Shared Channel Access Prediction for Collision-Aware Wireless Nodes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In wireless networks, simultaneous packet transmission by multiple nodes leads to communication collisions, reducing channel throughput and increasing node communication delay, especially in distributed scenarios where central node scheduling is absent.
Innovation Solution
A node employs a target neural network to predict channel access success by inputting network status information, including time segments and interfering node data, to determine whether to send packets over a shared channel.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If nodes randomly access the shared channel or use traditional prediction methods, then the channel access decision is simple, but the probability of successfully accessing the channel is low and channel throughput is reduced
Solution Approach 1:
The system performs preliminary spectrum sensing and network status detection before channel access decisions are made. Nodes collect network status information including busy/idle states, time segments, and interfering node data in advance, then input this pre-collected data into the neural network for prediction, enabling informed channel access decisions rather than random access
Solution Approach 2:
A neural network model serves as an intermediary between raw network status information and channel access decisions. The neural network processes complex multi-dimensional input data (network status information from multiple nodes and time points) and transforms it into predictive results that guide channel access, acting as an intelligent mediator that improves decision accuracy without requiring nodes to manually analyze complex patterns
2Productivity
If nodes simultaneously send packets on the shared channel, then channel utilization is high, but communication collisions occur and throughput is reduced
Solution Approach 1:
The system implements feedback mechanisms where nodes detect and share network status information including busy/idle states of the shared channel, time segments since last successful transmissions, and counts of interfering nodes. This feedback loop allows nodes to continuously update their understanding of channel conditions and adjust their access decisions accordingly, preventing simultaneous transmissions that would cause collisions
Solution Approach 2:
Nodes perform preliminary predictions using the neural network model before attempting channel access. By inputting current network status information into the trained model, nodes obtain predictive results that indicate the likelihood of successful transmission, allowing them to defer access when conditions suggest high collision probability and thus avoiding simultaneous packet transmissions
Data Source
Figure 1(A)~1(B)
Figure 2
Figure 3~4
AI summary
Embodiments of this application disclose a channel access method and apparatus. In the method, a first node obtains network status information at each moment in a target period. The network status information includes a busy/idle state of a shared channel detected by the first node, a first time segment, a second time segment, and a number of interfering nodes; and the interfering node is a node that is detected by a second node and that sends a packet through the shared channel other than the first node. The first node inputs the network status information at each moment into a target neural network, to obtain a first prediction value and a second prediction value. The first node sends a to-be-sent packet to the second node through the shared channel when the first prediction value is greater than the second prediction value. According to this technical solution, a node can decide, based on a current network status, whether to access a shared channel. This method can improve a success rate of accessing the channel by the node, and increase a throughput of the channel.