Session Manager for Real-Time Resource Re-allocation in Prepaid Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current communication systems face inefficiencies in managing concurrent sessions, particularly in prepaid networks, leading to network performance overhead, inaccurate accounting, and denial of service due to insufficient credit allocation, especially when handling scalar services like voice calls and concurrent services like background downloading.
Innovation Solution
A real-time service supply monitoring and control system that uses a session manager to dynamically allocate and re-allocate resources across concurrent sessions based on historical data, resource requirements, and charging policies, ensuring that all sessions end at approximately the same time, and interfaces with service supply entities using a reserve-capture scheme to optimize resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple reservation requests are made for scalar services to improve credit usage efficiency, then service flexibility is improved, but network server CPU requirements and response times increase
Solution Approach 1:
The system performs preliminary rating and estimation of resource requirements before actual service consumption. By predicting how much credit a scalar service will consume based on historical data and service parameters, the system can pre-allocate resources without requiring continuous reservation requests during service execution, thereby reducing server CPU load while maintaining service flexibility.
Solution Approach 2:
The system implements feedback mechanisms where actual service consumption is monitored and compared with predicted values. This feedback loop allows the system to adjust future resource allocations and improve prediction accuracy over time, enabling more efficient credit usage without increasing the number of reservation requests to the network server.
2Reliability
If all available credit is allocated to a single service, then service priority is maintained, but concurrent services are denied despite available credit
Solution Approach 1:
The system dynamically adjusts credit allocation across concurrent services based on real-time monitoring of service consumption rates and remaining credit balances. Rather than static allocation, the system continuously re-evaluates and redistributes available credit among active services, allowing multiple concurrent services to share the credit pool while maintaining priority for services that consume resources more rapidly.
Solution Approach 2:
The system changes the parameter of credit allocation from fixed or service-specific to dynamic and service-rate-dependent. By monitoring actual consumption rates and adjusting allocation based on real-time service behavior, the system enables concurrent services to access available credit proportionally to their actual usage patterns, preventing denial of service while maintaining priority for high-consumption services.
3Ease of manufacture
If static credit reservations are used, then accounting simplicity is maintained, but network performance overhead increases due to multiple small reservation requests
Solution Approach 1:
The system performs preliminary rating and estimation of resource requirements before actual service consumption. By predicting how much credit a scalar service will consume based on historical data and service parameters, the system can pre-allocate resources without requiring continuous reservation requests during service execution, thereby reducing server CPU load while maintaining service flexibility.
Solution Approach 2:
The system maintains continuous monitoring and estimation of service resource requirements throughout the service duration. Rather than discrete reservation requests, the system continuously tracks consumption patterns and adjusts allocations smoothly, eliminating the start-stop nature of traditional reservation requests and reducing network overhead while maintaining accurate accounting.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A real time monitoring and control system (101) interfaces with service supply entities such as SMSC, MSC, GGSN, SGSN, and IP probe entities. A data manager (104) maintains account resource data associating each of a plurality of accounts with one or more services, in which at least one account is associated with a plurality of services. Each account may have multiple pots, each associated with a type of resource such as bytes, packets, or time. A session manager (102) monitors concurrent service usage in real time and generates messages or control instructions for account modification and for account resource re-allocation across services in real time according to service usage. The session manager (102) dynamically re-distributes available account resources to sessions as the sessions occur, including resource re-allocation during a plurality of sessions including scalar and event-based sessions.