Proactive VoIP Congestion Control via Selective Packet Dropping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing VoIP congestion control methods, such as the 'leaky-bucket' scheme, indiscriminately degrade the quality of service for all VoIP calls when the bandwidth is exceeded, leading to unsatisfactory communication quality.
Innovation Solution
Implementing a Proactive Congestion Control System that monitors VoIP traffic and selectively drops packets from new calls or replaces terminated calls to maintain consistent quality of service by calculating the bandwidth required for new VoIP calls and comparing it to the available bandwidth at each router port.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If leaky-bucket packet policer is used to control VoIP traffic, then bandwidth compliance is improved, but VoIP call quality is worsened due to indiscriminate packet dropping
Solution Approach 1:
The patent segments the VoIP traffic control by identifying individual call flows using five-tuple identifiers (source IP, destination IP, source port, destination port, protocol). Instead of treating all traffic uniformly, the system divides traffic into discrete call-specific streams, allowing selective packet dropping per call rather than indiscriminate dropping across all calls.
Solution Approach 2:
The patent applies different quality treatment to different call flows based on their individual bandwidth consumption. Each call receives localized quality control - calls exceeding bandwidth limits have their packets selectively dropped, while calls within limits maintain full quality. This localizes the impact of congestion control to only the necessary calls.
2Quantity of substance
If pre-assigned bandwidth limit is enforced, then network resource allocation is improved, but service flexibility is worsened when new calls arrive
Solution Approach 1:
The patent implements dynamic bandwidth allocation where the available bandwidth for new calls is calculated in real-time based on current active calls and their consumption. When a new call arrives, the system dynamically determines if bandwidth is available by comparing total current consumption plus the new call's requirements against the pre-assigned limit. This allows flexible admission decisions rather than rigid rejection of all exceeding calls.
Solution Approach 2:
The system continuously monitors VoIP traffic volume and call active states, providing feedback to the congestion control mechanism. This feedback enables real-time adjustment of bandwidth allocation - when calls terminate, bandwidth becomes available for new calls; when calls are added, the system recalculates available capacity. This closed-loop feedback maintains both bandwidth compliance and admission flexibility.
Data Source
AI summary
VoIP traffic bandwidth through a router port is controlled so that it does not exceed a predetermined bandwidth to ensure a desired quality of service for all VoIP calls routed through the port. Information of a packet received by the port is extracted from a packet header. The information includes source and destination addresses and port numbers as well as the bandwidth used by the packet. If the packet is from an existing call, the packet is routed and the time the packet was received is updated. If the packet is from a new call, it is determined whether the port has sufficient bandwidth to route the new call. If the port has sufficient bandwidth, the packet is routed, bandwidth is allotted for subsequent packets from the new call, and the port information is updated to accept the subsequent packets. If the port does not have sufficient bandwidth, the packet is dropped.