Early Congestion Detection in Ad-Hoc Wireless Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for detecting congestion in Ad-Hoc wireless networks are inaccurate due to reliance on queue utilization, packet drop rate, link failure rate, and channel loading measurements, which can lead to packet drops and degraded throughput, and require significant overhead to bypass congested links.
Innovation Solution
Measuring congestion based on channel utilization for TDMA and CSMA messages, using non-uniform quantization to provide early warnings, and distributing quantized utilization levels through Link State Advertisements (LSA) to avoid congested links without additional overhead, utilizing ACK and HELLO messages for increased reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional congestion measurement techniques (queue utilization, packet drop rate, link failure rate, channel loading) are used, then congestion detection is performed, but measurement accuracy deteriorates leading to false congestion indications
Solution Approach 1:
The patent changes the measurement parameter from indirect metrics (queue utilization, packet drop rate) to direct channel utilization measurement. By measuring the actual time the channel is busy versus idle, the system obtains accurate congestion information without the false indications caused by queue size variations or transient packet drops.
Solution Approach 2:
The congestion measurement is performed autonomously by each node based on its own channel observations. Nodes independently measure channel utilization by monitoring their own transmission and reception activities, eliminating the need for external measurement mechanisms that could introduce errors.
2Adaptability or versatility
If secondary routes or multiple routes are determined to bypass congested areas, then route flexibility improves, but control packet overhead increases significantly
Solution Approach 1:
The patent incorporates congestion information (quantized utilization levels) into Link State Advertisements (LSAs) in advance, before routing decisions are made. This allows all nodes to have up-to-date congestion information available for route calculation, enabling adaptive routing without requiring additional control packets for congestion reporting.
Solution Approach 2:
The patent merges congestion information transmission with existing routing protocol traffic (LSAs). By piggybacking quantized utilization levels on regular LSA updates, the system conveys congestion information without generating separate control packets, thus avoiding additional overhead.
3Loss of time
If congestion information is transmitted frequently to ensure timely updates, then information freshness improves, but network bandwidth consumption increases
Solution Approach 1:
The patent uses periodic LSA updates that already occur in the network for congestion information dissemination. Rather than introducing continuous or event-driven congestion updates, the system leverages the existing periodic routing protocol updates, achieving a balance between information freshness and bandwidth efficiency.
Solution Approach 2:
The patent extracts only the essential congestion information (quantized utilization levels) and transmits it within existing LSA structures. By taking out only the necessary congestion data rather than transmitting complete channel state information, the system minimizes bandwidth consumption while maintaining useful information freshness.
4Measurement precision
If constant channel sampling is performed for congestion measurement, then measurement coverage improves, but processor power consumption increases
Solution Approach 1:
The congestion measurement leverages existing channel access protocol operations (TDMA/CSMA) that nodes already perform. By measuring channel utilization during normal protocol operation rather than through separate sampling activities, the system achieves accurate measurement without additional processor power consumption.
Solution Approach 2:
The patent continuously measures channel utilization as part of normal communication operations. Since nodes must continuously monitor the channel for TDMA/CSMA protocol functioning, the measurement is performed as a byproduct of essential communication activities, eliminating the need for separate power-consuming sampling operations.
Data Source
AI summary
The present invention embodiments bypass congested links in a multi-hop Ad-Hoc wireless network. Initially, congestion is measured at each network node based on channel utilization for both transmission and reception of TDMA and CSMA messages. The measured utilization is quantized to conserve transmission bandwidth. Non-uniform quantization is applied to enable the measured utilization to be quantized to the highest value within the quantization range prior to occurrence of congestion, thereby effectively providing early notification of the congestion. The quantized utilization is distributed with the original use costs of a communication link to the remaining network nodes by a Link State Advertisement (LSA) flood and supplemental ACK and/or HELLO packets. After a network node receives the quantized utilization, the link cost is updated and used to select a routing path that minimizes the total costs from the source node to the destination node for a multi-hop network.