Burst data flow mapping load capacity optimization method in sensing cloud environment

Abstract

The invention discloses a burst data flow mapping load capacity optimization method in a sensing cloud environment, which is characterized by comprising the following steps: (1) dividing all sensors into sensor clusters according to signal intensity; (2) carrying out time slot allocation according to the principle that the larger the received signal strength is, the less the time slots allocatedin one polling period are; (3) in a specific time slot, enabling the sensors to establish connections with the edge nodes as many as possible; when a burst data stream request which is large enough occurs, estimating a channel resource allocation interval, optimizing the updated channel resource allocation strategy by adopting a Q-learning algorithm, and updating channel resource allocation of thecorresponding sensor cluster at the next moment; and mapping the idle channel resources to the burst data stream. According to the method, an optimization strategy is learned by using a Q-learning algorithm, and load resources are dynamically configured in real time, so that data packet mapping conflicts are effectively reduced, and the mapping capacity of data streams is improved.

Description

technical field [0001] The invention belongs to the technical field of sensor cloud communication, and more specifically relates to a load capacity optimization method for burst data flow mapping in a sensor cloud environment. Background technique [0002] As an integrated technology of wireless sensor network and cloud computing, sensor cloud can significantly improve the utilization efficiency of network resources. In order to save the resources of wireless sensor network and improve the service quality of data distribution, wireless sensor network is mapped to a virtual sensor service network in the cloud platform through the aggregation of services by edge nodes. The sensor cloud edge node (SCE) carrying the mapping transmission task needs to provide reliable wireless access technology to ensure that a large number of physical sensor data streams in various application scenarios can be mapped to the virtual sensor service network according to time slots and maximize the ...

AI Technical Summary

Problems solved by technology

[0006] (1) The proposed method considers the routing strategy, cannot deal with the problem of burst data flow mapping, and cannot dynamically configure the network in real time and increase the network capacity

[0007] (2) With the increase of historical network state information data observed by SCE nodes, none of the above methods use multi-Q-learning game participants to learn collaboratively and configure loads to improve the mapping capacity

[0008] (3) The traditional mapping capacity optimization method makes the mapping load estimation inaccurate and poor in real-time due to interference and time delay.

Images