Ad Hoc Radio Network Scheduling via Probabilistic Interference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ad hoc wireless communication networks face interference challenges due to limited spectral resources, where existing methods like TDMA reduce throughput and newer methods like TIM require accurate identification of strong and weak interferers, which is difficult with fluctuating channel conditions.
Innovation Solution
A method that uses probabilistic criteria to determine strong interference based on SINR probability density, estimating statistical parameters of propagation channels to optimize scheduling and reduce the number of transmission cycles for interference graph learning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If TDMA is implemented between adjacent clusters to manage interference, then interference is reduced, but throughput is halved
Solution Approach 1:
The patent applies dynamic scheduling where transmission opportunities are allocated based on real-time channel conditions and interference levels. Instead of static TDMA time slots, the system dynamically determines which clusters can transmit simultaneously by evaluating probabilistic interference metrics, allowing flexible adaptation to changing network conditions while managing interference effectively.
Solution Approach 2:
The system changes the parameter used for interference management from deterministic time-based allocation (TDMA) to probabilistic SINR-based allocation. By using the parameter P(SINR ≥ Γ₀) to characterize interference strength, the system can make more informed decisions about simultaneous transmissions, potentially allowing more clusters to transmit at once while maintaining acceptable interference levels.
2Difficulty of detecting and measuring
If instantaneous SINR-based approach is used to identify strong interferers, then interference graph learning is enabled, but the approach is sensitive to channel time fluctuations
Solution Approach 1:
The patent performs preliminary characterization of interference by computing the probability P(SINR ≥ Γ₀) in advance, before making transmission decisions. This probabilistic metric captures the statistical behavior of interference over time, allowing the system to identify strong interferers based on their typical impact rather than instantaneous values, thereby reducing sensitivity to channel fluctuations.
Solution Approach 2:
Instead of relying on expensive and unreliable instantaneous SINR measurements that require precise timing and are sensitive to fading, the system uses a more robust probabilistic metric that can be estimated from multiple observations. This 'cheaper' metric in terms of measurement reliability provides stable interference characterization without requiring complex real-time measurements.
3Productivity
If the number of channels is kept low compared to the number of clusters, then spectral efficiency is improved, but nearby clusters may receive the same channels causing harmful interference
Solution Approach 1:
The system uses feedback from probabilistic interference measurements to dynamically adjust channel allocation decisions. By computing P(SINR ≥ Γ₀) for each potential transmission and using this information to determine whether to grant transmission opportunities, the system can safely reuse channels in nearby clusters when the probabilistic interference metric indicates low risk, thereby improving spectral efficiency while managing interference.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The invention relates to a method for scheduling transmissions in an ad hoc radio communication network of n nodes according to cycles such that (n) = {α1,α2,α3...} where cycle t includes the action αt corresponding to: "nodes j1t, ..., jmt of the network transmit simultaneously in cycle t", represented by the expression αt = (j1t, ..., jmt), said method comprising the following steps, in a first phase: - we determine α1=1,…,⌈n2⌉ and α2=⌈n2⌉+1,…,n; - we iterate the previous step on the subnetworks 1,…,⌈n2⌉ and ⌈n2⌉+1,…,n, we determine the action α3 as the combination of the transmissions of the first cycles determined for the two subnetworks and the action α4 as the combination of the transmissions of the second cycles determined for the two subnets; and - as long as the size of the subnets is strictly greater than a threshold Pmin, these steps are iterated; and when the size, n', of the considered subnets ≤ Pmin, the actions α1, ..., αT, of (n) having then been determined, a second phase is implemented.