A low-power data transmission method and system for narrowband internet of things

Abstract

The application discloses a low-power-consumption data transmission method and system of narrowband Internet of Things, and relates to the technical field of narrowband Internet of Things communication. The low-power-consumption data transmission method of narrowband Internet of Things comprises the following steps: after receiving a core network downlink data packet, a base station analyzes a terminal identifier, a service type, a data length and a downlink arrival time limit requirement, and acquires a current cell network load; when a terminal is in a sleep state, a target physical resource parameter group is generated, and a downlink data packet is sent through minimal coding; and a low-power-consumption sleep receiving unit of the terminal completes radio frequency energy segment detection, data decoding and buffering and feedback triggering. Through the cooperation of target resource allocation, minimal coding, low-power-consumption sleep receiving and minimal confirmation feedback, the sleep terminal can complete detection, buffering and confirmation feedback when downlink data arrives while reducing invalid monitoring of a main communication radio frequency link.

Description

Technical Field

[0001] This invention relates to the field of narrowband Internet of Things (IoT) communication technology, specifically to a low-power data transmission method and system for narrowband IoT. Background Technology

[0002] Narrowband IoT is a cellular IoT communication technology designed for low-speed, wide-coverage, and low-power terminal connections. It is commonly used in scenarios such as smart meters, environmental monitoring, security alarms, device status acquisition, and remote control. Because many narrowband IoT terminals rely heavily on battery power and are deployed in locations where maintenance is inconvenient, terminals typically need to shorten the online time of the main communication link through sleep mechanisms such as power-saving modes and extended discontinuous reception to extend the continuous operating time of the devices. However, in services such as remote control, alarm confirmation, parameter updates, and firmware upgrades, the network side still needs to send control commands or short data packets to the sleep terminals. The coordination capability between low-power sleep and timely downlink data delivery directly affects terminal battery life, service response speed, and network resource utilization efficiency.

[0003] The limitations of existing technologies include at least the following problems: Existing narrowband IoT low-power transmission solutions typically reduce the online time of the terminal's main communication link through power-saving modes or extended discontinuous reception mechanisms. After the terminal enters sleep mode, downlink data reception relies on a preset wake-up window in conjunction with main radio frequency link monitoring. The wake-up window adopts a fixed or semi-fixed period arrangement. In scenarios with sparse service data, data exhibits the characteristics of random burst arrival and long intervals of idle time. The fixed rhythm of the wake-up window is difficult to adjust with the actual downlink arrival events. The terminal still needs to start the radio frequency reception and baseband monitoring process when there is no downlink data window, resulting in invalid monitoring power consumption that is disconnected from effective data reception. If invalid wake-ups are reduced by extending the monitoring interval, the downlink command waiting latency will increase accordingly. Urgent commands and ordinary data still need to wait for the same period to be sent, making it difficult to balance terminal battery life, downlink delivery timeliness, and network resource utilization. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a low-power data transmission method and system for narrowband IoT, which solves the problem of mismatch between periodic wake-up monitoring and sparse downlink data arrival in existing narrowband IoT low-power transmission, resulting in increased power consumption from invalid monitoring, increased downlink command waiting latency, and difficulty in balancing terminal battery life and downlink delivery timeliness.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a low-power data transmission method for narrowband Internet of Things (IoT), comprising the following steps: a base station receives downlink data packets from the core network to the narrowband IoT terminal, parses the terminal identifier, service type, data length, and downlink delivery timeliness requirement, obtains the current cell network load, queries the sleep state of the narrowband IoT terminal, and when the narrowband IoT terminal is in a power-saving sleep mode or an extended discontinuous reception sleep state, performs dynamic resource allocation calculation driven by the terminal identifier based on the terminal identifier, service type, data length, downlink delivery timeliness requirement, and current cell network load to obtain a target physical resource parameter group including the target time-frequency location, target reference sequence, number of repeated transmissions, and listening duration; determines the encoding rate, payload field length, check field length, and wake-up discrimination field length according to the data length, downlink delivery timeliness requirement, and service type, encodes and reassembles the downlink data packets into a sleep downlink simplified data packet, and then assigns the data packet to the target time-frequency location and... The system repeatedly transmits simplified downlink data packets during sleep. When the main communication RF link and main processor of the narrowband IoT terminal are in deep sleep, the low-power sleep receiving unit collects the RF energy segment corresponding to the target time-frequency position during the listening duration. It performs matched filtering on the RF energy segment using the target reference sequence to obtain the matched filter peak value, extracts the noise floor value, calculates the threshold offset based on the historical false trigger count, and adjusts the detection threshold. It compares the matched filter peak value with the detection threshold to obtain the sleep downlink hit result. When the sleep downlink hit result meets the reception conditions, the low-power sleep receiving unit decodes the simplified downlink data packet, writes the payload field into the low-power buffer, and triggers the preset minimal feedback unit connected to the main processor to switch to a shallow working state. The preset minimal feedback unit reads the payload field, payload verification result, and buffer index, generates a feedback data packet according to the simplified confirmation feedback encoding logic, sends it to the base station through the main communication RF link, and controls the narrowband IoT terminal to return to deep sleep.

[0006] Furthermore, the specific steps for performing dynamic resource allocation calculation driven by terminal identifier are as follows: A resource seed value is generated based on the terminal identifier, and the terminal access duration is calculated; a service priority level is generated based on the service type; a downlink load requirement value is generated based on the data length and downlink delivery timeliness requirements; a resource congestion level is generated based on the current cell network load; candidate time-frequency resources are selected from the available downlink resources set of the base station according to the resource seed value, terminal access duration, service priority level, downlink load requirement value, and resource congestion level; a target time-frequency location is generated based on the candidate time-frequency resources; a target reference sequence is generated based on the resource seed value; and the number of repeated transmissions and the listening duration are generated based on the service priority level and resource congestion level to obtain the target physical resource parameter group.

[0007] Furthermore, the specific steps for conflict avoidance after obtaining the target physical resource parameter group are as follows: Read the occupied time-frequency resource records and allocated reference sequence records corresponding to the current cell network load; perform overlap comparison between the target time-frequency position in the target physical resource parameter group and the occupied time-frequency resource records to generate a time-frequency conflict marker; perform correlation comparison between the target reference sequence in the target physical resource parameter group and the allocated reference sequence records to generate a sequence conflict marker; adjust the target time-frequency position according to the time-frequency conflict marker, and adjust the target reference sequence according to the sequence conflict marker; update the target physical resource parameter group according to the adjusted target time-frequency position and the adjusted target reference sequence.

[0008] Furthermore, the specific steps for encoding and reassembling downlink data packets into dormant downlink simplified data packets are as follows: determine the payload field length based on the data length; determine the encoding rate based on downlink delivery timeliness requirements and service type; determine the wake-up discrimination field length based on the service type; determine the check field length based on the encoding rate and payload field length; when the ratio of the check field length to the payload field length exceeds the preset ratio range, adjust the encoding rate and re-determine the check field length; complete field compression and reassembly in the order of wake-up discrimination field, payload field, and check field to obtain the dormant downlink simplified data packet.

[0009] Furthermore, the specific steps for the low-power sleep receiver unit to acquire radio frequency energy segments are as follows: The low-power sleep receiver unit reads the target time-frequency position and listening duration from the target physical resource parameter group; during the listening duration, the low-power acquisition channel is turned on and the target time-frequency position is locked; energy sampling is performed on the downlink signal corresponding to the target time-frequency position to obtain the original energy sampling sequence; the original energy sampling sequence is slid segmented according to the length of the target reference sequence to obtain multiple candidate radio frequency energy segments; amplitude normalization processing is performed on the multiple candidate radio frequency energy segments to obtain the radio frequency energy segments.

[0010] Further, the specific steps to obtain the downlink hit result in sleep mode are as follows: The low-power sleep receiver unit performs matched filtering on the radio frequency energy segment using the target reference sequence to obtain the matched filter output sequence; the maximum output value is extracted from the matched filter output sequence as the matched filter peak value; the sampling points in the neighborhood of the peak value are removed from the matched filter output sequence, and the noise floor value is calculated based on the remaining sampling points; a threshold offset is generated based on the number of historical false triggers, and the detection threshold is adjusted using the threshold offset; the matched filter peak value is compared with the detection threshold, and when the matched filter peak value reaches the detection threshold, a downlink hit result in sleep mode that meets the reception conditions is generated.

[0011] Further, the specific steps for writing the payload field into the low-power buffer are as follows: The low-power sleep receiving unit decodes the sleep downlink simplified data packet to obtain the wake-up discrimination field, the payload field, and the verification field; the terminal identifier is verified based on the wake-up discrimination field to generate a terminal matching result; the payload field is verified based on the verification field to generate a payload verification result; when both the terminal matching result and the payload verification result meet the buffering conditions, the payload field and the payload verification result are written into the low-power buffer; a cache index is generated for the payload field written into the low-power buffer, and the cache index is passed to the preset minimal feedback unit.

[0012] Furthermore, the specific steps for generating feedback data packets according to the simplified confirmation feedback encoding logic are as follows: The preset minimal feedback unit reads the payload field, payload verification result, and cache index from the low-power buffer; generates a receive flag field based on the cache index; generates a receive status field based on the payload verification result; generates a terminal identifier compressed field based on the terminal identifier; encodes the terminal identifier compressed field, receive flag field, and receive status field into a feedback data packet, and sends it to the base station through the main communication radio frequency link.

[0013] Furthermore, the specific steps for controlling the narrowband IoT terminal to return to deep sleep state are as follows: the preset minimal feedback unit records the transmission completion flag of the feedback data packet; the transmission channel of the main communication RF link is closed according to the transmission completion flag; the payload field, payload verification result and cache index in the low power buffer are retained; the terminal identifier compression field, reception flag field and reception status field generated in the preset minimal feedback unit are cleared; the preset minimal feedback unit is controlled to switch to sleep state, and the narrowband IoT terminal is controlled to return to deep sleep state.

[0014] A low-power data transmission system for narrowband Internet of Things (IoT) includes: a base station parsing unit, used to receive downlink data packets sent from the core network to the narrowband IoT terminal, parse the terminal identifier, service type, data length, and downlink delivery timeliness requirement, obtain the current cell network load, and query the sleep state of the narrowband IoT terminal; a dynamic resource allocation unit, used to perform terminal identifier-driven dynamic resource allocation calculation based on the terminal identifier, service type, data length, downlink delivery timeliness requirement, and current cell network load to obtain a target physical resource parameter set; and a simplified encoding and transmission unit, used to encode and reassemble the downlink data packets into a sleep downlink simplified data packet according to the data length, downlink delivery timeliness requirement, and service type, and transmit the data packet according to the target physical resource parameter set. The system sends a simplified downlink data packet to the target time-frequency location and the number of repetitions. A low-power sleep receiver unit acquires the radio frequency energy segment corresponding to the target time-frequency location during the listening duration, performs matched filtering on the radio frequency energy segment using the target reference sequence, and performs noise floor extraction, threshold offset calculation, and detection threshold adjustment to obtain the sleep downlink hit result. A low-power buffer caches the payload field, payload verification result, and cache index written by the low-power sleep receiver unit. A preset minimal feedback unit reads the payload field, payload verification result, and cache index, generates a feedback data packet according to the simplified acknowledgment feedback encoding logic, sends it to the base station via the main communication radio frequency link, and controls the narrowband IoT terminal to return to deep sleep mode.

[0015] The present invention has the following beneficial effects:

[0016] (1) The low-power data transmission method of the narrowband Internet of Things (IoT) first queries the sleep state of the narrowband IoT terminal based on the terminal identifier after the base station receives the downlink data packet sent by the core network. When the narrowband IoT terminal is in the power-saving sleep state or the extended discontinuous reception sleep state, the terminal identifier-driven dynamic resource allocation calculation is performed in combination with the service type, data length, downlink delivery time requirement and current cell network load to obtain the target time-frequency position, target reference sequence, number of repeated transmissions and listening duration. This makes the downlink data packet no longer simply wait for the fixed wake-up window to accept, but completes the transmission of the sleep downlink simplified data packet according to the target physical resource parameter group, reducing the invalid power consumption caused by the main communication radio frequency link repeatedly starting to listen in order to check uncertain downlink data.

[0017] (2) The low-power data transmission method of this narrowband Internet of Things determines the encoding rate, payload field length, check field length and wake-up discrimination field length according to the data length, downlink delivery time requirements and service type. When the ratio of the check field length to the payload field length exceeds the preset ratio range, the encoding rate is adjusted and the check field length is re-determined. This enables the dormant downlink simplified data packet to be compressed and reassembled around the wake-up discrimination field, payload field and check field, avoiding the dormant terminal from receiving too much redundant content unrelated to the current downlink payload, and reducing the processing burden of the low-power dormant receiving unit in the decoding, verification and buffering process.

[0018] (3) The low-power data transmission method of this narrowband Internet of Things collects the radio frequency energy segment corresponding to the target time-frequency position through the low-power sleep receiving unit during the listening duration, performs matched filtering on the radio frequency energy segment using the target reference sequence to obtain the matched filtering output sequence, extracts the matched filtering peak from the matched filtering output sequence, and calculates the noise floor value based on the remaining sampling points after removing the sampling points in the neighboring area of ​​the peak. Then, it generates a threshold offset and adjusts the detection threshold by combining the historical false trigger count, so that the sleep downlink hit result no longer depends on the fixed threshold discrimination, reducing the impact of noise rise, transient interference and differences in the terminal's environment on downlink signal recognition.

[0019] (4) The low-power data transmission system of this narrowband Internet of Things completes downlink data packet parsing, sleep state query, target physical resource parameter group generation and sleep downlink simplified data packet transmission through the base station parsing unit, dynamic resource allocation unit and simplified coding transmission unit. When the main communication radio frequency link and the main processor are in deep sleep state, the low-power sleep receiving unit completes radio frequency energy segment acquisition, matched filtering and detection threshold adjustment. The low-power buffer receives the effective payload field, payload verification result and buffer index. The preset minimal feedback unit generates feedback data packets and controls the terminal to restore the deep sleep state, so that the downlink triggering on the base station side, the sleep reception on the terminal side, the buffer handover and the confirmation feedback form a continuous closed loop.

[0020] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0021] Figure 1 This is a flowchart of a low-power data transmission method for narrowband Internet of Things according to the present invention.

[0022] Figure 2 This is a flowchart illustrating the specific steps involved in obtaining a hibernation downlink hit result in a low-power data transmission method for narrowband Internet of Things according to the present invention.

[0023] Figure 3 This is a block diagram of a low-power data transmission system for narrowband Internet of Things according to the present invention. Detailed Implementation

[0024] Please see Figure 1 This invention provides a technical solution: a low-power data transmission method for narrowband Internet of Things (IoT), comprising the following steps: A base station receives downlink data packets from the core network to a narrowband IoT terminal, parses the terminal identifier, service type, data length, and downlink delivery timeliness requirement, obtains the current cell network load, queries the sleep state of the narrowband IoT terminal, and when the narrowband IoT terminal is in a power-saving sleep mode or an extended discontinuous reception sleep state, performs dynamic resource allocation calculation driven by the terminal identifier based on the terminal identifier, service type, data length, downlink delivery timeliness requirement, and current cell network load to obtain a target physical resource parameter group including the target time-frequency location, target reference sequence, number of repeated transmissions, and listening duration; determines the encoding rate, payload field length, check field length, and wake-up discrimination field length according to the data length, downlink delivery timeliness requirement, and service type, encodes and reassembles the downlink data packets into a sleep downlink simplified data packet, and performs the retransmission according to the target time-frequency location and number of repeated transmissions. The system sends a simplified downlink data packet during sleep. When the main communication RF link and main processor of the narrowband IoT terminal are in deep sleep, the low-power sleep receiving unit collects the RF energy segment corresponding to the target time-frequency position during the listening duration. It performs matched filtering on the RF energy segment using the target reference sequence to obtain the matched filter peak value, extracts the noise floor value, calculates the threshold offset based on the historical false trigger count, and adjusts the detection threshold. It compares the matched filter peak value with the detection threshold to obtain the sleep downlink hit result. When the sleep downlink hit result meets the reception conditions, the low-power sleep receiving unit decodes the simplified downlink data packet, writes the payload field into the low-power buffer, and triggers the preset minimal feedback unit connected to the main processor to switch to a shallow working state. The preset minimal feedback unit reads the payload field, payload verification result, and buffer index, generates a feedback data packet according to the simplified confirmation feedback encoding logic, sends it to the base station through the main communication RF link, and controls the narrowband IoT terminal to return to deep sleep.

[0025] In this embodiment, the base station generates a target physical resource parameter group based on the terminal identifier, service type, data length, downlink delivery time requirement, and current cell network load. The terminal detects the target signal through a low-power sleep receiving unit and triggers a preset minimal feedback unit to complete the feedback after a hit, so that downlink triggering, low-power detection, buffer writing, and acknowledgment feedback form a continuous processing flow.

[0026] In one implementation, the same target resource generation rules are pre-configured on both the base station side and the narrowband IoT terminal side. The target resource generation rules include resource seed value generation rules, target reference sequence generation rules, target time-frequency position candidate rules, repeated transmission count generation rules, and listening duration generation rules. The narrowband IoT terminal saves the target resource generation rules during network access, attachment update, or parameter update, and generates an updatable candidate listening resource set based on the terminal identifier. The candidate listening resource set includes multiple sets of selectable candidate time-frequency positions, candidate reference sequences, and candidate listening durations.

[0027] When the main communication RF link and the main processor are in deep sleep, the low-power sleep receiver unit collects RF energy segments within the range limited by the candidate listening resource set and matches the corresponding candidate reference sequence to complete signal detection.

[0028] After the base station sends downlink data packets from the core network, it selects and determines the target time-frequency location, target reference sequence, number of repeated transmissions, and listening duration from the candidate monitoring resource set corresponding to the terminal based on the terminal identifier, service type, data length, downlink delivery time requirement, and current cell network load. It then generates a target physical resource parameter set, enabling the terminal to identify the dormant downlink simplified data packets sent by the base station within the range limited by the candidate monitoring resource set.

[0029] Specifically, the steps for performing dynamic resource allocation calculation driven by the terminal identifier are as follows:

[0030] Resource seed values ​​are generated based on terminal identifiers, and terminal access duration is calculated, specifically as follows:

[0031] Extract the terminal identifier string used to distinguish narrowband IoT terminals, compress and map the terminal identifier to a fixed length using a hash mapping algorithm, and generate a resource seed value corresponding to the terminal identifier. The resource seed value is used for the differentiated generation of subsequent target resources.

[0032] At the same time, the base station counts the continuous online time and historical cumulative access time of the terminal in this cell, removes the offline intermittent time, and obtains the effective access time of the terminal. The terminal access time is used to characterize the terminal's camping status in the current cell and participates in the subsequent candidate time and frequency resource screening.

[0033] In one implementation, the terminal identifier is subjected to CRC-16 hash compression to generate an integer resource seed value within a preset range, and the resource collision caused by different terminals having the same or similar resource seed values ​​is reduced by subsequent conflict avoidance processing.

[0034] The business priority level is generated based on the business type, as follows:

[0035] The downlink services of narrowband IoT terminals are classified and categorized in advance, and multiple priority levels are divided according to the urgency and latency sensitivity of the services. Services with extremely high real-time requirements, such as emergency alarms and remote power failure control, are classified as high priority, device parameter queries and status inspections are classified as medium priority, and routine data reporting receipts and log synchronization are classified as low priority. Different priorities correspond to different resource scheduling weights and transmission configuration standards.

[0036] In one implementation, the business priority classification criteria are shown in Table 1:

[0037] Table 1 Example of Business Priority Grading Standards

[0038] Level 1 (highest) Anomaly alarms, remote control, and emergency command issuance. Extremely high, with a permissible latency of ≤1s. 1.0 Level 2 (Intermediate) Parameter update, firmware upgrade, device status query Medium latency, with a permissible delay of ≤5s 0.7 Level 3 (lowest) Regular receipts, log synchronization, heartbeat calibration Low latency, allowing for a delay of ≤10s 0.4

[0039] The downlink load requirement value is generated based on the data length and downlink delivery timeliness requirements, specifically as follows:

[0040] Downlink load demand is used to quantify the intensity of the current downlink data’s demand for network resources and transmission latency. The larger the data length and the higher the delivery time requirement, the higher the load demand value and the higher the resource scheduling priority.

[0041] A load demand calculation model is constructed, which combines data length coefficient and timeliness coefficient. The calculation formula is as follows:

[0042] ;

[0043] in, This represents the downlink load demand value. The current downlink data packet length, This refers to the maximum allowed downlink single packet data length for the cell. The maximum allowable delivery delay for the business. The maximum general latency is preset for the system. For adaptive weighting coefficients, in high-time-efficiency business scenarios Value greater than This allows the downlink delivery timeliness requirement to account for a higher proportion of the downlink load requirement value, and is used to adjust the proportion of the data length coefficient and the timeliness urgency coefficient in the downlink load requirement value.

[0044] In one implementation, routine business Emergency services .

[0045] The resource congestion level is generated based on the current network load of the community, specifically as follows:

[0046] The base station uses real-time statistics on the proportion of time-frequency resources already occupied in the cell, the number of online terminals, and the length of the downlink data buffer queue. The cell resource congestion level is determined by combining these three indicators. The higher the resource occupancy rate and the longer the queue, the higher the congestion level and the more cautious the resource scheduling.

[0047] In one implementation, the resource congestion level classification criteria are as follows:

[0048] Resource utilization rate <30% is Level 1 (Idle);

[0049] Resource utilization rate between 30% and 60% is Level 2 (normal);

[0050] Resource utilization between 60% and 85% is classified as Level 3 (Busy).

[0051] Resource utilization rate > 85% is classified as Level 4 (Congestion).

[0052] Based on the resource seed value, terminal access duration, service priority level, downlink load requirement value, and resource congestion level, candidate time-frequency resources are selected from the available downlink resource set of the base station. A target time-frequency location is generated based on the candidate time-frequency resources, a target reference sequence is generated based on the resource seed value, and the number of repeated transmissions and the listening duration are generated based on the service priority level and resource congestion level, resulting in the target physical resource parameter group, which is as follows:

[0053] First, the resource seed value is used as an index to initially screen the target time-frequency resource range that matches the terminal, and the resource offset is corrected in combination with the terminal access duration.

[0054] Then, combining business priorities and downlink load requirements, priority is given to matching candidate time-frequency resources with lower resource congestion levels and those that meet downlink load requirements, avoiding time-frequency resource ranges with high resource occupancy, and selecting matching candidate time-frequency resources and fixing them as target time-frequency positions.

[0055] By relying on the resource seed value to perform pseudo-random sequence iteration, a target reference sequence corresponding to the resource seed value and with a correlation lower than a preset threshold is generated, thereby reducing the correlation interference between reference sequences of different terminals.

[0056] This embodiment sets up a calculation model that couples listening duration with retransmission count, and dynamically quantifies configuration parameters based on service priority and resource congestion level. The calculation formula is as follows:

[0057] ;

[0058] ;

[0059] in, For the final listening duration, This is the system's basic listening duration. This is the resource congestion coefficient (the higher the congestion level, the larger the coefficient, with a value ranging from 0 to 0.8). This is the business priority weighting coefficient; the higher the business priority, the larger the value of the business priority weighting coefficient.

[0060] This is the final number of repeated transmissions. The base number of transmissions, and round is the rounding function;

[0061] In high-priority services and high-congestion scenarios, the formula adaptively increases the number of data packets repeatedly sent and extends the terminal listening time. In low-priority and idle scenarios, the number of sending and listening time are automatically reduced to coordinate the degree of transmission repetition and the terminal listening time. Finally, all parameters are integrated to generate a target physical resource parameter group that is adapted to the current terminal and services.

[0062] When the resource congestion level increases, the number of repeated transmissions is increased first, and the listening duration is adjusted only within the limited range of the candidate listening resource set to avoid the listening duration from expanding indefinitely with the congestion level.

[0063] In one implementation, the base listening duration Basic number of sends In Level 4 congestion scenarios and Level 1 high-priority services, the listening duration can be adaptively increased to 45ms, and the number of retransmissions can be increased to 4.

[0064] In this implementation plan, the terminal identifier, service priority level, downlink load demand value, and resource congestion level jointly participate in the generation of the target physical resource parameter group, so that the target time-frequency location, target reference sequence, number of repeated transmissions, and listening duration can be adjusted according to the terminal status, service demand, and cell resource occupancy, thereby reducing resource mismatch caused by dormant terminals waiting for downlink data at fixed intervals.

[0065] Specifically, the conflict avoidance steps after obtaining the target physical resource parameter set are as follows:

[0066] Read the occupied time-frequency resource records and allocated reference sequence records corresponding to the current cell network load, specifically as follows:

[0067] The base station maintains a real-time ledger of cell resource occupancy. The ledger is synchronized with the current cell network load status in real time, and records the time-frequency location, resource occupancy duration, and allocated reference sequence number and sequence parameters of all online and dormant terminals.

[0068] After generating a new target physical resource parameter group, the complete data of the ledger is retrieved in real time, so that the conflict comparison is based on the current resource occupancy record;

[0069] In one implementation, the base station refreshes the resource ledger every 10ms to update the dynamic resource matching criteria.

[0070] The target time-frequency location in the target physical resource parameter group is compared with the occupied time-frequency resource records to generate a time-frequency conflict marker, which is specifically as follows:

[0071] Extract the start and end positions of the time domain and the frequency domain resource range of the target time and frequency location, and compare them one by one with the range of all occupied time and frequency resources in the ledger to determine whether there is time domain overlap, frequency domain overlap or cross coverage.

[0072] If resource overlap exists in any dimension, it is determined to be a resource conflict, and a high-priority time-frequency conflict marker is generated; if no overlap is detected, a non-conflict marker is generated.

[0073] The target reference sequence in the target physical resource parameter group is compared with the allocated reference sequence records to generate a sequence conflict marker, which is specifically as follows:

[0074] Calculate the cross-correlation coefficient between the target reference sequence and all allocated reference sequences in the cell. The higher the cross-correlation coefficient, the higher the sequence similarity and the greater the risk of signal interference.

[0075] A preset correlation coefficient threshold is set. When the cross-correlation coefficient exceeds the threshold, it is determined that there is sequence interference conflict and a sequence conflict marker is generated. If the cross-correlation coefficient does not exceed the preset threshold, a non-conflict marker is generated.

[0076] In one implementation, a preset correlation coefficient threshold of 0.2 is set, and sequences below this threshold are considered independent and free from interference.

[0077] The target time-frequency position is adjusted according to the time-frequency conflict marker, and the target reference sequence is adjusted according to the sequence conflict marker, specifically as follows:

[0078] If there is a time-frequency conflict marker, within the base station's idle resource range, the resource offset is finely adjusted based on the original resource seed value, and idle time-frequency positions with the same priority and adaptability are rematched to avoid overlapping resources.

[0079] If a sequence conflict marker exists, a new reference sequence is generated iteratively based on the original resource seed value to reduce the correlation with the already assigned sequence until the correlation is lower than a preset threshold; if a non-conflict marker is generated, the original parameters are retained unchanged.

[0080] The target physical resource parameter group is updated based on the adjusted target time-frequency location and the adjusted target reference sequence, specifically as follows:

[0081] The adjusted target time-frequency location and target reference sequence are re-entered into the parameter group. The corresponding repeated transmission count and listening duration parameters are matched and adapted synchronously, overriding the original conflict parameters. A target physical resource parameter group after conflict avoidance is generated. At the same time, the base station resource ledger is updated synchronously, and the resources in this group are reserved for the current terminal's downlink transmission.

[0082] In this implementation scheme, before sending the hibernation downlink simplified data packet, the base station performs conflict avoidance on the target time-frequency location and the target reference sequence, so that the time-frequency resource occupation and reference sequence-related interference are included in the verification scope, reducing false hits and mutual interference when multiple terminals perform hibernation detection on similar resources.

[0083] Specifically, the steps for encoding and reassembling downlink data packets into dormant downlink simplified data packets are as follows:

[0084] The payload field length is determined based on the data length, specifically as follows:

[0085] The payload field is the core business data carrying area of ​​the data packet. The field length adapts adaptively to the data length of the original downlink data packet. It maps data content to short data packets, and performs data simplification and deduplication on ultra-long and segmented business data packets, removes redundant padding bits, retains business payload data, and determines the final payload field length to avoid resource waste or data loss caused by fixed-length fields.

[0086] The encoding rate is determined based on downlink delivery timeliness requirements and business type, specifically as follows:

[0087] High-time-sensitive and high-priority services employ high coding rates to shorten the time spent on coding processing and transmission.

[0088] Low-time-sensitivity and low-priority services use a lower coding rate than high-priority services, and redundant checks are used to improve transmission reliability.

[0089] Among them, the higher the coding rate, the lower the redundancy check ratio;

[0090] In one implementation, the coding rate for Level 1 emergency services is 1 / 2, for Level 2 medium services it is 1 / 3, and for Level 3 regular services it is 1 / 4.

[0091] The length of the wake-up discrimination field is determined based on the business type, specifically as follows:

[0092] The wake-up discrimination field is used by the terminal to quickly verify the compatibility of downlink data and reduce the probability of false wake-up; a longer wake-up discrimination field is set for high-priority emergency services to improve recognition accuracy.

[0093] Low-priority routine services use short fields to reduce data packet size and transmission overhead;

[0094] In one implementation, the emergency service wake-up discrimination field is 16 bits long, while the regular service field is 8 bits long.

[0095] The length of the check field is determined based on the encoding rate and the payload field length, specifically as follows:

[0096] The check field is used to verify the integrity of the transmitted payload field. The length of the check field is determined based on the encoding rate and the length of the payload field. When the length of the payload field increases, the base length of the check field is increased. When the encoding rate increases, the proportion of the check field is decreased. This provides a basis for data verification and reduces the risk of insufficient verification under high-speed encoding.

[0097] When the ratio of the check field length to the payload field length exceeds the preset range, the encoding rate is adjusted and the check field length is redefined. Field compression and reassembly are completed in the order of wake-up discrimination field, payload field, and check field to obtain a hibernation downlink simplified data packet, which is as follows:

[0098] The preset reasonable ratio range is 0.2-0.5, which can balance the reliability of verification and the simplicity of data packets;

[0099] If the ratio of the length of the check field to the length of the payload field is greater than the upper limit of the preset ratio range, the encoding rate will be increased and the length of the check field will be shortened to reduce data packet redundancy overhead.

[0100] If the ratio of the check field length to the payload field length is less than the lower limit of the preset ratio range, the encoding rate will be reduced and the check field length will be increased to enhance the data transmission error correction capability until the check field ratio falls into the preset range, thus correcting the problem of the field ratio deviating from the preset ratio range.

[0101] This embodiment sets up encoding field balancing adjustment logic to achieve adaptive matching and adjustment of encoding rate and field ratio. The specific adjustment method is as follows:

[0102] The formula for calculating the proportional deviation coefficient is:

[0103] ;

[0104] in, This is the proportional deviation coefficient. To validate the field length, For the payload field length, The average of the preset standard ratio is 0.35 in this embodiment.

[0105] when When the value is greater than zero, it means that the proportion of the verification field is higher than the average of the preset standard proportion, which improves the encoding speed and shortens the length of the verification field.

[0106] when When the ratio is less than zero, it means that the proportion of the verification field is lower than the preset standard average ratio. The encoding rate is reduced and the length of the verification field is increased until the ratio of the length of the verification field to the length of the payload field falls into the preset ratio range.

[0107] After optimization, the wake-up discrimination field, payload field, and check field are concatenated in a fixed order, invalid padding bits are removed, and the data packet is compressed and reassembled to generate a simplified hibernation downlink data packet adapted for low-power reception of hibernation terminals.

[0108] In one implementation, when the check load ratio reaches 0.6 and the deviation coefficient is 0.25, the encoding rate is increased by a preset level, and the length of the check field is redefined.

[0109] In this implementation scheme, the field length of the hibernating downlink simplified data packet is adjusted according to the data length, service type and downlink delivery time requirements. When the ratio of the verification field length to the payload field length exceeds the preset range, the encoding rate and verification field length are re-determined to keep the downlink data packet matched between payload carrying, wake-up discrimination and verification overhead.

[0110] Specifically, the steps for the low-power sleep receiver unit to acquire radio frequency energy segments are as follows:

[0111] The low-power sleep receiver unit reads the target time-frequency location and listening duration from the target physical resource parameter group, specifically as follows:

[0112] After the terminal enters deep sleep, only the low-power sleep receiver unit remains in low-power standby mode;

[0113] In the implementation where a candidate listening resource set is pre-stored on the terminal side, the low-power sleep receiving unit determines the candidate listening resource set based on the locally stored target resource generation rules and terminal identifier, and reads the candidate listening parameters corresponding to the target time-frequency position and listening duration from the candidate listening resource set to determine the time window and frequency range for signal acquisition.

[0114] During the listening duration, a low-power acquisition channel is activated, and the target time-frequency position is locked, specifically as follows:

[0115] Unlike the full-band monitoring of the main communication RF link, the low-power acquisition channel is a narrowband directional acquisition channel. It only enables signal acquisition for the narrowband frequency band corresponding to the target time-frequency position, while all other frequency bands are shielded to reduce acquisition power consumption.

[0116] Continuously identify target time-frequency resources during the monitoring period to reduce frequency band and time slot offset.

[0117] Energy sampling is performed on the downlink signal corresponding to the target time-frequency location to obtain the original energy sampling sequence, which is as follows:

[0118] According to the system's preset fixed sampling frequency, the downlink radio frequency signal at the locked target time-frequency position is continuously sampled for energy. The time-domain energy amplitude of the signal is collected and arranged in the order of sampling time to generate the original energy sampling sequence.

[0119] The original energy sampling sequence is segmented by sliding according to the length of the target reference sequence to obtain multiple candidate radio frequency energy segments, specifically:

[0120] Using the code length of the target reference sequence as a fixed segment length, the ultra-long original energy sampling sequence is segmented by a sliding window. The sliding step size of the window is set to 1 / 2 of the sequence code length to reduce the probability of signal segments being truncated at the segment boundaries, thereby splitting the sequence into multiple candidate radio frequency energy segments of equal length that cover the entire listening window.

[0121] Amplitude normalization is performed on multiple candidate radio frequency energy segments to obtain the radio frequency energy segments, specifically:

[0122] To reduce amplitude differences caused by wireless channel fading and signal strength fluctuations, the amplitudes of all candidate radio frequency energy segments are uniformly normalized, mapping all segment amplitudes to the 0-1 range, eliminating abnormal extreme sampling points, and retaining the signal amplitude characteristics used for matched filtering, thus obtaining standardized radio frequency energy segments and providing a unified data benchmark for subsequent matched filtering detection.

[0123] In this implementation, the low-power sleep receiver only samples energy at the target time-frequency position during the listening duration, and performs sliding segmentation and amplitude normalization processing on the original energy sampling sequence according to the target reference sequence length, so that the terminal can obtain the radio frequency energy segment that can be used for matched filtering without activating the main communication radio frequency link to listen to the entire frequency band.

[0124] Specifically, such as Figure 2 As shown, the specific steps to obtain the hibernation downlink hit result are as follows:

[0125] The low-power sleep receiver unit performs matched filtering on the radio frequency energy segment using the target reference sequence to obtain the matched filter output sequence, which is as follows:

[0126] The normalized radio frequency energy segment is subjected to sliding correlation matching operation with the locally stored target reference sequence. The signal component that matches the characteristics of the target sequence is selected by the matched filtering algorithm, and irrelevant noise signals are suppressed. A continuous matched filter output sequence is output, and the amplitude of the sequence corresponds to the degree of signal matching.

[0127] The maximum output value is extracted from the matched filter output sequence as the matched filter peak value, specifically as follows:

[0128] The complete matched filter output sequence is traversed, and the maximum amplitude value in the sequence is selected. This value is the maximum signal matching peak value in the current monitoring window, which represents the matching strength of the target downlink signal and serves as the core basis for valid signal determination.

[0129] The peak neighborhood sampling points are removed from the matched filter output sequence, and the noise floor is calculated based on the remaining sampling points. Specifically:

[0130] To avoid the effective signal peak interfering with the accuracy of noise statistics, the sampling point corresponding to the peak of the matched filter is taken as the center, and a preset number of neighboring sampling points (N is preset to 5) are removed. The remaining sampling points are used as noise estimation sampling points, and the noise floor value is calculated based on the noise estimation sampling points to characterize the current channel noise level.

[0131] A threshold offset is generated based on the historical number of false triggers, and the detection threshold is adjusted using the threshold offset. Specifically:

[0132] This embodiment sets up a dynamic threshold correction mechanism, which adaptively offsets the threshold based on the number of historical false triggers, reducing false triggers or missed detections caused by fixed thresholds, and lowering the probability of false triggers and missed detections in weak signal scenarios. The calculation formula is as follows:

[0133] ;

[0134] in, This is the threshold offset. For fixed correction factors, This represents the number of recent accidental triggers. The number of times the statistics director listened;

[0135] The more false triggers there are, the greater the threshold offset becomes. Therefore, it is advisable to raise the detection threshold appropriately to reduce repeated false triggers.

[0136] When there is no historical false triggering, the offset is 0, the basic detection threshold is retained, and the weak signal detection threshold is retained;

[0137] The final detection threshold is the sum of the base threshold and the threshold offset.

[0138] By comparing the matched filter peak value with the detection threshold, when the matched filter peak value reaches the detection threshold, a hibernation downlink hit result that meets the reception conditions is generated, specifically as follows:

[0139] The real-time matched filter peak value is compared with the adaptively adjusted detection threshold;

[0140] If the peak value is greater than or equal to the detection threshold, it is determined that a valid target downlink signal has been detected, a dormant downlink hit result is generated, and the subsequent decoding and reception process is triggered.

[0141] If the peak value is less than the detection threshold, it is determined to be noise interference. The terminal remains in deep sleep mode and does not trigger the preset minimal feedback unit to switch to shallow working mode.

[0142] In this implementation, the low-power sleep receiver removes peak neighborhood sampling points and calculates the noise floor after matched filtering. Then, it adjusts the detection threshold by combining the number of historical false triggers. This ensures that the sleep downlink hit result is constrained by both the current channel noise and the terminal's historical detection performance, thereby reducing false triggers or missed detections caused by fixed thresholds in noise fluctuation scenarios.

[0143] Specifically, the steps for writing the payload field to the low-power buffer are as follows:

[0144] The low-power sleep receiver unit decodes the sleep downlink simplified data packet to obtain the wake-up discrimination field, payload field, and check field, which are as follows:

[0145] After a downlink hit occurs during hibernation, the low-power hibernation receiver unit initiates a simplified decoding logic. Unlike the complete service decoding process executed by the main processor, it only performs rapid decoding on the three segments of the simplified data packet, parsing out the independent wake-up discrimination field, the core service data payload field, and the data verification field. The decoding process is executed by the low-power hibernation receiver unit without needing to start the main processor.

[0146] The terminal identifier is verified based on the wake-up discrimination field to generate a terminal matching result, which is as follows:

[0147] The wake-up discrimination field obtained by decoding is compared bit by bit with the terminal identification code stored locally to verify the matching relationship between the wake-up discrimination field and the terminal identification code;

[0148] If a match is found, the data is determined to be valid data sent to this terminal, and a terminal matching result is generated.

[0149] If there is no match, it is determined to be interference data from neighboring cells or false detection data, a mismatch result is generated, and the data packet is discarded.

[0150] The payload field is validated based on the validation field, and a payload validation result is generated, as follows:

[0151] The integrity and accuracy of the payload field are verified by the verification field obtained through decoding, and the data is checked for problems such as packet loss, bit flipping, and data distortion.

[0152] If the verification passes, a load verification qualified result is generated; if the verification fails, a load verification unqualified result is generated, and the data is marked as abnormal.

[0153] When both the terminal matching result and the payload verification result meet the caching conditions, the payload field and the payload verification result are written to the low-power cache, specifically as follows:

[0154] When both the terminal identity matching and data verification conditions are met simultaneously, data caching operations are performed to prevent invalid or erroneous data from occupying cache resources.

[0155] The payload service data and the corresponding qualified verification results are synchronously written to the low-power cache. The low-power cache adopts an independent low-power storage area, and the write and standby power consumption is lower than that of the main storage area.

[0156] A cache index is generated for the payload field written to the low-power cache, and the cache index is passed to the preset minimum feedback unit, specifically as follows:

[0157] For each set of cached payload data, an ordered cache index is generated. The index contains data cache time, business type, and data number information, which facilitates the location and reading of data by the preset minimum feedback unit.

[0158] After the index is generated, the index information is synchronously transmitted to the preset minimal feedback unit in standby mode to complete the data handover.

[0159] In this implementation, the low-power sleep receiving unit completes terminal matching verification and payload verification before writing to the low-power buffer. Only the valid payload field, payload verification result and cache index that meet the cache conditions are written to the low-power buffer, thereby reducing the cache space occupied by false detection data or verification abnormal data and triggering the feedback process.

[0160] Specifically, the steps for generating the feedback data packet according to the simplified confirmation feedback encoding logic are as follows:

[0161] The preset minimal feedback unit reads the payload field, payload verification result, and cache index from the low-power buffer, specifically as follows:

[0162] After receiving the cache index, the preset minimal feedback unit locates and reads the corresponding payload data and payload verification results in the low-power cache area based on the index. It does not need to traverse all cache data, reducing the number of traversal reads and reading power consumption. The reading process performs cache index addressing and field reading without starting the full business processing flow of the main processor.

[0163] The receive tag field is generated based on the cache index, specifically as follows:

[0164] Based on the data number and cache time information in the cache index, a receive tag field is generated to mark the downlink data packet corresponding to the current feedback, so that the base station can match the correspondence between uplink and downlink data and avoid confusion of feedback data packets.

[0165] The receive status field is generated based on the payload verification result, specifically as follows:

[0166] Based on the pass / fail result of the payload verification, a corresponding receive status field is generated. The receive status field can be set to 1 bit to represent the receive status with a shorter field, thereby reducing the length of the feedback data packet.

[0167] A compressed terminal identifier field is generated based on the terminal identifier, specifically as follows:

[0168] The original terminal identifier is compressed and encoded, redundant fixed prefixes are removed, core identification bits are retained, and a short-byte terminal identifier compressed field is generated. This reduces the length of the feedback data packet while retaining the information required by the base station to identify the terminal.

[0169] The terminal identifier compression field, receive tag field, and receive status field are encoded into a feedback data packet and transmitted to the base station via the main communication radio frequency link. Specifically:

[0170] The three types of simplified fields are concatenated in a fixed order and then subjected to simplified channel coding to generate short feedback data packets.

[0171] The main communication radio frequency link transmission channel is opened, and the feedback data packet is sent uplink to the base station to complete the downlink data reception receipt. After the transmission is completed, the main communication radio frequency link transmission channel is closed.

[0172] In this implementation scheme, the preset minimal feedback unit reads the payload field and payload verification result according to the cache index, and encodes the terminal identifier compression field, the receive flag field and the receive status field into a feedback data packet, so that the feedback content only retains the fields required by the base station to confirm the downlink receive status, thereby reducing the length of the feedback data packet and the amount of computation in the feedback processing.

[0173] Specifically, the steps for controlling a narrowband IoT terminal to resume deep sleep mode are as follows:

[0174] The preset minimal feedback unit records the completion marker of the feedback data packet transmission, specifically as follows:

[0175] The transmission status of the main communication radio frequency link is monitored in real time. When the feedback data packet is sent, a transmission completion flag is generated. The transmission completion flag is used to trigger the shutdown of the main communication radio frequency link transmission channel and the sleep switching of the preset minimal feedback unit.

[0176] The main communication radio frequency link's transmission channel is closed based on the transmission completion flag, specifically as follows:

[0177] After detecting the completion marker, the uplink transmission channel of the main communication RF link is shut down, RF transmission is stopped, and the power consumption generated by the idling of the transmission channel is reduced, while the underlying standby circuit is preserved.

[0178] The payload field, payload verification result, and cache index from the low-power cache are retained, specifically as follows:

[0179] To facilitate business data query and subsequent processing, the low-power buffer continuously retains the received payload data, verification results and cache index without actively clearing the data. It can support local data query on the terminal and subsequent business linkage, and the buffer has low standby power consumption.

[0180] Clear the terminal identifier compression field, receive flag field, and receive status field generated in the preset minimal feedback unit, specifically as follows:

[0181] After the feedback is sent, promptly clear the terminal identifier compression field, receive flag field, and receive status field generated in the minimal feedback unit to release the unit's running memory, avoid redundant data occupying resources and increasing the unit's standby power consumption, and reset the unit's operating status.

[0182] The system controls the preset minimal feedback unit to switch to sleep mode and controls the narrowband IoT terminal to return to deep sleep mode, specifically as follows:

[0183] After completing data feedback and resource reset, the preset minimal feedback unit in the shallow working state is switched back to the sleep state, and the transmission channel and feedback processing resources started during this feedback process are shut down. This restores the terminal's main processor and main communication RF link to a deep sleep state, reduces the overall power consumption of the terminal, and completes a complete low-power data transmission process.

[0184] In this implementation scheme, the preset minimal feedback unit closes the transmission channel of the main communication radio frequency link after the feedback data packet is sent, clears the feedback field generated this time and switches back to the sleep state, so that the terminal resumes the deep sleep process after completing the downlink reception confirmation, avoiding the transmission channel and feedback processing resources from continuing to occupy power consumption after the feedback is completed.

[0185] Please see Figure 3This invention provides a technical solution: a low-power data transmission system for narrowband Internet of Things (IoT), comprising: a base station parsing unit, used to receive downlink data packets sent from the core network to the narrowband IoT terminal, parse the terminal identifier, service type, data length, and downlink delivery timeliness requirement, obtain the current cell network load, and query the sleep state of the narrowband IoT terminal; a dynamic resource allocation unit, used to perform dynamic resource allocation calculation driven by the terminal identifier based on the terminal identifier, service type, data length, downlink delivery timeliness requirement, and current cell network load, to obtain a target physical resource parameter group; and a simplified encoding and transmission unit, used to encode and reassemble the downlink data packets into a sleep downlink simplified data packet according to the data length, downlink delivery timeliness requirement, and service type, and transmit the data packet according to the target physical resource parameter group. The system sends a simplified downlink data packet to the sleep mode based on the target time-frequency location and the number of repetitions in the resource parameter group. A low-power sleep receiver unit collects the radio frequency energy segment corresponding to the target time-frequency location during the listening duration, performs matched filtering on the radio frequency energy segment using the target reference sequence, and performs noise floor extraction, threshold offset calculation, and detection threshold adjustment to obtain the sleep downlink hit result. A low-power buffer caches the payload field, payload verification result, and cache index written by the low-power sleep receiver unit. A preset minimal feedback unit reads the payload field, payload verification result, and cache index, generates a feedback data packet according to the simplified acknowledgment feedback encoding logic, sends it to the base station via the main communication radio frequency link, and controls the narrowband IoT terminal to return to deep sleep mode.

[0186] In one implementation, the base station parsing unit, the dynamic resource allocation unit, and the minimal coding transmission unit are deployed on the base station side, while the low-power sleep receiving unit, the low-power buffer, and the preset minimal feedback unit are deployed on the narrowband IoT terminal side.

[0187] The base station generates a target physical resource parameter group based on the downlink data packets sent by the core network and sends a hibernation downlink simplified data packet. When the main communication radio frequency link and the main processor are in deep hibernation, the terminal completes the target signal detection through the low-power hibernation receiving unit, completes the payload field caching through the low-power buffer, and completes the feedback data packet sending and hibernation recovery control through the preset minimal feedback unit.

[0188] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0189] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A low-power data transmission method for narrowband Internet of Things, characterized in that, Includes the following steps: The base station receives downlink data packets sent from the core network to the narrowband IoT terminal, parses the terminal identifier, service type, data length, downlink delivery timeliness requirement, and current cell network load, queries the sleep state of the narrowband IoT terminal, and when the narrowband IoT terminal is in power-saving mode sleep state or extended discontinuous reception sleep state, it performs dynamic resource allocation calculation driven by the terminal identifier based on the terminal identifier, service type, data length, downlink delivery timeliness requirement, and current cell network load to obtain a target physical resource parameter group including target time-frequency location, target reference sequence, number of repeated transmissions, and listening duration; Based on the data length, downlink delivery time requirements, and service type, determine the encoding rate, payload field length, check field length, and wake-up discrimination field length, encode and reassemble the downlink data packets into dormant downlink simplified data packets, and send the dormant downlink simplified data packets according to the target time-frequency position and the number of repetitions; When the main communication RF link and main processor of the narrowband IoT terminal are in deep sleep, the low-power sleep receiver unit collects the RF energy segment corresponding to the target time-frequency position during the listening duration. The target reference sequence is used to perform matched filtering on the RF energy segment to obtain the matched filtering peak value, the noise floor value is extracted, the threshold offset is calculated by combining the historical false trigger count and the detection threshold is adjusted, and the matched filtering peak value is compared with the detection threshold to obtain the sleep downlink hit result. When the hibernation downlink hit result meets the reception conditions, the low-power hibernation receiver unit decodes the hibernation downlink simplified data packet, writes the payload field into the low-power buffer, and triggers the preset minimal feedback unit in the main processor to switch to shallow working state. The preset minimal feedback unit reads the payload field, generates a feedback data packet according to the minimal confirmation feedback encoding logic, sends it to the base station through the main communication radio frequency link, and controls the narrowband IoT terminal to return to deep sleep state.

2. The low-power data transmission method for narrowband Internet of Things according to claim 1, characterized in that, The specific steps for performing dynamic resource allocation calculation driven by the terminal identifier are as follows: Resource seed values ​​are generated based on terminal identifiers, and terminal access duration is counted. Generate business priority levels based on business type; Generate downlink load demand values ​​based on data length and downlink delivery timeliness requirements; Resource congestion levels are generated based on the current network load of the community; Candidate time-frequency resources are selected from the available downlink resources set of the base station according to the resource seed value, terminal access duration, service priority level, downlink load demand value and resource congestion level. The target time-frequency location is generated according to the candidate time-frequency resources, the target reference sequence is generated according to the resource seed value, and the number of repeated transmissions and the listening duration are generated according to the service priority level and resource congestion level to obtain the target physical resource parameter group.

3. The low-power data transmission method for narrowband Internet of Things according to claim 1, characterized in that, The specific steps for conflict avoidance after obtaining the target physical resource parameter set are as follows: Read the occupied time-frequency resource records and allocated reference sequence records corresponding to the current cell network load; The target time-frequency location in the target physical resource parameter group is compared with the occupied time-frequency resource record to generate a time-frequency conflict marker. The target reference sequence in the target physical resource parameter group is compared with the allocated reference sequence record to generate a sequence conflict marker; Adjust the target time-frequency position according to the time-frequency conflict marker, and adjust the target reference sequence according to the sequence conflict marker; Update the target physical resource parameter group based on the adjusted target time-frequency location and the adjusted target reference sequence.

4. The low-power data transmission method for narrowband Internet of Things according to claim 1, characterized in that, The specific steps for encoding and reassembling downlink data packets into dormant downlink simplified data packets are as follows: The length of the payload field is determined based on the data length; The encoding rate is determined based on the downlink delivery timeliness requirements and the type of business. The length of the wake-up detection field is determined based on the business type; The length of the check field is determined based on the encoding rate and the payload field length; When the ratio of the length of the check field to the length of the payload field exceeds the preset ratio range, the encoding rate is adjusted and the length of the check field is redefined. Field compression and reassembly are completed in the order of wake-up discrimination field, payload field, and check field to obtain a hibernation downlink simplified data packet.

5. The low-power data transmission method for narrowband Internet of Things according to claim 1, characterized in that, The specific steps for the low-power sleep receiver unit to acquire radio frequency energy segments are as follows: The low-power sleep receiver unit reads the target time-frequency location and listening duration from the target physical resource parameter group; During the listening duration, activate the low-power acquisition channel and lock the target time-frequency position; Energy sampling is performed on the downlink signal corresponding to the target time-frequency location to obtain the original energy sampling sequence; The original energy sampling sequence is segmented by sliding according to the length of the target reference sequence to obtain multiple candidate radio frequency energy segments; Amplitude normalization is performed on multiple candidate radio frequency energy segments to obtain the radio frequency energy segments.

6. The low-power data transmission method for narrowband Internet of Things according to claim 1, characterized in that, The specific steps to obtain the hibernation downlink hit result are as follows: The low-power sleep receiver unit performs matched filtering on the radio frequency energy segment using the target reference sequence to obtain the matched filter output sequence; Extract the maximum output value from the matched filter output sequence as the matched filter peak value; Remove peak neighborhood sampling points from the matched filter output sequence, and calculate the noise floor based on the remaining sampling points; A threshold offset is generated based on the number of historical false triggers, and the detection threshold is adjusted using the threshold offset. By comparing the peak value of the matched filter with the detection threshold, when the peak value of the matched filter reaches the detection threshold, a hibernation downlink hit result that meets the reception conditions is generated.

7. The low-power data transmission method for narrowband Internet of Things according to claim 1, characterized in that, The specific steps for writing the payload field to the low-power buffer are as follows: The low-power sleep receiver unit decodes the sleep downlink simplified data packet to obtain the wake-up discrimination field, payload field and check field; The terminal identifier is verified based on the wake-up discrimination field, and a terminal matching result is generated; The payload field is validated based on the validation field, and the payload validation result is generated. When both the terminal matching result and the payload verification result meet the caching conditions, the effective payload field and the payload verification result are written to the low-power cache. Generate a cache index for the payload field written to the low-power cache and pass the cache index to the preset minimal feedback unit.

8. The low-power data transmission method for narrowband Internet of Things according to claim 1, characterized in that, The specific steps for generating the feedback data packet using the simplified confirmation feedback encoding logic are as follows: The preset minimal feedback unit reads the payload field, payload verification result, and cache index from the low-power buffer; Generate a receive tag field based on the cache index; Generate the receive status field based on the payload verification results; Generate a compressed terminal identifier field based on the terminal identifier; The terminal identifier compression field, the receive tag field, and the receive status field are encoded into a feedback data packet and sent to the base station through the main communication radio frequency link.

9. The low-power data transmission method for narrowband Internet of Things according to claim 1, characterized in that, The specific steps for controlling a narrowband IoT terminal to resume deep sleep mode are as follows: A preset minimal feedback unit records the completion marker of the feedback data packet transmission; The main communication radio frequency link's transmission channel is closed based on the transmission completion flag; Retain the payload field, payload verification result, and cache index in the low-power cache; Clear the terminal identifier compression field, receive flag field, and receive status field generated in the preset minimal feedback unit; The preset minimal feedback unit is controlled to switch to sleep mode, and the narrowband IoT terminal is controlled to return to deep sleep mode.

10. A low-power data transmission system for narrowband Internet of Things (IoT), employing the low-power data transmission method for narrowband IoT as described in any one of claims 1 to 9, characterized in that, include: The base station parsing unit is used to receive downlink data packets sent from the core network to the narrowband IoT terminal, parse the terminal identifier, service type, data length, downlink delivery time requirement and current cell network load, and query the sleep status of the narrowband IoT terminal. The dynamic resource allocation unit is used to perform dynamic resource allocation calculation driven by the terminal identifier based on the terminal identifier, service type, data length, downlink delivery time requirement and current cell network load, to obtain the target physical resource parameter group; The simplified encoding and transmission unit is used to encode and reassemble downlink data packets into dormant downlink simplified data packets according to data length, downlink delivery time requirements and service type, and to send the dormant downlink simplified data packets according to the target time-frequency location and number of repeated transmissions in the target physical resource parameter group. The low-power sleep receiver unit is used to acquire the radio frequency energy segment corresponding to the time-frequency position of the target during the listening duration, perform matched filtering on the radio frequency energy segment using the target reference sequence, and perform noise floor extraction, threshold offset calculation and detection threshold adjustment to obtain the sleep downlink hit result; The low-power buffer is used to cache the payload field, payload verification result, and cache index written by the low-power sleep receiver unit; A preset minimal feedback unit is used to read the payload field, payload verification result and cache index, generate feedback data packets according to the minimal confirmation feedback encoding logic, send them to the base station through the main communication radio frequency link, and control the narrowband IoT terminal to return to deep sleep state.

Images