Adaptive optimization scheduling method for mobile terminal software based on deep reinforcement learning

A mobile terminal and reinforcement learning technology, applied in the field of computing, can solve the problems of increasing antenna transmission power loss, long unloading time, overshooting, etc., to improve user experience, optimize process scheduling and offloading, and reduce computing delay and energy loss Effect

Active Publication Date: 2018-12-14
XIAMEN UNIV
View PDF4 Cites 28 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, mobile devices usually must be connected to the edge computing network through a wireless network. The instability of the wireless channel has a great impact on the effect of the edge computing network. When the commun...

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] The technical solution of the present invention is further described below in conjunction with the examples, but the scope of protection is not limited to the description.

[0026] Embodiments of the present invention include the following steps:

[0027] Step 1: The mobile terminal device is connected to the surrounding edge computing devices through the wireless network.

[0028]Step 2: Construct a deep convolutional neural network with 4 layers. The first layer is a convolutional layer, the number of inputs is 21×21, it contains 20 convolution kernels of 10×10, the step is 1, and the number of outputs is 20×12×12; the second layer is a convolutional layer , the number of inputs is 20×12×12, including 40 convolution kernels of 5×5, the step is 1, and the number of outputs is 40×8×8; the third layer is a fully connected layer, and the number of inputs is 2560, the number of outputs is 1024; the last layer is a fully connected layer, the input size is 1024, and the nu...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention relates to an adaptive optimization scheduling method for mobile terminal software based on deep reinforcement learning, which relates to a computing technology. The method comprises thesteps of: automatically optimizing the number of software processes unloaded on the mobile device, prioritizing the local task processes, and allocating CPU computing resources and memory resources to reduce the energy consumption of the mobile device and reduce the task processing delay. By measuring the scale of the real-time threading tasks of the software on the mobile device, the bandwidth of the dynamic wireless link between the mobile device and the edge device is estimated, and the feedback information such as the time delay and the energy loss of each process is evaluated by using the depth reinforcement learning algorithm, so as to obtain the optimized scheduling scheme of the software. There is no need to predict the wireless channel model from mobile device to edge device andthe CPU computing resource and memory resource occupancy model of mobile device system, which can reduce the processing delay and energy consumption of software tasks on mobile device and improve theuser experience.

Description

technical field [0001] The invention relates to computing technology, in particular to a mobile terminal software adaptive optimization scheduling method based on deep reinforcement learning. Background technique [0002] With the development and innovation of various application software, mobile devices such as smart phones need to handle larger and larger computing tasks, and the computing power requirements for mobile device CPUs are getting higher and higher. Poor CPU performance or unreasonable calculation order of mobile devices will lead to unsmooth operation of application software, such as game freezes, etc., affecting user experience. Coordinating task processes through system-level optimization schemes under fixed and mobile device hardware conditions is of great significance for reducing task processing delays and improving user experience. [0003] In this regard, A.S.Wu et al. (A.S.Wu, H.Yu, S.Jin, et al, "An incremental geneticgorithm approach to multiprocess...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06F9/50G06N3/04
CPCG06F9/5016G06F9/5038G06N3/045Y02D10/00
Inventor 肖亮戴灿煌许冬瑾江东华唐余亮
Owner XIAMEN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap