Wireless RU Selection Using Feedback for Dense MU-OFDMA
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication methods in dense environments, such as MU OFDMA, suffer from high collision rates and low throughput due to random resource unit (RU) selection, which can lead to incompatibility with standard 802.11 devices and require additional infrastructure or centralized allocation unsuitable for dynamic networks.
Innovation Solution
A wireless communication device with a RU selection module that uses carrier sensing data and acknowledgement signals to determine optimal RUs for transmission, employing a data processing unit, feature extraction and scoring, reward computation, and a controller to adaptively select RUs based on past experiences, ensuring compatibility with standard 802.11 devices and improving throughput.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If random RU selection is used in MU OFDMA, then device complexity is reduced, but throughput decreases due to high collision rates
Solution Approach 1:
Each wireless communication device autonomously determines RUs by itself using carrier sensing data and acknowledgement signals from the network, without requiring centralized resource allocation. The device independently extracts features from device state information, generates scores for each RU, and selects RUs based on these scores, enabling self-service resource selection that improves throughput while maintaining distributed operation
Solution Approach 2:
The system uses acknowledgement signals from the network as feedback to update device state information. This feedback mechanism allows devices to learn from transmission outcomes and adjust their RU selection strategy over time, improving throughput through adaptive decision-making based on network conditions and past transmission results
2Productivity
If patterned resource allocation with additional group designation field is used, then throughput is improved, but compatibility with standard 802.11 devices deteriorates
Solution Approach 1:
The system maintains compatibility with standard 802.11 devices by using existing PHY header formats without adding group designation fields. The same carrier sensing data and acknowledgement signals serve multiple purposes: they enable both standard 802.11 operation and the enhanced distributed RU selection mechanism, allowing the system to work universally with both standard and enhanced devices
3Productivity
If centralized resource allocation is used, then throughput is improved, but adaptability to dynamic network conditions deteriorates
Solution Approach 1:
The system dynamically adapts to changing network conditions by continuously updating device state information based on fresh carrier sensing data and acknowledgement signals. Each device independently adjusts its RU selection strategy in real-time based on current network state, enabling dynamic adaptation without centralized control or periodic reconfiguration
4Productivity
If neural network based semi-distributed access is used, then throughput is improved, but device complexity and infrastructure requirements increase
Solution Approach 1:
The system uses simple, lightweight data structures (device state information, feature vectors, and scores) that can be easily stored and processed in device memory, replacing complex neural network models. These lightweight objects are updated iteratively using standard reinforcement learning updates, reducing computational requirements and device complexity while maintaining adaptive performance
Data Source
AI summary
A data processing unit generates device state information based on carrier sensing data and acknowledgement signals received from an access point. A feature extraction and scoring unit extracts features from the device state information and generates scores for each RU representative of possibility of successful transmission based on the extracted features. A reward computation unit computes reward for providing feedback on prediction accuracy of the scores. A controller determines a RU used to transmit data based on the scores. The controller stores experiences including the device state information and the reward in a memory unit. The controller updates parameters of the feature extraction and scoring unit based on the experiences stored in the memory unit.