Dynamic Compression Training With Offline Dictionaries for Mobile Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional dynamic compression techniques in wireless communication are expensive and inefficient, especially in constrained devices, as they require on-line adaptation and ignore potential optimizations using off-line resources.
Innovation Solution
A method for training dynamic compressors and creating dictionaries for both constrained and unconstrained devices, allowing for pre, during, or post-communication optimization, using a mobile state table and network state table to improve compression efficacy and reduce device demands, with an automated system for selecting appropriate dictionaries based on network traffic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If on-line adaptation is used for dynamic compression in constrained devices, then compression can be adjusted during communication, but device cost and complexity increase
Solution Approach 1:
The patent applies preliminary action by performing compression training offline before actual communication occurs. The trainer analyzes communication patterns and creates optimized compression dictionaries in advance, so that during runtime, the compressor only needs to apply pre-computed compression rules without performing complex adaptation, thereby reducing device complexity while maintaining adaptability
Solution Approach 2:
The patent introduces an intermediary trainer component that separates the complex adaptation logic from the constrained communication device. The trainer acts as a mediator that performs the heavy computational work of analyzing communication patterns and generating compression dictionaries, then delivers these to the compressor for simple application during communication, thus protecting the device from complexity
2Productivity
If on-line adaptation is used for dynamic compression, then compression can optimize during communication, but communication cost increases
Solution Approach 1:
The patent performs compression training and dictionary creation offline before communication occurs, transferring the computational burden to a separate training phase. During actual communication, the system only applies pre-computed compression rules, significantly reducing energy consumption and communication cost while maintaining compression efficiency
Solution Approach 2:
The patent implements periodic training where the trainer analyzes communication patterns at specific intervals and updates compression dictionaries accordingly. This periodic approach allows the system to maintain high compression efficiency through adaptation while avoiding continuous adaptation overhead, thereby reducing overall communication cost
3Device complexity
If dynamic compression training is performed offline, then device resources are saved, but training time increases
Solution Approach 1:
The patent segments the compression process into two distinct phases: an offline training phase where patterns are analyzed and dictionaries are created, and an online compression phase where pre-computed rules are applied. This segmentation allows resource-intensive training to occur separately from communication operations, saving device resources during runtime while accepting training time as a one-time upfront cost
4Productivity
If conventional dynamic compression is used in constrained devices, then communication can be compressed, but compression efficiency decreases
Solution Approach 1:
The patent applies preliminary action by pre-computing compression dictionaries offline that are optimized for specific communication patterns. When deployed in constrained devices, these pre-computed dictionaries enable high compression efficiency without requiring the device to perform complex real-time adaptation, thus achieving efficient compression within device constraints
Solution Approach 2:
The patent changes the operational parameters of the compression system by shifting from real-time adaptation to pre-computed compression rules. The trainer analyzes communication patterns and determines optimal compression parameters in advance, then these parameters are applied in constrained devices without requiring complex runtime adjustment mechanisms, thereby achieving high efficiency within device constraints
Data Source
AI summary
An apparatus for communicating between an access endpoint and a mobile device endpoint comprising a network service center attached to the access endpoint. A memory stores a prototype corpus representing historical communication information. The network service center includes a processor to train the prototype corpus and create a plurality of records including a protocol trigger, training data and a user dictionary. The access endpoint sends the training data to endpoints to produce training data specific to classes of endpoints. The access endpoint and the mobile device endpoint communicate with one another, and the network service center collects communication data related to the communications interaction and stores the communication data in the memory. The processor generates feedback data based at least in part of the communication data. Advantages of the invention include the ability for network operators to optimize communication techniques based on network traffic.