An aerobic exercise data transmission method and device

Abstract

The application discloses an aerobic exercise data transmission method and device, and belongs to the technical field of digital information transmission and protocol stack scheduling. The method is applied to a medium access control layer of a communication device, and comprises the following steps: receiving multiplexed heterogeneous service data streams issued by an application layer, identifying a first time-sensitive service stream and a second tolerant service stream containing specific physiological parameter payloads; converting a time sequence dynamic gradient of the first service stream into a medium access urgency parameter of the MAC layer based on a cross-layer mapping mechanism; when a threshold value is reached or exceeded, triggering flow queue preemption scheduling, dynamically calculating a target silence time length of a network allocation vector according to the gradient absolute value, and broadcasting a control frame, and sending the short payload data frame in the allocated contention-free exclusive window. Through cross-layer protocol parameter mapping and adaptive management of service stream queues, the application solves the uncontrollable problems of transmission congestion and queuing delay caused by heterogeneous data bursts in high-concurrency digital networks.

Description

Technical Field

[0001] This invention relates to the field of sports data transmission technology, and in particular to a wearable sports status indicator, training safety warning and sports data transmission method and device for aerobic exercise training. Background Technology

[0002] In aerobic exercise scenarios such as university physical education classes, endurance running training, group gymnastics classes, and stationary bike training, coaches typically need to monitor athletes' heart rate, muscle oxygen saturation, cadence, pace, and exercise load changes in real time to determine whether athletes have entered a critical state of aerobic endurance. Existing exercise wristbands, heart rate armbands, or physical education monitoring terminals can only collect and display data for a single athlete, lacking tiered indications and early warning mechanisms for synchronized multi-person training scenarios. When multiple athletes simultaneously approach the fatigue threshold, the coaching or training management terminal cannot obtain crucial early warning data in a timely manner, leading to delays in exercise training guidance and safety intervention.

[0003] The two types of data mentioned above are multiplexed and mixed in the same transmission queue, and are uniformly transmitted in a contention-based manner by the carrier sense multiple access / collision avoidance mechanism of the medium access control layer. To meet the low-latency requirements of time-sensitive services, the industry generally adopts the enhanced distributed channel access mechanism defined in the IEEE 802.11e standard, which assigns different access categories to different service flows. However, the static priority adjustment-based scheme has a fundamental flaw in scenarios with dense equipment and bursty service flows: when multiple high-priority service flows exist simultaneously, it can only shorten the queuing time of data packets, but cannot provide deterministic zero-collision transmission, resulting in the maximum transmission delay of high-priority services being completely uncontrollable. This is the core communication technology problem that this invention aims to solve.

[0004] The aforementioned technical issues are particularly prominent in the aerobic exercise scenario of digital physical education classes in universities. In this scenario, hundreds of wearable communication devices share the same wireless local area network, and the devices periodically report two types of service data: one is mechanical state-tolerant service, and the other is vital sign time-sensitive service with extremely high real-time requirements. As confirmed in the *Journal of Jilin Sport University*, muscle oxygen saturation is a leading warning indicator of the human body's aerobic endurance limit, and its sudden change occurs earlier than heart rate and exercise mechanical parameters. At this time, vital sign data is upgraded to extremely high-priority safety warning information. Even if the existing EDCA mechanism marks it as the highest priority, it cannot avoid channel collisions caused by concurrent transmission of multiple nodes. When the number of nodes increases and the service load is high, the probability of multiple nodes competing for the channel at the same time increases significantly, causing high-priority service flows to still collide and retransmit, further amplifying the inherent defects of the existing communication mechanism. Summary of the Invention

[0005] This invention overcomes the shortcomings of the prior art and provides a method and device for transmitting aerobic exercise data.

[0006] To achieve the above objectives, the technical solution adopted by this invention is: an aerobic exercise data transmission method, applied to the protocol stack media access control layer of a wearable communication device, comprising:

[0007] Receive heterogeneous service data streams sent by the application layer of the communication equipment. The heterogeneous service data streams include a first time-sensitive service stream with time-delay sensitive attributes and a second time-tolerant service stream with time-delay tolerant attributes.

[0008] Application layer features are extracted from the first time-sensitive service flow to obtain its payload time change rate parameter, and the time change rate parameter is converted into a media access urgency parameter of the media access control layer based on a preset cross-layer mapping strategy.

[0009] When the media access urgency parameter is lower than the preset channel intervention threshold, the first time-sensitive service flow and the second tolerance service flow are combined to generate a standard data frame, and the standard data frame is used for contention for channel access.

[0010] When the media access urgency parameter reaches or exceeds the channel intervention threshold, the MAC layer channel preemption mechanism is triggered, including: suspending the media access control layer transmission process of the second tolerable service flow, encapsulating the first time-sensitive service flow into a short payload emergency frame, generating a self-clearing transmission management control frame, dynamically calculating the silence duration of the target network allocation vector according to the time change rate parameter, and writing it into the self-clearing transmission management control frame; broadcasting the self-clearing transmission management control frame to force other communication nodes in the same wireless LAN to enter a sleep backoff state, and sending a short payload emergency frame within the contention-free time window corresponding to the silence duration.

[0011] In a preferred embodiment of the present invention, the time rate of change parameter is a first-order descent gradient, which characterizes the rate of change of the application layer payload per unit time.

[0012] In a preferred embodiment of the present invention, the silent duration of the first-order descent gradient dynamic calculation of the target network allocation vector specifically includes:

[0013] Obtain the base transmission duration of the short payload emergency rescue frame at the current physical layer transmission rate;

[0014] Based on a preset cross-layer mapping strategy, a compensation silence duration is superimposed on the basic transmission duration to generate the target silence duration.

[0015] The value of the compensation silence duration is positively correlated with the absolute value of the first-order descent gradient.

[0016] The compensation silence duration and the absolute value of the first-order descent gradient are linearly mapped, with the mapping ratio coefficient ranging from 0.1 to 10 ms / (% / s).

[0017] In a preferred embodiment of the present invention, the broadcast self-clearing transmission management control frame specifically includes:

[0018] After generating the self-clearing transmission management control frame, monitor the wireless physical channel status;

[0019] When the physical channel is detected to be idle for the time specified in the point coordination function inter-frame interval, any pending data frame transmission requests are intercepted, and a self-clearing management control frame is broadcast directly through the radio frequency channel.

[0020] In a preferred embodiment of the present invention, suspending the transmission process for the second tolerable service flow in the initial transmission queue includes:

[0021] Identify the protocol identifier of the second-tolerance traffic flow in the traffic classifier at the MAC layer;

[0022] Pause the delivery of data packets corresponding to the protocol identifier to the physical layer and clear the retransmission counter for that data packet in the current MAC layer cache;

[0023] Subsequent arrivals of the second-tolerance traffic flow data are transferred to the bypass buffer of the local non-volatile memory.

[0024] In a preferred embodiment of the present invention, the first-order descent gradient is calculated as follows:

[0025] Multiple sampling points are collected within the current sliding time window, and the slope value is obtained by linear fitting using the least squares method. The negative of the slope value is then used as the first-order descent gradient.

[0026] The length of the sliding time window is 1~3s, and the sampling frequency is 10~100Hz.

[0027] In a preferred embodiment of the present invention, a service flow recovery step is further included:

[0028] Continuously monitor media access urgency parameters;

[0029] When a first-order descent gradient fall is detected and the medium access urgency parameter is lower than the channel intervention threshold, the generation of the self-clearing transmission management control frame is stopped.

[0030] Resume the regular multiplexing process of the initial transmission queue, and extract the backlogged second-tolerance traffic flows from the bypass buffer in batches, and re-merge them with the current first-time traffic flows for payload retransmission.

[0031] In a preferred embodiment of the present invention, when retrieving the backlogged second tolerance traffic flow from the bypass buffer in batches, a gradually increasing transmission rate is used.

[0032] In a preferred embodiment of the present invention, the gradually increasing transmission rate is increased in a gradient of 10 to 25 packets per second.

[0033] An aerobic exercise data transmission device, comprising

[0034] The stream mapping module is used to receive heterogeneous service data streams sent by the application layer of the communication device and inject them into the initial sending queue. The heterogeneous service data streams include a first time-sensitive service stream and a second tolerance service stream.

[0035] The gradient extraction module is used to extract the time rate of change parameter and convert the first-order descent gradient into the medium access urgency parameter of the current MAC layer.

[0036] The conventional transmission module is used to combine the payloads of the two types of service flows to generate a standard data frame when the medium access urgency parameter is lower than the channel intervention threshold, and to transmit it using a carrier sense multiple access / collision avoidance mechanism.

[0037] The preemption module is used to suspend the transmission process of the second-tolerance service flow and generate a short payload emergency frame when the medium access urgency parameter is greater than or equal to the channel intervention threshold; generate a self-clearing transmission management control frame, dynamically calculate the target silence duration based on the time change rate parameter and write it into its network allocation vector field; broadcast the control frame with a higher priority than ordinary data frames to force other communication nodes to sleep, and send a short payload emergency frame during the sleep window period.

[0038] This invention addresses the shortcomings of the prior art and has the following beneficial effects:

[0039] (1) This invention maps the first-order descent gradient of the application layer across layers to the medium access urgency parameter of the MAC layer, and dynamically triggers the exclusive channel preemption mechanism based on CTS-to-Self frames accordingly. At the same time, it introduces a fast recovery strategy for service flows to solve the technical problems of unavoidable channel collisions and uncontrollable maximum delay in the existing distributed priority channel contention mechanism when multiple high-priority service flows coexist.

[0040] (2) When the urgency parameter of medium access reaches or exceeds the threshold, the present invention triggers the active channel preemption mechanism; the channel access decision of the MAC layer is upgraded from simple physical carrier perception to active channel reservation driven by cross-layer parameters; the first-order descent gradient reflects the urgent change trend of application layer services and is used as the dynamic calculation basis of NAV silent duration to realize adaptive exclusive protection that occupies the channel for longer when the situation is more critical; the PIFS inter-frame interval is used to broadcast CTS-to-Self frames, and the characteristics of PIFS being shorter than DIFS and any random backoff time slot are utilized to ensure that the preemption frame can always obtain the channel priority access right; this method enables wearable communication devices to still provide millisecond-level, zero-jitter deterministic transmission guarantee for critical service flows in dense network environments, meeting the latency requirements of safety early warning data in extreme aerobic exercise scenarios.

[0041] (3) After the preemption mode ends, this invention continuously monitors the medium access urgency parameter. When the first-order descent gradient is detected to have fallen back and the parameter is below the channel intervention threshold, the generation of CTS-to-Self frames is stopped, the normal multiplexing process of the initial transmission queue is restored, and the backlogged second-tolerance service streams are extracted from the bypass buffer in batches and retransmitted with the current first-time-sensitive service streams. This solves the problem of secondary congestion that is easy to cause after the emergency mode exits in the prior art. The bypass buffer completely saves the tolerance service streams that were suspended during the preemption period to avoid data loss. The gradually increasing transmission rate allows the backlogged data to be injected into the channel in a gradual manner, preventing new collisions caused by sudden traffic surges. In the existing emergency transmission scheme, the system directly returns to the normal competition state after the preemption ends. A large amount of backlogged data and real-time data flood into the channel at the same time, causing the system throughput to be unable to recover for a long time. However, this application can provide deterministic transmission guarantee in critical moments and maintain efficient channel utilization in normal conditions.

[0042] (4) This invention transforms the burst characteristics of the application layer into active channel preemption commands of the MAC layer through cross-layer parameter mapping. Compared with the traditional IEEE 802.11e EDCA mechanism, it solves the physical layer channel collision problem of high-priority data streams in high-concurrency scenarios and realizes deterministic zero-jitter transmission of time-sensitive service streams. The recovery mechanism quickly pulls the system back to steady-state competition mode after the crisis is resolved, avoiding long-term backlog of tolerant service streams due to excessive preemption. The organic combination of the two enables this invention to achieve zero collision in crisis and high-efficiency adaptive transmission in normal conditions in dense wireless network environments, and thoroughly solves the deterministic access problem of channels in CSMA / CA networks from the perspective of MAC layer protocol mechanism.

[0043] (5) The non-contention window defined by the dynamically calculated NAV silent duration provides a deterministic emptying time window for the bypass buffer; the dynamic adjustment of the minimum contention window value during the recovery phase avoids the channel contention caused by the backlog of tolerant service flows from adversely affecting the normal transmission of time-sensitive services, thus improving the transmission robustness and protocol stack adaptability of wearable communication devices in complex network environments. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a system flowchart of the present invention;

[0046] Figure 2 This is a system module diagram of the present invention. Detailed Implementation

[0047] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein. Therefore, the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0049] Application Overview:

[0050] The aerobic exercise data transmission mentioned in the title of this invention is a typical application scenario of the technical solution of this invention. The core technical contribution of this invention to the prior art belongs to the field of digital information transmission technology, specifically involving the heterogeneous service flow adaptive scheduling and deterministic channel access control method of the wireless local area network protocol stack media access control layer. The data transmission of existing wearable communication devices in intensive aerobic exercise scenarios has the following fundamental defects: Although the enhanced distributed channel access (hereinafter referred to as: EDCA) mechanism based on static priority adjustment can shorten the queuing time of high-priority services, when multiple devices generate high-priority data at the same time, channel collisions at the physical layer are still unavoidable, and the maximum latency is in an uncontrollable state; when calculating the channel occupancy time or backoff parameters, the protocol stack media access control (hereinafter referred to as: MAC layer) only refers to the bit length of the data frame and the physical transmission rate, severing the association with the application layer service content; the existing protocol stack layers are independent of each other, lacking cross-layer scheduling means that can sense the time-sensitive data changes of the application layer and actively implement exclusive intervention on the physical channel.

[0051] This invention proposes a cross-layer driven adaptive transmission method for aerobic exercise data streams. Its core concept lies in mapping the first-order descent gradient to a medium access urgency parameter at the MAC layer. When this parameter exceeds the channel intervention threshold, a channel active preemption mechanism is triggered, suspending tolerable service streams and generating a short payload emergency frame. Simultaneously, a lightweight channel protection mechanism is constructed to manage control frames and dynamically calculate the silence duration based on the absolute value of the gradient. This frame is broadcast to force other nodes to suspend active channel access. Then, an emergency frame is sent within a contention-free window to avoid channel collisions and achieve deterministic zero-jitter transmission of critical service streams.

[0052] Example 1:

[0053] like Figure 1 As shown, an aerobic exercise data transmission method, applied to the protocol stack media access control layer of a wearable communication device, is characterized by comprising:

[0054] Receive heterogeneous service data streams sent by the application layer of the communication equipment. The heterogeneous service data streams include a first time-sensitive service stream with time-delay sensitive attributes and a second time-tolerant service stream with time-delay tolerant attributes.

[0055] Application layer features are extracted from the first time-sensitive business flow to obtain the time change rate parameter of its payload. Based on the preset cross-layer mapping strategy, the time change rate parameter is converted into the media access urgency parameter of the media access control layer.

[0056] When the media access urgency parameter is lower than the preset channel intervention threshold, the first time-sensitive service flow and the second tolerance service flow are combined to generate a standard data frame, and the standard data frame is used for contention for channel access.

[0057] When the media access urgency parameter reaches or exceeds the channel intervention threshold, the MAC layer channel preemption mechanism is triggered, including: suspending the media access control layer transmission process of the second tolerant service flow, encapsulating the first time-sensitive service flow into a short payload emergency frame, generating a self-clearing transmission management control frame, dynamically calculating the silence duration of the target network allocation vector based on the time change rate parameter, and writing it into the self-clearing transmission management control frame; broadcasting the self-clearing transmission management control frame to force other communication nodes in the same wireless LAN to enter a sleep backoff state, and sending a short payload emergency frame within the contention-free time window corresponding to the silence duration.

[0058] Key challenges include: a cross-scale time mismatch between sampling frequency and MAC layer channel state monitoring period; how to extract reliable gradient features from low-frequency physiological signals to drive millisecond-level channel preemption while avoiding false triggers caused by sensor noise or motion artifacts; the cross-layer transformation function mapping physiological gradients to NAV silence duration must simultaneously consider the basic transmission duration of short payload emergency frames, physical layer retransmission protection requirements, and channel occupancy fairness boundaries; in dense WLAN environments, multiple wearable devices may simultaneously detect excessive physiological gradients and attempt to broadcast CTS-to-Self frames, and without an effective conflict resolution mechanism, this can lead to collisions within the control frames themselves; when releasing backlogged tolerable traffic flows in batches from the bypass buffer after preemption, improper recovery rate control can easily induce instantaneous congestion and secondary collisions, and the response time of traditional TCP congestion control algorithms is much longer than the MAC layer frame interval, making direct reuse impossible.

[0059] Specifically, it receives heterogeneous service data streams from the application layer of the communication equipment. These heterogeneous service data streams include a first time-sensitive service stream and a second tolerance-based service stream.

[0060] The application layer of wearable devices collects physiological and exercise data at a fixed frequency, such as heart rate armbands and muscle oxygenation monitoring bracelets. The application layer sends raw data packets to the stream mapping module of the MAC layer through callback functions. Each data packet header carries a service type identifier: type 0x01 represents heart rate data, 0x02 represents muscle oxygenation saturation data, and 0x03 represents cadence or pace data collected by the accelerometer.

[0061] After parsing the identifier, the stream mapping module assigns data of type 0x01 and 0x02 to the first time-sensitive service stream, i.e., core vital sign data, and data of type 0x03 to the second tolerance service stream, i.e., mechanical status data. The two types of service streams are injected into the first-in-first-out initial transmission queue. Each entry in the queue contains: data payload pointer, service stream type label, arrival timestamp, and data length. The initial transmission queue depth is set to 128 data packets, which can be adjusted according to the device memory size.

[0062] The first time-sensitive service flow includes at least muscle oxygen saturation data; when both heart rate data and muscle oxygen saturation data are included, muscle oxygen saturation is used as the primary criterion for triggering channel preemption, and heart rate data is used as the secondary criterion; the muscle oxygen saturation descent gradient is used to calculate the media access urgency parameter; when the heart rate change rate exceeds the preset heart rate threshold, a correction term is added to the media access urgency parameter.

[0063] The callback function is a predefined cross-layer asynchronous communication interface function between the application layer and the MAC layer. It belongs to the standardized calling mechanism for data interaction between protocol stack layers. After the application layer completes the collection, encapsulation and marking of raw data such as heart rate, muscle oxygen saturation and mechanical motion status, and business type identifier, it actively calls the callback function to push the heterogeneous business data packet carrying the identifier field from top to bottom to the stream mapping module of the MAC layer, thus completing the inter-layer data distribution and transmission.

[0064] 0x01, 0x02, and 0x03 are business type identifiers defined in hexadecimal format, and are preset cross-layer protocol identifier fields between the application layer and the MAC layer.

[0065] Heterogeneous service data packets that enter the queue first in a first-in, first-out (FIFO) manner are preferentially retrieved, encapsulated, and sent by the media access control layer. Data packets that enter the queue later are queued and await scheduling in sequence. This ensures the orderliness and traceability of the service data stream transmission sequence and avoids transmission anomalies caused by out-of-order data packets.

[0066] The data payload pointer is an addressable parameter embedded in the queue entry. It is used to point to the starting physical or logical address of the valid data payload of the corresponding service data packet in the device memory. When the MAC layer performs frame encapsulation and transmission operations, it uses this pointer to quickly locate and read the valid service data, thereby improving the efficiency of data interaction between layers and reducing system storage and computing overhead.

[0067] The business flow type label is a pre-defined classification identifier field when the application layer sends it to the media access control layer. It corresponds to the aforementioned business type identifiers such as 0x01, 0x02, and 0x03, and is used to distinguish between time-sensitive business flows and tolerance business flows to which heart rate data, muscle oxygen saturation data, and mechanical motion state data belong.

[0068] The arrival timestamp is a timing marker that records the moment when heterogeneous service data packets are sent and injected into the initial sending queue. It is generated by the device system clock and fixed in the queue entry. It is used for data packet transmission delay statistics, sliding time window calculation, channel access timing determination and subsequent data retransmission, traffic recovery scheduling and other operations, providing a timing reference for the media access control layer to realize adaptive transmission and timing management.

[0069] Data length is a quantitative parameter that represents the number of bytes occupied by the effective payload of a single service data packet. It is sent synchronously with the data packet and recorded in the initial transmission queue entry. The MAC layer uses this parameter to complete the legal encapsulation of wireless data frames, buffer space allocation, transmission duration estimation, and short payload rescue frame pruning and encapsulation, ensuring frame structure compliance and reasonable utilization of channel resources.

[0070] Specifically, the time rate of change parameter is the first-order descent gradient, which characterizes the rate of change of the application layer payload per unit time. The first-order descent gradient is extracted and mapped to the media access urgency parameter of the media access control layer.

[0071] The first-order descent gradient is calculated as follows:

[0072] Multiple muscle oxygen saturation sampling points are collected within the current sliding time window. The slope value is obtained by linear fitting using the least squares method, and the negative of the slope value is used as the first-order descent gradient.

[0073] The length of the sliding time window is 1~3s, and the sampling frequency is 10~100Hz.

[0074] This step is performed by the gradient extraction module, which reads historical time-series data of muscle oxygen saturation from the initial sending queue every 0.5 seconds in a polling manner.

[0075] Among them, muscle oxygen saturation is a physiological parameter characteristic collected by wearable physiological sensors that can characterize the oxygenation level of human muscle tissue and its dynamic changes. It is a core physiological indicator that reflects the fatigue state and criticality of vital signs of athletes.

[0076] The first-order descent gradient is a quantitative feature parameter characterizing the rate of decrease in muscle oxygen saturation per unit time. It is obtained by linearly fitting the time-series muscle oxygen saturation data within a sliding time window and then calculating the inverse of the slope. It is used to reflect the magnitude and trend of the sudden drop in muscle oxygen saturation. It is set as a gradient because the absolute value of human muscle oxygen saturation varies from person to person. The first-order descent gradient reflects a sudden acceleration in tissue oxygen consumption. Least squares fitting is used to filter out random spikes caused by sensor motion artifacts, ensuring that channel preemption is triggered only when a trend-driven sudden drop occurs, thus improving triggering accuracy.

[0077] The media access urgency parameter is a cross-layer mapping parameter used by the media access control layer to quantify the channel access priority and transmission urgency. Its value is obtained by transforming the first-order descending gradient of muscle oxygen saturation and is used as the basis for determining the switching between normal transmission mode and emergency preemptive transmission mode.

[0078] The sliding time window is set to a length of 1-3 seconds and slides forward continuously with a fixed step size. Gradient calculations are performed only on the most recently acquired continuous sampling points within the window to ensure the real-time and timely extraction of physiological features. The sampling points are discrete muscle oxygen saturation values ​​periodically collected and output by physiological sensors, with a sampling frequency set to 10-100Hz. Multiple time-series continuous sampling points together constitute the original dataset used for linear fitting.

[0079] Least squares linear fitting is a standard numerical analysis method for linearizing time-series sampled data. By minimizing the sum of squares of errors between each muscle oxygen saturation sampling point and the fitted line, an optimal linear equation that can characterize the overall trend of the sampled data is constructed, thus obtaining a fitted line that reflects the overall rising and falling trend of muscle oxygen saturation.

[0080] The slope value refers to the degree of inclination of the optimal straight line obtained after linear fitting using the least squares method. It is used to quantitatively characterize the overall rate of change of muscle oxygen saturation over time within the sliding time window. A negative slope indicates that muscle oxygen saturation is decreasing.

[0081] The inverse number is a mathematical operation that performs a numerical transformation on the slope value obtained from linear fitting. By inverting the sign of the slope value, a first-order descent gradient is obtained, so that the gradient value can intuitively correspond to the severity of the decrease in muscle oxygen saturation, which is convenient for subsequent numerical comparison with the channel intervention threshold.

[0082] Specifically, the length of the sliding time window is L. w The unit is seconds, L w The value range is 1 to 3 seconds; the sampling frequency is f. s The unit is Hz, f s The value range is 10 to 100 Hz, and the number of sampling points N=L within the window. w ×f s N is an integer.

[0083] Let the i-th sampling point within the sliding window be (t) i ,S i ), where S i Let be the muscle oxygen saturation, expressed as %, i = 1, 2, ..., N. A straight line is fitted using the least squares method: The formula for calculating the slope 'a' is:

[0084] ;

[0085] Since muscle oxygen saturation tends to decrease at the edge of fatigue, 'a' is negative, indicating a first-order descent gradient. Defined as the inverse of the slope, this value represents the rate of decrease in muscle oxygen saturation per unit time; during resting or low-intensity exercise, Between 0 and 0.5% / s; when the application layer detects muscle oxygen saturation. The rate rises to 2.0–5.0% / s; the first-order descent gradient is mapped to a medium access urgency parameter, which ranges from 0 to 10% / s, with a larger value indicating a more critical physiological state.

[0086] The channel intervention threshold is set based on empirical research results: during low-intensity exercise, the rate of decrease in human muscle oxygen saturation is stable between 0 and 0.5% / s; when the exerciser approaches their aerobic endurance limit, the rate of decrease in muscle oxygen saturation can reach 2.0 to 5.0% / s. Therefore, this invention sets the channel intervention threshold at 1.5% / s, which can accurately identify the exerciser's critical fatigue state and trigger the channel preemption process in a timely manner. Furthermore, this threshold can be personalized according to the exercise capacity of different users; professional athletes can have it raised to 2.0% / s, while ordinary users with weaker physical conditions can have it lowered to 1.0% / s.

[0087] When the media access urgency parameter is lower than the channel intervention threshold, the standard carrier sense multiple access / collision avoidance contention mechanism is used to send heterogeneous service data streams.

[0088] Specifically, when the media access urgency parameter is lower than the channel intervention threshold, the device operates in steady-state contention mode, and the workflow is as follows:

[0089] Data packets to be sent are retrieved from the initial sending queue in chronological order, and the first time-sensitive service data and the second tolerance service data are combined into a payload.

[0090] The merged standard data frame conforms to the IEEE 802.11g specification. The frame header contains the frame type, destination address, source address, and sequence number, and the frame body contains one or more service data units. The payload merging strategy is to prioritize ensuring that the first time-sensitive service data is not split, and to append the subsequent second-tolerance service data to the same frame to reduce frame overhead. For example, a typical standard data frame contains 4 bytes of heart rate value, 4 bytes of muscle oxygen saturation value, and 8 bytes of cadence / pace data, with a total payload of 16 bytes.

[0091] The frame is submitted to the contention transmission engine of the MAC layer to implement the standard Carrier Sense Multiple Access / Collision Avoidance Mechanism (hereinafter referred to as CSMA / CA mechanism): it listens to the physical channel, and if the channel idle time is greater than the Distributed Inter-Frame Interval (hereinafter referred to as DIFS), it starts a random backoff timer and sends the data frame after the backoff is completed; if the channel is busy, it postpones the transmission, does not distinguish between the access categories of the two types of service flows, and uses the default parameters uniformly.

[0092] Among them, the steady-state contention mode is the conventional channel access mode in which wearable communication devices operate when the medium access urgency parameter does not reach the critical condition. Various heterogeneous service flows uniformly adopt the standard contention mechanism to share wireless channel resources, which is suitable for data flow transmission under the user's normal aerobic exercise state.

[0093] Payload splicing refers to the media access control layer's operation of merging and encapsulating valid data segments for multiple types of service data. It involves orderly splicing and integrating the valid data payloads of the first time-sensitive service data and the second tolerance service data to form a single continuous data segment, in order to adapt to the encapsulation format requirements of wireless data frames.

[0094] The IEEE 802.11g specification is a standard protocol for the physical layer and media access control layer of wireless local area networks developed by IEEE. It specifies the modulation method, frame structure, channel access mechanism and interaction timing in the 2.4GHz band.

[0095] The frame header is the header control field of a standard data frame, used by wireless communication nodes to identify, address, verify, and manage the timing of the data frame. The frame body is the effective data carrying section of the standard data frame, used to encapsulate business data units such as heart rate, muscle oxygen saturation, cadence, and pace.

[0096] The CSMA / CA mechanism is a standardized wireless channel sharing access mechanism defined by the IEEE 802.11 series of protocols. Through operations such as channel listening, idle waiting, and random backoff, it avoids channel collisions caused by multiple communication nodes transmitting data simultaneously, and realizes orderly contention and sharing of the wireless channel by multiple nodes.

[0097] A physical channel is a physical transmission medium channel used to transmit radio frequency signals in wireless communication.

[0098] The random backoff timer is a timing unit within the contention transmission engine. It starts random timing when the channel meets the idle condition, and further reduces the collision probability of multiple nodes transmitting data simultaneously by randomizing the backoff duration.

[0099] When the media access urgency parameter reaches or exceeds the channel intervention threshold: the transmission of the second tolerance service flow is suspended, the first timeliness service flow is encapsulated into a short payload emergency frame, and a self-clearing transmission management control frame is generated. The target silence duration is dynamically calculated based on the absolute value of the first-order descent gradient. The self-clearing transmission management control frame is broadcast to make other communication nodes sleep for the target silence duration and send a short payload emergency frame during the sleep period.

[0100] Specifically, when the media access urgency parameter reaches or exceeds the threshold, the system determines that the user has entered a physiological emergency state and immediately triggers the preemption module, including the following steps;

[0101] A: Suspend the transmission of the second tolerance traffic flow and encapsulate the first timeliness traffic flow into a short payload emergency frame;

[0102] B: Generate a self-clearing transmission management control frame and dynamically calculate the target silence duration based on the absolute value of the first-order descent gradient;

[0103] C: Broadcast self-clearing sends management control frames to cause other communication nodes to sleep for the target silence duration;

[0104] D: Send a short payload emergency rescue frame during hibernation.

[0105] In step A, the transmission of the second tolerable service flow is suspended, and the first time-sensitive service flow is encapsulated into a short payload emergency frame. The preemption module first identifies the protocol identifier 0x03 of the second tolerable service flow in the traffic classifier of the MAC layer.

[0106] The delivery of all data packets corresponding to identifier 0x03 to the physical layer is suspended by mapping the output port of the traffic classifier to an empty processing function; the retransmission counter for this type of data packet in the current MAC layer buffer is cleared. The retransmission counter originally recorded the number of times a data frame was attempted to be retransmitted due to lack of acknowledgment. The clearing operation avoids wasting channel resources retransmitting low-priority services in critical moments; the subsequent arriving second-tolerance service stream data is temporarily stored in the bypass buffer of the local non-volatile memory to avoid data loss.

[0107] The bypass buffer adopts a circular queue structure with a capacity of M data packets, where M is preferably 256 to 1024. Each buffer entry includes the service type, arrival timestamp, data length, and payload pointer. When the buffer is full, the oldest data for the second tolerance service flow is discarded according to the timestamp, or the mechanical state data is downsampled and merged. When resuming transmission, data is retrieved in a first-in-first-out manner according to the timestamp.

[0108] Meanwhile, only the first time-sensitive service flow in the initial transmission queue is extracted and encapsulated into a short payload emergency frame. The payload length of this frame is much smaller than that of a standard data frame: for example, it contains only 8 bytes of data: 2 bytes of heart rate value, 2 bytes of muscle oxygen saturation value, and 4 bytes of timestamp. The frame header is compressed to the minimum, and the total frame length is 10 bytes. The overall frame length and payload length are much smaller than that of a regular standard data frame, thereby shortening the transmission time of a single frame and improving the efficiency of emergency data transmission.

[0109] Among them, the traffic classifier is a data packet identification and diversion component integrated into the MAC layer. By parsing the protocol identifier in the header of the data packet, it can accurately identify, classify and divert heterogeneous service flows, and realize differentiated scheduling of time-sensitive service flows and tolerance-sensitive service flows.

[0110] The empty processing function is a predefined no-operation logic function in the MAC layer. After mapping the output port of the traffic classifier to this function, the corresponding business flow data packets are only received but no encapsulation or delivery operations are performed, thereby pausing the business flow transmission.

[0111] The retransmission counter is a measurement parameter that records the number of times a data frame is retransmitted. It is used to count the number of times a data frame is retransmitted due to channel collision or failure to receive an acknowledgment frame from the receiver. It serves as the basis for the MAC layer to perform retransmission control and packet loss judgment.

[0112] Non-volatile memory is a local storage medium that will not lose data after the device loses power. It is used to temporarily store suspended secondary tolerable business flow data to avoid data loss.

[0113] Step B generates a self-clearing transmission management control frame and dynamically calculates the target silence duration based on the absolute value of the first-order descent gradient, specifically including:

[0114] Obtain the base transmission duration of the short payload emergency rescue frame at the current physical layer transmission rate;

[0115] Based on a preset cross-layer mapping strategy, a compensation silence duration is superimposed on the basic transmission duration to generate the target silence duration.

[0116] Among them, the value of the compensation silence duration is positively correlated with the absolute value of the first-order descent gradient.

[0117] The cross-layer mapping strategy assumes the first-order descent gradient of muscle oxygen saturation is G, in units of % / s. The media access urgency parameter U is calculated as U = min(max(G,0),10); when U ≥ U th Channel preemption is triggered at any time, U th The default value is 1.5% / s, and it can be calibrated according to the user's physical fitness level within the range of 1.0 to 2.0% / s.

[0118] Using normalized values, it can be written as: U=clip(G / G max ,0,1), where G max =10% / s, Channel Interference Threshold U th =0.15.

[0119] Basic transmission duration T b The calculation method, for example, if the frame length is 80 bits and the physical layer transmission rate is 54 Mbps in 802.11g, then T b =80 / (54×10 6 )≈1.48μs, but 2μs is used in actual engineering.

[0120] Silence duration T c The calculation method is as follows:

[0121] ,

[0122] The scaling factor K ranges from 0.1 to 10 ms / (% / s); for example, K is set to 5 ms / (% / s). =2.5% / s, =5 × 2.5 = 12.5 ms; when When it rises sharply to 4.0% / s, =20ms, the target silence duration is ultimately between 2 and 30ms, in the network allocation vector field of the self-clearing transmission control frame, the unit is microseconds, for example 12500μs.

[0123] In an IEEE 802.11g wireless LAN environment, 50-200 wearable communication devices were deployed with a physical layer transmission rate of 54 Mbps and a short payload emergency rescue frame length of 40 bytes. By changing the value of K, the transmission success rate and average latency of emergency rescue frames under different muscle oxygen saturation decrease gradients G were measured. When K < 0.1 ms / (% / s), the compensation silence duration was too short, making it difficult to cover the time window required for physical layer retransmission. When the channel load was higher than 70%, the transmission success rate of emergency rescue frames was less than 95%. When K > 10 ms / (% / s), the NAV silence duration was too long, resulting in the channel being monopolized by a single device for a long time, causing severe backlog of tolerance traffic for other devices, and a decrease in the overall system throughput of more than 40%.

[0124] The MAC layer preemption module assigns and stores the target silence duration, which is dynamically calculated using the first-order descent gradient, into the network allocation vector field of the self-clearing transmission control frame. After the self-clearing transmission control frame carrying this network allocation vector field is broadcast, it can declare the exclusive occupancy duration of the wireless channel to all other wearable communication nodes in the same wireless LAN, so that all nodes in the network keep the channel dormant during the time period specified in the network allocation vector field, thereby providing a collision-free and deterministic transmission window for short payload emergency frames.

[0125] The self-clearing transmission frame (hereinafter referred to as the CTS-to-Self frame) is a dedicated MAC control frame defined by the IEEE 802.11 protocol family. Its RA field is set to the MAC address of the transmitting device itself, and the duration field is written with the target NAV silent duration. Nodes that support 802.11 virtual carrier sensing within the same BSS update their local NAV after receiving this frame. This frame is actively sent by the device waiting to preempt the channel. The destination address and source address in the frame both point to the transmitting device itself. It is used to broadcast the channel occupancy declaration to the entire network and force other nodes to give way.

[0126] The Network Allocation Vector (NAV) field is a duration-type control field embedded in the control frame. It is used to declare to other communication nodes in the network the duration for which the device will occupy the channel. Nodes that receive this field will update their own network allocation vector according to the duration parameter carried in it, and will stop initiating any channel access and data transmission behavior before the vector timer reaches zero, thereby realizing virtual carrier sensing and channel avoidance.

[0127] Step C involves broadcasting a self-clearing management control frame to cause other communication nodes to sleep for the target silence duration.

[0128] Specifically, after generating a CTS-to-Self frame, the preemption module monitors the status of the wireless physical channel. When it detects that the physical channel has been idle for a period of time until the Point Coordination Function Inter-Frame Interval (hereinafter referred to as PIFS), the module intercepts the ordinary data frame transmission requests that have not yet been delivered to the physical layer in the local queue of the device, including standard data frames and acknowledgment frames of the regular transmission module. Without starting a random backoff timer, the module directly broadcasts the CTS-to-Self frame through the radio frequency channel.

[0129] By using PIFS to broadcast CTS-to-Self frames, PIFS has a higher priority than all ordinary data frames at the physical channel contention level. When the channel changes from busy to idle, it preemptively seizes the bus, achieving deterministic channel access and solving the deadlock problem caused by collisions that prevent the data from being sent in high-concurrency scenarios.

[0130] The prerequisite for each communication node in a wireless local area network to initiate channel access is that the detected physical channel remains idle for a duration that reaches the inter-frame interval threshold corresponding to its own service. There is a fixed difference in the inter-frame interval duration adapted to different service types. The smaller the timing threshold, the earlier the corresponding service initiates channel access. In this application, PIFS is set to 25μs, which is a short interval with a higher priority than the conventional distributed inter-frame interval and is specifically used for channel preemption determination of high-priority control frames. DIFS is set to 50μs, which is longer than PIFS and is the standard interval parameter in steady-state contention mode. Since PIFS is shorter than DIFS and any random backoff time slot, the CTS-to-Self frame has priority in obtaining channel access rights. After receiving the CTS-to-Self frame, other communication nodes in the same basic service set parse its NAV field, update their local network allocation vector counter, prohibit any transmission operation during the target quiet period, and enter a backoff sleep state.

[0131] To address the control frame collision issue caused by multiple devices simultaneously detecting physiological emergency states and synchronously broadcasting CTS-to-Self frames, a backoff mechanism based on the degree of physiological emergency is implemented: When a device is preparing to broadcast a CTS-to-Self frame, if another CTS-to-Self frame signal is detected on the channel within the PIFS interval, the NAV field duration of that frame is immediately parsed, and the absolute value of the first-order descent gradient of the device and the transmitting node is compared: If the absolute value of the gradient of the device is lower than that of the transmitting node, the broadcast of this CTS-to-Self frame is terminated, and the device enters a sleep state with the corresponding NAV duration; if the absolute value of the gradient of the device is higher than that of the transmitting node, the device waits for the current frame transmission to complete, listens for channel idle time until the PIFS interval is reached, and then broadcasts its own CTS-to-Self frame, ensuring that devices with more critical physiological states have priority in obtaining channel access rights.

[0132] Specifically, each node that triggers preemption calculates a preemption backoff count B=max(0,B) based on the medium access urgency parameter U. max −β U )+H(ID), where H(ID) is a small random disturbance generated by the node identifier; the larger U is, the smaller the backoff count; when U is the same or close, arbitration is performed based on the low bits of the node's MAC address; the node only sends CTS-to-Self after the backoff count is zero and the channel is idle PIFS.

[0133] Step D involves sending a short payload emergency rescue frame during the sleep period. After the short inter-frame interval (SIFS) following the completion of the CTS-to-Self broadcast, the preemption module does not need to perform channel sniffing, random backoff, or other contention access procedures again. Instead, it directly delivers the encapsulated short payload emergency rescue frame to the physical layer and completes the transmission of radio frequency signals within the contention-free transmission window defined by the NAV field. All other communication nodes within the same basic service set are in a sleep state where channel access is prohibited. The wireless physical channel is exclusively used by the current device, and there is no multi-node signal collision conflict. This ensures that, within the same controlled wireless LAN and under the condition that each node supports NAV resolution, the probability of the emergency rescue frame competing and colliding with other ordinary data frames is reduced, thereby reducing transmission delay jitter.

[0134] SIFS is the highest priority and shortest duration standardized inter-frame interval parameter defined by the IEEE 802.11 protocol. It is used to control the minimum timing interval between a frame and the high-priority data frame that follows, ensuring the compactness and continuity of inter-frame transmission.

[0135] By utilizing SIFS compact timing, seamless connection between control frames and emergency frames is achieved, maximizing the use of exclusive channel resources and compressing the total transmission latency of early warning data. Through the contention-free window period of NAV bundles, channel collisions are eliminated at the protocol mechanism level, solving the defects of collisions and uncontrollable latency in high-priority early warning data transmission under dense wearable device networking. The real-time performance and integrity of life safety early warning data transmission for sudden drop in muscle oxygenation are guaranteed, meeting the safety monitoring transmission needs in extreme aerobic exercise scenarios.

[0136] For example, suppose the wearable communication device detects a first-order descent gradient of application-layer muscle oxygen saturation of 2.0% / s, which exceeds the preset channel intervention threshold of 1.5% / s; the mapping ratio coefficient K is taken as an empirical value of 5ms / (% / s), and the compensation silence duration is calculated to be 10ms according to the formula. Adding the basic transmission duration of the short payload emergency frame at a physical rate of 54Mbps of 20μs, the target silence duration is obtained and converted into the NAV field value, which is 10020μs.

[0137] The 10020μs contention-free window ensures that even in extreme situations where the short payload emergency frame is 40 bytes in size, there is still full dedicated bandwidth. Even if hundreds of devices in the same wireless LAN attempt to initiate the transmission of standard data frames up to 1500 bytes, they will be forced to enter sleep mode because they have detected the NAV field, and the emergency frame transmission of this device will not be interfered with, thus ensuring latency determinism.

[0138] Example 2:

[0139] Based on Example 1, an additional service flow recovery step is added:

[0140] Continuously monitor media access urgency parameters;

[0141] When the first-order descent gradient falls back and the medium access urgency parameter is lower than the channel intervention threshold, the generation of self-clearing transmission management control frames is stopped.

[0142] Resume the regular multiplexing process of the initial transmission queue, and extract the backlogged second-tolerance traffic flows from the bypass buffer in batches, and re-merge them with the current first-time traffic flows for payload retransmission.

[0143] When retrieving backlogged second-tolerance traffic flows from the bypass buffer in batches, a gradually increasing transmission rate is used, and the initial minimum contention window is set to half of the standard value.

[0144] Specifically, during the operation of the preemption module, the medium access urgency parameter is continuously monitored. When the first-order descent gradient is detected to fall back, and the medium access urgency parameter is below the channel intervention threshold for three consecutive sliding windows, and the stabilization time exceeds 0.5s, it is determined that the user's physiological state has stabilized, and the service flow recovery step is executed. A single window gradient fall back does not trigger the recovery process, thus avoiding channel oscillations caused by frequent mode switching.

[0145] The service flow recovery steps are as follows: stop generating new CTS-to-Self management control frames, the preemption module exits the active state, and no longer intercepts the regular transmission process; resume the regular multiplexing process of the initial transmission queue, re-enable the transmission channel of the second-tolerant service flow in the traffic classifier, and allow its data packets to enter the physical layer transmission scheduler; extract the backlogged second-tolerant service flows in batches from the bypass buffer, and re-merge them with the currently arriving first time-sensitive service flows for payload retransmission. To avoid secondary congestion caused by sudden traffic, the transmission rate is gradually increased during transmission recovery, and the initial minimum contention window value is set to half of the standard value.

[0146] For example, if the normal transmission rate is 50 packets / second, after the recovery begins, only 25 packets / second will be transmitted in the first second, 35 packets / second in the second, 45 packets / second in the third second, and the rate will recover to 50 packets / second in the fourth second. During this process, the minimum contention window value will remain at 7 until the backlog queue depth drops to less than 20% of the initial depth, and then the minimum contention window value will be restored to 15.

[0147] Example 3:

[0148] like Figure 2 As shown, an adaptive transmission device for aerobic exercise data streams is used to implement the above method. This device is deployed in the protocol stack of a wearable communication device and includes the following functional modules:

[0149] The stream mapping module receives heterogeneous service data streams from the application layer of the communication device and injects them into the initial transmission queue. By parsing the packet header identifiers, it classifies data containing heart rate and muscle oxygen saturation features into the first time-sensitive service stream, and data containing mechanical state features into the second tolerance service stream. The initial transmission queue adopts a first-in-first-out structure with a depth of 128.

[0150] The gradient extraction module is used to extract the first-order descent gradient and convert it into the medium access urgency parameter of the current MAC layer. The module maintains a sliding time window, preferably set to the default of 2 seconds, with a sampling frequency of 50Hz. It calculates the least squares fitting slope in real time and outputs the first-order descent gradient.

[0151] The conventional transmission module is used to merge the payloads of the two types of service flows to generate standard data frames when the medium access urgency parameter is lower than the channel intervention threshold (in this application, the default channel intervention threshold is 1.5% / s), and then transmits them using a CSMA / CA contention mechanism.

[0152] The preemption module is used to perform active channel preemption when the medium access urgency parameter is greater than or equal to the threshold channel intervention threshold. This module further includes the following sub-modules:

[0153] Service flow suspension submodule: Identifies the second-tolerable service flow, suspends its transmission, clears the retransmission counter, and transfers subsequent data to the bypass buffer.

[0154] Emergency Frame Encapsulation Submodule: Extracts only the first time-sensitive business flow and encapsulates it into a short payload emergency frame.

[0155] CTS-to-Self Generation and NAV Calculation Submodule: Constructs CTS-to-Self frames, calculates based on the proportional controller, and writes the target silence duration into the NAV field.

[0156] Priority Broadcast Submodule: After the listening channel is idle for PIFS, the PIFS value of this application is 25μs. No random backoff is required, and CTS-to-Self frames are broadcast directly.

[0157] Contention-free transmission submodule: Transmits a short payload emergency frame after SIFS. In this application, SIFS is set to 10μs.

[0158] The proportional controller submodule is built into the preemption module. It linearly calculates the compensation silence time based on the absolute value of the first-order descent gradient and limits the output amplitude within the range of 0 to 30ms to prevent the system from being paralyzed for a long time due to excessive NAV.

[0159] The recovery control module continuously monitors the media access urgency parameter. When the gradient drops and falls below the threshold for more than 0.5 seconds, it stops preemption, resumes normal transmission, and empties the bypass buffer by gradually increasing the rate with the minimum contention window value being half of the standard value.

[0160] Example 4:

[0161] This embodiment provides a wearable communication device for implementing the above-described method. The device includes: a processor, such as an ARM Cortex-M4 with a clock frequency of 100MHz; a memory, including RAM and flash memory; and a wireless communication interface supporting IEEE 802.11b / g / n, whose MAC layer supports CTS-to-Self frame generation, NAV field writing, and frame interval control. A computer program is stored in the memory, and when executed by the processor, the computer program implements the steps of the method as described in any one of claims 1-7; the wireless communication interface has a built-in MAC layer hardware accelerator for performing frame encapsulation, CSMA / CA contention transmission, and CTS-to-Self frame priority broadcast operations.

[0162] Based on the preferred embodiments of the present invention described above, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A method for transmitting data during aerobic exercise, applied to the media access control layer of the protocol stack of a wearable communication device, characterized in that, include: Receive heterogeneous service data streams sent by the application layer of the communication equipment. The heterogeneous service data streams include a first time-sensitive service stream with time-delay sensitive attributes and a second time-tolerant service stream with time-delay tolerant attributes. Application layer features are extracted from the first time-sensitive service flow to obtain the time change rate parameter of its payload, and the time change rate parameter is converted into the media access urgency parameter of the media access control layer based on a preset cross-layer mapping strategy. When the media access urgency parameter is lower than the preset channel intervention threshold, the first time-sensitive service flow and the second tolerance service flow are combined to generate a standard data frame, and the standard data frame is used for contention for channel access. When the media access urgency parameter reaches or exceeds the channel intervention threshold, the MAC layer channel preemption mechanism is triggered, including: suspending the media access control layer transmission process of the second tolerable service flow, encapsulating the first time-sensitive service flow into a short payload emergency frame, generating a self-clearing transmission management control frame, dynamically calculating the silent duration of the target network allocation vector according to the time change rate parameter, and writing it into the self-clearing transmission management control frame. The broadcast self-clears the sending of management control frames to force other communication nodes in the same wireless LAN to enter a sleep backoff state, and sends short payload emergency frames within the contention-free time window corresponding to the silence duration.

2. The aerobic exercise data transmission method according to claim 1, characterized in that: The time rate of change parameter is a first-order descent gradient, which characterizes the rate of change of the application layer payload per unit time.

3. The aerobic exercise data transmission method according to claim 2, characterized in that: The silent duration for dynamically calculating the target network allocation vector using the first-order descent gradient specifically includes: Obtain the base transmission duration of the short payload emergency rescue frame at the current physical layer transmission rate; Based on a preset cross-layer mapping strategy, a compensation silence duration is superimposed on the basic transmission duration to generate the target silence duration. The value of the compensation silence duration is positively correlated with the absolute value of the first-order descent gradient. The compensation silence duration and the absolute value of the first-order descent gradient are linearly mapped, with the mapping ratio coefficient ranging from 0.1 to 10 ms / (% / s).

4. The aerobic exercise data transmission method according to claim 1, characterized in that: Broadcast self-cleaning transmission management control frames, specifically including: After generating the self-clearing transmission management control frame, monitor the wireless physical channel status; When the physical channel is detected to be idle for the time specified in the point coordination function inter-frame interval, any pending data frame transmission requests are intercepted, and a self-clearing management control frame is broadcast directly through the radio frequency channel.

5. The aerobic exercise data transmission method according to claim 1, characterized in that: Suspend the sending process for the second-tolerance traffic flow in the initial sending queue, including: Identify the protocol identifier of the second-tolerance traffic flow in the traffic classifier at the MAC layer; Pause the delivery of data packets corresponding to the protocol identifier to the physical layer and clear the retransmission counter for that data packet in the current MAC layer cache; Subsequent arrivals of the second-tolerance traffic flow data are transferred to the bypass buffer of the local non-volatile memory.

6. The aerobic exercise data transmission method according to claim 2, characterized in that: The first-order descent gradient is calculated as follows: Multiple sampling points are collected within the current sliding time window, and the slope value is obtained by linear fitting using the least squares method. The negative of the slope value is then used as the first-order descent gradient. The length of the sliding time window is 1~3s, and the sampling frequency is 10~100Hz.

7. The aerobic exercise data transmission method according to claim 1, characterized in that: It also includes business flow recovery steps: Continuously monitor media access urgency parameters; When the first-order descent gradient falls back and the medium access urgency parameter is lower than the channel intervention threshold, the generation of the self-clearing transmission management control frame is stopped. Resume the regular multiplexing process of the initial transmission queue, and extract the backlogged second-tolerance traffic flows from the bypass buffer in batches, and re-merge them with the current first-time traffic flows for payload retransmission.

8. The aerobic exercise data transmission method according to claim 7, characterized in that: When retrieving backlogged second-tolerance traffic flows from the bypass buffer in batches, a gradually increasing transmission rate is used.

9. The aerobic exercise data transmission method according to claim 8, characterized in that: The gradually increasing transmission rate is increased in increments of 10 to 25 packets per second.

10. An aerobic exercise data transmission device, used to implement the aerobic exercise data transmission method according to any one of claims 1-9, characterized in that: include The stream mapping module is used to receive heterogeneous service data streams sent by the application layer of the communication device and inject them into the initial sending queue. The heterogeneous service data streams include a first time-sensitive service stream and a second tolerance service stream. The gradient extraction module is used to extract the time rate of change parameter and convert the first-order descent gradient into the medium access urgency parameter of the current MAC layer. The conventional transmission module is used to combine the payloads of the two types of service flows to generate a standard data frame when the medium access urgency parameter is lower than the channel intervention threshold, and to transmit it using a carrier sense multiple access / collision avoidance mechanism. The preemption module is used to suspend the transmission process of the second-tolerance service flow and generate a short payload emergency frame when the medium access urgency parameter is greater than or equal to the channel intervention threshold. A self-clearing transmission management control frame is generated, and the target silence duration is dynamically calculated based on the time change rate parameter and written into its network allocation vector field. The control frame is broadcast with a higher priority than ordinary data frames to force other communication nodes to hibernate, and a short payload emergency frame is sent during the hibernation window.

Images