A Bluetooth communication prediction method and system for smart wearable devices

Abstract

The invention discloses an intelligent wearable device Bluetooth communication prediction method and system, and the method comprises a training process, a prediction process, and a communication process. The training process comprises the steps: enabling a server to sample an RSSI value of an intelligent wearable device, generating a training sample, and carrying out the training of a neural network model, so as to obtain a trained neural network model; the prediction process comprises the steps that the server side samples an RSSI value of the intelligent wearable device to generate an analysis sample, the analysis sample is analyzed through the neural network model, and the peak moment of the next RSSI value is predicted; Wherein the communication process comprises the step that the intelligent wearable device carries out Bluetooth communication with the server side according to the peak moment. The method has the advantages that the energy consumption of the intelligent wearable device in the Bluetooth communication prediction process can be effectively reduced, the prediction accuracy is good, and the Bluetooth communication quality can be effectively improved.

Description

technical field [0001] The invention relates to the technical field of Bluetooth communication, in particular to a Bluetooth communication prediction method and system for smart wearable devices. Background technique [0002] In the existing methods for reducing the energy consumption of Bluetooth communication of wearable devices, how to reduce the path loss of the signal transmitted by the radio frequency module in the periodic motion as much as possible, and predict the signal minimum path loss point of the device in the periodic motion is one of the important technologies. one. The technology of predicting the signal minimum path loss point in periodic motion is an opportunistic prediction technology with high real-time requirements. Minimization, this technology combines communication and prediction methods, and makes full use of the redundant computing resources of smartphones and edge computing nodes in the edge computing scenario. In recent years, with the vigorous...

AI Technical Summary

Problems solved by technology

This method is relatively simple to predict the moment when the RSSI value reaches the peak, and can achieve better results in the short term, but its disadvantage is that the movement of the human body is not strictly periodic. The result will gradually accumulate into a large offset error; when the offset error is large, perform FFT transformation again to find the new fundamental frequency f 0 , frequent calculations consume a lot of energy on wearable devices

Its disadvantages are: wearable devices are required to have accelerometers, and the cost is too high; in addition, the clustering idea is not enough to predict the loose periodic motion of the human body. When the motion cycle drifts, the prediction error will gradually increase. It is solved by calibration, but frequent calibration operations will also greatly consume the power of wearable devices

Images