Sending rate adjusting method of poultry farming monitoring wireless sensor network based on environmental perception learning strategy

Abstract

The invention discloses a sending rate adjusting method of a poultry farming monitoring wireless sensor network based on an environmental perception learning strategy. The method comprises the following steps: establishing a model for an environment in which a wireless sensor network is located, and establishing a prediction model between an environment variable (such as temperature and humidity)and the growth state (weight growth rate, egg laying rate and death rate) of poultry; sensing the environment data by each sensor node, reasoning the poultry farming environment state by using a fuzzysystem, determining the mode of the node according to different environment states, and determining the transmission rate limit of the node; the optimal state of the network is further searched, andenabling the sensor node to use a Q learning algorithm to carry out adaptive adjustment on the own sending rate, wherein the sensor node counts the packet loss rate, the data effective rate, the dataintegrity and the energy consumption level of the sensor node as the input of a return function of Q learning, and selects the rate adjustment action of the current state. According to the embodimentof the invention, the network congestion is reduced, the overall service life of the network is optimized, and the method has a wide application value.

Description

technical field [0001] The invention belongs to the technical field of wireless sensor networks, in particular to a method for adjusting the transmission rate of wireless sensor networks based on an environment perception learning strategy. Background technique [0002] Poultry is an important source of meat and protein for humans. Animals are prone to heat stress in high temperature and high humidity environments, and animals experiencing heat stress often reduce heat production by restricting feed intake, which can adversely affect performance and health. Heat stress is usually caused by changes in indoor temperature and relative humidity. Therefore, when raising poultry in farming enterprises, wireless sensor networks are usually used to monitor the growth environment of poultry. Unlike traditional wired networks, wireless sensor network nodes are susceptible to interference, and due to their small size, the stored power is extremely limited. Therefore, the data transmis...

AI Technical Summary

Problems solved by technology

But in a star network, this algorithm is not applicable

Images