Drone Provisioning System for Network Load Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mobile networks face challenges in managing sudden spikes in demand during unusual situations, leading to network overload, and existing systems struggle with unexpected events due to a lack of redundancy and comfort issues with ubiquitous surveillance.
Innovation Solution
A system comprising a communication network and a drone provisioning system that uses predictive modeling, deep learning for anomaly detection, and optimal deployment of surveillance and communication drones to balance network load and provide on-demand capacity while minimizing invasive surveillance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If continual video surveillance is deployed to identify situations where unusual demand may be needed, then network load prediction accuracy is improved, but public comfort and privacy are compromised due to ubiquitous surveillance
Solution Approach 1:
The system performs preliminary video surveillance analysis to detect crowd anomalies and predict future network demand spikes before they occur. By identifying unusual crowd behavior patterns in advance, the system can proactively deploy drones to provide additional network capacity, avoiding the need for continual surveillance while maintaining accurate load prediction.
Solution Approach 2:
The invention introduces drones as intermediary devices that bridge the gap between surveillance needs and network capacity provision. Instead of using ubiquitous surveillance for both monitoring and capacity planning, the system uses minimal surveillance to detect anomalies, then deploys drones as intermediaries to provide both additional network capacity and targeted surveillance only where and when needed.
2Reliability
If drones are deployed to provide on-demand network capacity, then network service quality is improved, but system complexity and deployment overhead increase
Solution Approach 1:
The system merges multiple functions into the drone platform: network capacity provision, targeted surveillance, and real-time data collection. By combining these functions in a single mobile unit, the system reduces overall complexity compared to having separate fixed infrastructure for each function, while improving service quality through flexible on-demand deployment.
Solution Approach 2:
The drone deployment system is designed to be dynamic and adaptive, automatically adjusting drone positions, capacities, and surveillance focus based on real-time network conditions and detected anomalies. This dynamic approach simplifies management compared to static systems, as the drones self-organize to meet changing demands without requiring complex manual coordination.
3Productivity
If highly efficient systems without redundancy are used, then productivity and resource utilization are improved, but the system struggles to cope with unexpected events
Solution Approach 1:
The drones are designed as universal, multi-functional units that can provide network capacity, perform surveillance, and adapt to various unexpected situations. This multi-functionality allows the system to maintain high resource utilization while being resilient to unexpected events, as the same drones can be rapidly reconfigured for different tasks without requiring dedicated redundant infrastructure for each scenario.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system comprising a communication network and a drone provisioning system, the communication network comprising a network modelling sub-system and a network behavior comparison algorithm, the drone provisioning system comprising a drone platoon sub-system comprising a set of drones, wherein the set of drones comprises at least a surveillance drone and a communication drone, a surveillance drone routing algorithm, an anomalous behavior detection and prediction algorithm and a communication drone placement algorithm, wherein: the network modelling sub-system is configured to create a predictive model of subscriber behavior in terms of network traffic; the network behavior comparison algorithm is configured to compare the predicted model of subscriber behavior, with real network traffic and output a result of the comparison to the drone provisioning system; the surveillance drone routing algorithm is configured to send one or more surveillance drone routing requests to the drone platoon sub-system, based on the result of the comparison; the drone platoon sub-system is configured to deploy one or more surveillance drones to an anomaly location specified by the one or more surveillance drone routing requests, wherein the deployed drone or drones are configured to: perform surveillance at the anomaly location, collect surveillance information, assess the reason for the anomaly, provide temporary additional service capabilities, and send the surveillance information and reason for the anomaly to the anomalous behavior detection and prediction algorithm; the anomalous behavior detection and prediction algorithm is configured to: detect anomalous behavior based on the surveillance information and the reason for the anomaly, predict future capacity requirements and future subscriber behavior, based on the above information, and send the predicted future capacity requirements to the communication drone placement algorithm; the communication drone placement algorithm is configured to: calculate an optimal placement of one or more communication drones to meet current and predicted network capacity requirements, based on available resources and the predicted future network capacity requirements, send one or more communication drone routing requests, for communication drone routing and placement for provision of network capacity, to the drone platoon sub-system; the drone platoon sub-system is configured to optimally deploy communication drones to an anomaly location specified by the one or more communication drone routing requests.