An air conditioning system intelligent internet of things module control method and system based on LTE-Cat.1

Abstract

The application discloses an air conditioner system intelligent internet of things module control method and system based on LTE-Cat.1, which comprises the following steps: establishing a bidirectional communication link between an air conditioner host and a remote server and a user terminal; improving a transmission mechanism based on a UDP protocol to perform packet processing on air conditioner operation data, so as to obtain real-time data packets and non-real-time batch packets; storing the real-time data packets and the non-real-time batch packets in real-time buffer areas and non-real-time buffer areas respectively, and adopting an improved FIFO priority scheduling mechanism to perform data scheduling transmission; and monitoring network signal strength, a packet loss rate and network delay in real time through an LTE-Cat.1 module, and triggering an adaptive adjustment strategy when any one of the following conditions is met: the signal strength is lower than a threshold value, the packet loss rate is higher than a threshold value or the network delay is higher than a threshold value. The application solves the technical problems that real-time performance and reliability are difficult to balance, data transmission efficiency is low and network state adaptability is poor in the existing air conditioner internet of things system.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, and specifically to a smart IoT module control method and system for an air conditioning system based on LTE-Cat.1. Background Technology

[0002] With the widespread adoption of IoT technology in smart homes, the intelligent upgrading of air conditioning systems has become an industry trend, and users' demands for functions such as remote control, energy consumption monitoring, and fault early warning are increasing. However, existing air conditioning IoT systems still face the following key technical bottlenecks in data transmission and network adaptation:

[0003] (1) It is difficult to balance real-time performance and reliability.

[0004] Traditional air conditioning IoT systems mostly use the TCP protocol for data transmission, requiring a "three-way handshake" to establish a connection. This results in a long link readiness time (typically >300ms), which is gradually failing to meet the rapid response requirements for real-time control commands (such as temperature adjustment and mode switching). At the same time, for non-real-time data (such as energy consumption statistics and fault logs), due to their large data volume and low real-time requirements, using the TCP protocol can easily lead to wasted bandwidth resources and reduced transmission efficiency.

[0005] (2) Inefficient data packetization and scheduling mechanisms

[0006] Existing technologies mostly adopt a uniform packet distribution strategy (such as fixed packet size), without distinguishing between real-time and non-real-time data characteristics:

[0007] If real-time control commands are transmitted in large packets, network latency can easily lead to delayed responses.

[0008] If non-real-time data is transmitted in small packets, it will increase packet header redundancy and the number of transmissions, wasting bandwidth.

[0009] (3) Poor network adaptability and insufficient stability

[0010] Air conditioner installation scenarios are complex (such as wall obstruction, signal interference, remote areas), and the network environment fluctuates greatly, but existing systems lack dynamic monitoring and adaptive adjustment capabilities.

[0011] To address the aforementioned issues, there is an urgent need for a smart IoT module control method and system for air conditioning systems that can balance real-time performance and reliability, dynamically adapt to network conditions, and efficiently schedule data transmission, in order to improve the stability and user experience of IoT-enabled air conditioning systems. Summary of the Invention

[0012] In order to overcome the shortcomings of the prior art, the purpose of this invention is to provide a control method and system for an intelligent Internet of Things module of an air conditioning system based on LTE-Cat.1, so as to solve the technical problems of difficulty in balancing real-time performance and reliability, low data transmission efficiency and poor network status adaptability in the existing air conditioning Internet of Things system.

[0013] To solve the above problems, the technical solution adopted by the present invention is as follows:

[0014] A method for controlling an intelligent IoT module of an air conditioning system based on LTE-Cat.1 includes the following steps:

[0015] By accessing the mobile network through the LTE-Cat.1 module and the built-in eSIM card, a two-way communication link is established between the air conditioner host and the remote server and user terminal;

[0016] Based on the improved transmission mechanism of UDP protocol, air conditioner operation data is processed into packets to obtain real-time data packets and non-real-time batch packets.

[0017] The main control MCU stores real-time data packets and non-real-time batch packets in its built-in real-time buffer and non-real-time buffer respectively, and uses an improved FIFO priority scheduling mechanism for data scheduling and transmission.

[0018] The LTE-Cat.1 module monitors network signal strength, packet loss rate, and network latency in real time. When any of the following conditions are met, such as signal strength below a threshold, packet loss rate above a threshold, or network latency above a threshold, an adaptive adjustment strategy is triggered.

[0019] Preferably, when performing subcontracting, the following are included:

[0020] Establish a data link based on the connectionless nature of the UDP protocol;

[0021] Real-time control command data and non-real-time data are classified and identified in real time through an air conditioning data feature library;

[0022] The real-time control command data is processed using the smallest packet unit. The packet header is embedded with a priority flag and a timestamp to obtain real-time data packets. Independent channel transmission is achieved through dynamic port mapping.

[0023] For non-real-time data, adaptive batch packetization is adopted. The packet size is dynamically adjusted according to the total amount of data. After compression and merging by the data compression algorithm, the packet header embeds priority flag, timestamp and data block verification information to obtain non-real-time batch packetization.

[0024] Add a 16-bit CRC checksum to real-time data packets, and add a 32-bit CRC checksum plus a data block index to non-real-time batch packets. If a checksum error is detected in a real-time data packet, an immediate retransmission is triggered. If a checksum error is detected in a non-real-time batch packet, a delayed retransmission is triggered by marking the error with an index.

[0025] Preferably, when performing data scheduling and transmission, it includes:

[0026] Dynamic priority weights are introduced into the FIFO scheduling logic, setting real-time buffer data as a high-priority queue and non-real-time buffer data as a low-priority queue. When real-time data packets arrive, low-priority transmission is interrupted and the channel is occupied first.

[0027] Utilize the idle time slots of the high-priority queue to transmit non-real-time batch packets in FIFO order, monitor the buffer occupancy rate in real time, and trigger the overflow protection mechanism when the occupancy rate meets the preset threshold condition.

[0028] The transmission duration of low-priority queues is limited by time-slice round-robin isolation, and channel resources are forcibly released upon timeout.

[0029] Preferably, when monitoring network signal strength, packet loss rate, and network latency in real time, the following are included:

[0030] Network status parameters are periodically collected via the built-in AT command interface of the LTE-Cat.1 module, including:

[0031] Signal strength: Quantized based on the received signal power value reported by the module;

[0032] Network latency: The round-trip time to the remote server is tested using ICMP echo requests, with multiple samples taken per cycle and the average value calculated.

[0033] Packet loss rate: The percentage of packets lost is calculated by sending multiple consecutive UDP probe packets to a remote server and receiving feedback from the receiving end.

[0034] Preferably, when the signal strength is below a threshold, the adaptive adjustment strategy includes re-searching for a base station with a stronger signal, as follows:

[0035] Retrieve historical performance database, calculate comprehensive score using weighted algorithm to generate priority list, and prioritize searching base stations with scores greater than or equal to the set value;

[0036] The top 3 base stations in the priority list are scanned, and the real-time signal strength is obtained through AT commands. If a base station with a signal strength of ≥-85dBm is found, the signal is switched immediately. If no base station is found, the scan is expanded to all operator base stations, and frequency hopping scanning and real-time CRC verification are performed.

[0037] During base station handover, real-time control command data is temporarily stored in the real-time buffer, and non-real-time data transmission is suspended; after handover is completed, non-real-time data is restored by resuming interrupted transmission and timestamp alignment.

[0038] The handover performance data is written to the historical database, and the base station priority weights are updated through an incremental learning algorithm to dynamically optimize the selection strategy.

[0039] Preferably, when the signal strength is below a threshold, the adaptive adjustment strategy includes switching to a backup network channel, as follows:

[0040] Based on a multi-dimensional evaluation model that considers channel type priority, historical communication quality score, and real-time bandwidth capacity, a candidate list is generated by sorting backup network channels.

[0041] During the switchover, the main control MCU writes data to the primary and backup channel buffers simultaneously, and confirms the transmission link using a first-come, first-served principle.

[0042] The hardware acceleration module compresses the handover signaling processing time to ≤30ms, and combined with the data fragmentation and retransmission mechanism, the real-time control command transmission interruption time is ≤50ms.

[0043] The target backup channel receiver detects data through dual verification including timestamp continuity check. If the timestamp jumps by more than 100ms, it triggers retransmission.

[0044] Dynamically adjust the transmission protocol based on the target backup channel type;

[0045] Feedback is provided every 30 seconds on the signal strength and packet loss rate of the target backup channel. If the quality is better than that of the main channel for three consecutive times, a primary / backup switch is triggered.

[0046] The target backup channel is marked as faulty and removed after three consecutive failed switching attempts. The recovery status is detected by low power detection every hour. After two consecutive successful attempts, it is reinstated into the candidate list.

[0047] Preferably, when the packet loss rate is higher than the threshold, the adaptive adjustment strategy includes adjusting the data packet size, strategic retransmission, cross-layer collaboration, and adaptive rollback.

[0048] Adjusting the data packet size includes:

[0049] Three packet loss rate threshold ranges are preset for mild, moderate and severe, and the original packet size is reduced by 20%, 50% and 70% respectively, with the minimum packet unit limited to 128 bytes.

[0050] Moderate packet loss triggers the packet header extension field, while severe packet loss triggers the data fragmentation redundancy mechanism.

[0051] The packet size is dynamically updated every 200ms based on the latest packet loss rate sample value.

[0052] Preferred strategic retransmissions include:

[0053] The lost data packets reported by the receiving end are retransmitted using the sliding window protocol;

[0054] For real-time data packets, a base number of retransmissions is set based on the current packet loss rate, and an exponential backoff algorithm is used to determine the retransmission interval. For non-real-time batch packets, the number of retransmissions is fixed at 2. If no acknowledgment is received after the timeout, the packet is marked as waiting for idle retransmission and the high-priority queue gap is used to make up for the retransmission.

[0055] Pre-retransmission of critical data packets is initiated based on the slope of the packet loss rate change.

[0056] Cross-layer collaboration includes: increasing the compression ratio of timeout data packets in the non-real-time buffer to 1:8 and suspending the writing of non-critical data when the packet loss rate is higher than the threshold; adjusting PDCP layer parameters through the AT command interface of the LTE-Cat.1 module; enabling RLC layer confirmation mode when there is severe packet loss; and increasing the maximum number of retransmissions to 8.

[0057] Adaptive rollback includes: when the packet loss rate drops below the threshold for three consecutive cycles, the packet size is gradually restored by 50% of the original adjustment range and the number of retransmissions is reduced synchronously.

[0058] Preferably, when network latency exceeds a threshold, the adaptive adjustment strategy includes reducing the data transmission rate, as follows:

[0059] Based on the average round-trip time of ICMP echo request tests, network latency is divided into three levels: mild, moderate, and severe.

[0060] Mild latency reduces the baud rate of non-real-time data transmission by 20% and extends the packet transmission interval to 1.5 times the original interval; moderate latency reduces the baud rate of non-real-time data transmission by 50%, suspends non-critical data transmission, and compresses the size of real-time data packets to ≤64 bytes; severe latency reduces the baud rate of real-time control command transmission by 30%, suspends non-real-time data transmission, and temporarily stores it.

[0061] Predicting delay level changes by the slope of average RTT change allows for early initiation of target rate adjustments.

[0062] Increase the real-time control command data scheduling weight by 20% for moderate or higher latency.

[0063] Switch to a low-latency frequency band via the LTE-Cat.1 module and disable unnecessary network-side functions;

[0064] Once the delay drops below the threshold, the rate is gradually restored by 50% of the original adjustment range; the adjustment effect data is written into the historical database, and the rate adjustment coefficient is updated through incremental learning.

[0065] A smart IoT module control system for an air conditioning system based on LTE-Cat.1, during operation, executes the above-mentioned method, including:

[0066] LTE-Cat.1 module: wirelessly interacts with remote servers and user terminals via an external antenna;

[0067] Main control MCU: Connects to the LTE-Cat.1 module and the air conditioner main unit respectively, and coordinates data transmission and control logic;

[0068] Built-in eSIM card: Integrated into the LTE-Cat.1 module, providing network access authentication and data transmission services;

[0069] Homebus communication interface: Located between the main control MCU and the air conditioning unit, it collects air conditioning operation data and sends control commands;

[0070] AC-DC power module: Converts alternating current to direct current to power the entire control system;

[0071] Data caching and scheduling unit: configured in the main control MCU, it performs priority sorting and congestion control on transmitted data;

[0072] Network status monitoring unit: Configured in the main control MCU, it monitors network signal strength, packet loss rate and network latency in real time and triggers adaptive adjustment strategies.

[0073] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0074] Establish an efficient and stable communication link: Through the LTE-Cat.1 module and the built-in eSIM card, two-way communication between the air conditioner host and the remote server and user terminal is realized, solving the problems of easy loss and poor contact of traditional physical SIM cards and improving the reliability of network connection.

[0075] Optimize data transmission and scheduling mechanisms: Improve the transmission mechanism based on the UDP protocol, distinguish between real-time control commands and non-real-time data, adopt differentiated packet segmentation strategies (minimum packet unit + adaptive batch packet segmentation) and improved FIFO priority scheduling to ensure low-latency transmission of real-time commands (≤100ms) and efficient batch transmission of non-real-time data.

[0076] Achieve adaptive network status adjustment: By monitoring network signal strength, packet loss rate and latency in real time, trigger dynamic adjustment strategies (such as base station switching, backup channel switching, packet size optimization, rate adjustment, etc.) to improve data transmission stability in complex network environments (such as signal obstruction and interference).

[0077] Improve system resource utilization: By separating storage in dual buffers (real-time / non-real-time) and using dynamic priority scheduling, data congestion and resource waste are avoided. Combined with data compression, verification mechanisms and overflow protection, a balance between data integrity and transmission efficiency is ensured.

[0078] This invention achieves intelligent remote control, efficient data interaction, and strong network adaptability of air conditioning systems through the above-mentioned technical means, thereby improving user experience and system operation stability.

[0079] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0080] Figure 1 This is a flowchart illustrating the steps of the intelligent IoT module control method for an air conditioning system according to an embodiment of the present invention.

[0081] Figure 2 This is a logic diagram of the improved transmission mechanism based on the UDP protocol according to an embodiment of the present invention;

[0082] Figure 3 This is a logic diagram of the improved FIFO priority scheduling mechanism according to an embodiment of the present invention;

[0083] Figure 4 This is an interaction diagram of the control system module of the intelligent Internet of Things module for the air conditioning system according to an embodiment of the present invention.

[0084] Reference numerals: 20. Intelligent IoT module control system for air conditioning system; 21. Remote server; 22. User terminal; 23. Air conditioning main unit; 24. External antenna; 201. LTE-Cat.1 module; 202. Main control MCU; 203. Built-in eSIM card; 204. Homebus communication interface; 205. AC-DC power supply module; 206. Data buffer and scheduling unit; 207. Network status monitoring unit. Detailed Implementation

[0085] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0086] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0087] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the scope of this application and its application or use.

[0088] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0089] Example 1, see Figure 1 The present invention provides a step-by-step diagram of a smart IoT module control method for an air conditioning system. Figure 1 The method for controlling an intelligent IoT module of an air conditioning system based on LTE-Cat.1, as shown, includes the following steps:

[0090] S101. Establish communication connection: Access the mobile network through LTE-Cat.1 module 201 and built-in eSIM card 203 to establish a two-way communication link between air conditioner host 23 and remote server 21 and user terminal 22;

[0091] S102. Optimize data transmission protocol: Improve the transmission mechanism based on UDP protocol to perform packet processing on air conditioner operation data to obtain real-time data packets and non-real-time batch packets;

[0092] S103. Data caching and scheduling: The main control MCU 202 temporarily stores the data to be transmitted through its built-in buffer and uses an improved FIFO priority scheduling mechanism for data scheduling and transmission.

[0093] S104. Network Status Monitoring and Adaptive Adjustment: The network signal strength, packet loss rate and network latency are monitored in real time through the LTE-Cat.1 module 201. When any of the following conditions are met, such as signal strength below the threshold, packet loss rate above the threshold or network latency above the threshold, the adaptive adjustment strategy is triggered.

[0094] Background Description: In traditional air conditioning IoT systems, data transmission faces the challenge of balancing real-time performance and reliability. TCP establishes a connection through a three-way handshake, a cumbersome process that results in excessively long link readiness times for real-time control commands (such as temperature adjustment), making it increasingly difficult to meet the air conditioner's need for rapid response to control commands.

[0095] Air conditioner operation data includes two key types of data: real-time control commands (such as temperature adjustment and mode switching): must be responded to within ≤100ms, otherwise it will affect the user experience; non-real-time data (such as energy consumption statistics and fault logs): have low real-time requirements (>100ms), but large data volume, and need to be transmitted in batches efficiently.

[0096] Traditional packet splitting strategies are inefficient: Using a uniform packet splitting method (such as fixed-size packets) can lead to delays in real-time data transmission or redundancy in non-real-time data, and lacks targeted verification mechanisms, making data loss or errors likely. Therefore:

[0097] See Figure 2 The logic diagram of the improved transmission mechanism based on the UDP protocol is as follows: In step S102 above, when the air conditioner operation data is processed by packet segmentation based on the improved transmission mechanism based on the UDP protocol, it includes:

[0098] UDP connectionless optimization: Establishing a data link based on the connectionless transmission characteristics of the UDP protocol (by stripping away the traditional TCP three-way handshake connection establishment process, the connectionless transmission characteristics of the UDP protocol are used to reduce network connection establishment time, compressing the time from the initiation of real-time control command data transmission to the link readiness to less than 50ms).

[0099] Dynamic data type identification: The air conditioner operation data is classified in real time through a preset air conditioner data feature library (including control command opcode and historical data field identifiers). It identifies real-time control command data with a real-time requirement of less than or equal to 100ms (such as temperature adjustment and mode switching commands) and non-real-time data with a real-time requirement of more than 100ms (such as daily operating energy consumption statistics and historical fault logs).

[0100] Differentiated subcontracting strategy:

[0101] The real-time control command data is divided into the smallest packet unit (≤128 bytes / packet), and a priority flag bit (such as 0x01) and a timestamp are embedded in the packet header to obtain real-time data packets. Independent channel transmission is achieved through dynamic port mapping (such as dedicated control port 10001).

[0102] For non-real-time data, adaptive batch packetization is adopted, and the packet size is dynamically adjusted according to the total amount of data (1KB-8KB / packet). The continuous time series data is compressed and merged by a data compression algorithm (such as LZ77 algorithm), and priority flag (such as 0x02), timestamp and data block verification information are embedded in the packet header to obtain non-real-time batch packetization.

[0103] Enhanced packet verification: A 16-bit CRC checksum is added to real-time data packets, and a 32-bit CRC checksum + data block index is used in non-real-time batch packets. When the receiving end detects a real-time data packet verification error, it triggers immediate retransmission. For non-real-time batch packet verification errors, the index marks delay retransmission.

[0104] In this embodiment of the invention, it is necessary to further explain that the improved transmission mechanism based on the UDP protocol achieves efficient transmission of air conditioning data through connectionless optimization, dynamic data classification, differentiated packetization, and enhanced verification.

[0105] UDP's connectionless design reduces link establishment time and improves real-time performance. It eliminates the TCP three-way handshake process, directly establishing data links using UDP's connectionless nature. This reduces the time from initiating a real-time control command to link readiness to less than 50ms. For example, when a user sends a command to "raise the temperature by 1°C," there's no need to wait for connection confirmation; the command is transmitted directly through the UDP link, ensuring a rapid response.

[0106] The system dynamically identifies and distinguishes between real-time and non-real-time data, laying the foundation for differentiated processing. It uses a pre-defined air conditioning data feature library (including control command opcodes and historical data field identifiers) to classify data in real-time: Real-time control command data: real-time requirement ≤100ms, such as temperature adjustment and mode switching commands; Non-real-time data: real-time requirement >100ms, such as daily energy consumption statistics and historical fault logs. For example, the feature library identifies the opcode "0x03" as a "mode switching command" and classifies it as real-time data.

[0107] Differentiated packet splitting strategies optimize transmission efficiency for different data types, ensuring a balance between real-time performance and bandwidth utilization.

[0108] Real-time control command data: Employs the smallest packet unit (≤128 bytes / packet), embedding a priority flag (e.g., 0x01) and a timestamp in the packet header. Independent channel transmission is achieved through dynamic port mapping (e.g., dedicated control port 10001) to avoid conflicts with other data. For example, a temperature adjustment command (64 bytes) is encapsulated into a single real-time packet and transmitted preferentially through dedicated control port 10001.

[0109] Non-real-time data: Adaptive batch packetization is adopted, dynamically adjusting the packet size (1KB-8KB / packet) according to the total amount of data. After compressing continuous time series data using the LZ77 algorithm, the data is merged into batch packets. The packet header embeds a priority flag (such as 0x02), a timestamp, and data block verification information. For example, daily energy consumption statistics (5KB) are compressed into 2KB / packet, generating 3 non-real-time batch packets.

[0110] Enhanced packet assembly verification improves data transmission reliability and reduces the impact of error retransmissions on real-time performance.

[0111] Real-time data packetization: A 16-bit CRC checksum is added, and the receiver retransmits immediately when an error is detected, ensuring that control commands are corrected without delay.

[0112] Non-real-time batch packet splitting: 32-bit CRC checksum + data block index is used. In case of error, the retransmission is delayed by marking the index, so as to avoid occupying real-time channel resources.

[0113] For example, when a non-real-time packet verification error occurs, only the erroneous data block marked "index 3" is retransmitted, instead of the entire batch packet.

[0114] Background Description: In traditional air conditioning IoT systems, using a single cache to store all data can easily lead to real-time commands being blocked by non-real-time data, or the loss of critical control commands due to cache overflow. Furthermore, if non-real-time data is not compressed during storage, it will consume excessive cache resources, reducing transmission efficiency. Therefore:

[0115] In step S103 above, the process of temporarily storing data to be transmitted and scheduling data transmission includes:

[0116] Dynamic partitioning and temporary storage of cache

[0117] Dual-partition cache design: The main control MCU 202 has an independent real-time cache area (capacity ≥ 512 bytes) and a non-real-time cache area (capacity ≥ 32KB), which respectively store real-time data packets and non-real-time batch packets;

[0118] Data classification and writing: Based on the priority flag bit in the data packet header (e.g., 0x01 for real-time control command data), the packet data processed by the improved UDP protocol transmission mechanism is automatically routed to the corresponding buffer area. The real-time buffer area adopts a ring storage structure to avoid overflow, and the non-real-time buffer area supports data block compressed storage (based on the LZ77 algorithm).

[0119] See Figure 3 The logic diagram of the improved FIFO priority scheduling mechanism is as follows:

[0120] Priority queue partitioning: Dynamic priority weights are introduced into the FIFO scheduling logic to divide real-time buffer data into high-priority queues (control instructions with real-time requirements ≤100ms, such as temperature adjustment and mode switching instructions), and non-real-time buffer data into low-priority queues (non-real-time data with real-time requirements >100ms, such as daily operating energy consumption statistics and fault logs).

[0121] Real-time data priority processing: The scheduling algorithm prioritizes scanning high-priority queues and adopts a preemptive scheduling strategy: when a real-time data packet arrives, the current low-priority queue data transmission process is immediately interrupted, and the transmission channel of the LTE-Cat.1 module 201 is occupied first to ensure that the delay from the real-time data packet entering the buffer to the initiation of transmission is ≤20ms.

[0122] Low-priority data idle time slot filling: The low-priority queue adopts an adaptive time slot filling mechanism: the main control MCU 202 monitors the idle status of the high-priority queue in real time (idle is determined when there is no real-time data transmission for 50ms consecutively), and uses the idle time slot to transmit non-real-time batch packets in FIFO order. A maximum of 4 non-real-time batch packets (total size ≤32KB) can be transmitted in a single idle time slot.

[0123] Congestion control and cache management

[0124] Dynamic threshold judgment: Real-time monitoring of buffer occupancy rate. When the real-time buffer occupancy rate is ≥80% or the non-real-time buffer occupancy rate is ≥90%, the data overflow protection mechanism is triggered: the compression ratio of newly entered non-real-time data is temporarily increased, and duplicate historical redundant data (such as 3 consecutive identical temperature sampling values) are discarded.

[0125] Priority inversion avoidance: Time-slice round-robin isolation ensures that low-priority queue data will not block high-priority queue data. The maximum transmission time of low-priority queue data is set to ≤100ms. If the timeout occurs, the channel resources are forcibly released for use by high-priority queue data.

[0126] In this embodiment of the invention, it is necessary to further explain that dynamic priority weights are introduced into the FIFO logic:

[0127] High-priority queue: Real-time buffer data (real-time requirement ≤100ms, such as temperature adjustment instructions);

[0128] Low-priority queue: Non-real-time buffer data (real-time requirements >100ms, such as energy consumption statistics).

[0129] A preemptive scheduling strategy is adopted: when real-time data packets arrive, low-priority data transmission is immediately interrupted, and the LTE-Cat.1 channel is occupied first to ensure that the delay from the buffer to the initiation of transmission is ≤20ms.

[0130] The idle status of high-priority queues is monitored in real time (idle is determined when there is no real-time data transmission for 50ms consecutively), and non-real-time packets are transmitted in batches in FIFO order using idle time slots. Limitation: A maximum of 4 non-real-time batch packets (total size ≤ 32KB) can be transmitted in a single idle time slot to avoid consuming too many channel resources.

[0131] Real-time monitoring of cache occupancy: When the real-time cache occupancy is ≥80% or the non-real-time cache occupancy is ≥90%, the overflow protection mechanism is triggered: the compression ratio of newly entered non-real-time data is increased (based on the LZ77 algorithm); duplicate and redundant data (such as 3 consecutive identical temperature sampling values) are discarded, and cache resources are released.

[0132] The time-slice round-robin isolation mechanism sets the maximum transmission time of low-priority data to ≤100ms. If the timeout occurs, the channel is forcibly released to ensure that high-priority queue data (real-time instructions) is transmitted first, avoiding long-term blocking by low-priority data.

[0133] Background Description: In smart IoT applications for air conditioning systems, the stability of the network environment directly affects the real-time performance of control commands (e.g., temperature adjustment and mode switching require a response time of ≤100ms) and the reliability of data transmission (e.g., energy consumption statistics and fault logs require complete upload). Traditional IoT modules often face the following problems:

[0134] Network fluctuation sensitivity: In complex environments (such as building obstruction, signal interference), the network is prone to sudden drops in signal strength, increases in packet loss rate or increases in latency, which can lead to control command lag or data loss.

[0135] Lack of real-time monitoring mechanism: Traditional modules mostly use fixed-period or event-triggered monitoring, which cannot capture network status changes in a timely manner and are lagging behind in adaptive adjustment;

[0136] Insufficient parameter acquisition accuracy: Inconsistent quantification methods for key indicators such as signal strength and latency lead to biases in network quality assessment, affecting the effectiveness of adjustment strategies. Based on this:

[0137] In step S104 above, real-time monitoring of network signal strength, packet loss rate, and network latency includes:

[0138] Network status parameters are collected periodically every 500ms via the AT command interface built into the LTE-Cat.1 module 201 (such as AT+CSQ to obtain signal strength and AT+CPING to test network latency), specifically including:

[0139] Signal Strength (RSRP): Quantized from the received signal power value (in dBm) reported by the LTE-Cat.1 module 201, ranging from -110dBm to -50dBm;

[0140] Network latency: The round-trip time (RTT) with the remote server 21 was tested using ICMP Ping, with a sample size of 5 times per cycle, and the average value was taken as the current latency value;

[0141] Packet loss rate: The percentage of packets lost reported by the receiver is calculated by sending 10 consecutive UDP probe packets (each packet is 128 bytes in size, including timestamp and sequence number) to the remote server 21. (Packet loss rate = (number of packets sent - number of packets received) / number of packets sent × 100%).

[0142] In this embodiment of the invention, it is necessary to further explain that the network status parameter is collected every 500ms to ensure a rapid response to network fluctuations (such as sudden signal attenuation can be detected within 1 second); parameter synchronization: signal strength, packet loss rate and latency are collected in parallel to avoid decision-making bias caused by monitoring a single indicator.

[0143] It adopts the RSRP (Reference Signal Received Power) index of the LTE standard, covering a range from -110dBm (communication edge, prone to disconnection) to -50dBm (excellent signal, stable transmission).

[0144] The ICMP Ping mechanism is used to send five consecutive probe packets to the remote server 21, and the round-trip time (RTT) of each packet is recorded. The average of the five RTTs is taken as the current network latency value to reduce the random error of a single test (such as instantaneous network jitter) and ensure the stability of latency assessment.

[0145] Probe packet design: Send 10 consecutive UDP probe packets (128 bytes each, including timestamp and sequence number) to remote server 21 to simulate real-time data transmission scenario using the connectionless nature of UDP; for example, if 10 packets are sent and 8 are received, the packet loss rate is 20%, which directly reflects the integrity of the link data transmission.

[0146] The three indicators are linked by an AND / OR logic to trigger adaptive adjustment: any abnormality in any indicator will trigger an immediate response: when the signal strength is below the threshold (e.g., -95dBm), the packet loss rate is above the threshold (e.g., 10%), or the latency is above the threshold (e.g., 200ms), the corresponding adjustment strategy (e.g., base station reselection, packet size optimization) will be immediately activated to ensure that network problems are handled in a timely manner.

[0147] Background Description: In smart IoT applications for air conditioning systems, network signal strength is a key indicator for ensuring stable data transmission. However, in complex scenarios (such as building obstructions, multi-device interference, and remote areas), signal strength may suddenly drop (e.g., RSRP below -95dBm), causing delays or loss in the transmission of real-time control commands (such as temperature adjustment and mode switching), impacting user experience. Based on this:

[0148] In step S104 above, when the signal strength is below the threshold, the adaptive adjustment strategy includes re-searching for a base station with a stronger signal.

[0149] When searching for a base station with a stronger signal, this includes:

[0150] Base station priority dynamic sorting

[0151] Retrieve the preset historical performance database of base stations (including the average signal strength, packet loss rate and handover success rate of each base station in the past 72 hours), calculate the comprehensive score of the base station (average signal strength weight 40%, packet loss rate weight 30%, handover success rate weight 30%) using the weighted algorithm built into the main control MCU 202, generate a priority sorting list, and prioritize searching for base stations with a score higher than 85 points (out of 100 points).

[0152] Phased rapid scanning mechanism

[0153] Phase 1 (rapid screening, time ≤ 1.5 seconds): Only scan the top 3 base stations in the priority ranking list, and send AT commands (such as AT+COPS=?) through the LTE-Cat.1 module 201 to obtain the real-time signal strength. If there is a base station with a signal strength ≥ -85dBm, immediately stop scanning and initiate a handover request.

[0154] Second stage (deep scan, time ≤ 3 seconds): If no base station that meets the conditions is found in the first stage, the scanning range is expanded to all available base stations of all operators (through multi-operator configuration supported by eSIM card), and a frequency hopping scanning strategy is adopted (switching a frequency band every 0.3 seconds) to reduce interference. At the same time, real-time CRC verification is performed on the scan data to ensure accuracy.

[0155] Data transmission coordination during handover

[0156] During base station handover, the main control MCU 202 temporarily stores real-time control command data (such as temperature adjustment and mode switching) in an independent real-time buffer (capacity ≥ 256 bytes) and suspends the transmission of non-real-time data (such as operating energy consumption statistics) through an interrupt mechanism. After the handover is completed, data transmission is restored using the "breakpoint resume + timestamp alignment" method, so that the transmission delay of real-time control command data is ≤ 50ms and the time series continuity error of non-real-time data is ≤ 1 second.

[0157] Search result feedback and self-learning optimization

[0158] The actual performance data of this base station handover (handover success rate, handover time, average signal strength and packet loss rate within 1 minute after handover) is written into the historical performance database. The base station priority weight is updated through an incremental learning algorithm (e.g., for every 1% increase in handover success rate, the corresponding base station priority weight increases by 0.5%), thereby realizing the dynamic optimization of the base station selection strategy.

[0159] In this embodiment of the invention, it is necessary to further explain that this embodiment adopts a closed-loop design of "historical data-driven priority + phased fast scanning + data transmission coordination + self-learning optimization" to quickly switch to the optimal base station when the signal strength is insufficient, ensuring the real-time performance of air conditioning control commands (delay ≤ 50ms) and the integrity of non-real-time data (time series error ≤ 1 second), while continuously improving the network adaptability through self-learning.

[0160] In step S104 above, when the signal strength is below the threshold, the adaptive adjustment strategy includes switching to a backup network channel.

[0161] When switching to the backup network channel, the following applies:

[0162] Backup network channel priority dynamic sorting

[0163] The backup network channels are dynamically ranked based on a multi-dimensional evaluation model that combines channel type priority, historical communication quality scores, and real-time bandwidth capacity.

[0164] Channel type priority: Prioritize the selection of backup channels (such as Wi-Fi, NB-IoT) that are heterogeneous with the main channel (LTE-Cat.1) technology to avoid interference from the same frequency band;

[0165] Historical communication quality score: retrieve the average signal strength (weight 30%), packet loss rate (weight 40%), and handover success rate (weight 30%) of each backup channel over the past 72 hours, and generate a comprehensive score (out of 100) through a weighted algorithm.

[0166] Real-time bandwidth capacity pre-detection: 500ms before the switchover is triggered, a probe packet (128-byte UDP packet) is sent to each backup channel through the main control MCU 202 to obtain the real-time available bandwidth (in Mbps), and only channels with bandwidth ≥ 2Mbps are included in the candidate list;

[0167] Seamless switching and data consistency assurance

[0168] Dual-buffered parallel transmission mechanism: During the handover process, the main control MCU 202 writes the data to be transmitted into the main channel buffer (LTE-Cat.1) and the target backup channel buffer (such as Wi-Fi) at the same time. It selects the first successfully transmitted link to confirm the data and sends it according to the "first-come, first-served" principle to ensure that there is no duplication or loss of data.

[0169] Switching delay control: The channel switching signaling processing time is compressed to ≤30ms by hardware acceleration modules (such as integrated network coprocessors), and with the data fragmentation retransmission mechanism (based on sequence number comparison), the transmission interruption time of real-time control command data during the switching process is ≤50ms;

[0170] Data verification and timestamp alignment: The target backup channel receiver performs dual verification on the data packets (CRC verification + timestamp continuity verification). If a timestamp jump of more than 100ms is detected, the main control MCU 202 is immediately triggered to resend the lost intermediate data packets.

[0171] Adaptive parameter optimization after switching

[0172] Dynamic adaptation of transmission protocols: Adjusting the data transmission protocol according to the target backup channel type, including:

[0173] When switching to Wi-Fi, the TCP protocol is used to transmit non-real-time data (such as historical operation logs) to improve reliability, while the UDP protocol is reserved for transmitting real-time control command data.

[0174] When switching to NB-IoT, enable dynamic adjustment of data compression ratio (adaptive from 1:3 to 1:8) and extend the heartbeat interval to 30s to reduce power consumption;

[0175] Channel quality feedback closed loop: Every 30 seconds, the real-time signal strength and packet loss rate of the target backup channel are fed back to the main control MCU 202. If the feedback quality is better than the current state of the main channel for 3 consecutive times, the role of the main and backup channels will be switched and the target backup channel will be set as the main channel.

[0176] Fault channel isolation and recovery mechanism

[0177] Fault marking: For a target backup channel that fails to switch over three consecutive times, it is automatically marked as a fault channel and removed from the candidate list;

[0178] Periodic self-healing detection: For faulty channels, low-power probes (sending only 1 probe packet) are used every hour to check whether the faulty channel has recovered. If the quality meets the standard for two consecutive detections (signal strength ≥ -95dBm and packet loss rate ≤ 5%), the channel is re-included in the candidate list for dynamic priority sorting.

[0179] In this embodiment of the invention, it needs to be further explained that during handover, the data to be transmitted is simultaneously written to both the primary channel buffer (LTE-Cat.1) and the target backup channel buffer (such as Wi-Fi), and the transmission link is confirmed using a "first-come, first-served" principle. For example, if the Wi-Fi link completes transmission first, the Wi-Fi data takes precedence, and the data in the primary channel buffer is automatically discarded to avoid duplicate transmission. This embodiment ensures optimal backup channel selection through multi-dimensional priority sorting, guarantees no data loss and low latency through a seamless handover mechanism, dynamically optimizes parameters to adapt to different channel characteristics, and enhances system robustness through a fault self-healing mechanism, ultimately achieving stable communication of the air conditioning IoT module in weak signal scenarios.

[0180] Background Description: Traditional IoT modules suffer from the following problems in network packet loss scenarios:

[0181] Fixed packet size strategy has poor adaptability: When using a uniform packet size (such as a fixed 1KB / packet), in a high packet loss environment, the overall data is easily lost due to a single packet error, while in a low packet loss environment, the small packet size causes bandwidth waste.

[0182] The retransmission mechanism is inefficient: Traditional batch retransmission (such as retransmitting the entire data block when a packet is lost) will occupy a lot of channel resources and does not distinguish the retransmission priority of real-time and non-real-time data, resulting in control command delay.

[0183] Lack of cross-layer coordination: The retransmission strategies of the data link layer (such as the PDCP layer) and the application layer are independent, and there is no linkage between buffer scheduling and network status, resulting in delayed packet loss recovery and low resource utilization. Based on this:

[0184] In step S104 above, when the packet loss rate is higher than the threshold, the adaptive adjustment strategy includes adjusting the data packet size, strategic retransmission, cross-layer collaboration, and adaptive rollback.

[0185] When adjusting the size of data packets, the following are included:

[0186] Packet loss rate classification and dynamic packet size adjustment

[0187] Packet loss rate threshold classification: three preset packet loss rate threshold ranges, including mild packet loss (5% < packet loss rate ≤ 10%), moderate packet loss (10% < packet loss rate ≤ 20%), and severe packet loss (packet loss rate > 20%), with corresponding differentiated packet adjustment coefficients for each level;

[0188] Dynamic packet size calculation: Based on the current packet loss rate level, the data packet size is adjusted according to the following rules:

[0189] In the event of minor packet loss, reduce the original packet size (e.g., 1KB) by 20% (adjust to 800 bytes) and limit the minimum packet unit to 128 bytes;

[0190] When there is moderate packet loss, the original packet size (e.g., 1KB) is reduced by 50% (adjusted to 500 bytes), and the minimum packet unit is limited to 128 bytes. The packet header extension field is enabled (adding a 2-byte packet loss statistics identifier).

[0191] In the event of severe packet loss, the original packet size (e.g., 1KB) is reduced by 70% (adjusted to 300 bytes), and the minimum packet unit is limited to 128 bytes, triggering the data fragmentation redundancy mechanism (one check redundancy packet is added for every 3 packets).

[0192] Real-time feedback calibration: The packet size is dynamically updated every 200ms based on the latest packet loss rate sample value (calculated via UDP probe packets) to ensure that the adjustment response delay is ≤50ms;

[0193] Among these, strategic retransmission, cross-layer coordination, and adaptive backoff include:

[0194] Intelligent retransmission strategy based on packet loss pattern

[0195] Selective retransmission mechanism: adopts a sliding window protocol (the window size dynamically adapts to the packet loss rate, with a window size of 8 for mild packet loss, shrinking to 4 for moderate packet loss, and shrinking to 2 for severe packet loss), and only retransmits the lost data packets reported by the receiver (confirmed by sequence number comparison), instead of the traditional batch retransmission.

[0196] Dynamic allocation of retransmission count:

[0197] For real-time data packets (such as control commands): Set the basic number of retransmissions = 3 + 10 × current packet loss rate % (rounded down), with a maximum of 8 times, and use the exponential backoff algorithm for each retransmission interval (100ms for the first time, doubled for subsequent times, with a maximum of 500ms).

[0198] For non-real-time batch packet processing (such as historical logs): the number of retransmissions is fixed at 2. If no acknowledgment is received after a timeout, it is marked as "awaiting idle retransmission" and the high-priority queue data transmission gap is used to make up for the retransmission.

[0199] Packet loss trend prediction and pre-retransmission: By analyzing the slope of packet loss rate changes over 5 consecutive cycles (2.5 seconds) (e.g., if the packet loss rate continues to rise and the slope is >0.02% / ms), pre-retransmission (sending an extra copy) is initiated for critical data packets (such as fault alarm commands) that are about to be transmitted, thereby reducing the risk of packet loss.

[0200] Enhanced congestion control through cross-layer collaboration

[0201] In conjunction with the cache scheduling algorithm: When the packet loss rate exceeds the threshold, the "packet loss priority degradation" mechanism of the data cache is triggered: the compression ratio of data packets that have timed out and have not been transmitted (more than 5 seconds) in the non-real-time cache is increased to 1:8 (based on the LZ77 algorithm), and the writing of non-critical data (such as repeated temperature sampling data) is suspended, and cache resources are released first for retransmission queue.

[0202] Physical layer parameter adaptation: The PDCP layer retransmission parameters are dynamically adjusted through the AT command interface of the LTE-Cat.1 module 201 (such as AT+QCFG="pdcp"). In the case of heavy packet loss, the "RLC layer confirmation mode" (AM) is enabled to increase the maximum number of retransmissions to 8, ensuring the coordination of retransmission strategies between the data link layer and the application layer.

[0203] Adaptive rollback after packet loss recovery

[0204] When the packet loss rate drops below the threshold for three consecutive cycles (1.5 seconds), a smooth rollback mechanism is activated: the packet size is gradually restored by 50% of the original adjustment range (e.g., from 300 bytes → 400 bytes → 500 bytes), and the number of retransmissions is reduced synchronously to avoid repeated adjustments triggered by network fluctuations and ensure a balance between transmission stability and efficiency.

[0205] In this embodiment of the invention, it is necessary to further explain that when the packet loss rate exceeds a threshold, the adaptive adjustment strategy achieves reliable data transmission through packet loss rate grading, dynamic packet adjustment, intelligent retransmission optimization, and cross-layer collaborative control. Three preset threshold ranges correspond to differentiated adjustment strategies, and the packet size is adjusted in real time according to the current packet loss level to balance transmission efficiency and packet loss resistance.

[0206] Moderate packet loss: Enable header extension field (add a 2-byte packet loss statistics identifier to record packet transmission status); Severe packet loss: Trigger data fragmentation redundancy mechanism (add 1 check redundancy packet for every 3 packets to improve fault tolerance through redundant data).

[0207] The sliding window protocol is adopted to retransmit only lost data packets and reduce invalid transmissions: the window size is dynamically adapted to the packet loss rate: for light packet loss (window=8), medium packet loss (window=4), and heavy packet loss (window=2), lost packets are identified by comparing sequence numbers. After the receiving end reports back, only the corresponding sub-packet is retransmitted (e.g., if the 5th packet is lost, only the 5th packet is retransmitted, not the entire data block).

[0208] The retransmission interval adopts an exponential backoff algorithm (100ms initially, doubling thereafter, with a maximum of 500ms) to ensure that high-priority data is restored first.

[0209] Utilize high-priority queues to fill idle time slots (such as gaps in real-time command transmission) to avoid occupying critical channel resources.

[0210] This embodiment employs a closed-loop design of "hierarchical adjustment - intelligent retransmission - cross-layer collaboration - smooth recovery". This strategy can dynamically optimize transmission parameters under different packet loss scenarios, ensuring the real-time performance of air conditioning control commands and the integrity of non-real-time data (no loss, low redundancy), while reducing network resource occupancy.

[0211] Background Description: In smart IoT applications for air conditioning systems, network latency is a key indicator affecting data transmission efficiency and user experience. Traditional IoT modules suffer from the following problems when dealing with network latency: poor adaptability to fixed-rate transmission, no difference between real-time and non-real-time data processing, a vicious cycle of adjustment lag and network degradation, and a lack of cross-layer collaboration. Based on this:

[0212] In step S104 above, when the network latency is higher than the threshold, the adaptive adjustment strategy includes reducing the data transmission rate to reduce data traffic.

[0213] Reducing the data transmission rate includes:

[0214] Based on the average round-trip time (RTT) with remote server 21 obtained using ICMP Ping tests, network latency is divided into three levels:

[0215] Mild delay: Average RTT > threshold and ≤ threshold × 1.5 (e.g., when the threshold is 200ms, the corresponding value is 200ms < average RTT ≤ 300ms).

[0216] Moderate latency: Average RTT > threshold × 1.5 and ≤ threshold × 3 (300ms < average RTT ≤ 600ms).

[0217] Severe latency: Average RTT > threshold × 3 (average RTT > 600ms);

[0218] Differentiation rate adjustment rules:

[0219] Mild delay: Reduce the transmission baud rate of non-real-time data (such as daily operating energy consumption statistics) by 20%, and dynamically extend its packet transmission interval to 1.5 times the original interval through the main control MCU 202 (the original interval is based on the buffer idle time slot filling mechanism, such as 50ms / packet adjusted to 75ms / packet).

[0220] Medium latency: Reduce the transmission baud rate of non-real-time data by 50% and suspend the transmission of non-critical data (such as repeated temperature sampling values); compress the size of real-time data packets from ≤128 bytes to ≤64 bytes to reduce transmission time by reducing the amount of data per packet;

[0221] Severe latency: Trigger minimum transmission mode: Reduce the transmission baud rate of real-time control command data by 30% and suspend the transmission of all non-real-time data, temporarily storing it in the non-real-time buffer (capacity ≥32KB). After the latency is recovered, the data will be retransmitted through the "breakpoint resume + timestamp alignment" mechanism.

[0222] Delayed trend prediction and pre-adjustment mechanism

[0223] Trend prediction algorithm: Predicts the level change of network latency by the slope of the average RTT change over 5 consecutive monitoring periods (2.5 seconds, based on network latency data collected every 500ms);

[0224] Pre-adjustment trigger: Start the target level rate adjustment strategy one cycle (500ms) in advance. For example, when it is predicted that the network will move from mild latency to moderate latency, the transmission baud rate of non-real-time data will be further reduced by 30% in advance to reduce the adjustment lag after the network latency deteriorates.

[0225] Cross-layer coordination with cache scheduling and network parameters

[0226] Cache scheduling coordination: When the latency is moderate or higher, the "latency priority weighting" mechanism of the real-time buffer is triggered, which increases the weight of real-time control command data in the preemptive scheduling strategy by 20%, ensuring that it can still occupy the transmission channel first at low speeds, and the latency from data entering the buffer to initiating transmission is ≤30ms.

[0227] Physical layer parameter adaptation: Dynamically switch to low-latency frequency bands (such as prioritizing mainstream frequency bands such as LTE Band 3 / 7) through the AT command interface of LTE-Cat.1 module 201 (such as AT+QCFG="lte / band"), and disable unnecessary network-side functions (such as DRX power-saving mode) to reduce network-side processing latency.

[0228] Rate recovery and self-learning optimization

[0229] Smooth recovery mechanism: When the delay drops below the threshold for three consecutive cycles (1.5 seconds), the transmission rate is gradually restored at 50% of the original adjustment range to avoid a sudden increase in rate that could cause a delay rebound.

[0230] Self-learning optimization: Write the effect data of each delay adjustment (average RTT change before and after adjustment, data transmission success rate) into the historical database, and update the rate adjustment coefficient of each level of delay through incremental learning algorithm (e.g., if reducing the transmission baud rate of real-time control command data by 30% is not effective when there is severe delay in a certain area, it will be automatically optimized to reduce it by 40%).

[0231] In this embodiment of the invention, it is necessary to further explain that, through a dynamic mechanism of "hierarchical control - predictive pre-adjustment - cross-layer collaboration - self-learning optimization", when the network latency is higher than the threshold, it can reduce traffic pressure by reducing the non-real-time data transmission rate, and prioritize the low-latency transmission of real-time control commands. At the same time, through trend prediction and self-learning, it continuously optimizes and adjusts the strategy to ensure the stable operation of the air conditioning IoT module in complex network environments.

[0232] Example 2, see Figure 4 The present invention provides an interactive diagram of the intelligent IoT module control system module for an air conditioning system. Figure 4 The air conditioning system intelligent IoT module control system 20 based on LTE-Cat.1 shown includes: LTE-Cat.1 module 201, main control MCU 202, built-in eSIM card 203, Homebus communication interface 204, AC-DC power supply module 205, data buffer and scheduling unit 206, and network status monitoring unit 207.

[0233] The LTE-Cat.1 module 201 is configured to achieve wireless data interaction with the remote server 21 and the user terminal 22 via an external antenna 24;

[0234] The main control MCU 202 is electrically connected to the LTE-Cat.1 module 201 and the air conditioning unit 23 respectively, and is used to coordinate data transmission and control logic;

[0235] The built-in eSIM card 203 is integrated into the LTE-Cat.1 module 201, providing network access authentication services and data transmission services for the LTE-Cat.1 module 201;

[0236] Homebus communication interface 204 is located between the main control MCU 202 and the air conditioning unit 23, and is used to collect air conditioning operation data and send control commands.

[0237] AC-DC power module 205 is used to convert AC power to DC power and provide a stable power supply for the entire control system;

[0238] The data caching and scheduling unit 206 is configured in the main control MCU 202 and is used to prioritize and control the congestion of transmitted data.

[0239] The network status monitoring unit 207 is configured in the main control MCU 202 to monitor network signal strength, packet loss rate and network latency parameters in real time, and trigger adaptive adjustment strategies.

[0240] In this embodiment of the invention, it should be further explained that the built-in eSIM card 203 avoids the problems of easy loss and poor contact associated with traditional physical SIM cards, ensuring the reliability of network connectivity. The AC-DC power module 205 converts alternating current into direct current, providing a stable power supply for the entire intelligent IoT module control system and ensuring the normal operation of all components. All parts work closely together to realize the intelligent communication and control functions of the air conditioning system.

[0241] In one possible embodiment, the LTE-Cat.1 module 201 is model number NT26KCNB00NNA.

[0242] In one possible embodiment, the main control MCU 202 is model R5F566TEBDFP.

[0243] In one possible embodiment, the LTE-Cat.1 module 201 is connected to the external antenna 24 via a standard radio frequency interface.

[0244] In this embodiment of the invention, it should be further explained that connecting via a standard radio frequency interface can effectively reduce signal loss during transmission and ensure signal strength and stability. For example, using an IPEX radio frequency mount to connect the external antenna 24 allows the antenna to be flexibly installed in a suitable position on the air conditioner casing to obtain the best signal reception effect.

[0245] In one possible embodiment, the built-in eSIM card 203 is integrated onto the circuit board of the LTE-Cat.1 module 201 by soldering and connected via an electrical interface to ensure stable data transmission and electrical connection.

[0246] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for controlling an intelligent IoT module of an air conditioning system based on LTE-Cat.1, characterized in that, Includes the following steps: By accessing the mobile network through the LTE-Cat.1 module and the built-in eSIM card, a two-way communication link is established between the air conditioner host and the remote server and user terminal; Based on the improved transmission mechanism of UDP protocol, air conditioner operation data is processed into packets to obtain real-time data packets and non-real-time batch packets. The main control MCU stores real-time data packets and non-real-time batch packets in its built-in real-time buffer and non-real-time buffer respectively, and uses an improved FIFO priority scheduling mechanism for data scheduling and transmission. The network signal strength, packet loss rate and network latency are monitored in real time by the LTE-Cat.1 module. When any of the following conditions are met, such as signal strength below the threshold, packet loss rate above the threshold or network latency above the threshold, an adaptive adjustment strategy is triggered. When performing subcontracting, the following are included: Establish a data link based on the connectionless nature of the UDP protocol; Real-time control command data and non-real-time data are classified and identified in real time through an air conditioning data feature library; The real-time control command data is processed using the smallest packet unit. The packet header is embedded with a priority flag and a timestamp to obtain real-time data packets. Independent channel transmission is achieved through dynamic port mapping. For non-real-time data, adaptive batch packetization is adopted. The packet size is dynamically adjusted according to the total amount of data. After compression and merging by the data compression algorithm, the packet header embeds priority flag, timestamp and data block verification information to obtain non-real-time batch packetization. Add a 16-bit CRC checksum to real-time data packets, and add a 32-bit CRC checksum plus a data block index to non-real-time batch packets. If a checksum error is detected in a real-time data packet, an immediate retransmission is triggered. If a checksum error is detected in a non-real-time batch packet, a delayed retransmission is triggered by marking the error with an index.

2. The method according to claim 1, characterized in that, When performing data scheduling and transmission, the following are included: Dynamic priority weights are introduced into the FIFO scheduling logic, setting real-time buffer data as a high-priority queue and non-real-time buffer data as a low-priority queue. When real-time data packets arrive, low-priority transmission is interrupted and the channel is occupied first. Utilize the idle time slots of the high-priority queue to transmit non-real-time batch packets in FIFO order, monitor the buffer occupancy rate in real time, and trigger the overflow protection mechanism when the occupancy rate meets the preset threshold condition. The transmission duration of low-priority queues is limited by time-slice round-robin isolation, and channel resources are forcibly released upon timeout.

3. The method according to claim 1, characterized in that, When monitoring network signal strength, packet loss rate, and network latency in real time, including: Network status parameters are periodically collected via the built-in AT command interface of the LTE-Cat.1 module, including: Signal strength: Quantized based on the received signal power value reported by the module; Network latency: The round-trip time to the remote server is tested using ICMP echo requests, with multiple samples taken per cycle and the average value calculated. Packet loss rate: The percentage of packets lost is calculated by sending multiple consecutive UDP probe packets to a remote server and receiving feedback from the receiving end.

4. The method according to claim 3, characterized in that, When the signal strength is below a threshold, the adaptive adjustment strategy includes re-searching for base stations with stronger signals, as follows: Retrieve historical performance database, calculate comprehensive score using weighted algorithm to generate priority list, and prioritize searching base stations with scores greater than or equal to the set value; The top 3 base stations in the priority list are scanned, and the real-time signal strength is obtained through AT commands. If a base station with a signal strength of ≥-85dBm is found, the signal is switched immediately. If no base station is found, the scan is expanded to all operator base stations, and frequency hopping scanning and real-time CRC verification are performed. During base station handover, real-time control command data is temporarily stored in the real-time buffer, and non-real-time data transmission is suspended. After the switch is completed, non-real-time data is restored by resuming interrupted downloads and aligning timestamps. The handover performance data is written to the historical database, and the base station priority weights are updated through an incremental learning algorithm to dynamically optimize the selection strategy.

5. The method according to claim 3, characterized in that, When the signal strength is below a threshold, the adaptive adjustment strategy includes switching to a backup network channel, as follows: Based on a multi-dimensional evaluation model that considers channel type priority, historical communication quality score, and real-time bandwidth capacity, a candidate list is generated by sorting backup network channels. During the switchover, the main control MCU writes data to the primary and backup channel buffers simultaneously, and confirms the transmission link using a first-come, first-served principle. The hardware acceleration module compresses the handover signaling processing time to ≤30ms, and combined with the data fragmentation and retransmission mechanism, the real-time control command transmission interruption time is ≤50ms. The target backup channel receiver detects data through dual verification including timestamp continuity check. If the timestamp jumps by more than 100ms, it triggers retransmission. Dynamically adjust the transmission protocol based on the target backup channel type; Feedback is provided every 30 seconds on the signal strength and packet loss rate of the target backup channel. If the quality is better than that of the main channel for three consecutive times, a primary / backup switch is triggered. The target backup channel is marked as faulty and removed after three consecutive failed switching attempts. The recovery status is detected by low power detection every hour. After two consecutive successful attempts, it is reinstated into the candidate list.

6. The method according to claim 3, characterized in that, When the packet loss rate exceeds the threshold, the adaptive adjustment strategy includes adjusting the data packet size, strategic retransmission, cross-layer collaboration, and adaptive rollback. Adjusting the data packet size includes: Three packet loss rate threshold ranges are preset for mild, moderate and severe, and the original packet size is reduced by 20%, 50% and 70% respectively, with the minimum packet unit limited to 128 bytes. Moderate packet loss triggers the packet header extension field, while severe packet loss triggers the data fragmentation redundancy mechanism. The packet size is dynamically updated every 200ms based on the latest packet loss rate sample value.

7. The method according to claim 6, characterized in that, Strategic retransmissions include: The lost data packets reported by the receiving end are retransmitted using the sliding window protocol; For real-time data packets, a base number of retransmissions is set based on the current packet loss rate, and an exponential backoff algorithm is used to determine the retransmission interval. For non-real-time batch packets, the number of retransmissions is fixed at 2. If no acknowledgment is received after the timeout, the packet is marked as waiting for idle retransmission and the high-priority queue gap is used to make up for the retransmission. Pre-retransmission of critical data packets is initiated based on the slope of the packet loss rate change. Cross-layer collaboration includes: increasing the compression ratio of timeout data packets in the non-real-time buffer to 1:8 and suspending the writing of non-critical data when the packet loss rate is higher than the threshold; adjusting PDCP layer parameters through the AT command interface of the LTE-Cat.1 module; enabling RLC layer confirmation mode when there is severe packet loss; and increasing the maximum number of retransmissions to 8. Adaptive rollback includes: when the packet loss rate drops below the threshold for three consecutive cycles, the packet size is gradually restored by 50% of the original adjustment range and the number of retransmissions is reduced synchronously.

8. The method according to claim 3, characterized in that, When network latency exceeds a threshold, the adaptive adjustment strategy includes reducing the data transmission rate, as follows: Based on the average round-trip time of ICMP echo request tests, network latency is divided into three levels: mild, moderate, and severe. Mild latency reduces the baud rate of non-real-time data transmission by 20% and extends the packet transmission interval to 1.5 times the original interval; moderate latency reduces the baud rate of non-real-time data transmission by 50%, suspends non-critical data transmission, and compresses the size of real-time data packets to ≤64 bytes; severe latency reduces the baud rate of real-time control command transmission by 30%, suspends non-real-time data transmission, and temporarily stores it. Predicting delay level changes by the slope of average RTT change allows for early initiation of target rate adjustments. Increase the real-time control command data scheduling weight by 20% for moderate or higher latency. Switch to a low-latency frequency band via the LTE-Cat.1 module and disable unnecessary network-side functions; Once the delay drops below the threshold, the rate is gradually restored by 50% of the original adjustment range; the adjustment effect data is written into the historical database, and the rate adjustment coefficient is updated through incremental learning.

9. A smart IoT module control system for an air conditioning system based on LTE-Cat.1, characterized in that, During runtime, the method of claim 1 is executed, comprising: LTE-Cat.1 module: wirelessly interacts with remote servers and user terminals via an external antenna; Main control MCU: Connects to the LTE-Cat.1 module and the air conditioner main unit respectively, and coordinates data transmission and control logic; Built-in eSIM card: Integrated into the LTE-Cat.1 module, providing network access authentication and data transmission services; Homebus communication interface: Located between the main control MCU and the air conditioning unit, it collects air conditioning operation data and sends control commands; AC-DC power module: Converts alternating current to direct current to power the entire control system; Data caching and scheduling unit: configured in the main control MCU, it performs priority sorting and congestion control on transmitted data; Network status monitoring unit: Configured in the main control MCU, it monitors network signal strength, packet loss rate and network latency in real time and triggers adaptive adjustment strategies.

Images