Network Server Sparse Feedback for Wireless Buffer Overflow Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current transport-layer protocols, such as TCP and UDP, perform poorly in wireless telecommunications due to their inability to rapidly adapt to changes in wireless channel capacity and delay, leading to underutilization of link capacity, buffer overflows, and reduced interactivity between servers and clients.
Innovation Solution
A method involving a network server and mobile user terminal that generates and uses sparse feedback signals, such as a repetitive heartbeat signal, to adjust characteristics like Forward Error Correction overhead, ensuring efficient data transmission by maintaining optimal buffer occupancy and quality of service.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If loss-based congestion control (e.g., TCP Cubic) is used to maximize link throughput, then throughput is improved, but buffer overflows occur and data losses increase due to misinterpreting channel fluctuations as congestion
Solution Approach 1:
The patent implements a feedback mechanism where the receiver sends explicit congestion feedback signals to the sender based on actual buffer status and channel conditions. This feedback loop allows the sender to adjust transmission rate dynamically, distinguishing between true congestion and temporary channel fluctuations, thereby preventing buffer overflows while maintaining high throughput.
Solution Approach 2:
The system performs preliminary actions by pre-configuring buffer management parameters and congestion control thresholds before data transmission begins. The receiver prepares buffer status reports and congestion indicators in advance, enabling the sender to proactively adjust transmission rates before buffer overflows occur, rather than reacting after data loss has happened.
2Loss of time
If delay-based congestion control mechanisms (e.g., TCP Vegas) are used to improve latency performance, then latency is reduced, but throughput is substantially reduced
Solution Approach 1:
The patent employs dynamic congestion control that adapts transmission parameters in real-time based on current network conditions. The system dynamically adjusts between loss-based and delay-based control strategies, switching mechanisms and parameters flexibly to optimize both latency and throughput according to instantaneous channel state, buffer status, and traffic patterns.
Solution Approach 2:
The system changes key parameters such as congestion window size, transmission interval, and buffer thresholds dynamically based on measured performance metrics. By adjusting these parameters in response to feedback about actual network conditions, the system achieves low latency without the severe throughput penalties of traditional delay-based approaches.
3Reliability
If feedback from receiving client is used for congestion control at sending server, then congestion control is achieved, but additional delays occur reducing interactivity
Solution Approach 1:
The patent implements periodic feedback mechanisms where the receiver sends congestion status updates at regular intervals rather than continuously. This periodic action reduces the frequency of feedback transmissions, minimizing the cumulative delay impact on interactivity while still providing sufficient information for effective congestion control. The period is optimized to balance control accuracy with latency requirements.
Data Source
Figure 1
Figure 2~4
AI summary
A method is provided of controlling at a network server a characteristic of data to be sent to a cellular telecommunications base station for transmission to a mobile user terminal by: generating in the mobile user terminal a feedback signal representing sparse feedback, sending the feedback signal representing sparse feedback to the base station, forwarding the signal to the network server, and in the network server determining the sparse feedback from the feedback signal, and adjusting a characteristic of data to be sent dependent on the sparse feedback.