Crosshaul Network Function Splitting and Routing Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for deploying Cloud Radio Access Networks (C-RAN) face challenges in optimizing the functional split and routing of base station functions across a network, leading to inefficient use of resources and high operational costs due to rigid fronthaul and backhaul separation, and the inability to dynamically adjust to varying traffic demands.
Innovation Solution
A method that determines network information, entity locations, and function splitting capabilities to jointly optimize routing, assignments of connecting entities to computing entities, and functional splits, enabling flexible and efficient management of transport resources across an integrated backhaul/fronthaul packet-based network, known as crosshaul.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional fronthaul interface relying on fiber optics is used, then bandwidth and synchronization requirements are met, but bandwidth usage is constant and independent of user traffic which is highly inefficient
Solution Approach 1:
The patent applies dynamic bandwidth allocation in packet-based fronthaul networks, where resource allocation adapts to varying user traffic demands rather than maintaining constant dedicated bandwidth. This enables efficient statistical multiplexing while meeting service requirements through dynamic adjustment of allocated resources.
Solution Approach 2:
The patent changes the fronthaul interface from traditional fixed bandwidth allocation to packet-based dynamic bandwidth allocation, transforming the bandwidth parameter from a static physical layer guarantee to a flexible network layer resource that can be dynamically adjusted based on traffic conditions.
2Reliability
If point-to-point optical links are used for fronthaul, then synchronization requirements are met, but only point-to-point connections are allowed which results in low path diversity
Solution Approach 1:
The patent makes the fronthaul network universal by using packet-based networks that can simultaneously support multiple functions including fronthaul transport, backhaul connectivity, and various service types. This enables path diversity and flexible routing while maintaining synchronization through protocol-level mechanisms rather than physical layer constraints.
Solution Approach 2:
The patent introduces packet-based network protocols as intermediaries that enable multiple connecting entities to access centralized computing entities through shared network infrastructure, replacing direct point-to-point optical links with protocol-mediated packet switching that provides path diversity while maintaining service requirements.
3Ease of manufacture
If fixed and static split of base station function between RRH and BBU is used, then implementation simplicity is maintained, but flexibility to adjust to varying traffic demands is lost
Solution Approach 1:
The patent enables dynamic functional splitting where the division of base station functions between RRH and BBU can be adjusted based on traffic demands and network conditions. Centralized computing entities can dynamically allocate processing functions, transforming the static functional split into a dynamic, adaptable configuration.
Solution Approach 2:
The patent segments base station functions into modular, virtualizable units that can be independently allocated and distributed across multiple entities. This functional segmentation enables flexible assignment of specific functions to centralized or distributed locations based on operational requirements rather than fixed architectural constraints.
4Ease of operation
If fronthaul is a separate network segment incompatible to backhaul, then clear functional separation is maintained, but network complexity increases and resource utilization efficiency decreases
Solution Approach 1:
The patent merges fronthaul and backhaul into a unified packet-based network infrastructure, eliminating the need for separate physical networks while maintaining clear functional separation through protocol and logical layer distinctions. This reduces network complexity and enables more efficient resource utilization across the entire transport network.
Solution Approach 2:
The patent resolves the contradiction by moving functional separation from the physical network layer to the protocol and logical layer, allowing fronthaul and backhaul to share physical infrastructure while maintaining clear functional boundaries through software-defined networking and protocol differentiation.
5Productivity
If extensive centralization is implemented to support 5G data rates, then capacity and coverage are increased, but operational costs increase due to high-speed low-latency network requirements
Solution Approach 1:
The patent changes the network parameter from dedicated high-speed optical links to packet-based networks with dynamic resource allocation, enabling centralized processing to support 5G capacity requirements while reducing operational costs through more efficient bandwidth utilization and statistical multiplexing.
Solution Approach 2:
The patent implements dynamic resource allocation in the fronthaul network, where bandwidth and resources are allocated based on actual traffic demands rather than being permanently reserved, enabling high capacity when needed while reducing operational costs during lower traffic periods.
Data Source
AI summary
A method provides operating information for a network having a set of computing entities (CPE) for performing functions on data; a set of connecting entities (CNE) for connecting users and performing functions on data; and a set of forwarding entities (SFE) for forwarding data between the CPE and CNE. CNE are assigned to the CPE. At least part of the functions performed by the CNE are split and offloaded to the assigned CPE for performing the part of functions. The method includes: determining: a) network information, including network topology information and network capability information of the network, b) entity location information, including network locations of the entities, and c) function splitting information, including possible configurable functions which can be split; and, based on the determined information of a)-c), jointly optimizing: routing between the network entities, assignments of the CNE to the CPE, and function splits.