Predictive Disconnect for Offline Charging System Bottlenecks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Offline Charging Systems (OFCS) in telecommunications networks face resource bottlenecks, leading to incomplete Charging Data Records (CDRs) and significant revenue leakage due to conventional disconnect mechanisms that split charging messages and prevent reconnection, causing inefficiencies and increased processing strain on the billing domain.
Innovation Solution
The system determines resource bottlenecks and transmits an enhanced Disconnect Peer Request (eDPR) message with an expected recovery time to Charging Trigger Functions (CTFs), allowing them to reconnect after the specified time, thereby preventing resource overload and maintaining session continuity without splitting charging messages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If the OFCS implements conventional disconnect mechanisms when experiencing resource bottlenecks, then the processing strain on the billing domain is reduced, but charging messages are split and session continuity is lost, leading to incomplete CDRs and revenue leakage
Solution Approach 1:
The system performs preliminary actions by sending an enhanced disconnect message that includes an expected recovery time to the CTF before the resource bottleneck becomes critical. This allows the CTF to proactively disconnect and pause message transmission, preventing overload while maintaining the ability to reconnect automatically when resources are available, thus avoiding both message splitting and revenue leakage
Solution Approach 2:
The system implements feedback by including the expected recovery time in the disconnect message, which provides information to the CTF about when to reconnect. This feedback mechanism enables the CTF to resume transmission at the appropriate time, ensuring session continuity and complete CDR generation while preventing processing strain through coordinated pause and resume operations
2Reliability
If the OFCS maintains connection with CTFs during resource bottlenecks, then session continuity is maintained, but the processing strain increases and incomplete CDRs are generated
Solution Approach 1:
The system applies dynamics by making the connection state adaptive rather than static. The CTF dynamically transitions between connected and disconnected states based on the expected recovery time received from the OFCS. This dynamic approach allows the system to maintain session continuity through automatic reconnection while reducing processing strain by pausing message transmission during resource bottlenecks
3Reliability
If the CTF continuously transmits charging messages to the OFCS, then complete CDRs can be generated, but the OFCS experiences resource bottlenecks and revenue leakage occurs
Solution Approach 1:
The system applies preliminary anti-action by having the CTF disconnect and pause message transmission in advance when the OFCS indicates resource bottlenecks through the expected recovery time. This preventive measure stops the generation of incomplete CDRs and associated revenue leakage before they can occur, while the automatic reconnection mechanism ensures CDR completeness is restored once resources are available
Data Source
AI summary
System and method for implementing a predictive disconnect in an Offline Charging System of a telecommunications network are provided. In one aspect, an apparatus for implementing a charging function in the OFCS is configured to determine that a resource allocated to the OFCS for processing messages received from one or more Charging Trigger Functions (CTFs) connected to the OFCS is experiencing or about to experience a bottleneck condition. The apparatus is configured to heuristically determine an expected duration of time for the resource to return to a normal condition, and is further configured to initiate a disconnection between the OFCS and one or more of the CTFs connected to the OFCS for at least the determined expected period of time.