Mobile Network Load Prediction for Proactive Small Cell Sleep Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current energy saving solutions for cellular networks are reactive, leading to suboptimal switching of cell states due to dynamic traffic conditions, incur latency in cell wake-up times, impractical cell discovery, and conflict-prone designs, failing to meet 5G requirements for low latency and efficient energy consumption.
Innovation Solution
AURORA, a proactive mobile network optimization framework, uses semi-Markov based spatio-temporal mobility prediction to anticipate user behavior, optimizing small cell sleep cycles and resource allocation to minimize energy consumption while ensuring QoS, leveraging CIOs for load balancing and incorporating heuristics to solve the non-convex optimization problem.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If small cells are deployed in ultra-dense networks to achieve capacity gain, then network capacity increases, but overall network energy consumption increases due to load-independent power consumption
Solution Approach 1:
The system performs mobility prediction in advance to forecast future user locations and traffic patterns. Based on these predictions, it proactively configures small cell states (active/sleep) before traffic actually occurs, allowing the network to prepare optimal energy-saving configurations ahead of time without reacting to already-happening traffic conditions
Solution Approach 2:
The system dynamically adjusts small cell states (active or sleep mode) based on predicted traffic conditions. Instead of static always-on deployment, small cells transition between states according to forecasted user presence and data demand, making the network configuration adaptive to anticipated conditions rather than fixed
2Use of energy by stationary object
If reactive energy saving solutions are used to switch cell states, then energy consumption is reduced, but latency occurs in cell wake-up times and user experience degrades
Solution Approach 1:
The system predicts user mobility and traffic patterns in advance, then proactively activates small cells before users actually arrive or traffic demands occur. This preliminary activation eliminates wake-up latency because cells are already in active state when needed, rather than being switched on reactively after traffic conditions are detected
Solution Approach 2:
The system takes preliminary action to prevent the harmful effect of latency. By predicting when small cells will be needed and activating them in advance, the system counteracts the potential delay that would occur with reactive switching, ensuring seamless user experience without perceptible interruption
3Use of energy by stationary object
If small cells are switched to sleep mode to save energy, then energy consumption decreases, but network coverage and quality of service may be compromised
Solution Approach 1:
The system uses mobility prediction to forecast future traffic conditions and proactively configures small cell states accordingly. By predicting when cells will be needed, the system ensures QoS requirements are met while maximizing energy savings during periods when cells can be safely placed in sleep mode without impacting service quality
Solution Approach 2:
The system changes the operational parameters of small cells (power levels, transmission settings) based on predicted conditions. Instead of binary on/off switching alone, the system adjusts multiple parameters to optimize the balance between energy consumption and QoS, allowing cells to operate at reduced power levels when appropriate while maintaining service quality
Data Source
AI summary
An apparatus comprises: a memory; and a processor coupled to the memory and configured to: build a prediction model that predicts next cells of UEs in a future time step of a mobile network; map the next cells to future user locations; determine future loads of BSs in the mobile network based on the future user locations; determine an optimization of the mobile network using the future loads; and implement the optimization by instructing the BSs to adjust a parameter in the future time step.