CN-122002056-A - Moving camera real-time live broadcast system based on wireless networking and data packet forwarding method

CN122002056ACN 122002056 ACN122002056 ACN 122002056ACN-122002056-A

Abstract

The invention relates to the technical field of network communication and streaming media transmission, in particular to a moving camera real-time live broadcast system based on wireless networking and a data packet forwarding method, comprising a moving state sensing step of acquiring inertial measurement data and calculating a moving state change rate representing the physical moving state of a camera; the method comprises a channel trend prediction step of inputting the change rate into a preset mapping model to predict the channel instability trend in a future time period, a transmission strategy decision step of comparing the trend with a threshold value and dynamically determining the fragment granularity parameter and the forward error correction coding density of a video data stream, a data packet encapsulation forwarding step of encapsulating the video stream based on the parameter to generate a target data packet and mapping the target data packet to a transmission queue for forwarding, and a network fluctuation is predicted by using physical motion priori information.

Inventors

CHEN KUI
Xing purun
MA LI
CHEN HAICHAO
ZHANG BO
DENG HAIBO

Assignees

深圳市永迦电子科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260309

Claims (9)

1. The method for forwarding the live broadcast data packet of the moving camera based on the wireless networking is characterized by comprising the following steps of: S1, under a wireless networking framework, acquiring inertial measurement data acquired by an inertial measurement unit of a motion camera, and calculating a motion state change rate representing a physical motion state of the camera based on the inertial measurement data; S2, inputting the motion state change rate into a preset motion-channel state mapping model to predict the channel instability trend of the current transmission channel in a future preset time period; s3, comparing the channel instability trend with a preset instability threshold value, and dynamically determining a fragmentation granularity parameter and a forward error correction coding density of a video data stream to be transmitted according to a comparison result; S4, based on the fragment granularity parameter and the forward error correction coding density, carrying out encapsulation processing on the video data stream to generate a target data packet, and mapping the target data packet to a sending queue for forwarding.
2. The method for forwarding live broadcast data packets of a motion camera based on wireless networking according to claim 1, wherein the step S1 comprises the following substeps: S11, acquiring triaxial acceleration data and triaxial angular velocity data of the motion camera in real time; s12, carrying out fusion calculation on the triaxial acceleration data and the triaxial angular velocity data to extract a composite acceleration vector and an angular velocity vector of the camera; S13, calculating the motion state change rate based on the time differentiation of the synthesized acceleration vector and the angular velocity vector.
3. The method for forwarding live broadcast data packets of a motion camera based on wireless networking according to claim 1, wherein the step S2 comprises the following substeps: S21, acquiring round trip delay data of a current network link, and taking the round trip delay data and the motion state change rate as input feature vectors; S22, constructing the motion-channel state mapping model by using a long-term memory network or a Kalman filtering algorithm, and performing time sequence analysis on the input feature vector; s23, outputting a numerical value representing the variation gradient of the channel instability as the channel instability trend.
4. The method for forwarding live broadcast data packets of a motion camera based on wireless networking according to claim 1, wherein the step S3 comprises the following substeps: S31, judging whether the channel instability trend is larger than the preset instability threshold value; S32, if the channel instability trend is judged to be greater than the instability threshold, entering a micronization mode, adjusting the fragment granularity parameter to be smaller than the first byte length of a preset standard maximum transmission unit, and increasing the forward error correction coding density; S33, if the channel instability trend is judged to be smaller than or equal to the instability threshold, entering an aggregation mode, adjusting the fragment granularity parameter to be equal to the second byte length of a preset standard maximum transmission unit, and reducing the forward error correction coding density; Wherein the first byte length is less than the second byte length.
5. The method for forwarding live broadcast data packets of a motion camera based on wireless networking according to claim 4, wherein the target data packet generated in S32 is a micronized data packet, and the construction process of the micronized data packet includes: Encoding the video data stream by adopting a fountain code to generate a group of particle data blocks with redundancy check information; and encapsulating the particle data block into UDP data packets with the payload length conforming to the first byte length so as to support error correction decoding of a receiving end without retransmission request.
6. The method for forwarding live broadcast data packets of a motion camera based on wireless networking according to claim 1, wherein the step S4 comprises the following substeps: S41, analyzing the frame type of the video data stream, and identifying I frame data particles and P frame data particles; S42, distributing forwarding priority higher than that of the P frame data particles for the I frame data particles in the transmission queue; s43, according to the distributed forwarding priority, the target data packet is scheduled to be sent to a receiving end through a network interface controller.
7. The method for forwarding live broadcast data packets of a motion camera based on wireless networking according to claim 1, wherein the method further comprises: S5, monitoring the packet loss space distribution characteristics of the target data packet; s6, reversely correcting the weight parameter of the motion-channel state mapping model based on the correlation between the packet loss space distribution characteristics and the motion state change rate so as to complete closed loop feedback.
8. The motion camera real-time live broadcast system based on wireless networking is characterized by comprising: The motion sensing module is used for acquiring inertial measurement data of the motion camera and calculating the change rate of the motion state; the channel state prediction module is used for predicting the channel instability trend based on the motion state change rate and a preset motion-channel state mapping model; The self-adaptive decision module is used for dynamically adjusting the fragment granularity parameter and the forward error correction coding density of the video data stream according to the comparison result of the channel instability trend and a preset instability threshold; And the atomization forwarding engine is used for packaging the video data stream into atomization data packets or aggregation data packets based on the adjusted parameters and executing forwarding operation.
9. The wireless networking based motion camera live broadcast system of claim 8, wherein the micronized forwarding engine is run in an embedded processor with a network interface controller and is configured with a memory buffer for performing data slicing and reassembly.

Description

Moving camera real-time live broadcast system based on wireless networking and data packet forwarding method Technical Field The invention relates to the technical field of network communication and streaming media transmission, in particular to a moving camera real-time live broadcast system based on wireless networking and a data packet forwarding method. Background In the existing real-time live broadcast system of the moving camera, the system comprises a video acquisition unit and a wireless transmission unit, wherein the video acquisition unit is connected with a video encoder and a network communication module through a data bus to form a mobile terminal capable of carrying out outdoor live broadcast, the terminal is called a plug-flow device, and the plug-flow device is in communication connection with a network side through a wireless networking mode; The wireless transmission unit comprises a network protocol stack, a sending buffer zone and a wireless radio frequency front end, wherein the network protocol stack is connected with the sending buffer zone, the generated video data packet is transmitted to the base station through the wireless radio frequency front end, and the adjustment of the transmission parameters of the video stream is realized by analyzing the acknowledgement character or the packet loss statistical information fed back by the receiving end, so that the push stream equipment can maintain certain transmission quality when the network fluctuates. But the video data packet is greatly lost in a transmission link due to the fact that the video data packet cannot respond to the fast fading of a physical channel in time when a moving camera is in a high-speed movement or violent shaking state due to the fact that the video data packet is regulated by depending on feedback information of a receiving end, so that the problems of picture pause, image artifact and end-to-end delay and rapid increase of a live picture are caused. Disclosure of Invention In order to solve the technical problems, the invention provides a moving camera real-time live broadcast system based on wireless networking and a data packet forwarding method, and specifically, the technical scheme of the invention is as follows: a method for forwarding a live broadcast data packet of a moving camera based on wireless networking comprises the following steps: S1, under a wireless networking framework, acquiring inertial measurement data acquired by an inertial measurement unit of a motion camera, and calculating a motion state change rate representing a physical motion state of the camera based on the inertial measurement data; S2, inputting the motion state change rate into a preset motion-channel state mapping model to predict the channel instability trend of the current transmission channel in a future preset time period; s3, comparing the channel instability trend with a preset instability threshold value, and dynamically determining a fragmentation granularity parameter and a forward error correction coding density of a video data stream to be transmitted according to a comparison result; S4, based on the fragment granularity parameter and the forward error correction coding density, carrying out encapsulation processing on the video data stream to generate a target data packet, and mapping the target data packet to a sending queue for forwarding. Preferably, the step S1 includes the following substeps: S11, acquiring triaxial acceleration data and triaxial angular velocity data of the motion camera in real time; s12, carrying out fusion calculation on the triaxial acceleration data and the triaxial angular velocity data to extract a composite acceleration vector and an angular velocity vector of the camera; S13, calculating the motion state change rate based on the time differentiation of the synthesized acceleration vector and the angular velocity vector. Preferably, the step S2 includes the following substeps: S21, acquiring round trip delay data of a current network link, and taking the round trip delay data and the motion state change rate as input feature vectors; S22, constructing the motion-channel state mapping model by using a long-term memory network or a Kalman filtering algorithm, and performing time sequence analysis on the input feature vector; s23, outputting a numerical value representing the variation gradient of the channel instability as the channel instability trend. Preferably, the step S3 includes the following substeps: S31, judging whether the channel instability trend is larger than the preset instability threshold value; S32, if the channel instability trend is judged to be greater than the instability threshold, entering a micronization mode, adjusting the fragment granularity parameter to be smaller than the first byte length of a preset standard maximum transmission unit, and increasing the forward error correction coding density; S33, if the channel instability trend is judged to be