Skiing system high-speed transmission method based on 5G communication

Abstract

The invention belongs to the technical field of crossing of wireless communication and intelligent sports equipment, and particularly relates to a skiing system high-speed transmission method based on 5G communication, which comprises the following steps: terminal data acquisition and edge preprocessing: acquiring motion states, physiological indexes, environmental parameters and visual data through a multi-type sensor terminal; and the edge calculation module carries out denoising, feature extraction and key event screening on the original data and then compresses and transmits the data. According to the dynamic self-adaptive code rate control technology, the coding rate of each type of data is dynamically adjusted by combining the data type priority and the real-time channel quality (SINR / BLER). For example, when the athlete enters the forest (SINRlt; when the bandwidth is limited (10dB), the reliability of the Level1 data (heart rate / position) is guaranteed through a redundant packet (retransmission 20%), the Level2 data (video) is reduced from 4K to 1080P to reduce the code rate requirement, and the transmission of the Level3 data (environmental parameters) is suspended, so that the key data is preferentially guaranteed under the limited bandwidth.

Description

TECHNICAL FIELD

[0001] The application belongs to the cross technical field of wireless communication and intelligent sports equipment, and particularly relates to a high-speed transmission method for a skiing system based on 5G communication. BACKGROUND

[0002] As a high-speed and high-risk outdoor sport, skiing has distinctive scene specificity in data transmission requirements: strong dynamics: the speed of athletes can reach 100-130 km / h or 20-60 km / h, and they need to frequently deal with actions such as jumping, turning, and sudden braking, resulting in high data collection frequency and complex data types; high environmental complexity: the race track is mostly distributed in mountainous, forest, and snow-covered areas, and traditional Wi-Fi signals are easily blocked by trees, 4G base station coverage is limited, and signal blind spots often occur in tunnels, under jumps, and other areas; strict real-time requirements: coaches need to adjust training strategies in real time based on the real-time speed, turning angle, and heart rate of athletes; the collision warning system must issue an alarm within 20 ms when the distance between athletes is less than 3 meters and the relative speed is greater than 10 m / s; and outstanding multi-terminal concurrency: single field training / race needs to transmit data of dozens of athletes simultaneously and support multiple device access such as coach terminals, referee systems, and audience live streaming platforms.

[0003] Traditional skiing data transmission solutions mainly rely on Wi-Fi (2.4 GHz / 5 GHz) and 4G LTE networks, but both have significant drawbacks: Wi-Fi technology: the theoretical bandwidth can reach 300 Mbps-1 Gbps in open snow tracks, but it is actually limited by multipath interference and the number of concurrent devices, and has poor mobility; 4G LTE network: air interface delay is about 30-50 ms, and uplink bandwidth is generally lower than 50 Mbps, and base station coverage is sparse in mountainous terrain; and non-adaptability of general 5G solutions: existing 5G applications are not optimized for the special needs of skiing scenarios, which is manifested in the following aspects: dynamic code rate control strategies are not designed in combination with IMU data characteristics, there is a lack of a special mobility management mechanism for high-speed movement, and multi-source data is not scheduled with hierarchical QoS, so it is necessary to design a high-speed transmission method for a skiing system based on 5G communication to solve the above problems. SUMMARY

[0004] The purpose of the present application is to provide a high-speed transmission method for a skiing system based on 5G communication to solve the problems of high delay, low bandwidth, weak coverage, and poor mobility non-adaptability raised in the background.

[0005] A high-speed transmission method for a skiing system based on 5G communication, comprising the following steps:

[0006] (1) Terminal data acquisition and edge preprocessing: Collect motion state, physiological indicators, environmental parameters and visual data through multiple types of sensor terminals, and compress transmission after denoising, feature extraction and key event screening of raw data by edge computing module;

[0007] (2) 5G network dynamic adaptive transmission: Based on real-time channel state, dynamically adjust the encoding rate of each type of data (distinguish Level1 key data / Level2 medium priority data / Level3 low priority data), and through multi-priority QoS scheduling (5QI identifier) to ensure low delay and high reliability of high priority data;

[0008] (3) Network slice resource isolation: Allocate independent virtual network slice for skiing system through 5G core network, customize bandwidth, delay and reliability parameters for different scenarios (competitive training / mass safety / event live broadcast);

[0009] (4) Edge node cooperative processing: Use MEC server to analyze uploaded data in real time, extract key features and filter non-key data, and reduce backhaul traffic.

[0010] Preferably, the terminal data acquisition includes: IMU inertial measurement unit (acceleration / angular velocity), GPS / Beidou positioning module, barometer, heart rate band, snowboard pressure sensor and camera, the edge preprocessing includes Kalman filter denoising, event trigger mechanism (such as detecting jump action when vertical acceleration >8g) and ROI video encoding.

[0011] Preferably, the specific strategy of dynamic adaptive rate control is:

[0012] (1) Level1 data (position / heart rate / acceleration) always guarantees minimum bandwidth 5Mbps, improves to 10Mbps when channel is good, and reduces packet loss rate through redundant packets when channel is poor;

[0013] (2) Level2 data (video stream / pose data) code rate is 100Mbps when channel is good, 50Mbps when channel is medium, and 20Mbps when channel is poor;

[0014] (3) Level3 data (environmental parameters / logs) is only transmitted when channel is good, and code rate is ≤5Mbps.

[0015] Preferably, in the multi-priority QoS scheduling: Level1 data corresponds to URLLC (air interface delay <1ms, packet loss rate <0.1%), Level2 data corresponds to eMBB (air interface delay <20ms, packet loss rate <1%), and Level3 data corresponds to BestEffort (best effort).

[0016] Preferably, the network slices include: a competition training dedicated slice (symmetrical uplink and downlink bandwidth 100-500 Mbps, delay <10 ms), a mass safety dedicated slice (uplink 20 Mbps / downlink 50 Mbps, delay <20 ms), and a live broadcast dedicated slice (uplink 1 Gbps / downlink 200 Mbps, delay <50 ms).

[0017] Preferably, the edge node cooperative processing includes: real-time analysis of video / sensor data to extract key features (such as action posture angle, sliding trajectory), and pushing to the coach terminal through a low-delay channel (<10 ms), while screening non-key data to reduce backhaul traffic.

[0018] Preferably, the high-speed transmission system of the 5G communication-based skiing system is: (1) a terminal subsystem: deployed at the sensor terminal of the athlete and the edge computing module, used for data acquisition and preprocessing;

[0019] (2) a network subsystem: composed of a Sub-6GHz macro base station, a millimeter wave micro base station, and a core network, supporting dynamic network slicing and multi-priority QoS scheduling;

[0020] (3) a cloud terminal subsystem: including a real-time processing platform, a storage database, and a visual terminal;

[0021] (4) an edge subsystem: an MEC server, used for local data screening and real-time analysis.

[0022] Compared with the prior art, the application has the following beneficial effects:

[0023] (1) Dynamic adaptive code rate control technology: the traditional scheme adopts a fixed code rate (such as 4K video constant 50 Mbps), while the application dynamically adjusts the coding code rate of each type of data in combination with the data type priority and real-time channel quality (SINR / BLER). For example, when the athlete enters the forest (SINR <10 dB), Level 1 data (heart rate / position) is transmitted through redundant packets (retransmission 20%) to ensure reliability, Level 2 data (video) is reduced from 4K to 1080P to reduce the code rate requirement, and Level 3 data (environmental parameters) is temporarily suspended, so as to prioritize the key data under limited bandwidth.

[0024] (2) Multi-priority QoS hierarchical transmission mechanism: Different data is assigned with differentiated scheduling priority (URLLC / eMBB / Best Effort) through 5G 5QI identifier, ensuring the absolute transmission reliability of high-priority data (such as collision warning). For example, Level 1 data (heart rate > 180 bpm) is transmitted through the URLLC slice (air interface delay < 1 ms), and even if the network is congested, it will not be squeezed by Level 2 / 3 data; while Level 2 data (video stream) enjoys high bandwidth (100 Mbps) when the channel is good, and automatically degrades (20 Mbps) but guarantees basic availability when the channel is poor.

[0025] (3) 5G network slice resource isolation: Custom independent virtual network slices for different scenarios such as competitive training, public safety, and live streaming, and achieve resource isolation through parameter customization (bandwidth / delay / reliability). For example, the competitive training slice exclusively occupies 100-500 Mbps bandwidth (to avoid being occupied by Level 3 data of tourists), ensuring that coaches can receive real-time 4K video and sensor data of athletes; the live streaming slice supports 1 Gbps uplink bandwidth (to meet 8K video streaming), while ensuring the concurrent access of hundreds of viewers.

[0026] (4) Edge intelligence preprocessing cooperation: Deploy lightweight AI models (such as TinyML) on terminals and MEC nodes to achieve intelligent processing through "end-edge" cooperation. For example, the terminal uploads only the data during the jump key period through the event trigger mechanism (reducing 70% of redundant transmission), and the MEC server further encodes the ROI of the video stream (increasing the code rate of the athlete's body by 30% and reducing the background by 50%), ultimately reducing the data volume returned to the cloud by 30%-50%, while the real-time performance of key information is not affected (analysis result delay < 20 ms). BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 The figure is a schematic diagram of the overall architecture of the system of the present application;

[0028] Figure 2 The figure is a decision flowchart of dynamic adaptive bit rate control of the present application;

[0029] Figure 3 The figure is a 5QI mapping table of multi-priority QoS scheduling of the present application;

[0030] Figure 4 The figure is a parameter configuration comparison table of network slices of the present application;

[0031] Figure 5 The figure is a flowchart of edge node cooperative processing of the present application. DETAILED DESCRIPTION

[0032] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative work are within the protection scope of the present application.

[0033] Please refer to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative work are within the protection scope of the present application. Figures 1-5 The present application provides several embodiments:

[0034] A high-speed transmission method of a skiing system based on 5G communication, comprising the following steps:

[0035] (1) Terminal data acquisition and edge preprocessing: collecting motion state, physiological index, environmental parameter and visual data through multi-type sensor terminal, and compressing transmission after denoising, feature extraction and key event screening of the original data by an edge computing module;

[0036] (2) 5G network dynamic adaptive transmission: dynamically adjusting the coding rate of each type of data (distinguishing Level 1 key data / Level 2 priority data / Level 3 low priority data) based on real-time channel state, and guaranteeing low delay and high reliability of high priority data through multi-priority QoS scheduling (5QI identifier);

[0037] (3) Network slice resource isolation: allocating an independent virtual network slice for the skiing system through the 5G core network, and customizing the bandwidth, delay and reliability parameters of different scenes (competitive training / mass safety / event live broadcast);

[0038] (4) Edge node cooperative processing: using the MEC server to perform real-time analysis on the uploaded data, extract key features and screen non-key data, and reduce the backhaul traffic.

[0039] Further, the terminal data collection includes: IMU inertial measurement unit (acceleration / angular velocity), GPS / Beidou positioning module, barometer, heart rate band, snowboard pressure sensor and camera, the edge preprocessing includes Kalman filter denoising, event trigger mechanism (such as detecting jump action when vertical acceleration> 8g) and ROI video encoding; Multi-source data collection: Motion state data: IMU (6-axis / 16 channels, sampling rate 100Hz-1kHz) collects acceleration (±200g), angular velocity (±2000° / s), gravity direction; GPS / Beidou (positioning accuracy ±0.1m) obtains longitude / latitude / altitude; Barometer (accuracy ±0.1hPa) measures snow road altitude change (estimated slope). Physiological indicators: heart rate band (optical PPG sensor, sampling rate 5-10Hz) monitors heart rate / oxygen; Body temperature sensor (contact type, accuracy ±0.2℃) collects body surface temperature. Environmental parameters: snow road slope sensor (inclinometer, accuracy ±0.5°), air temperature / humidity sensor (accuracy ±1℃ / ±5%RH) are deployed at fixed points. Visual data: helmet camera (4K@60fps or 8K@30fps) records the first perspective sliding picture; Panoramic camera (1080P@30fps) monitors the overall snow road; Edge preprocessing: data filtering: Kalman filter (eliminate high-frequency noise) + low-pass filter (separate gravity component, extract dynamic acceleration) is used for IMU raw data to improve signal quality (SNR is improved by 10-15dB). Feature extraction: key parameters (such as instantaneous speed = Δ position / Δ time, turning radius = angular velocity × speed / lateral acceleration) are directly calculated at the terminal, reducing redundant data transmission (such as uploading only characteristic values instead of original sampling sequence). Event trigger mechanism: through threshold judgment (such as vertical acceleration> 8g triggers "jump event", heart rate> 180bpm triggers "abnormal warning"), only the data of key event period (such as 3-5 seconds of data from take-off to landing) is retained, and the redundant information of non-key period (such as IMU data sampling rate from 1kHz to 100Hz during uniform sliding) is compressed. Video encoding optimization: ROI (region of interest) encoding is used (assign higher code rate to athlete's body, reduce code rate for background), and resolution is dynamically adjusted based on current channel quality (such as reducing from 4K to 1080P and frame rate from 60fps to 30fps when channel is poor).

[0040] Further, the specific strategy of dynamic adaptive code rate control is: according to the data type priority (divided into three levels) and the real-time channel state (through the signal-to-noise ratio SINR and block error rate BLER of CSI feedback of 5GNR).

[0041] (1) Level 1 data (position / heart rate / acceleration) always guarantees a minimum bandwidth of 5Mbps, which is increased to 10Mbps when the channel is good, and is reduced by redundant packets to reduce the packet loss rate when the channel is poor;

[0042] (2) Level2 data (video stream / pose data) channel optimal time code rate 100 Mbps, medium time 50 Mbps, poor time 20 Mbps;

[0043] (3) Level3 data (environmental parameters / logs) is transmitted only in the channel optimal time, code rate ≤5 Mbps.

[0044] Further, in the multi-priority QoS scheduling: Level1 data corresponds to URLLC (air interface delay <1ms, packet loss rate <0.1%), mapped to 5QI=20 (URLLC, Ultra-Reliable Low-Latency Communication), using fixed GBR (guaranteed bandwidth) and fast HARQ retransmission (retransmission interval <2ms), Level2 data corresponds to eMBB (air interface delay <20ms, packet loss rate <1%), mapped to 5QI=40 (eMBB, Enhanced Mobile Broadband), using dynamic GBR (bandwidth allocation on demand) and MCS (modulation and coding scheme) adaptive (selecting 256QAM / 64QAM / QPSK according to SINR), Level3 data corresponds to BestEffort (BestEffort), mapped to 5QI=9 (BestEffort), no strict delay / reliability guarantee, transmitted only when the network is idle.

[0045] Further, the network slice includes: a competitive training dedicated slice with symmetric uplink and downlink bandwidth of 100-500Mbps (guaranteeing 4K video + full sensor data), air interface delay <10ms, reliability >99.9% (packet loss rate <0.1%), priority higher than other slices, a public safety dedicated slice with uplink bandwidth of 20Mbps (for collision warning data), downlink bandwidth of 50Mbps (for safety prompt push), air interface delay <20ms, reliability >99% (packet loss rate <1%), and a live broadcast dedicated slice with uplink bandwidth of 1Gbps (supporting 8K@60fps video stream, code rate about 80-100Mbps per road), downlink bandwidth of 200Mbps (supporting hundreds of concurrent viewers), air interface delay <50ms, supporting high concurrency connection (single base station ≥100 streams).

[0046] Further, the edge node cooperative processing includes: real-time analysis of video / sensor data to extract key features (such as action posture angle, sliding trajectory), and push to the coach terminal through a low-delay channel (<10ms), while screening non-key data to reduce backhaul traffic; real-time analysis: detecting athlete posture (such as jumping, turning) on uploaded video stream (through YOLOv5 model), predicting sliding trajectory on sensor data (through LSTM model), and extracting key features (such as "current turning angle deviation 5°", "insufficient take-off height"); data filtering: only uploading data segments determined as "key" (such as IMU data of jump start and end period, position data of collision risk period) to reduce traffic back to the cloud (by 30%-50%); local early warning: pushing real-time analysis results (such as "there are tourists 3 meters ahead, suggest slowing down") to the coach / athlete terminal through a low-delay channel (<10ms).

[0047] Further, the high-speed transmission system of the 5G communication-based skiing system is: (1) terminal subsystem: deployed in the athlete's multiple types of sensor terminals (IMU inertial measurement unit, GPS / Beidou positioning module, barometer, heart rate band, snowboard pressure sensor) and cameras (first perspective / panoramic high-definition camera), as well as an embedded edge computing module (low-power SoC chip, such as Qualcomm QCM5430, integrated with a 5G modem and an AI inference unit);

[0048] (2) network subsystem: composed of Sub-6GHz macro base stations (coverage radius 1-3km, ensuring wide-area connectivity), millimeter wave micro base stations (coverage radius 50-300m, providing ultra-high speed), and 5G core network (supporting network slicing and QoS scheduling) deployed by the ski resort;

[0049] (3) cloud subsystem: including real-time processing platform (stream computing engine + AI analysis model), storage database (time series database storing high-frequency sensor data, object storage saving high-definition video), and visualization terminal (coach AR / VR interface, audience live streaming platform);

[0050] (4) edge subsystem: MEC (multi-access edge computing) server deployed at the edge of the ski resort (close to the base station, delay <10ms), used for local data filtering, feature extraction, and real-time early warning.

[0051] Embodiment 1: Competitive skiing training scenario (ski jumping)

[0052] Scenario description: the athlete jumps from the ski jump (speed about 90km / h), flies in the air for 2-3 seconds, and lands, requiring real-time monitoring of take-off angle (accuracy ±0.5°), in-flight posture (angular velocity vector), landing impact force (peak >500N), and heart rate change (to avoid excessive tension leading to heart rate >180bpm).

[0053] Implementation process: Terminal configuration: Athlete wears smart helmet (integrates 6-axis IMU + front camera, resolution 4K@60fps), waist sensor vest (16-channel IMU, sampling rate 1kHz), heart rate band and snowboard pressure sensor (distributed, sampling rate 500Hz); edge computing module (high pass QCM5430) calculates take-off angle (through acceleration / angular velocity fusion of IMU, error <0.5°), in-air posture (angular velocity vector analysis) in real time, and performs Kalman filtering on the original data to remove noise.

[0054] Transmission process: Before take-off (uniform sliding stage): IMU data (Level 1) is uploaded at a code rate of 10Mbps (delay <5ms) through URLLC slice; heart rate data (Level 1) is transmitted at a code rate of 5Mbps. Take-off moment: the camera (4K@60fps) starts recording, and the video stream is transmitted at a code rate of 80Mbps through eMBB slice (channel optimization, SINR>20dB); at the same time, key data such as take-off angle and initial speed are uploaded through Level 1 channel in priority.

[0055] Air stage (high-speed movement, about 90km / h): The system ensures stable connection through dual connection (macro base station control plane + micro base station user plane), with switching delay <30ms; IMU data is continuously uploaded (Level 1), and video stream is dynamically adjusted according to channel quality (if entering cloud cover area, SINR drops to 10dB, code rate drops from 80Mbps to 50Mbps, resolution remains 4K).

[0056] Landing stage: Snowboard pressure sensor collects impact force distribution (Level 2), extracts "maximum impact force value" through edge node and uploads in priority (Level 1); coach terminal receives real-time analysis results (such as "insufficient air time in the air" "landing angle deviation 3°") through AR glasses, with delay <10ms.

[0057] Example 2: Mass skiing safety scenario (daily operation of ski resort)

[0058] Scenario description: There are hundreds of tourists sliding in the ski resort at the same time (speed 20-60km / h), and it is necessary to monitor the collision risk between tourists (distance <3m and relative speed >10m / s) and emergency help signal (such as fall detection).

[0059] Implementation process: Terminal configuration: Tourists wear lightweight bracelet (GPS + accelerometer, sampling rate 1Hz; fall detection sensor, judges through acceleration mutation threshold), only uploads position data (accuracy ±3m) and motion direction (calculated through acceleration vector)

[0060] Transmission procedure: MEC server calculates relative distance and speed of all visitors every 500ms (based on GPS data), if collision risk is detected (distance < 3m and relative speed > 10m / s), an alert is sent to visitor's wristband (vibration + LED flashing) through Level 1 priority channel (URLLC) with <20ms latency. Environment sensor (slope meter, air temperature / humidity sensor) data is uploaded through Best Effort slice low priority channel (transmitted only when channel is optimal), for ski resort operator to optimize slope planning. If visitor actively triggers fall detection (acceleration mutation > 5g), terminal immediately uploads location data (5Mbps) through Level 1 channel, and rescue personnel terminal (instructor / patroller) receives alert within 5 seconds.

[0061] It will be apparent to those skilled in the art that the application is not limited to the details of the foregoing exemplary embodiments and that the present application can be implemented in other specific forms without departing from the spirit or essential characteristics thereof. The foregoing embodiments are therefore to be considered in all respects as illustrative only and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. No feature of the application is to be construed as a limitation thereon unless it is expressly stated to be such.

Claims

1. A high-speed transmission method for a ski system based on 5G communication, characterized in that, Comprise the following steps: (1) Terminal data acquisition and edge preprocessing: Collect motion state, physiological indicators, environmental parameters and visual data through multiple types of sensor terminals, and compress transmission after denoising, feature extraction and key event screening of the original data by the edge computing module; (2) 5G network dynamic adaptive transmission: Based on real-time channel state, dynamically adjust the encoding rate of each type of data (distinguish Level1 key data / Level2 medium priority data / Level3 low priority data), and through multi-priority QoS scheduling (5QI identifier) to guarantee the low delay and high reliability of high priority data; (3) Network slice resource isolation: Allocate independent virtual network slices for the skiing system through the 5G core network, customize the bandwidth, delay and reliability parameters of different scenarios (competitive training / mass safety / event live broadcast); (4) Edge node cooperative processing: Use MEC server to analyze uploaded data in real time, extract key features and filter non-key data to reduce backhaul traffic.

2. The high-speed transmission method of a ski system based on 5G communication according to claim 1, characterized in that: The terminal data acquisition includes: IMU inertial measurement unit (acceleration / angular velocity), GPS / Beidou positioning module, barometer, heart rate band, snowboard pressure sensor and camera, the edge preprocessing includes Kalman filter denoising, event trigger mechanism (such as detecting jumping action when vertical acceleration>8g) and ROI video encoding.

3. The high-speed transmission method of a ski system based on 5G communication according to claim 2, characterized in that: The specific strategy of dynamic adaptive code rate control is: (1) Level1 data (position / heart rate / acceleration) always guarantees the minimum bandwidth of 5Mbps, improves to 10Mbps when the channel is good, and reduces the packet loss rate through redundant packets when the channel is poor; (2) Level2 data (video stream / pose data) code rate is 100Mbps when the channel is good, 50Mbps when the channel is medium, and 20Mbps when the channel is poor; (3) Level3 data (environmental parameters / logs) is only transmitted when the channel is good, and the code rate is ≤5Mbps.

4. The high-speed transmission method of a ski system based on 5G communication according to claim 1, characterized in that: In the multi-priority QoS scheduling: Level1 data corresponds to URLLC (air interface delay<1ms, packet loss rate<0.1%), Level2 data corresponds to eMBB (air interface delay<20ms, packet loss rate<1%), and Level3 data corresponds to BestEffort (best effort).

5. The high-speed transmission method of a ski system based on 5G communication according to claim 1, characterized in that: The network slice includes: competitive training dedicated slice (symmetrical uplink and downlink bandwidth 100-500Mbps, delay<10ms), mass safety dedicated slice (uplink 20Mbps / downlink 50Mbps, delay<20ms) and event live broadcast dedicated slice (uplink 1Gbps / downlink 200Mbps, delay<50ms).

6. The high-speed transmission method of a ski system based on 5G communication according to claim 1, characterized in that: The edge node cooperative processing includes: real-time analysis of video / sensor data to extract key features (such as action pose angle, sliding trajectory), and push to the coach terminal through a low-delay channel (<10ms), while filtering non-key data to reduce backhaul traffic.

7. The high-speed transmission method of a ski system based on 5G communication according to claim 1, characterized in that: The high-speed transmission system of a skiing system based on 5G communication is: (1) Terminal subsystem: Deployed in the sensor terminal and edge computing module of the athlete, used for data acquisition and preprocessing; (2) Network subsystem: composed of Sub-6GHz macro base station, millimeter wave micro base station and core network, supporting dynamic network slicing and multi-priority QoS scheduling; (3) Cloud subsystem: including real-time processing platform, storage database and visual terminal; (4) Edge subsystem: MEC server, used for local data filtering and real-time analysis.

Images