Intelligent control method and system for refrigeration house lamp

Abstract

The invention discloses a cold storage lamp intelligent control method and system, and relates to the technical field of cold storage lamp intelligent control, and the method comprises the following steps: building a dynamic channel observation layer based on electromagnetic propagation characteristics, collecting multipath reflection waveforms and phase drift information generated in a moving process of a metal goods shelf in a cold storage in real time, path loss fingerprints changing along with time are generated; and inputting the path loss fingerprints into a frequency shift adaptive prediction model, identifying key frequency points causing route time delay accumulation according to phase deviation trajectory comparison of continuous time periods, and generating a frequency shift pre-judgment window on the basis of the key frequency points. According to the invention, through dynamic channel observation, frequency shift prediction and time delay correction, stable transmission of the illumination instruction in the refrigeration house environment is realized, millisecond-level response is realized under the support of cooperative beam shaping and priority triggering, and the operation safety and the efficient stability of illumination control are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for cold storage lights, specifically to an intelligent control method and system for cold storage lights. Background Technology

[0002] Cold storage facilities, as crucial infrastructure for cold chain logistics and food warehousing, operate in environments characterized by consistently low temperatures and high humidity, placing higher demands on lighting, temperature and humidity control, and safety monitoring. With the development of the Internet of Things (IoT) and wireless communication technologies, cold storage facilities are increasingly incorporating wireless networking for sensing and control, enabling real-time transmission and sharing of data on temperature, humidity, personnel activity, and energy consumption among distributed nodes. By deploying multi-source sensor nodes within the storage area and utilizing wireless networking to construct a comprehensive, latency-controlled communication network, managers can achieve dynamic sensing and remote monitoring of the cold storage's operating environment. This type of network structure offers advantages such as flexible installation, high scalability, and low maintenance costs, making it suitable for applications in cold storage scenarios with dense shelving and complex wiring. Simultaneously, real-time data acquisition and coordinated response provide a solid foundation for energy-efficient operation, operational safety, and information management in cold storage facilities, driving the development of cold chain infrastructure towards intelligence and efficiency.

[0003] The existing technology has the following shortcomings: During wireless network operation, when the metal shelves inside the cold storage are frequently moved, the electromagnetic reflection path will undergo dynamic frequency shift, causing the lighting control commands to gradually accumulate delays in multi-hop routing. When this delay occurs during emergency operations, the lighting area cannot be illuminated immediately, which may lead to falls by personnel in low-light environments.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent control method and system for cold storage lights to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for intelligent control of cold storage lights, comprising the following steps: S100. Establish a dynamic channel observation layer based on electromagnetic propagation characteristics, collect multipath reflection waveforms and phase drift information generated during the movement of metal shelves inside the cold storage in real time, and generate path loss fingerprints that change over time. S200: Input the path loss fingerprint into the frequency shift adaptive prediction model, identify the key frequency points that cause the accumulation of routing delay based on the comparison of phase offset trajectories in continuous time periods, and generate a frequency shift prediction window based on this. S300. Construct a time delay correction table for distributed nodes within the frequency shift prediction window. Utilize the high-precision timestamp exchange mechanism between nodes to calibrate the time delay offset in multi-hop routes hop by hop and form a dynamic correction mapping. The dynamic correction mapping is directly associated with the frequency shift prediction window. S400 performs real-time route adjustment based on dynamic correction mapping, switches paths affected by electromagnetic reflection interference to backup frequency bands, and maintains time synchronization during data transmission to ensure the stability of control command transmission. S500 introduces a cooperative beamforming strategy on the stable path after real-time routing adjustment. It utilizes the transmit phase modulation of distributed nodes and combines it with the stable path to form directional electromagnetic coverage, eliminating local fading areas caused by shelf movement and enhancing link energy. S600: Based on the signal of enhanced link energy, a priority weighted triggering mechanism is introduced in the lighting control link. The result of enhanced link energy is used to realize the millisecond-level response of the lamp control command, so as to avoid lighting failure caused by the accumulation of time delay.

[0007] Preferably, the steps for establishing a dynamic channel observation layer based on electromagnetic propagation characteristics include: Wireless communication nodes with transmitting and receiving functions are deployed in the main aisle, shelf intersection area and warehouse entrance of the cold storage. The wireless communication nodes are equipped with multi-band radio frequency transceivers and external low temperature resistant high gain antennas. The transmitting end periodically sends reference signals with pseudo-random codes, and the receiving end collects the reflected waveform and phase drift with high time resolution and adds timestamps. The acquired raw waveform and phase data are subjected to fast Fourier transform and phase unwrapping processing. Combined with the amplitude attenuation curve, a feature set that can reflect the multipath evolution law is formed. Sliding window averaging and median filtering are used to remove noise interference. The amplitude attenuation value is normalized based on the feature set and superimposed on the time axis to form a path loss fingerprint. The path loss fingerprint is embedded with a timestamp and stored in a segmented dynamic database to reveal the frequency shift pattern caused by shelf movement and provide data input for subsequent prediction.

[0008] Preferably, the steps of inputting the path loss fingerprint into the frequency shift adaptive prediction model and generating the frequency shift prediction window include: The generated path loss fingerprint is segmented into observation segments in chronological order, and the phase shift trajectory and amplitude attenuation curve of the main reflection path are extracted to form a feature combination. The phase offset trajectory is compared point by point and the phase change process is matched by dynamic time warping method, so as to obtain a set of phase offset patterns that can reflect the movement law of the shelf; Based on the phase offset pattern set, key frequency points related to the cumulative routing delay in the spectrum distribution are identified by discrete Fourier transform, and effective frequency points are screened by combining correlation calculation and inter-node spectrum consistency. A frequency shift prediction window is generated based on the distribution of key frequency points, and the path loss fingerprint is continuously monitored within the window to serve as a reference framework for subsequent time delay correction.

[0009] Preferably, the step of calibrating the delay offset in multi-hop routing hop-by-hop and forming a dynamically corrected mapping includes: Within the frequency shift prediction window, path loss fingerprint data is selected as input, and the time difference between nodes is obtained through timestamp exchange between distributed nodes. Based on the timestamp exchange results, the transmission delay of adjacent nodes is calibrated hop by hop, and abnormal delays are eliminated by sliding window averaging and median filtering. At the same time, a timestamp is added to each calibration result to correspond to the frequency shift prediction window. Based on hop-by-hop calibration, a transmission path delay sequence is formed and compared with a frequency shift prediction window to identify hops related to key frequency points. The delay deviation of each hop is recorded as a correction amount and bound to a timestamp to form a dynamic correction mapping table. The dynamic correction mapping is integrated with the frequency shift prediction window, the correction amount is weighted and accumulated and cross-validated with key frequency points to obtain the total correction value for subsequent real-time routing adjustments.

[0010] Preferably, the steps for performing real-time route adjustments based on dynamically corrected mappings include: Based on dynamic correction mapping, the transmission status of each node in the multi-hop route is monitored in real time. When the deviation exceeds the threshold in multiple time segments, the interfered path is identified. The interfered path is cross-validated with the frequency shift prediction window to determine the time of interference. After confirming the interfered path, a backup frequency band is selected based on the dynamic correction mapping and frequency shift prediction window. The transmitter and receiver are synchronously switched under the trigger of a unified timestamp, and the synchronization accuracy of the new path is confirmed by exchanging timestamps. After the handover is completed, the stability of the routing link is verified. The reliability of the path is confirmed by comparing the round-trip delay with the expected value of the dynamically corrected mapping. When an anomaly is detected, a fast reselection of the backup frequency band is triggered. At the same time, the candidate frequency band is updated in combination with the frequency shift prediction window to ensure millisecond-level transmission of control commands.

[0011] Preferably, the steps of introducing a cooperative beamforming strategy based on the stable path after real-time route adjustment include: When a stable path is established, the source node sends a signal with a phase reference sequence and a timestamp. All relay nodes calculate the difference between the received phase and the theoretical phase and feed it back. The source node generates a full-path phase offset distribution table to achieve a unified phase reference. The phase compensation adjustment of the node transmitter is performed based on the phase offset distribution table so that the signal can be phase superimposed in the target direction, and the frequency shift prediction window is periodically updated to form a cooperative beam. By combining the feedback of signal strength from the receiving node, the direction of the cooperative beam is finely adjusted to cover the fading area, and the relay nodes in the path synchronously adjust the phase to ensure beam continuity. At the end of each frequency shift prediction window period, a full-link power measurement is triggered, the energy gain is compared, and the phase compensation parameters are updated again when the gain is below the threshold, so that the cooperative beam can maintain link energy enhancement for a long time and provide a stable signal basis for lighting control commands.

[0012] Preferably, during the full-link power measurement process at the end of each frequency shift prediction window period, the receiving node compares the measured average power with the reference value when cooperative beamforming is not used, and when the energy gain is lower than the threshold, the source node re-collects the phase offset data and updates the phase compensation parameters to ensure the continuous stability of the link energy enhancement effect.

[0013] Preferably, the steps of introducing a priority-weighted triggering mechanism in the lighting control process include: With the support of link energy enhancement results, a priority classification rule for lighting commands is established, and priority weights and maximum allowable delays are set for different types of commands; Priority classification rules are bound to link energy enhancement signals in real time to form a weighted scheduling table, and high-priority instructions are transmitted by prioritizing the path with the highest energy and stable phase based on link quality indicators. After the transmission path is confirmed in the scheduling table, the source node encapsulates the high-priority instruction into a high-speed trigger data packet and adds a timestamp and acknowledgment mark. The receiving node immediately triggers the lighting equipment to respond when the parallel verification passes. If the verification fails, a second retransmission is triggered in the spare frequency band to ensure millisecond-level execution. After the instruction transmission is completed, the receiving node provides feedback on the actual response time and energy utilization. Based on this, the source node dynamically adjusts the priority weights and energy allocation to maintain the stability and efficiency of lighting control.

[0014] A smart control system for cold storage lights includes a dynamic channel observation module, a frequency shift prediction module, a time delay correction module, a real-time route adjustment module, a cooperative beamforming module, and a priority triggering module. The dynamic channel observation module establishes a dynamic channel observation layer based on electromagnetic propagation characteristics, which collects multipath reflection waveforms and phase drift information generated during the movement of metal shelves inside the cold storage in real time, and generates path loss fingerprints that change over time. The frequency shift prediction module inputs the path loss fingerprint into the frequency shift adaptive prediction model, identifies the key frequency points that cause the accumulation of routing delay based on the comparison of phase offset trajectories in continuous time periods, and generates a frequency shift prediction window based on this. The delay correction module constructs a delay correction table for distributed nodes within the frequency shift prediction window. It utilizes a high-precision timestamp exchange mechanism between nodes to calibrate the delay offset in multi-hop routes hop by hop and form a dynamic correction mapping. The dynamic correction mapping is directly associated with the frequency shift prediction window. The real-time routing adjustment module performs real-time routing adjustments based on the dynamic correction mapping, switching paths affected by electromagnetic reflection interference to backup frequency bands and maintaining time synchronization during data transmission. The cooperative beamforming module introduces a cooperative beamforming strategy based on the path after real-time routing adjustment. It utilizes the transmit phase modulation of distributed nodes and combines it with a stable path to form directional electromagnetic coverage, eliminating local fading areas caused by shelf movement and enhancing link energy. The priority triggering module introduces a priority-weighted triggering mechanism in the lighting control process based on the signal enhanced by the link energy, and uses the link energy enhancement result to achieve millisecond-level response to lighting control commands.

[0015] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention establishes a dynamic channel observation layer based on electromagnetic propagation characteristics in a cold storage environment, and combines a frequency shift adaptive prediction model with a distributed delay correction mechanism to achieve proactive control of dynamic frequency shift and routing delay accumulation caused by shelf movement. By identifying key frequency points within the frequency shift prediction window and forming a dynamic correction map, the uncertainty of hop-by-hop delay can be eliminated in multi-hop transmission in a timely manner. This ensures stable transmission of control commands even in complex metal reflection environments, effectively solving the problem of uncontrollable delay caused by passive retransmission in existing wireless networks. This guarantees that lighting control has immediate response capabilities in emergency situations, significantly improving the safety of the cold storage operating environment.

[0016] This invention achieves millisecond-level response for lighting control commands by introducing cooperative beamforming on a stable path after real-time routing adjustments and establishing a priority-weighted triggering mechanism supported by link energy enhancement results. Through phase modulation of distributed nodes, a directional electromagnetic beam covering fading regions is formed, effectively improving link energy concentration and anti-interference capabilities. Based on this, lighting commands are triggered and scheduled according to priority weights, ensuring that emergency lighting needs are prioritized under conditions of limited link resources. This not only significantly reduces safety hazards caused by lighting delays during cold storage operations but also maintains efficient and stable lighting control through dynamic optimization during long-term operation. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0018] Figure 1 This is a flowchart of a method for intelligent control of cold storage lights according to the present invention.

[0019] Figure 2 This is a schematic diagram of a module of an intelligent control system for cold storage lights according to the present invention. Detailed Implementation

[0020] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0021] This invention provides, for example Figure 1 The intelligent control method for cold storage lights shown includes the following steps: S100. Establish a dynamic channel observation layer based on electromagnetic propagation characteristics, collect multipath reflection waveforms and phase drift information generated during the movement of metal shelves inside the cold storage in real time, and generate path loss fingerprints that change over time. The specific steps for generating a path loss fingerprint that varies over time are as follows: Active electromagnetic signal transmission and reception are deployed inside the cold storage facility to collect multipath reflection waveforms and phase drift information caused by shelf movement in real time. Specifically, several wireless communication nodes with transmission and reception capabilities are installed at key locations prone to shelf movement, such as the main aisles, shelf intersection areas, and entrances. Each node is equipped with a multi-band RF transceiver that can switch between 2.4 GHz and 5 GHz and an external low-temperature resistant high-gain antenna to ensure stable operation in environments as low as -20 degrees Celsius. The transmitter periodically sends a reference signal with a known training sequence, which includes a fixed-length pseudo-random code to ensure the receiver can perform waveform alignment and phase calculation. The receiver samples the received signal at 1-millisecond intervals with a sampling rate of 20 MHz to ensure complete capture of multipath signal details. Each acquisition records both amplitude and phase parameters simultaneously and includes a high-precision clock timestamp. Because the rack may move slowly at a speed of 0.2 to 0.5 meters per second under forklift traction, the instantaneous changes in the channel are relatively large. Therefore, high temporal resolution sampling can ensure complete tracking of the reflection path change process. After this step, a large amount of raw reflection waveform data and phase drift data that dynamically change with the movement of the rack are obtained.

[0022] Feature extraction and regularization processing are performed on the raw waveform and phase data obtained in the previous step to form stable and reliable channel observation results. Specifically, the acquired time-domain signal is first subjected to a Fast Fourier Transform (FFT) to convert it into a frequency-domain energy distribution, and the signal amplitude and phase corresponding to each frequency component are calculated. Then, a phase unwrapping algorithm is used to eliminate phase jumps caused by sampling interval truncation, thereby obtaining a continuous phase trajectory over time. To improve the accuracy of feature extraction, the time-domain and frequency-domain data are jointly analyzed. By tracking the energy peak position on the frequency axis, the main path information of multipath reflection is obtained, and the phase changes of these main paths are correlated with the amplitude attenuation. Furthermore, considering the random noise from refrigeration equipment, motor operation, and personnel operations in the cold storage environment, a sliding window averaging algorithm is applied to the phase trajectory and amplitude curve to eliminate the interference of instantaneous jitter on the overall trend. For outliers with fluctuation amplitudes exceeding 3 dB within the window, median filtering is used for smoothing to preserve the signal change characteristics caused by the actual movement of the shelving. Through the above processing, a set of features that can reflect the dynamic evolution of multipath effects is finally formed. This set of features includes the time delay, amplitude attenuation curves, and continuous phase shift trajectory of each main reflection path over time.

[0023] Based on the aforementioned feature set, path loss fingerprints are generated and constructed into a dynamic database that evolves over time. Specifically, the amplitude attenuation values ​​of all major reflection paths are first normalized, mapping data from different observation points to a unified reference standard to eliminate deviations caused by differences in node antenna gain or placement. Then, the normalized amplitude curves and phase trajectories are superimposed point-by-point on the time axis to form a two-dimensional fingerprint map that intuitively reflects changes in the propagation environment inside the cold storage. The horizontal axis of this map represents time, and the vertical axis represents the reflection path number; the color intensity or numerical value corresponds to the instantaneous loss degree and phase shift amplitude of that path. To facilitate subsequent use by the frequency shift adaptive prediction model, a high-precision timestamp is embedded in the generated path loss fingerprint, ensuring that fingerprint data at any given time corresponds one-to-one with the actual acquisition time. Furthermore, considering the periodic characteristics of cold storage shelf movement—for example, high-frequency movement for tens of minutes during loading and unloading operations, while the shelves remain largely stationary during storage—the path loss fingerprint database adopts a segmented storage method, storing data from different operational stages in corresponding index areas. When a new shelf movement event occurs, the database adds new records in real time and deletes expired data that has exceeded a preset time threshold to maintain data freshness and applicability. The resulting path loss fingerprint not only reflects the current electromagnetic propagation state of the cold storage but also reveals the frequency shift patterns caused by shelf movement through historical trajectories, providing complete and high-quality data input for frequency shift prediction and time delay correction in subsequent steps.

[0024] Through the above steps, a dynamic channel observation layer based on electromagnetic propagation characteristics is concretely realized inside the cold storage. This observation layer can continuously collect multipath reflection waveforms and phase drift data when the shelves are frequently moved, and after feature extraction and path loss fingerprint generation processes, it finally forms a database that is dynamically updated over time. The specific advantages of this implementation method are: first, by deploying multi-band transceiver nodes at key locations, high-resolution acquisition of the entire process of shelf movement is achieved; second, the use of multiple data processing methods combining frequency and time domains, phase unwrapping, and sliding window smoothing ensures the accuracy and stability of feature extraction results; third, a path loss fingerprint database with timestamps and segmented storage mechanisms is established, allowing the channel observation results to be directly used in subsequent prediction and correction stages, thereby significantly improving the real-time performance and reliability of intelligent control of cold storage lights.

[0025] S200: Input the path loss fingerprint into the frequency shift adaptive prediction model, identify the key frequency points that cause the accumulation of routing delay based on the comparison of phase offset trajectories in continuous time periods, and generate a frequency shift prediction window based on this. The key frequency points that cause routing delay accumulation are identified, and a frequency shift prediction window is generated based on these points. The specific steps are as follows: First, the generated path loss fingerprint is imported into the prediction processing as input data. The path loss fingerprint contains the amplitude attenuation curves and phase shift trajectories of the multipath reflection paths during the movement of the cold storage rack, along with high-precision timestamps. In practice, the path loss fingerprint is first divided into several fixed-length observation segments in chronological order, for example, each segment is set to a length of five to ten seconds to ensure coverage of the continuous phase changes caused by the slow movement of the rack. Within each segment, the phase shift trajectory of the main reflection path over time is extracted and correlated with the amplitude attenuation curve within the same segment, thus forming a feature combination that describes the electromagnetic propagation state under rack movement. In this step, the path loss fingerprint is not only stored as static data but also converted into a continuous time-series input signal, providing a data foundation for subsequent phase trajectory comparison.

[0026] The phase shift trajectories in the aforementioned continuous segments are compared point-by-point to identify potential frequency shift trends. Specifically, for each major reflection path, the derivative curve of the phase versus time is first calculated to characterize the instantaneous rate of change of the phase shift. Then, the phase change rate curves in different time segments are compared to identify phase shift patterns that recur in multiple segments. During the comparison process, to reduce interference from noise or instantaneous jitter, a dynamic time warping method is used to match the phase trajectories, allowing phase change processes of different lengths or speeds to be stretched or compressed to a uniform time scale for comparison. This method identifies the main trajectory features causing frequency shifts during shelf movement, and these features are combined with corresponding amplitude attenuation curves to further confirm which frequency shift changes are indeed related to channel fading and delay fluctuations. At the end of this step, a set of phase shift patterns reflecting the shelf movement pattern is obtained, laying the foundation for the extraction of key frequency points.

[0027] Based on the aforementioned set of phase shift patterns, key frequency points causing route delay accumulation are identified. Specifically, a discrete Fourier transform is first performed on the phase shift trajectory in the frequency domain to obtain the spectral distribution curve of phase change over time. Then, on this spectral distribution curve, frequency locations exhibiting energy peaks in multiple time segments are identified; these locations are the frequency shift characteristic points closely related to shelf movement. To further determine whether these frequency points have a causal relationship with route delay accumulation, the correlation between the phase change rate at the corresponding frequency point and the actually observed multi-hop transmission delay data is calculated. When the correlation coefficient exceeds a preset threshold, the frequency point is identified as a key frequency point. During this process, spurious frequency components caused by environmental random noise or non-shelf factors must be eliminated; effective screening is achieved by comparing the spectral consistency between different observation nodes. The final set of key frequency points not only accurately locates the main frequency shift components caused by shelf movement but also explains the root cause of route delay accumulation.

[0028] Finally, a frequency shift prediction window is generated based on the identified key frequency points, serving as a reference framework for subsequent delay correction. Specifically, firstly, based on the distribution of key frequency points, their period of influence in the time domain is determined, i.e., within which time intervals frequency shift fluctuations caused by shelf movement are likely to occur. Then, a buffer zone is extended before and after this time interval, for example, adding 20% ​​of the time before and after the corresponding period of the key frequency point, forming a time window that encompasses the complete frequency shift evolution process. This time window is the frequency shift prediction window. Within this window, the system continuously monitors the update of the path loss fingerprint and uses the key frequency points generated in the previous step as a reference to predict upcoming rapid phase shifts and routing delay fluctuations. Simultaneously, the frequency shift prediction window records the timestamp and amplitude change of each predicted event for direct reference in the next stage of delay correction. In this way, frequency shift in the cold storage environment is no longer merely passively observed but transformed into a predictable behavioral pattern, thus providing an accurate time reference for subsequent dynamic correction mapping and routing adjustments.

[0029] Through the above steps, the path loss fingerprint is effectively input into the frequency shift adaptive prediction process. After continuous processing including phase trajectory comparison, key frequency point identification, and frequency shift prediction window generation, it ultimately achieves forward-looking prediction of frequency shift phenomena caused by the movement of cold storage shelves. This method directly introduces the time-evolving path loss fingerprint into the frequency shift prediction process, achieving a close integration of channel characteristics and time delay data. Through dynamic time warping and multi-node spectrum consistency comparison, interference from environmental noise and instantaneous jitter is effectively eliminated, ensuring high reliability of key frequency point identification results. By introducing the frequency shift prediction window structure, random frequency shift fluctuations are transformed into controllable prediction intervals, enabling subsequent time delay correction and routing adjustments to be deployed in advance, avoiding uncontrollable cumulative delays in lighting control commands during emergency operations.

[0030] S300. Construct a time delay correction table for distributed nodes within the frequency shift prediction window. Utilize the high-precision timestamp exchange mechanism between nodes to calibrate the time delay offset in multi-hop routes hop by hop and form a dynamic correction mapping. The dynamic correction mapping is directly associated with the frequency shift prediction window. The delay offset in multi-hop routes is calibrated hop-by-hop and a dynamically corrected mapping is formed. The specific steps are as follows: Within the generated frequency shift prediction window, all path loss fingerprint data falling within that time window are selected as input, and combined with the prediction results of key frequency points, the transmission paths requiring focused monitoring are determined. In the specific implementation, each distributed node participating in communication is assigned a unique identifier, and a unified local clock reference is provided in a cold storage environment using a low-temperature resistant, highly stable crystal oscillator. To achieve strict time alignment, each node, upon entering the prediction window, periodically sends timestamp-marked data packets in a preset order. These packets record a high-precision time value of the transmission moment. Upon receiving the data packet, the receiving node immediately compares its local timestamp with the transmission timestamp and sends the result back to the sending node. In this way, a large amount of measurement data on the time differences between nodes can be obtained within the frequency shift prediction window, providing a foundation for subsequent hop-by-hop delay calibration.

[0031] Using the data obtained from the aforementioned timestamp exchange, the delay offset in the multi-hop route is calibrated hop-by-hop. Specifically, for any two adjacent nodes, the difference between their sending and receiving timestamps is first calculated to obtain the actual transmission delay of that hop. During multiple exchanges, considering that shelf movement may cause instantaneous link quality degradation or changes in reflection paths, statistical processing of multiple sets of timestamp differences is required. A combination of sliding window averaging and median filtering is used to remove abnormal delays that significantly deviate from the average value, retaining results with a stable distribution. Furthermore, to ensure that the calibration process reflects the dynamic characteristics within the frequency shift prediction window, a corresponding timestamp is added to each calibration result, so that each delay data can be accurately mapped to the specific time interval of the frequency shift prediction window. Through this hop-by-hop calibration process, a complete multi-hop route delay distribution map can be obtained, providing input for the next step of dynamic correction mapping generation.

[0032] Based on hop-by-hop calibration, a dynamic correction mapping directly associated with the frequency shift prediction window is formed. Specifically: First, the delay calibration results for each hop are arranged in order of node identifier, forming a complete transmission path delay sequence. Then, this delay sequence is compared with the frequency shift prediction window within the same time period to identify which hop segments' delay fluctuations are highly correlated with the predicted key frequency points. When the correlation exceeds a preset threshold, the hop segment is determined to be subject to frequency shift interference and needs to be highlighted in the dynamic correction mapping. Based on this, the deviation between the actual delay value and the expected standard delay value for each hop is recorded as a correction amount, and this correction amount is bound to the corresponding timestamp, forming a correction mapping table that can be dynamically updated over time. This mapping table not only stores the delay correction amount for each hop but also retains its direct correspondence with the frequency shift prediction window, thus ensuring that when frequency shift occurs, the correction value can be quickly invoked to adjust the path.

[0033] Finally, the aforementioned dynamic correction mapping is integrated with the frequency shift prediction window to form a complete correction framework for real-time route adjustment. In its implementation, the correction amount for each hop in the dynamic correction mapping is first weighted and accumulated to obtain the total correction value for the entire path. This correction value is then cross-validated with key frequency point information in the prediction window to ensure that the correction direction aligns with the frequency shift trend. This total correction value is then input into the routing scheduling logic to guide the selection of backup frequency bands and the switching of transmission paths in subsequent steps. Throughout this process, whenever a new frequency shift or shelf movement event occurs within the frequency shift prediction window, the dynamic correction mapping is updated synchronously, ensuring the real-time nature and effectiveness of the correction results. In this way, multi-hop transmission within the cold storage no longer relies on static fixed routes but can be dynamically corrected based on actual frequency shift conditions. This effectively eliminates the cumulative delay caused by shelf movement while ensuring time synchronization, providing millisecond-level stable response for lighting control in emergency situations.

[0034] Through the above steps, the construction of the distributed node delay correction table within the frequency shift prediction window was completed. This method not only achieves hop-by-hop calibration of multi-hop links through high-precision timestamp exchange, but also realizes proactive compensation for frequency shift phenomena in the cold storage environment by establishing a dynamic correction mapping directly associated with the frequency shift prediction window. It tightly integrates delay calibration and frequency shift prediction, forming a cross-level dynamic association; introduces a timestamp binding mechanism to ensure that the correction results remain consistent with the prediction window in the time dimension; and through real-time updates of the dynamic correction mapping, routing correction can quickly respond to link fluctuations caused by shelf movement, thereby avoiding safety risks caused by accumulated delays in lighting control.

[0035] S400 performs real-time route adjustment based on dynamic correction mapping, switches paths affected by electromagnetic reflection interference to backup frequency bands, and maintains time synchronization during data transmission to ensure the stability of control command transmission. Real-time route adjustments are performed based on dynamically corrected mappings. The specific steps are as follows: First, based on the generated dynamically corrected mapping, the transmission status between nodes in a multi-hop route is monitored in real time to identify paths affected by electromagnetic reflection interference. In the specific implementation, the transmission delay correction for each hop is timestamped and directly corresponds to a key frequency point in the frequency shift prediction window. Therefore, when entering a new time segment, the current actual transmission delay is compared with the reference value in the dynamically corrected mapping. When the deviation exceeds a set threshold, the path is determined to be affected by reflection interference. To avoid misjudgment due to a single anomaly, the path is only confirmed as an interfered path when deviations exceeding the threshold occur in multiple consecutive time segments. Simultaneously, the identifier of the interfered path and its corresponding delay offset curve are recorded and immediately cross-validated with the frequency shift prediction window to confirm that the time of interference occurrence matches the predicted frequency point. In this way, rapid location of interfered links in a multi-hop transmission path can be achieved.

[0036] After confirming the interfered path, it is switched to a backup frequency band, ensuring time synchronization during the handover process. Specifically, the backup frequency band is selected based on reference data from a dynamic correction map and a frequency shift prediction window. The dynamic correction map stores historical correction values ​​and alternative frequency band information for each path, while the frequency shift prediction window includes a predicted range for potential future frequency shifts. By simultaneously referencing both types of data, the most suitable backup frequency band can be determined, ensuring the switched path will not be subject to the same type of interference again. During the handover, all nodes need to perform synchronized frequency adjustments; that is, under a unified timestamp trigger, the transmitter and receiver simultaneously switch to the target frequency band to avoid short-term link interruptions caused by asynchronous handover. Furthermore, after the handover, a rapid timestamp exchange is required to confirm the synchronization accuracy of the new path. If the deviation exceeds the allowable range, a secondary alignment is immediately performed to ensure that the entire multi-hop link maintains a consistent time base after the handover. This method ensures the continuity and stability of data transmission even in scenarios where frequent shelf movement causes dynamic channel changes.

[0037] After path switching and time synchronization are completed, the entire routing link undergoes stability verification to ensure control commands can be transmitted in milliseconds. Specifically, in the initial stage after the new path is established, each node sends a test data packet at specified time intervals. This packet carries a high-precision timestamp and is immediately returned by the next node. By comparing the round-trip latency of the returned data packet with the expected correction value in the dynamic correction mapping, it can be determined whether the new path meets the latency control requirements. If an abnormal latency is detected at a certain hop, a rapid reselection of a backup frequency band is triggered until the entire routing link meets the stability threshold. During this process, the frequency shift prediction window is continuously updated. When the next possible frequency shift event is predicted, a new list of backup frequency band candidates is generated in advance so that they can be immediately invoked when needed. This method ensures that route adjustment is not a one-time passive response, but a continuous optimization process formed by combining dynamic correction mapping and frequency shift prediction window, thereby enabling cold storage lighting control commands to maintain stable transmission in a dynamic environment. Finally, after this verification is completed, the control commands are transmitted to the target lighting node, achieving a millisecond-level lighting response and avoiding the latency accumulation problem caused by path interference.

[0038] By implementing the above steps, the entire process of real-time route adjustment based on dynamic correction mapping is completed. During this process, interfered paths can be quickly identified, backup frequency bands can be activated promptly, time synchronization can be maintained during handover, and routing links can undergo stability verification after handover, thus ensuring the transmission reliability of cold storage lighting control.

[0039] S500 introduces a cooperative beamforming strategy on the stable path after real-time routing adjustment. It utilizes the transmit phase modulation of distributed nodes and combines it with the stable path to form directional electromagnetic coverage, eliminating local fading areas caused by shelf movement and enhancing link energy. Based on the stable path after real-time route adjustment, a cooperative beamforming strategy is introduced to enhance link energy. The specific steps are as follows: First, after real-time route adjustment is completed and a stable path is obtained, a unified phase reference alignment is performed on all distributed nodes participating in the path. Specifically, at the moment the stable path is established, the source node sends a signal with a phase reference sequence. This signal uses a fixed pseudo-random code modulation format and includes timestamp information so that receiving nodes can align with their local clocks. After receiving this phase reference sequence, all relay nodes in the path calculate the difference between the locally received phase and the theoretical reference phase and feed the difference back to the source node. The source node statistically analyzes all feedback results and generates a phase offset distribution table for the entire path. This process ensures that every node in the path uses a unified phase reference during transmission, providing conditions for subsequent cooperative beamforming. Since this step directly inherits the time synchronization results from real-time route adjustment, it ensures that the transmission of the phase reference is not affected by frequency shifts or route switching.

[0040] After obtaining the phase offset distribution table for the entire path, the node transmission phase is adjusted to achieve the initial formation of a cooperative beam. Specifically, each node adjusts the phase of its transmitter based on the difference recorded in the phase offset distribution table, ensuring that signals transmitted at the same time can achieve phase superposition in the target direction. To improve directional coverage, the coherent gain value after superposition is calculated in multiple candidate target directions, and the direction with the highest gain is selected as the final beam direction. During this process, considering the dynamic reflection environment caused by the movement of shelves inside the cold storage, the transmission phase adjustment is not set all at once, but is periodically adjusted according to the update frequency of the frequency shift prediction window, allowing the beam direction to be dynamically optimized as the environment changes. This method enables multiple distributed nodes on a stable path to form a directional joint transmission, thereby significantly improving the energy concentration of the link.

[0041] After the initial formation of the cooperative beam, active elimination of local fading areas is achieved by combining a stable path. Specifically, in densely packed areas of cold storage shelves or frequently moving work aisles, receiving nodes may experience momentary signal fading. To address this issue, the receiving node first continuously measures the received signal strength and compares it with reference values ​​in a dynamic correction mapping. When the signal strength in a certain area is detected to be below a preset threshold, local fading is determined to have occurred in that area. Subsequently, the source node fine-tunes the directional parameters of the cooperative beam based on feedback from the receiving node, ensuring that the main lobe of the superimposed signal covers the fading area, thereby achieving spatial energy compensation. During this process, other relay nodes in the path also synchronously perform phase adjustments to ensure the spatial continuity of the entire beamform. In this way, not only can signal loss caused by shelf movement be compensated in real time, but a stable energy distribution can also be maintained throughout the entire path, avoiding the loss or delay of control commands during propagation.

[0042] After eliminating local fading and enhancing link energy, the effectiveness of cooperative beamforming is dynamically verified and optimized. Specifically, at the end of each frequency shift prediction window period, the source node triggers a full-link power measurement process. The receiving node compares the received average power value with the baseline value before beamforming to calculate the energy gain. If the energy gain is lower than a preset target threshold, it indicates a deviation in the phase modulation of the cooperative beam. In this case, the source node re-collects phase offset data and updates the phase compensation parameters to form a new beam direction. Simultaneously, guided by the dynamic correction mapping, the direction of the cooperative beam automatically avoids the interference area predicted by the key frequency point, ensuring that the beam gain always applies to the transmission path that most needs compensation. Through this periodic verification and optimization process, it is ensured that cooperative beamforming is not only effective in the short term but also plays a continuous role in the long-term operation of the cold storage. Ultimately, the energy-enhanced link provides a stable signal basis for the priority triggering of subsequent lighting control commands, thereby ensuring that the cold storage lights achieve millisecond-level response in emergency situations.

[0043] Through the above steps, cooperative beamforming is achieved on the stable path after real-time routing adjustment. It can establish a unified phase reference among multiple distributed nodes, form directional electromagnetic coverage through dynamic phase modulation, and actively eliminate local fading areas by combining reception feedback, ultimately achieving the goal of link energy enhancement.

[0044] S600: Based on the signal of enhanced link energy, a priority weighted triggering mechanism is introduced in the lighting control link. The result of enhanced link energy is used to realize the millisecond-level response of the lamp control command, so as to avoid lighting failure caused by the accumulation of time delay. A priority-weighted triggering mechanism is introduced into the lighting control process, utilizing the link energy enhancement result to achieve millisecond-level response to lighting control commands. The specific steps are as follows: Supported by the link energy enhancement results, a priority hierarchy rule for lighting commands is established. Specifically, the link energy enhancement step provides a highly stable signal optimized by cooperative beamforming, ensuring that command data packets are not lost due to local fading during multi-node transmission. Based on this stable transmission environment, different priority weights are assigned to different types of commands in the lighting control stage. For example, emergency lighting activation commands related to personnel safety are assigned the highest weight, ordinary lighting adjustment commands related to daily temperature and humidity adjustment are assigned a secondary weight, and energy-saving lighting deactivation commands triggered during energy consumption optimization are assigned the lowest weight. When establishing the hierarchy rule, each priority not only defines the triggering order but also binds a maximum allowable delay value. The maximum allowable delay corresponding to the highest priority can be limited to within five milliseconds to ensure immediate response in emergency situations. The establishment of this hierarchy rule is the basis of the priority-weighted triggering mechanism.

[0045] Secondly, priority ranking rules are bound to link energy enhancement signals in real time to form an executable weighted scheduling table. Specifically, after cooperative beamforming enhances link energy, each hop node reports its current signal strength and phase stability. These parameters are input into the scheduling table as dynamically available link quality indicators. When a lighting command enters the scheduling process, it is first inserted into the corresponding position in the scheduling table according to its priority weight. The scheduling table then calculates the trigger time of the command in real time, combining the link quality indicators. For example, when the highest priority lighting turn-on command arrives, the scheduling table immediately calls the path with the highest current signal energy and the most stable phase for transmission, thus ensuring the command is executed within milliseconds. For low-priority commands, if a high-priority task exists within the same time period, the scheduling table will postpone its execution until link resources are available. This method of binding priority with link energy ensures that limited channel resources are prioritized for critical tasks, thereby avoiding control command congestion in emergency situations.

[0046] The priority-weighted triggering process employs a specific execution mechanism for millisecond-level response. Specifically, after the scheduling table confirms the transmission path of the highest-priority instruction, the source node immediately encapsulates the instruction into a high-speed trigger data packet, attaching a timestamp and acknowledgment flag so that the receiving node can identify it instantly. Upon receiving this data packet, the receiving node prioritizes parsing the highest-priority flag and directly triggers the lighting device's response, without waiting for other parallel instructions to complete. To ensure millisecond-level response speed, a parallel verification mechanism is used in this process. While parsing the data packet, the receiving node initiates a local fast phase alignment verification to confirm that the received signal matches the link energy enhancement result. When the verification passes, the lighting device immediately illuminates; if the verification fails, a second retransmission is immediately triggered on the spare frequency band to ensure that the response latency remains within the millisecond range. This millisecond-level response mechanism combines priority-weighted triggering with link energy enhancement, enabling the execution of lighting instructions to be completed in the shortest possible time.

[0047] After completing priority-weighted triggering and millisecond-level response, the execution results of the lighting control process are dynamically fed back and optimized. Specifically, after each lighting command is executed, the receiving node feeds back the actual response time, link energy consumption, and execution result to the source node. Based on the feedback information, the source node updates the latency parameters and energy utilization efficiency indicators in the priority-weighted triggering mechanism. For example, when the actual response time of an emergency lighting command approaches the five-millisecond limit, the system automatically increases its priority weight or reserves more link energy resources in the next prediction window to further reduce the response time. Simultaneously, for low-priority commands, if too many delays are detected, the source node will temporarily increase its priority if energy allows, to ensure the overall balance of the lighting function. Through this feedback and optimization mechanism, priority-weighted triggering not only guarantees millisecond-level response in a single task but also dynamically adapts to changes in the cold storage environment during long-term operation, maintaining the stability and efficiency of lighting control.

[0048] Through the above steps, the priority-weighted triggering mechanism is realized based on the enhanced link energy. This mechanism ensures the immediate response of cold storage lighting control in emergency situations and its continuous stability in long-term operation through four stages: hierarchical rule establishment, scheduling table binding, millisecond-level response execution, and feedback optimization.

[0049] This invention establishes a dynamic channel observation layer based on electromagnetic propagation characteristics in a cold storage environment, and combines a frequency shift adaptive prediction model with a distributed delay correction mechanism to achieve proactive control of dynamic frequency shift and routing delay accumulation caused by shelf movement. By identifying key frequency points within the frequency shift prediction window and forming a dynamic correction map, the uncertainty of hop-by-hop delay can be eliminated in multi-hop transmission, ensuring stable transmission of control commands even in complex metal reflection environments. This comprehensive channel observation, prediction, and correction mechanism effectively solves the problem of uncontrollable delay caused by passive retransmission in existing wireless networks, thereby ensuring that lighting control has immediate response capabilities in emergency situations and significantly improving the safety of the cold storage operating environment.

[0050] This invention achieves millisecond-level response for lighting control commands by introducing cooperative beamforming on a stable path after real-time routing adjustments and establishing a priority-weighted triggering mechanism supported by link energy enhancement results. Through phase modulation of distributed nodes, a directional electromagnetic beam covering fading regions is formed, effectively improving link energy concentration and anti-interference capabilities. Based on this, lighting commands are triggered and scheduled according to priority weights, ensuring that emergency lighting needs are prioritized under conditions of limited link resources. This method not only significantly reduces safety hazards caused by lighting delays during cold storage operations but also maintains efficient and stable lighting control through dynamic optimization in long-term operation, promoting the development of cold storage lighting management towards high reliability and intelligence.

[0051] This invention provides, for example Figure 2 The intelligent control system for cold storage lights shown includes a dynamic channel observation module, a frequency shift prediction module, a time delay correction module, a real-time route adjustment module, a cooperative beamforming module, and a priority triggering module. The dynamic channel observation module establishes a dynamic channel observation layer based on electromagnetic propagation characteristics, which collects multipath reflection waveforms and phase drift information generated during the movement of metal shelves inside the cold storage in real time, and generates path loss fingerprints that change over time. The frequency shift prediction module inputs the path loss fingerprint into the frequency shift adaptive prediction model, identifies the key frequency points that cause the accumulation of routing delay based on the comparison of phase offset trajectories in continuous time periods, and generates a frequency shift prediction window based on this. The delay correction module constructs a delay correction table for distributed nodes within the frequency shift prediction window. It utilizes a high-precision timestamp exchange mechanism between nodes to calibrate the delay offset in multi-hop routes hop by hop and form a dynamic correction mapping. The dynamic correction mapping is directly associated with the frequency shift prediction window. The real-time routing adjustment module performs real-time routing adjustments based on the dynamic correction mapping, switching paths affected by electromagnetic reflection interference to backup frequency bands and maintaining time synchronization during data transmission. The cooperative beamforming module introduces a cooperative beamforming strategy based on the path after real-time routing adjustment. It utilizes the transmit phase modulation of distributed nodes and combines it with a stable path to form directional electromagnetic coverage, eliminating local fading areas caused by shelf movement and enhancing link energy. The priority triggering module introduces a priority-weighted triggering mechanism in the lighting control process based on the signal enhanced by the link energy, and uses the link energy enhancement result to achieve millisecond-level response to lighting control commands.

[0052] The present invention provides an intelligent control method for cold storage lights, which is implemented through the above-mentioned intelligent control system for cold storage lights. For details of the specific method and process of the intelligent control system for cold storage lights, please refer to the embodiment of the above-mentioned intelligent control method for cold storage lights, which will not be repeated here.

[0053] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for intelligent control of cold storage lights, characterized in that, Includes the following steps: A dynamic channel observation layer based on electromagnetic propagation characteristics is established to collect multipath reflection waveforms and phase drift information generated during the movement of metal shelves inside the cold storage in real time, and to generate a path loss fingerprint that varies with time. The path loss fingerprint is input into the frequency shift adaptive prediction model. Based on the comparison of phase offset trajectories in continuous time periods, the key frequency points that cause the accumulation of routing delay are identified, and a frequency shift prediction window is generated on this basis. Within the frequency shift prediction window, a time delay correction table for distributed nodes is constructed. Using a high-precision timestamp exchange mechanism between nodes, the time delay offset in multi-hop routes is calibrated hop by hop, and a dynamic correction mapping is formed. The dynamic correction mapping is directly associated with the frequency shift prediction window. Real-time routing adjustments are performed based on dynamic correction mapping, switching paths affected by electromagnetic reflection interference to backup frequency bands, and maintaining time synchronization during data transmission; Based on the real-time route adjustment path, a cooperative beamforming strategy is introduced. By utilizing the transmit phase modulation of distributed nodes and combining it with a stable path, directional electromagnetic coverage is formed, eliminating the local fading area caused by shelf movement and achieving link energy enhancement. Based on the signal enhanced by link energy, a priority-weighted triggering mechanism is introduced into the lighting control process to achieve millisecond-level response of lighting control commands by utilizing the link energy enhancement results.

2. The intelligent control method for cold storage lights according to claim 1, characterized in that, The steps to establish a dynamic channel observation layer based on electromagnetic propagation characteristics include: Wireless communication nodes are deployed in the cold storage. The transmitting end of the wireless communication node periodically sends a reference signal with a pseudo-random code, and the receiving end collects the reflected waveform and phase drift with high time resolution and adds a timestamp. The acquired raw waveform and phase data are subjected to fast Fourier transform and phase unwrapping processing. Combined with the amplitude attenuation curve, a feature set that can reflect the multipath evolution law is formed. Sliding window averaging and median filtering are used to remove noise interference. The amplitude attenuation value is normalized based on the feature set and superimposed on the time axis to form a path loss fingerprint. The path loss fingerprint is embedded with the timestamp and stored in a segmented dynamic database to reveal the frequency shift pattern caused by shelf movement.

3. The intelligent control method for cold storage lights according to claim 1, characterized in that, The steps of inputting the path loss fingerprint into the frequency shift adaptive prediction model and generating the frequency shift prediction window include: The generated path loss fingerprint is segmented into observation segments in chronological order, and the phase shift trajectory and amplitude attenuation curve of the main reflection path are extracted to form a feature combination. By comparing the phase offset trajectory point by point and matching the phase change process using the dynamic time warping method, a set of phase offset patterns reflecting the movement law of the shelf is obtained. Based on the phase offset pattern set, key frequency points related to the cumulative routing delay in the spectrum distribution are identified by discrete Fourier transform, and effective frequency points are screened by combining correlation calculation and inter-node spectrum consistency. A frequency shift prediction window is generated based on the distribution of key frequency points, and the update of the path loss fingerprint is continuously monitored within the window as a reference framework for subsequent time delay correction.

4. The intelligent control method for cold storage lights according to claim 3, characterized in that, The steps for hop-by-hop calibration of delay offsets in multi-hop routes and the formation of dynamically corrected mappings include: Within the frequency shift prediction window, path loss fingerprint data is selected as input, and the time difference between nodes is obtained through timestamp exchange between distributed nodes. Based on the timestamp exchange results, the transmission delay of adjacent nodes is calibrated hop by hop, and abnormal delays are eliminated by sliding window averaging and median filtering. At the same time, a timestamp is added to each calibration result to correspond to the frequency shift prediction window. Based on hop-by-hop calibration, a transmission path delay sequence is formed and compared with a frequency shift prediction window to identify hops related to key frequency points. The delay deviation of each hop is recorded as a correction amount and bound to a timestamp to form a dynamic correction mapping table. The dynamic correction mapping is integrated with the frequency shift prediction window, the correction amount is weighted and accumulated and cross-validated with key frequency points to obtain the total correction value for subsequent real-time routing adjustments.

5. The intelligent control method for cold storage lights according to claim 4, characterized in that, The steps for performing real-time route adjustments based on dynamically corrected mappings include: Based on dynamic correction mapping, the transmission status of each node in the multi-hop route is monitored in real time. When the deviation exceeds the threshold in multiple time segments, the interfered path is identified. The interfered path is cross-validated with the frequency shift prediction window to determine the time of interference. After confirming the interfered path, a backup frequency band is selected based on the dynamic correction mapping and frequency shift prediction window. The transmitter and receiver are synchronously switched under the trigger of a unified timestamp, and the synchronization accuracy of the new path is confirmed by exchanging timestamps. After the handover is completed, the stability of the routing link is verified. The reliability of the path is confirmed by comparing the round-trip delay with the expected value of the dynamically corrected mapping. When an anomaly is detected, a fast reselection of the backup frequency band is triggered. At the same time, the candidate frequency band is updated in combination with the frequency shift prediction window to ensure millisecond-level transmission of control commands.

6. The intelligent control method for cold storage lights according to claim 5, characterized in that, The steps for introducing a cooperative beamforming strategy based on a stable path after real-time route adjustment include: When a stable path is established, the source node sends a signal with a phase reference sequence and a timestamp. All relay nodes calculate the difference between the received phase and the theoretical phase and feed it back. The source node generates a full-path phase offset distribution table to achieve a unified phase reference. The phase compensation adjustment of the node transmitter is performed based on the phase offset distribution table so that the signal can be phase superimposed in the target direction, and the frequency shift prediction window is periodically updated to form a cooperative beam. By combining the feedback of signal strength from the receiving node, the direction of the cooperative beam is finely adjusted to cover the fading area, and the relay nodes in the path synchronously adjust the phase to ensure beam continuity. At the end of each frequency shift prediction window period, a full-link power measurement is triggered, the energy gain is compared, and the phase compensation parameters are updated again when the gain is below the threshold, so that the cooperative beam can maintain the link energy enhancement for a long time.

7. The intelligent control method for cold storage lights according to claim 6, characterized in that, During the end-of-link power measurement process at the end of each frequency shift prediction window period, the receiving node compares the measured average power with the reference value when cooperative beamforming is not used, and the source node re-collects phase offset data and updates the phase compensation parameters when the energy gain is lower than the threshold.

8. The intelligent control method for cold storage lights according to claim 6, characterized in that, The steps involved in introducing a priority-weighted triggering mechanism in lighting control include: With the support of link energy enhancement results, a priority classification rule for lighting commands is established, and priority weights and maximum allowable delays are set for different types of commands; Priority classification rules are bound to link energy enhancement signals in real time to form a weighted scheduling table, and high-priority instructions are transmitted by prioritizing the path with the highest energy and stable phase based on link quality indicators. After the transmission path is confirmed in the scheduling table, the source node encapsulates the high-priority instruction into a high-speed trigger data packet and adds a timestamp and acknowledgment mark. The receiving node immediately triggers the lighting equipment to respond when the parallel verification passes. If the verification fails, a second retransmission is triggered in the spare frequency band to ensure millisecond-level execution. After the instruction transmission is completed, the receiving node provides feedback on the actual response time and energy utilization, and the source node dynamically adjusts the priority weight and energy allocation accordingly.

9. A smart control system for cold storage lights, used to implement the smart control method for cold storage lights according to any one of claims 1-8, characterized in that, It includes a dynamic channel observation module, a frequency shift prediction module, a time delay correction module, a real-time route adjustment module, a cooperative beamforming module, and a priority triggering module; The dynamic channel observation module establishes a dynamic channel observation layer based on electromagnetic propagation characteristics, which collects multipath reflection waveforms and phase drift information generated during the movement of metal shelves inside the cold storage in real time, and generates path loss fingerprints that change over time. The frequency shift prediction module inputs the path loss fingerprint into the frequency shift adaptive prediction model, identifies the key frequency points that cause the accumulation of routing delay based on the comparison of phase offset trajectories in continuous time periods, and generates a frequency shift prediction window based on this. The delay correction module constructs a delay correction table for distributed nodes within the frequency shift prediction window. It utilizes a high-precision timestamp exchange mechanism between nodes to calibrate the delay offset in multi-hop routes hop by hop and form a dynamic correction mapping. The dynamic correction mapping is directly associated with the frequency shift prediction window. The real-time routing adjustment module performs real-time routing adjustments based on the dynamic correction mapping, switching paths affected by electromagnetic reflection interference to backup frequency bands and maintaining time synchronization during data transmission. The cooperative beamforming module introduces a cooperative beamforming strategy based on the path after real-time routing adjustment. It utilizes the transmit phase modulation of distributed nodes and combines it with a stable path to form directional electromagnetic coverage, eliminating local fading areas caused by shelf movement and enhancing link energy. The priority triggering module introduces a priority-weighted triggering mechanism in the lighting control process based on the signal enhanced by the link energy, and uses the link energy enhancement result to achieve millisecond-level response to lighting control commands.

Images