network.

4. network as claimed in claim 1 is it is characterised in that described passback node is used for converging the data that access network sends Information, described data message is sent to Metro core network.

5. network according to claim 1 is it is characterised in that root between described baseband processing unit and remote radio units Descend the corresponding topological structure of scene setting according to this:

At a distance, under sparse deployment scenario, using ring topology；

Under short distance, dense deployment scene, using tree topology.

6. a kind of baseband processing unit resource regulating method of the arbitrary network based on claim 1-5, comprising:

Remote radio units receive user solicited message, described solicited message is sent to baseband processing unit；

Baseband processing unit receives solicited message, and described solicited message is sent to controller；

Controller calculates load needed for described request, is compared with the load threshold value of local baseband processing unit；Confirm When required load is more than local baseband processing unit load threshold value, local baseband processing unit is according to scheduling mechanism by user's request Other baseband processing units being mapped in this Base-Band Processing pond carry out resource process.

7. method as claimed in claim 6 it is characterised in that described step " local baseband processing unit is according to scheduling mechanism Other baseband processing units that user's request is mapped in this Base-Band Processing pond carry out resource process " further include:

Step 1, selects the purpose baseband processing unit of Optimization route, determines its position；

Step 2, controller passes according to the load information that scheduling mechanism executes source baseband processing unit to purpose baseband processing unit Defeated.

8. method as claimed in claim 7 is it is characterised in that described step 1 also includes: based on improved dijkstra algorithm To select the purpose baseband processing unit of optimum route.

9. method as claimed in claim 8 is it is characterised in that described improved dijkstra algorithm includes: with bandwidth usage Rate u (i, j) and bbu traffic usage ω as weight, determine node between shortest path and set up link；Link is remained The link that remaining bandwidth is less than loading demand bandwidth is designated as " busy ", obtains rest network g ', bring follow-up dijkstra algorithm into.

10. method as claimed in claim 9 is it is characterised in that the scheduling mechanism in described step 2 also comprises determining that needs After scheduling, demand signal information transmission to controller, is found optimum purpose bbu for it, is established the link by local bbu；Local bbu Load information is passed to purpose bbu；Delete link, update remaining bandwidth.