Network Request Prioritization via Crowdsource Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
User devices face challenges in prioritizing network requests efficiently, leading to unnecessary long response times and increased power consumption due to unclear priority levels and inefficient power management, especially when multiple applications compete for network resources.
Innovation Solution
A system where user devices send crowdsource data to a server, which generates filtered data based on network response times and total channel throughput, allowing the device to prioritize network requests and adjust power modes accordingly, ensuring that requests with shorter estimated response times are sent first and power consumption is minimized.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If the operating system sends network requests without clear priority levels, then all applications can access network resources, but network response times become unnecessarily long and power consumption increases
Solution Approach 1:
The patent introduces a network request determination component as an intermediary between applications and the network stack. This component receives network requests from multiple applications, determines their priority levels based on application characteristics and network conditions, and queues them accordingly. By inserting this intermediary layer, the system achieves efficient priority-based request handling without requiring complex modifications to individual applications or the core operating system network stack.
Solution Approach 2:
The system implements feedback mechanisms where the network request determination component continuously monitors network response times, channel throughput, and application performance metrics. Based on this feedback, it dynamically adjusts priority assignments and queue management strategies. For example, if certain applications consistently experience long response times, the system can elevate their priority levels, creating a closed-loop control system that optimizes network resource allocation over time.
2Use of energy by moving object
If the processor operates in normal power mode while waiting for network responses, then processing speed is maintained, but power consumption increases and battery runtime decreases
Solution Approach 1:
The patent implements dynamic power mode switching for the processor based on its state and current tasks. When the processor is idle or performing low-priority operations, it transitions to a lower power mode to conserve energy. When high-priority network requests arrive or the processor needs to perform computationally intensive tasks, it transitions back to normal power mode. This dynamic adjustment allows the system to optimize the trade-off between power consumption and processing speed based on real-time conditions.
Solution Approach 2:
The system performs preliminary actions by pre-processing network requests and preparing them for transmission before actual network operations begin. The network request determination component queues and prioritizes requests in advance, so when the processor becomes available, it can immediately process high-priority requests without delay. This preliminary preparation reduces the duration the processor needs to remain in high-power states, thereby reducing overall power consumption.
3Productivity
If applications assume maximum channel throughput is available, then download quality is high, but download times become excessively long when throughput is actually insufficient
Solution Approach 1:
The patent replaces the traditional mechanical approach of assuming maximum theoretical throughput with a data-driven estimation system. Instead of relying on theoretical maximums or simple historical averages, the system uses machine learning models trained on crowdsource data from multiple devices to predict actual achievable throughput. This substitution of the estimation mechanism with an intelligent predictive system allows the application to make more accurate decisions about download quality and timing, avoiding both excessive wait times and unnecessary quality reductions.
4Ease of operation
If the operating system cannot determine priority order among high-priority requests, then all requests are treated equally, but this leads to poor user experience due to long response times
Solution Approach 1:
The patent applies local quality by assigning different priority levels to different network requests based on their specific characteristics and the application's needs. Rather than treating all requests uniformly or using a single global priority scheme, the system evaluates each request individually, considering factors such as application type, request urgency, and current network conditions. This localized approach to priority assignment allows the system to optimize user experience for each specific request while maintaining overall system simplicity.
Data Source
AI summary
A user device may receive network application requests from multiple applications, such as a music application, news application, and social media application. The user device may send to a server for crowdsource data. The crowdsource data may be based on response data received from other user devices. The response data may indicate the location of the user device, type of application seeking to send a network application request, the response time to receive a network response, a network, and so forth. The server may filter the crowdsource data to correspond to the location of the user device, the applications seeking to send a network application request, and the network the user device is currently using. The user device based on estimated response times may determine a prioritization order in which to send the network application requests.