Adaptive Admission Control Using Stochastic Network Measurements
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional admission control methods in wireless networks struggle to efficiently manage network resources and quality of service (QoS) in heterogeneous and aperiodic traffic scenarios, often leading to suboptimal user experiences due to repeated denial of service or inadequate QoS satisfaction.
Innovation Solution
Employing a deep learning-based admission control policy that considers stochastic measurements, such as RSRP and SINR, and a configurable priority between QoS satisfaction and RRC rejection ratio to dynamically decide on user equipment admission, balancing network load and QoS satisfaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional admission control methods are used, then network resources can be managed, but quality of service and user experience deteriorate due to repeated denial of service or inadequate QoS satisfaction
Solution Approach 1:
The patent implements dynamic admission control by continuously monitoring network conditions (throughput, latency, packet loss) and adjusting admission decisions in real-time. The system dynamically updates acceptance thresholds based on current network state, transitioning from static to adaptive control to resolve the contradiction between maintaining QoS and optimizing resource utilization.
Solution Approach 2:
The system incorporates feedback mechanisms where performance metrics from admitted users are continuously collected and fed back to the admission control algorithm. This feedback loop enables the system to learn from actual user experiences and adjust future admission decisions, improving both QoS satisfaction and network efficiency through iterative optimization.
2Productivity
If admission control thresholds are lowered to admit more users, then network productivity increases, but quality of service deteriorates due to increased congestion and denial of service
Solution Approach 1:
The system dynamically adjusts admission thresholds based on real-time network conditions rather than using fixed thresholds. When network conditions are good, thresholds allow higher admission to maximize productivity. When congestion is detected, thresholds automatically tighten to protect QoS, resolving the contradiction between maximizing user admission and maintaining service quality.
Solution Approach 2:
The patent changes the parameter of admission thresholds from static values to dynamic values that adapt to network conditions. By making thresholds a function of current network state (throughput, latency, packet loss), the system can optimize the balance between admitting more users and maintaining QoS satisfaction under varying load conditions.
3Reliability
If admission control is strict to ensure QoS, then quality of service improves, but network efficiency deteriorates due to excessive denial of service
Solution Approach 1:
The system implements dynamic threshold adjustment that adapts to current network load and conditions. During low-congestion periods, the system adopts more permissive thresholds to maximize resource utilization and admit more users. During high-congestion periods, thresholds automatically become stricter to protect QoS, thereby resolving the contradiction between QoS satisfaction and network efficiency.
Solution Approach 2:
The feedback mechanism continuously monitors QoS metrics and network utilization, using this information to adjust admission thresholds. This enables the system to distinguish between temporary and persistent QoS degradation, making informed decisions about when to be strict and when to be permissive, thus optimizing both QoS satisfaction and network resource utilization.
Data Source
AI summary
Facilitating admission control in advanced communication networks is provided. A method includes, based on receipt of a request from a user equipment for admission into a communication network, analyzing, by a system comprising at least one processor, information indicative of stochastic measurements of the communication network, resulting in an adaptive threshold. The method also includes, based on the adaptive threshold being determined to satisfy a defined admission control criterion, facilitating, by the system, admission of the user equipment into the communication network. Further, the method includes, based on the adaptive threshold being determined to fail to satisfy the defined admission control criterion, denying, by the system, the admission of the user equipment into the communication network.