A low-power control method for intercoms

Abstract

This invention relates to the field of wireless communication technology, and more particularly to a low-power control method for walkie-talkies, comprising the following steps: In idle state, the radio frequency transceiver alternately performs discontinuous channel monitoring according to sleep duration and wake-up window; when the signal strength reaches a trigger threshold, the baseband processor is woken up; the baseband processor sequentially performs multi-level signal decision-making, progressively activating functional units as needed; when a decision fails, only the newly activated functional units of this node are reverted to sleep state; the channel interference statistics are updated according to the decision results, the sleep duration of the next round is adaptively adjusted, and a forced wake-up counter ensures call response capability. This invention enables the average power consumption of the walkie-talkie in idle state to dynamically adjust with changes in channel interference intensity, effectively suppressing the cumulative impact of invalid wake-up events on the overall standby power consumption without affecting normal call response capability.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a low-power control method for walkie-talkies. Background Technology

[0002] Walkie-talkies need to maintain channel monitoring capability for extended periods in standby mode while minimizing static power consumption to ensure battery life. The operating states of a walkie-talkie can be categorized into four types: transmitting mode (sending voice signals to other terminals, highest power consumption); receiving mode (receiving voice signals from the source terminal, audio circuitry is active, relatively high power consumption); idle mode (waiting to initiate or receive calls and continuously monitoring the channel, medium power consumption); and sleep mode (major components are off, extremely low power consumption, unable to transmit or receive). The idle mode is the core scenario for power optimization. Its fundamental challenge lies in balancing the need for complete system shutdown to maintain channel monitoring capability with the requirement to minimize continuous power consumption. Therefore, periodic sleep-wake mechanisms are commonly used in engineering to address this issue.

[0003] In the periodic sleep-wake mechanism, the RF transceiver briefly wakes up at fixed intervals to detect channel signal power and decide whether to wake up the baseband processor for further decision-making. The system remains in deep sleep for the rest of the time. Building on this, a multi-level wake-up decision framework further divides the baseband processor's wake-up decision into two consecutive stages: valid signal detection and terminal address matching. Normal reception only begins after both stages pass, reducing audio circuit misfiring caused by noise and irrelevant signals. Essentially, the periodic sleep-wake mechanism is a discontinuous channel monitoring scheduling mechanism, where the walkie-talkie terminal saves power during standby by periodically shortening the channel monitoring window. However, this mechanism has systemic flaws in both the wake-up decision execution and cross-cycle scheduling levels, leading to the accumulation of ineffective power consumption during discontinuous monitoring.

[0004] Chinese patent CN111294833A discloses a terminal and signal processing method, using a walkie-talkie as a typical application scenario. This patent discloses a collaborative energy-saving control scheme for an RF transceiver and a baseband processor: The RF transceiver, in a wake-up state, performs power detection on the received RF signal. The baseband processor is only awakened for subsequent processing when the power detection result reaches a preset power threshold; otherwise, the RF transceiver directly enters sleep mode. After being awakened, the baseband processor sequentially performs valid signal detection and terminal address matching detection. If any stage fails, both the baseband processor and the RF transceiver enter sleep mode. Only after all stages pass does the audio circuit activate for voice output. This scheme effectively reduces the probability of false triggering of the audio circuit through a multi-level early exit mechanism. However, in actual interference environments, a failed wake-up event still triggers the complete power-on process of the baseband link, causing a significant increase in invalid power consumption as interference events accumulate. Simultaneously, the fixed monitoring scheduling cycle means that the wake-up frequency cannot be reduced during periods of persistent channel interference, and the overall idle power consumption of the system remains constrained by the interference intensity, making it difficult to maintain an ideal low-power level in complex electromagnetic environments. Summary of the Invention

[0005] In view of this, the present invention provides a low-power control method for walkie-talkies to solve the problems in the prior art where the baseband link is fully powered on due to a failed wake-up event in the idle state, and the wake-up frequency is too high in interference environments due to the fixed monitoring scheduling cycle.

[0006] The technical solution of this invention is implemented as follows: On one hand, the present invention provides a low-power control method for a walkie-talkie, comprising the following steps: S1. When the walkie-talkie is in idle state, it alternates between the current sleep duration and the wake-up window to perform discontinuous channel monitoring. In the wake-up window, the radio frequency transceiver detects the signal strength of the current channel. If the trigger threshold is not reached, the wake-up event ends and the radio frequency transceiver returns to sleep state, proceeding to S3. If the trigger threshold is reached, the baseband processor is woken up and proceeding to S2. S2. After the baseband processor is woken up, it executes multi-level signal decisions sequentially according to the preset decision node order. For each decision node, it gradually activates the corresponding subset of functional units as needed, and uses the activated functional units to execute the decision task of this node. If a decision node fails, it only rolls back the newly activated functional units of this node to the sleep state, records the decision failure level of this wake-up event, exits the multi-level signal decision, and returns the RF transceiver to the sleep state, proceeding to S3. If all decision nodes pass, it enters the normal receiving state. After the call ends, it shuts down all functional units and returns the RF transceiver to the sleep state, proceeding to S3. S3. Based on the multi-level decision results of this wake-up event, update the statistics reflecting the channel interference status, and adaptively adjust the next round of sleep duration according to the updated statistics; when the cumulative number of sleep extensions reaches the preset upper limit, force a listening cycle with the default sleep duration to be executed once.

[0007] Based on the above technical solutions, preferably, step S1 specifically includes: S11. When the walkie-talkie is idle, it enters deep sleep according to the current sleep duration. The initial value of the sleep duration is the preset default sleep duration. The wake-up window duration is fixed for each round. After the sleep ends, the radio frequency transceiver automatically enters the wake-up window. S12. Within the wake-up window, the RF transceiver uses the received signal strength indicator energy detection method to detect the signal strength of the current channel and compares the detected value with a preset squelch threshold. If the detected value is lower than the preset squelch threshold, the RF transceiver directly returns to the sleep state, the wake-up event ends, and proceeds to S3. If the detected value is not lower than the preset squelch threshold, the RF transceiver sends a wake-up request to the baseband processor, the baseband processor exits the deep sleep state, and proceeds to S2.

[0008] Based on the above technical solutions, preferably, step S2 specifically includes: S21. The forced activation set and optional activation set pre-configured in the design phase for each decision node are used as inputs to this step, and the two satisfy mutual exclusion constraints. When the baseband processor is woken up, the node counter is initialized to the first-level decision node, and the set of activated functional units is set to empty. For the current decision node, all the functional units that have not yet been activated in its forced activation set are activated, and the received signal quality is lightly estimated using the currently activated functional units to obtain the normalized signal quality estimate. S22. Based on the normalized signal quality estimate, calculate the activation cost-effectiveness index for each functional unit in the optional activation set of the current decision node, and activate the optional functional units whose cost-effectiveness index is not lower than the preset activation decision threshold; the forced activation set and the selected optional functional units together constitute the actual activation set of this node. S23. The new portion of the actual activated set relative to the set of activated functional units of the previous node is recorded as the activation increment set. All functional units in the actual activated set execute the decision task corresponding to this node. If the decision fails, only the functional units in the activation increment set are rolled back to the sleep state, while the state of the activated functional units of the previous node remains unchanged. The decision failure level of this wake-up event is recorded, and the multi-level signal decision is exited. If the decision passes, the node counter is incremented to the next level decision node, and S21 to S23 are repeated until all decision nodes pass.

[0009] Based on the above technical solution, preferably, the forced activation set of each decision node in step S21 satisfies the monotonically increasing constraint, that is, the forced activation set of the k-th decision node... satisfy Where K is the total number of decision nodes; the normalized signal quality estimate is obtained by performing a lightweight channel signal-to-noise ratio estimate on the current received signal: ; in, Let be the normalized signal quality estimate for the k-th decision node. Forcible activation set The signal-to-noise ratio estimate output after lightweight feature extraction, in dB; Set a minimum usable signal-to-noise ratio threshold for the system; Set a preset upper limit for signal-to-noise ratio saturation for the system; This means that the result is limited to the range [0,1].

[0010] Based on the above technical solution, preferably, in step S22, for each functional unit in the optional activation set... Calculate the activation cost-effectiveness index The method is as follows: ; in, The current signal quality is insufficient; Functional unit Operating power consumption; The nominal duration of the decision for decision node k; Functional unit The contribution of the decision node k to improving the decision reliability is determined by the following formula: ; in, For containing functional units The probability of decision pass at decision node k at a typical operating point with low signal-to-noise ratio. For using only the forced activation set The probability of a decision passing at the same working point for decision node k; With preset activation decision threshold Comparison, When Included in the activation set of this node. hour Keep dormant.

[0011] Based on the above technical solutions, preferably, in step S23, the decision tasks corresponding to each decision node are determined according to the node level: when the decision node is at level 1, it performs valid signal detection, detecting whether the received signal has a valid modulation format and valid synchronization information in digital communication mode, and detecting whether the received signal is a valid analog frequency modulation signal in analog communication mode, so as to distinguish between real call signals and channel noise; when the decision node is at level 2, it performs terminal address matching, parsing the destination address field in the signal frame in digital communication mode and comparing it with the local preset address, and detecting and extracting the sub-tone signaling in the received signal in analog communication mode with sub-tone signaling configured and matching it with the local preset sub-tone parameters; subsequent decision nodes perform corresponding decision tasks by analogy according to service requirements.

[0012] Based on the above technical solutions, preferably, after each round of wake-up events, the scheduling logic on the RF transceiver side updates the scheduling parameters according to the multi-level decision results of this wake-up event. S3 further includes: S31. Using the judgment result information of this wake-up event as input, construct a binary observation, and use the exponentially weighted moving average method to update the channel interference probability estimate to obtain the updated interference probability estimate. S32. Based on the updated disturbance probability estimate, perform a linear mapping between the default sleep duration and the maximum allowed sleep duration to calculate the next round of adaptive sleep duration. S33. Maintain the forced wake-up counter. Each time step S32 is completed and the resulting adaptive sleep duration exceeds the default sleep duration, the counter is incremented once. When the accumulated value of the counter reaches the preset forced wake-up interval threshold, a standard wake-up detection cycle using the default sleep duration is forcibly executed once and the counter is cleared to zero. Then, the adaptive sleep duration of this round is written to the scheduling register and the process returns to S1.

[0013] Based on the above technical solutions, preferably, the binary observation in step S31... The definition is: when the first-level decision node fails to make a decision in this wake-up event. The system returns to sleep directly if the RSSI does not reach the trigger threshold; the decision nodes at level 2 and subsequent levels fail; or the system enters normal reception after all decision nodes pass. The update method for the channel interference probability estimate is as follows: ; in, For the first n The channel interference probability estimate obtained after the next wake-up event. Initial value; This is the exponential forgetting factor.

[0014] Based on the above technical solutions, the preferred method for calculating the adaptive sleep duration in step S32 is as follows: ; in, The default sleep duration is preset. The maximum sleep duration extension factor is determined during the design phase based on the ratio of the maximum allowed sleep duration to the default sleep duration. Follow exist Continuous variation within a range.

[0015] Based on the above technical solutions, preferably, the preset forced wake-up interval threshold in step S33 is determined in the design stage according to the maximum call response delay allowed by the service layer, so that when the adaptive sleep duration is configured with the default sleep duration and the maximum sleep duration extension multiple, the time interval between two adjacent forced wake-ups does not exceed the maximum allowed call response delay under the worst-case scenario, wherein the maximum allowed call response delay is determined according to the requirements of the walkie-talkie protocol standard for call setup delay.

[0016] The present invention has the following advantages over the prior art: This invention provides a low-power control method for walkie-talkies. By introducing a node-level progressive activation control mechanism and an adaptive scheduling mechanism within a discontinuous channel monitoring framework, the average power consumption of the walkie-talkie in idle state can be dynamically adjusted according to changes in channel interference intensity. Without affecting normal call response capability, it effectively suppresses the cumulative impact of invalid wake-up events on the standby power consumption of the entire device.

[0017] This invention dynamically determines the activation range of selectable functional units at the entry point of each decision node based on the current channel signal-to-noise ratio estimation result, constraining the actual activation set size of each decision node to the minimum range required to complete the decision task of this node. When a decision fails, only the newly activated functional units of this node are rolled back to a dormant state. Subsequent functional units and high-power independent power supply components such as vocoders and audio codecs remain powered off until all multi-level decisions are passed, thus making the baseband link activation power consumption in an invalid wake-up event significantly lower than the power consumption level of the entire link when fully powered on.

[0018] This invention uses the failure result of the first-level decision node as the observation of the channel interference probability, and employs an exponentially weighted moving average method to perform a lightweight estimation at the RF transceiver side, linearly mapping the estimation result to the next round of sleep duration. This process does not require waking up the baseband processor and introduces no additional power consumption. During the period of persistent interference, the sleep duration automatically increases with the increase of the interference probability, and the wake-up frequency decreases accordingly, enabling the system to proactively reduce the number of invalid wake-ups in interference environments. When the interference disappears, the sleep duration automatically recovers to the default value as the estimated value decays, without affecting normal monitoring response capabilities.

[0019] This invention establishes deterministic constraints on the maximum call response delay by using a forced wake-up counter and a forced wake-up interval threshold. This ensures that even under continuous strong interference scenarios and when the sleep duration is extended to its maximum value, the maximum time interval between two adjacent forced standard wake-ups still does not exceed the call setup delay requirement specified in the protocol standard. This allows the adaptive sleep extension mechanism to have a controllable worst-case response guarantee in actual deployment. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of the low-power control method for the walkie-talkie of the present invention; Figure 2 This is a schematic diagram of the timing of discontinuous channel monitoring according to the present invention; Figure 3 This is a schematic diagram of the multi-level signal decision node structure of the present invention; Figure 4 This is a schematic diagram of the adaptive sleep duration adjustment feedback mechanism of the present invention. Detailed Implementation

[0022] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0023] like Figure 1 As shown, the present invention provides a low-power control method for walkie-talkies, comprising the following steps: S1. When the walkie-talkie is in idle state, it alternates between the current sleep duration and the wake-up window to perform discontinuous channel monitoring. In the wake-up window, the radio frequency transceiver detects the signal strength of the current channel. If the trigger threshold is not reached, the wake-up event ends and the radio frequency transceiver returns to sleep state, proceeding to S3. If the trigger threshold is reached, the baseband processor is woken up and proceeding to S2. S2. After the baseband processor is woken up, it executes multi-level signal decisions sequentially according to the preset decision node order. For each decision node, it gradually activates the corresponding subset of functional units as needed, and uses the activated functional units to execute the decision task of this node. If a decision node fails, it only rolls back the newly activated functional units of this node to the sleep state, records the decision failure level of this wake-up event, exits the multi-level signal decision, and returns the RF transceiver to the sleep state, proceeding to S3. If all decision nodes pass, it enters the normal receiving state. After the call ends, it shuts down all functional units and returns the RF transceiver to the sleep state, proceeding to S3. S3. Based on the multi-level decision results of this wake-up event, update the statistics reflecting the channel interference status, and adaptively adjust the next round of sleep duration according to the updated statistics; when the cumulative number of sleep extensions reaches the preset upper limit, force a listening cycle with the default sleep duration to be executed once.

[0024] The low-power control logic of this invention uses discontinuous channel monitoring scheduling as its main framework, consisting of two cooperating layers with a cross-layer feedback channel established between them. The first layer is a system-level discontinuous channel monitoring scheduling mechanism, responsible for scheduling the sleep-wake cycle of the RF transceiver. The second layer is a node-level progressive activation control mechanism, which, after the baseband processor is woken up, gradually activates the required set of functional units based on the functional dependencies of each decision node. The cross-layer feedback channel transmits the multi-level decision results from step S2 to step S3, driving dynamic updates of the scheduling parameters. Together, these three elements constitute a complete discontinuous monitoring power-saving control method.

[0025] The functional unit referred to in this invention is the smallest processing module within a baseband processor that can independently control its power supply and clock, including but not limited to a channel demodulation unit, frame synchronization detection unit, address matching unit, channel equalization unit, and forward error correction decoding unit. The set of all functional units within the baseband processor is denoted as […]. The enumeration and embedding of functional units are completed by the firmware during system initialization; the set of functional units that are active at any given time is denoted as... When the system is powered on .

[0026] In one embodiment of the present invention, such as Figure 2 As shown, Figure 2The dual-line timing axis represents the operating states of the RF transceiver and the baseband processor, clearly presenting two typical branches: "Sleep → Wake-up Window → RSSI Judgment → Sleep / Continue Reception". Step S1 includes: S11. When the walkie-talkie is idle, it enters deep sleep according to the current sleep duration. The initial value of the sleep duration is the preset default sleep duration. The wake-up window duration is fixed for each round. After the sleep ends, the radio frequency transceiver automatically enters the wake-up window. When the walkie-talkie is idle, it will remain in sleep mode for the current duration. Entering deep sleep, The initial value is the preset default sleep duration. (Unit: ms), subsequently updated dynamically by the cross-layer feedback mechanism in step S3; duration of each wake-up window. (Unit: ms) Fixed and determined during the design phase; current complete sleep-wake cycle duration is During sleep, the RF transceiver shuts down the receiving link and is in an extremely low power state; the baseband processor remains in a deep sleep state until the RF transceiver sends a wake-up request; after the sleep duration ends, the RF transceiver automatically enters the wake-up window.

[0027] S12. Within the wake-up window, the RF transceiver uses Received Signal Strength Indication (RSSI) energy detection to detect the signal strength of the current channel and obtain the received signal strength measurement value. (Unit: dBm), compare it with the preset squelch threshold. (Unit: dBm) is compared; if the detected value is lower than the preset squelch threshold, the RF transceiver directly returns to sleep mode, the current wake-up event ends, and the current binary observation is recorded as... If the detected value is not lower than the preset squelch threshold, the RF transceiver sends a wake-up request to the baseband processor, and the baseband processor exits the deep sleep state and proceeds to S2.

[0028] In one embodiment of the present invention, such as Figure 3 As shown, step S2 includes: S21. The forced activation set and optional activation set pre-configured in the design phase for each decision node are used as inputs for this step, and the two satisfy mutual exclusion constraints. When the baseband processor is woken up, the node counter is initialized to the first-level decision node, and the set of activated functional units is set to empty. For the current decision node, all the functional units that have not yet been activated in its forced activation set are activated, and the received signal quality is lightly estimated using the currently activated functional units to obtain the normalized signal quality estimate.

[0029] Specifically, during the design phase, for each decision node k (K is the total number of decision nodes) A ​​forced activation set is pre-configured. With an optional activation set Both are then permanently stored in the system firmware. Forced activation set It is the set of the smallest functional units that decision node k depends on to complete its decision task, and all of them must be activated regardless of the current channel conditions. Taking decision node 1 (valid signal detection) as an example, It typically includes a channel demodulation unit and a frame synchronization detection unit; taking decision node 2 (terminal address matching) as an example, exist This typically includes an address matching unit. The forced activation set of each decision node satisfies a monotonically increasing constraint. This constraint reflects the engineering characteristic of functional dependencies in the walkie-talkie's forward pipeline decision architecture, which only increases and never decreases, and is a design premise for simplifying incremental activation management logic. Optional activation set This is a set of auxiliary functional units that can improve the decision reliability of decision node k under the current low signal-to-noise ratio (SNR) conditions, but do not need to be activated when the SNR is sufficiently good. Typical units in the optional activation set include the channel equalization unit (which can significantly improve symbol decision accuracy when activated under low SNR conditions, and can be bypassed under high SNR conditions) and the soft-decision forward error correction decoding unit (which has significant soft-decision gain under low SNR conditions, and hard decision can meet the frame error rate requirements under high SNR conditions). The forced activation set and the optional activation set satisfy a mutual exclusion constraint. .

[0030] When the baseband processor is woken up, the node counter is initialized. Activated functional unit set The previous node has already activated the snapshot set. For the current decision node, the set will be forcibly activated. Activate all previously inactive functional units and update. The received signal quality is lightly estimated using the currently activated functional units, specifically by performing a coarse signal-to-noise ratio measurement of the channel to obtain a normalized signal quality estimate. The calculation method is as follows: ; in, Forcible activation set The signal-to-noise ratio estimate output after lightweight feature extraction, in dB; Set a minimum usable signal-to-noise ratio threshold for the system, in dB; The system has a preset upper limit for signal-to-noise ratio saturation, in dB. This means that the result is limited to the range [0, 1]. This stage only performs fast signal-to-noise ratio estimation and does not perform the signal analysis processing required for complete decision tasks such as frame synchronization detection, in order to minimize the power consumption of the signal quality estimation stage.

[0031] S22. Based on the normalized signal quality estimate, calculate the activation cost-effectiveness index for each functional unit in the optional activation set of the current decision node, and activate the optional functional units whose cost-effectiveness index is not lower than the preset activation decision threshold; the forced activation set and the selected optional functional units together constitute the actual activation set of this node.

[0032] For each functional unit within the optional activation set Calculate the activation cost-effectiveness index The method is as follows: ; in, Functional unit The contribution of decision node k to improving decision reliability is obtained during the design phase through channel simulation or actual measurement: measurements are taken at typical operating points with low signal-to-noise ratios. Participation with "and" The probability of a decision being passed by node k under the two conditions is calculated by the following formula, then normalized and stored: ; in, For containing functional units The probability of decision pass at decision node k at a typical operating point with low signal-to-noise ratio. For using only the forced activation set The probability of a decision passing when decision node k is at the same working point; The current signal quality is insufficient; when the signal-to-noise ratio is sufficiently good... , The activation cost-effectiveness of optional units is close to zero, so the system will not activate them automatically. Functional unit The operating power consumption (unit: mW) is determined and stored in the chip specifications during the design phase. The nominal decision duration (in milliseconds) for decision node k is determined and stored during the design phase. The unit is mJ -1 In physical terms, this refers to the improvement in decision reliability resulting from unit activation energy consumption. With preset activation decision threshold (Unit: mJ) -1 Compare: If Then Activate and include in the activation set of this node; if ,but Remain dormant. The forced activation set and the filtered optional functional units together constitute the actual activation set of this node. During the design phase, the reliability and power consumption of the decision are determined and solidified based on a comprehensive trade-off between the walkie-talkie service and the requirements. Reasonable values ​​can be determined by testing different parameters under typical interference environments. The trade-off between power saving and the decrease in decision pass rate under the given value is obtained through simulation or actual measurement.

[0033] S23. The new portion of the actual activated set relative to the set of activated functional units of the previous node is recorded as the activation increment set. All functional units in the actual activated set execute the decision task corresponding to this node. If the decision fails, only the functional units in the activation increment set are rolled back to the sleep state, while the state of the activated functional units of the previous node remains unchanged. The decision failure level of this wake-up event is recorded, and the multi-level signal decision is exited. If the decision passes, the node counter is incremented to the next level decision node, and S21 to S23 are repeated until all decision nodes pass.

[0034] The portion of the actual activated set that is newer than the set of activated functional units in the previous node is denoted as the activation increment set: ; in, To activate the incremental set, This is a snapshot of the active set before entering the current node. Functional units already activated in the preceding nodes will not be repeated. The decision task corresponding to this node will be executed by all functional units in the actually activated set. The decision task is determined by node level: when the decision node is level 1, it performs valid signal detection, checking whether the received signal has a valid modulation format and valid synchronization information in digital communication mode, and whether the received signal is a valid analog FM signal in analog communication mode, to distinguish between real call signals and channel interference noise; when the decision node is level 2, it performs terminal address matching, parsing the destination address field in the signal frame in digital communication mode and comparing it with the local preset address, and detecting and extracting the Continuous Audio Coded Squelch System (CTCSS) signal or Digital Coded Squelch (DCS) signal in the received signal in analog communication mode with sub-tone signaling configured, and matching it with the local preset sub-tone parameters; subsequent decision nodes will execute corresponding decision tasks analogously according to service requirements.

[0035] If the judgment fails, only the incremental set will be activated. The functional units in the middle fall back to the sleep state, that is The process maintains the state of the activated functional units in the preceding node, records the failure level of the current wake-up event, and then exits the multi-level signal decision process. The essence of power saving here lies in the fact that when the decision fails, the subsequent functional units and high-power independent power supply components such as the vocoder and audio codec remain unpowered, saving the activation energy itself, not the operating energy after power-on. If the decision succeeds, a snapshot of the current node is recorded. The node counter is incremented to the next level decision node, and S21 to S23 are repeated until all decision nodes pass.

[0036] After all decision nodes pass, the system determines that the current call is a valid signal specific to the local machine, and sequentially activates the vocoder and audio codec circuits, while simultaneously activating all remaining functional units within the baseband chip. The complete signal chain (RF front-end → baseband → vocoder → audio codec → speaker) is ready and enters normal reception state. The binary observation corresponding to this wake-up is denoted as... After the call ends, if the system detects channel silence or a signal released by the other end, it will perform a status reset in the following order: the audio circuit is turned off; all functional units inside the baseband processor sequentially enter sleep mode. The baseband processor returns to deep sleep mode; the RF transceiver carries the current effective sleep duration. Resume the periodic sleep-wake process of step S1 and wait for the next call event.

[0037] In this invention, the node-level progressive activation control mechanism is based on the hardware functional dependencies of each decision node. Channel condition assessment and functional unit activation decisions are pre-positioned at the entry point of each decision node. Through cost-effectiveness filtering, the actual activation set size is compressed to the range required to complete the decision task of this node. The incremental backoff design in case of decision failure ensures that activated functional units of preceding nodes do not need to be powered on repeatedly, while high-power independent power supply components such as vocoders and audio codecs remain off until multiple decision stages are passed. This effectively reduces the average power consumption of a single wake-up in scenarios with high interference false trigger rates.

[0038] In one embodiment of the present invention, such as Figure 4 As shown, step S3 includes: S31. Using the decision result information of this wake-up event as input, construct a binary observation and use the exponentially weighted moving average method to update the channel interference probability estimate to obtain the updated interference probability estimate.

[0039] Using the judgment result information of this wake-up event as input, determine the binary observable. (n is the sequence number of the current wake-up event): When the first-level decision node fails to make a decision in this wake-up event. The system returns to sleep directly if the RSSI does not reach the trigger threshold; the decision nodes at level 2 and subsequent levels fail; or the system enters normal reception after all decision nodes pass. . The physical meaning is as follows: the RF transceiver detects a signal strength exceeding the squelch threshold, but the baseband processor, after being woken up, determines that the signal does not have a valid modulation format. This is a channel interference-induced false trigger, the root cause of the accumulation of invalid power consumption across cycles during discontinuous listening. Although the failure of the second-level and subsequent decision nodes is an invalid wake-up, it has passed the first-level valid signal detection, indicating that there is a real valid format signal in the channel, and it is not due to simple channel interference. Therefore, it is taken as... To avoid misjudging normal phenomena such as call / selection mismatch as interference events, an exponentially weighted moving average method is used to update the channel interference probability estimate. ; in, This is the channel interference probability estimate obtained after the nth wake-up event, with an initial value. ; The exponential forgetting factor controls the decay rate of historical observation weights. It is determined and fixed during the design phase based on the typical timescales of the occurrence and disappearance of interference in the electromagnetic environment of the walkie-talkie's actual use. The typical value range is... When the interference persists, Take 1 consecutively, Monotonically increasing towards 1; when the disturbance disappears, Take 0 consecutively, It decays at an exponential rate toward zero, and can automatically track the dynamic changes in channel interference status.

[0040] S32. Based on the updated disturbance probability estimate, perform a linear mapping between the default sleep duration and the maximum allowed sleep duration to calculate the next round of adaptive sleep duration.

[0041] With the updated disturbance probability estimate Based on this, calculate the next round of adaptive sleep duration: ; in, The adaptively adjusted sleep duration in the nth round of monitoring cycle. The default sleep duration is set in milliseconds (ms). To extend the maximum sleep duration multiple, the following conditions must be met: ,in This represents the maximum allowed duration of a single sleep session. Follow exist Continuous variation within the range: Sleep duration remains at default value This corresponds to the shortest sleep duration and the highest wake-up response capability in a non-interference scenario; Sleep duration extended to The wake-up frequency is reduced to a minimum to minimize the number of invalid wake-ups in interference environments. The changes are gradual due to the smoothing effect of EWMA, and will not cause drastic jumps due to fluctuations in a single observation.

[0042] S33. Maintain the forced wake-up counter. Each time step S32 is completed and the resulting adaptive sleep duration exceeds the default sleep duration, the counter is incremented once. When the accumulated value of the counter reaches the preset forced wake-up interval threshold, a standard wake-up detection cycle using the default sleep duration is forcibly executed once and the counter is cleared to zero. Then, the adaptive sleep duration of this round is written to the scheduling register and the process returns to S1.

[0043] The scheduling logic on the RF transceiver side maintains the forced wake-up counter. The initial value is 0. Each time S32 is executed and (That is, when the calculation result indicates a prolonged sleep state) Add 1; whenever a wake-up event passes through the first-level decision node, Reset to 0 (indicating a valid call signal has appeared in the channel, and the interference estimate has been accurately verified). When Reaching the preset forced wake-up interval threshold At any time, regardless of the current situation Regardless of the value, the system will force a single execution using the default sleep duration. The standard wake-up detection cycle, and Reset to 0. During the design phase, based on the maximum allowable call response delay (Determined based on the call setup delay requirements of the walkie-talkie protocol standard) It must meet the following requirements: ; This constraint guarantees that, in the worst-case scenario (using the longest sleep duration each time), the maximum time interval between two consecutive forced awakenings does not exceed [a certain value]. This ensures that the response latency of real calls under continuous strong interference scenarios meets the requirements of walkie-talkie services. After completing the above updates, Write to the RF transceiver's scheduling register, replacing the current effective sleep duration, and update the current complete cycle duration. ,in The duration of each wake-up window for the RF transceiver is specified. The system returns to S1 to begin the next sleep-wake cycle.

[0044] In this invention, the failure result of the first-level decision node is used as the observation signal of the channel interference state. The tracking estimation of the interference probability and the adaptive adjustment of the sleep duration are completed on the RF transceiver side with extremely low computational overhead, without waking up the baseband processor and without generating additional power consumption. The forced wake-up protection mechanism incorporates the maximum call response delay into the design space of the scheduling parameters in the form of constraints, maintaining a deterministic engineering constraint relationship between interference suppression and call response capability.

[0045] 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 low power control method of a intercom, characterized by, Includes the following steps: S1. When the walkie-talkie is in idle state, it alternates between the current sleep duration and the wake-up window to perform discontinuous channel monitoring. In the wake-up window, the radio frequency transceiver detects the signal strength of the current channel. If the trigger threshold is not reached, the wake-up event ends and the radio frequency transceiver returns to sleep state, proceeding to S3. If the trigger threshold is reached, the baseband processor is woken up and proceeding to S2. S2. After the baseband processor is woken up, it executes multi-level signal decisions in sequence according to the preset decision node order. For each decision node, it gradually activates the corresponding functional unit subset as needed and uses the activated functional units to execute the decision task of this node. If a decision node fails, only the newly activated functional unit of this node is rolled back to the sleep state, the failure level of this wake-up event is recorded, the multi-level signal decision is exited, and the RF transceiver is returned to the sleep state, and then proceeds to S3; if all decision nodes pass, the normal receiving state is entered, all functional units are turned off after the call ends, and the RF transceiver is returned to the sleep state, and then proceeds to S3. S3. Based on the multi-level decision results of this wake-up event, update the statistics reflecting the channel interference status, and adaptively adjust the next round of sleep duration according to the updated statistics; when the cumulative number of sleep extensions reaches the preset upper limit, force a listening cycle with the default sleep duration to be executed once.

2. The low power control method of a intercom according to claim 1, wherein, Step S1 specifically includes: S11. When the walkie-talkie is idle, it enters deep sleep according to the current sleep duration. The initial value of the sleep duration is the preset default sleep duration. The wake-up window duration is fixed for each round. After the sleep ends, the radio frequency transceiver automatically enters the wake-up window. S12. Within the wake-up window, the RF transceiver uses the received signal strength indicator energy detection method to detect the signal strength of the current channel and compares the detected value with a preset squelch threshold. If the detected value is lower than the preset squelch threshold, the RF transceiver directly returns to the sleep state, the wake-up event ends, and proceeds to S3. If the detected value is not lower than the preset squelch threshold, the RF transceiver sends a wake-up request to the baseband processor, the baseband processor exits the deep sleep state, and proceeds to S2.

3. The low power control method of a intercom according to claim 1, wherein, Step S2 specifically includes: S21. The forced activation set and optional activation set pre-configured in the design phase for each decision node are used as inputs to this step, and the two satisfy mutual exclusion constraints. When the baseband processor is woken up, the node counter is initialized to the first-level decision node, and the set of activated functional units is set to empty. For the current decision node, all the functional units that have not yet been activated in its forced activation set are activated, and the received signal quality is lightly estimated using the currently activated functional units to obtain the normalized signal quality estimate. S22. Based on the normalized signal quality estimate, calculate the activation cost-effectiveness index for each functional unit in the optional activation set of the current decision node, and activate the optional functional units whose cost-effectiveness index is not lower than the preset activation decision threshold; the forced activation set and the selected optional functional units together constitute the actual activation set of this node. S23. The new portion of the actual activated set relative to the set of activated functional units of the previous node is recorded as the activation increment set. All functional units in the actual activated set execute the decision task corresponding to this node. If the decision fails, only the functional units in the activation increment set are rolled back to the sleep state, while the state of the activated functional units of the previous node remains unchanged. The decision failure level of this wake-up event is recorded, and the multi-level signal decision is exited. If the decision passes, the node counter is incremented to the next level decision node, and S21 to S23 are repeated until all decision nodes pass.

4. The low power control method of a intercom according to claim 3, wherein The forced activation set of each decision node in step S21 satisfies a monotone increasing constraint, that is, the forced activation set of the kth decision node satisfies satisfies where K is the total number of decision nodes; The normalized signal quality estimate is obtained by performing a lightweight channel signal-to-noise ratio estimate on the current received signal: ； in, Let be the normalized signal quality estimate for the k-th decision node. Forcible activation set The signal-to-noise ratio estimate output after lightweight feature extraction, in dB; Set a minimum usable signal-to-noise ratio threshold for the system; Set a preset upper limit for signal-to-noise ratio saturation for the system; This means that the result is limited to the range [0,1].

5. The low power control method of a intercom according to claim 4, wherein, In step S22, the function units in the optional active set are activated The method for calculating the activation cost performance index is as follows: ； in, To activate the cost-performance ratio indicator, This is the normalized signal quality estimate for the k-th decision node; Functional unit Operating power consumption; The nominal duration of the decision for decision node k; Functional unit The contribution of the decision node k to improving the decision reliability is determined by the following formula: ； in, For containing functional units The probability of decision pass at decision node k at a typical operating point with low signal-to-noise ratio. For using only the forced activation set The probability of a decision passing at the same working point for decision node k; With preset activation decision threshold Comparison, When Included in the activation set of this node. hour Keep dormant.

6. The low power control method of a intercom according to claim 3, wherein, In step S23, the decision task corresponding to each decision node is determined according to the node level: when the decision node is level 1, it performs valid signal detection, detects whether the received signal has a valid modulation format and valid synchronization information in digital communication mode, and detects whether the received signal is a valid analog frequency modulation signal in analog communication mode, so as to distinguish between real call signals and channel noise. When the decision node is at level 2, terminal address matching is performed. In digital communication mode, the destination address field in the signal frame is parsed and compared with the local preset address. In analog communication mode with sub-tone signaling configured, sub-tone signaling is detected and extracted from the received signal and matched with the local preset sub-tone parameters. Subsequent decision nodes will perform corresponding decision tasks according to business requirements.

7. The low power control method of a intercom according to claim 1, wherein, After each wake-up event, the scheduling logic on the RF transceiver side updates the scheduling parameters based on the multi-level decision results of this wake-up event. S3 further includes: S31. Using the judgment result information of this wake-up event as input, construct a binary observation, and use the exponentially weighted moving average method to update the channel interference probability estimate to obtain the updated interference probability estimate. S32. Based on the updated disturbance probability estimate, perform a linear mapping between the default sleep duration and the maximum allowed sleep duration to calculate the next round of adaptive sleep duration. S33. Maintain the forced wake-up counter. Each time step S32 is completed and the resulting adaptive sleep duration exceeds the default sleep duration, the counter is incremented once. When the accumulated value of the counter reaches the preset forced wake-up interval threshold, a standard wake-up detection cycle using the default sleep duration is forcibly executed once and the counter is cleared to zero. Then, the adaptive sleep duration of this round is written to the scheduling register and the process returns to S1.

8. The low power control method of a intercom according to claim 7, wherein, The definition of the binary observation quantity in step S31 is: when the first-level decision node fails to make a decision in the current wake-up event , the RSSI does not reach the trigger threshold to return to sleep, the second-level and subsequent decision nodes fail to make a decision, or all decision nodes pass, in which case the device enters normal reception. The channel interference probability estimate is updated as follows: ； wherein is the channel interference probability estimate value obtained after the n first wake-up event, , is the initial value; is the exponential forgetting factor, , is the binary observation, and n is the sequence number of the current wake-up event.

9. The low power control method of a intercom according to claim 8, wherein, The adaptive sleep duration is calculated in step S32 as follows: ； in, The adaptively adjusted sleep duration in the nth round of monitoring cycle. The default sleep duration is preset. The maximum sleep duration extension factor is determined during the design phase based on the ratio of the maximum allowed sleep duration to the default sleep duration. Follow exist Continuous variation within a range.

10. The low-power control method for a walkie-talkie as described in claim 7, characterized in that, The preset forced wake-up interval threshold in step S33 is determined during the design phase based on the maximum call response delay allowed by the service layer. This ensures that, under the worst-case scenario, the time interval between two adjacent forced wake-ups does not exceed the maximum allowed call response delay when the adaptive sleep duration is configured using the default sleep duration and the maximum sleep duration extension multiple. The maximum allowed call response delay is determined based on the call setup delay requirements of the walkie-talkie protocol standard.

Images