A method and system for detecting a sync header that adapts to dynamic power fluctuations

Abstract

The application belongs to the technical field of synchronous head detection, and particularly relates to a synchronous head detection method and system suitable for dynamic power fluctuation, which receives a baseband I / Q signal, calculates the instantaneous power of each sampling point in real time, performs smoothing processing on the instantaneous power to obtain an in-band average power, and generates a basic threshold based on the in-band average power; a convolution correlation result of the received signal and a synchronous head template is calculated, a dynamic weighting item with time decay is constructed based on the convolution correlation result and the in-band average power, the basic threshold and the dynamic weighting item are superimposed to obtain a real-time adaptive decision threshold; a three-state machine containing a noise / search state, a peak tracking state and a decision output state is used to perform peak backtracking tracking, the peak tracking is started when the convolution correlation value first exceeds the adaptive decision threshold, the real peak position is locked and the validity decision is completed, and the synchronous position of the synchronous head detection is output.

Description

Technical Field

[0001] This invention belongs to the field of synchronization head detection technology, and particularly relates to a synchronization head detection method and system that adapts to dynamic power fluctuations. Background Technology

[0002] Existing wireless communication synchronization head detection methods mostly use fixed decision thresholds or simple threshold schemes based on average power. Fixed thresholds are difficult to adapt to dynamic changes in signal power. In weak signal scenarios, a high threshold can easily lead to missed detections, while in strong signal scenarios, a low threshold can easily cause noise misjudgments. Although simple power thresholds can adapt to some power fluctuations, they do not take into account the time-domain characteristics of the relevant peaks and cannot suppress the premature triggering of the relevant peak leading edge in strong signals, which in turn leads to deviations in the synchronization position.

[0003] In wireless transmission environments with rapidly fluctuating power, synchronization head detection always faces the core contradiction of high detection probability for weak signals and high synchronization position accuracy for strong signals. Existing adaptive threshold schemes have high computational complexity, large hardware resource consumption, and lack peak backtracking verification mechanisms. They are difficult to balance detection performance and synchronization accuracy under conditions of low signal-to-noise ratio and strong power fluctuations, and cannot meet the engineering application requirements of high-speed real-time communication systems. Summary of the Invention

[0004] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a synchronization head detection method and system adapted to dynamic power fluctuations, the method comprising: Receive baseband I / Q signals, calculate the instantaneous power at each sampling point in real time, smooth the instantaneous power to obtain the in-band average power, and generate a basic threshold based on the in-band average power; The convolutional correlation result between the received signal and the synchronization header template is calculated. Based on the convolutional correlation result and the in-band average power, a dynamic weighting term with band time decay is constructed. The base threshold and the dynamic weighting term are superimposed to obtain the real-time adaptive decision threshold. A three-state machine including noise / search state, peak tracking state, and decision output state is used to perform peak backtracking. When the convolution correlation value exceeds the adaptive decision threshold for the first time, peak tracking is started, the true peak position is locked, the validity decision is completed, and the synchronization position detected by the synchronization head is output.

[0005] Preferably, the formula for calculating the instantaneous power is: ; in This represents the sampled value of the baseband I-channel signal. This represents the sampled value of the baseband Q-channel signal. This is the sampling point number.

[0006] Preferably, the smoothing process is a first-order low-pass filter or a process with a length of [missing information]. Moving average processing; The formula for calculating the basic threshold is: ; in Based on the basic threshold, To accommodate the lowest detection signal-to-noise ratio, a fixed coefficient is used. This represents the average power within the band.

[0007] Preferably, constructing the dynamic weighting term includes: calculating the difference between the convolution correlation result and the average power after the shift. ,in For convolution-related results, This is the preset offset coefficient; Weighted increments are generated based on differences. ,when hour ,otherwise , These are weighting coefficients; The weighted increment is subjected to time decay filtering to obtain the decayed weighted term. , It is an attenuation factor less than 1.

[0008] Preferably, the calculation formula for the real-time adaptive decision threshold is as follows: .

[0009] Preferably, the workflow of the three-state machine is as follows: Noise / Search State: Continuously monitor convolution correlation values ​​and adaptive decision threshold, when Record temporary peak values ​​and peak positions, start the timer and switch to peak tracking mode; Peak tracking state: Compare the current convolutional correlation value with the temporary peak value, update the peak value and position or accumulate the timer, and switch to the decision output state when the timer reaches the preset hold window length; Decision output state: Verify the validity of the peak value. If the condition is met, output the peak value position as the synchronization position, and then reset to the noise / search state.

[0010] Preferably, the length of the holding window is slightly greater than the length of the synchronization header or the width of the main lobe of the relevant peak, and more preferably 1.5 times the width of the main lobe of the relevant peak.

[0011] A synchronization head detection system adaptable to dynamic power fluctuations, comprising: The power and basic threshold module is used to receive baseband I / Q signals, calculate instantaneous power and smooth it to obtain in-band average power, and generate basic threshold. The adaptive threshold module is used to calculate the convolution-related results, construct a dynamic weighting term with time decay, and superimpose them to obtain the real-time adaptive decision threshold; The peak backtracking state machine module is configured with three states: noise / search state, peak tracking state, and decision output state. It performs peak tracking, validity decision, and outputs the synchronization position.

[0012] Preferably, the power and basic threshold module uses a first-order low-pass filter to achieve instantaneous power smoothing, and the basic threshold is generated based on a fixed coefficient that adapts to the lowest detection signal-to-noise ratio and the in-band average power.

[0013] Preferably, the peak backtracking state machine module includes a peak storage unit, a timer, and a comparator, and is implemented using an FPGA.

[0014] Based on the above, by using real-time power smoothing processing and dynamic weighted threshold design with time decay, it can quickly adapt to dynamic fluctuations in signal power. In weak signal scenarios, the decision threshold is close to the basic power threshold, ensuring a high detection probability under low signal-to-noise ratio and eliminating missed detections of weak signals. In strong signal scenarios, the dynamic weighting term raises the threshold in real time, effectively suppressing premature triggering of the correlation peak front, avoiding noise misjudgment and false synchronization, and significantly improving detection robustness.

[0015] The three-state machine peak backtracking mechanism is adopted, which abandons the direct judgment of the first threshold crossing point, accurately locks the real peak position, completely eliminates the strong signal synchronization position deviation, and significantly improves the synchronization positioning accuracy. The overall computational complexity of the solution is low. The core processing only relies on shift, addition and simple comparison operations. The hardware can be directly adapted to FPGA implementation, with low resource consumption and low processing latency, which can meet the synchronization head detection engineering requirements of high-speed real-time wireless communication systems. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the execution flow of the synchronization head detection method adapted to dynamic power fluctuations provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a synchronization head detection system adapted to dynamic power fluctuations provided in an embodiment of the present invention; Figure 3 This is a system flowchart for synchronization head detection adapted to dynamic power fluctuations provided in an embodiment of the present invention. Detailed Implementation

[0017] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating a synchronization head detection method adapted to dynamic power fluctuations provided in an embodiment of the present invention. The following is a detailed description of this synchronization head detection method adapted to dynamic power fluctuations.

[0018] This technical solution addresses the core contradiction in wireless synchronization head detection: the threshold needs to be low enough to detect weak signals, while avoiding premature triggering of the rising edge of the correlation peak for strong signals. Based on the original three-step threshold adjustment approach, a low-complexity peak backtracking confirmation mechanism is added. This achieves a balance between high detection probability for weak signals and high accuracy in synchronization position for strong signals, while maintaining low overall complexity and ease of FPGA implementation. The specific technical solution is as follows: The base threshold is based on the real-time power of the received signal to ensure weak signal detection capability in low signal-to-noise ratio scenarios. The specific steps are as follows: For the received baseband I / Q signal, the instantaneous power at each sampling point is calculated in real time using the following formula: ,in This represents the sampled value of the baseband I-channel signal. This represents the sampled value of the baseband Q-channel signal. This is the sampling point number.

[0019] For instantaneous power Perform a length of The in-band average power can be obtained by either using a moving average or a first-order low-pass filter (preferably a first-order low-pass filter to reduce FPGA implementation complexity and resource consumption). ; where the moving average length Alternatively, the low-pass filter coefficient can be set reasonably according to the sampling rate and signal bandwidth of the actual communication system to ensure that instantaneous power fluctuations can be effectively smoothed and the overall power level of the signal can be reflected.

[0020] Based on the in-band average power generation threshold, the calculation formula is as follows: ,in It is a fixed coefficient.

[0021] Note: Fixed coefficient The threshold should be set according to the weakest signal the system needs to detect (e.g., signal-to-noise ratio SNR = -10dB). The core purpose is to ensure a high detection probability in low SNR scenarios and avoid missing weak signals. This setting will result in a relatively low base threshold in strong signal scenarios, which may lead to premature triggering when receiving high SNR signals (e.g., SNR > 3dB) before the rising edge of the relevant peak reaches the true peak value, thus providing an optimization target for subsequent dynamic weighted compensation and peak backtracking mechanisms.

[0022] By utilizing the difference between the convolution correlation result and the average power, a weighted value with time decay is constructed to dynamically compensate the basic threshold in real time, suppressing premature triggering of the correlation peak leading edge of strong signals without affecting the detection of weak signals. The specific steps are as follows: The difference between the convolution correlation result and the average power after the shift is calculated using the following formula: ; in To determine the convolution correlation results between the received signal and the synchronization header template, This is a preset offset coefficient used to adjust the sensitivity of the difference and avoid the weighting term being mistakenly increased due to noise fluctuations.

[0023] Constructing a weighted increment: Set a conditional function, if This indicates that the convolution correlation result is significantly higher than the current average power level, suggesting the possible arrival of a signal, at which point a weighted increment is generated. ;like This indicates that there is currently no valid signal or only noise exists; in this case, the weighted increment... ;in The weighting coefficient is used to control the magnitude of the weighting increment, preventing the threshold from being raised too high and causing weak signals to be missed.

[0024] Weighted term time decay processing: The weighted increment is subjected to time decay filtering, and the calculation formula is as follows: ;in The attenuation factor is less than 1, preferably ranging from 0.9 to 0.95, to achieve a slow attenuation of the weighting term.

[0025] Function Description: When a strong signal (high signal-to-noise ratio) arrives, the in-band power increases significantly, affecting the convolution correlation results. It can suddenly spike, causing the difference to... Rapidly increasing, weighted increment This increases accordingly, thus affecting the weighted terms after decay. The threshold is rapidly raised in real time, acting as a "brake" to prevent synchronous decision-making before the rising edge of the correlation peak reaches its peak value. Once the correlation peak exceeds its peak value, the convolutional correlation result gradually decreases, and the weighted increment... When the weighted term becomes 0, the weighted term... According to the decay factor The threshold should gradually decrease to ensure that subsequent signal detection is not affected, while also preventing the threshold from remaining high for an extended period and causing missed detections.

[0026] The base threshold is added to the attenuated dynamic weighted term to obtain the real-time adaptive decision threshold, enabling the threshold to respond quickly to changes in signal power. The specific calculation formula is as follows: ; The core advantage of this adaptive threshold lies in: in weak signal scenarios, the convolution-related results... Lower, difference ≤0, weighted term When the threshold is close to 0, the decision threshold is approximately equal to the base threshold. This ensures that weak signals can successfully pass the threshold and be detected; in strong signal scenarios, the weighting term... The rapid rise in threshold increases the decision threshold, suppressing premature triggering of the relevant peak front, while the attenuation characteristics prevent the threshold from being too high and affecting subsequent detection.

[0027] By employing a state machine and peak backtracking design, and implementing it within an FPGA with extremely low hardware resources, the contradiction between "weak signal detection and accurate synchronization of strong signal positioning" is completely resolved. The specific implementation is as follows: The state machine consists of three states: noise / search state (state 0), peak tracking state (state 1), and decision output state (state 2). The workflow and logic of each state are as follows: State 0 (Noise / Search State); This is the system's default initial state, and its core function is to continuously monitor convolution-related results. With real-time judgment threshold Size relationship: When detected Instead of immediately determining the synchronization position (to avoid position deviation caused by premature triggering of the leading edge of a strong signal), the following operations are performed: Record the convolution correlation value of the current sampling point as a temporary peak value. Simultaneously record the current sampling point position. ; Start timer and initialize it to 0; The system switches from state 0 to state 1 (peak tracking state).

[0028] If it is always detected If so, the system remains in state 0 and continues to perform search and monitoring.

[0029] State 1 (Peak Tracking State); The core function of this state is to track changes in convolution-related peaks and pinpoint the true peak locations. The specific operation is as follows: At each new sampling point The system continuously compares the current convolutional correlation values. With temporary peak : like This indicates that the current sampling point is a new peak point, and the temporary peak value is updated. Update peak position At the same time, the timer Reset (restart timing to ensure complete tracking of the falling edge after the peak); like This indicates that the current sampling point has either passed the peak value or has not yet reached the peak value, and the timer will stop. Accumulate (increment the timer by 1 for each sampled point); When the timer Reach the preset window length When the time is reached, it indicates that the relevant peak has been fully experienced (the peak value has passed, and the subsequent relevant value continues to decrease), the system exits state 1 and switches to state 2 (decision output state).

[0030] State 2 (Decision Output State); The core function of this state is to determine the validity of the tracked peak and output the final synchronization position. The specific operation is as follows: Peak validity check: Checks the final tracked temporary peaks. Does it meet the requirements? ;in Selectable peak position The corresponding real-time threshold value can also be the last threshold value when exiting state 1. Prioritize the threshold value corresponding to the peak position to ensure the accuracy of the judgment.

[0031] Synchronization position output: If the peak value meets the validity condition, output the synchronization position. The system detects the synchronization head by displaying a valid synchronization flag. If the validity condition is not met, the peak value is discarded and no synchronization signal is output.

[0032] State Reset: Regardless of whether a synchronization signal is output, after the decision is completed, the system returns to state 0 (noise / search state) to prepare for the next synchronization head detection.

[0033] Maintain window length The setting should be slightly larger than the length of the synchronization head or the width of the main lobe of the relevant peak (preferably 1.5 times the width of the main lobe of the relevant peak) to ensure complete coverage of the rising edge, peak value and falling edge of the relevant peak, and to avoid failure to track the true peak value due to the window being too short, or excessive detection delay due to the window being too long.

[0034] Position deviation elimination: This mechanism allows the state machine to be triggered when the relevant peak first crosses the threshold on the rising edge (peak ramp), but does not rely on the first crossing point as the synchronization position. Instead, it locks the true maximum value position by continuously tracking, completely eliminating the synchronization position deviation caused by crossing the threshold in advance, and ensuring the accuracy of the synchronization position in strong signal scenarios.

[0035] Weak signal detection guarantee: In low signal-to-noise ratio scenarios, the probability of missed detection is mainly determined by a fixed coefficient of the basic threshold. The decision is made because the peak backtracking mechanism can correct the synchronization position, so there is no need to worry. The issue of premature triggering caused by setting the value too low can be avoided by setting it too low. Set the value to adapt to the weakest signal detection (e.g., adapt to SNR=-10dB) to ensure a high detection probability for weak signals.

[0036] The overall processing flow of this method is clear and the steps are coherent, enabling real-time high-speed detection. The specific process is as follows: System initialization: Set fixed coefficients Offset coefficient Weighting coefficients Attenuation factor Maintain window length , moving average length (or low-pass filter coefficients), the state machine is initialized to state 0 (noise / search state), and the timer... Zeroing out, temporary peak and peak position initialization.

[0037] Signal reception and preprocessing: Receive baseband I / Q samples and calculate the instantaneous power of each sample point in real time. Simultaneously, the convolution correlation value between the received signal and the synchronization header template is calculated. .

[0038] Average power and baseline threshold update: for instantaneous power Perform a moving average or first-order low-pass filter to update the in-band average power. and according to Generate the basic threshold.

[0039] Dynamic weighting and adaptive threshold update: Calculating the difference Weighted increment Obtained through time decay The real-time decision threshold is obtained by adding the basic threshold to the weighted terms. .

[0040] Peak backtracking decision: convolution correlation values and real-time judgment threshold The peak backtracking state machine is fed into the peak tracking state machine, and peak tracking and decision-making are performed according to the logic of state 0, state 1, and state 2.

[0041] Synchronization output and reset: After the relevant peak ends, the state machine outputs the final synchronization position and valid indication (if the peak value is valid). Then the state machine is reset to state 0, and the above steps are repeated to achieve continuous synchronization head detection.

[0042] This technical solution fundamentally solves the core contradiction in wireless synchronization head detection through a three-layer design of basic threshold, dynamic weighting, and peak backtracking. The principles for resolving each contradiction are as follows: The core solution is a peak backtracking mechanism: it does not use "the first time the convolution correlation value exceeds the threshold" as a synchronization decision criterion, but only as a signal to start peak tracking; it continuously tracks the changes in the convolution correlation peak through a state machine to lock the true peak position, and the final output of the synchronization position is the peak position. It is independent of the threshold change and the time of the first threshold crossing, so the accuracy of the synchronization position is minimally affected by the threshold change, completely solving the position deviation problem caused by the early triggering of strong signals.

[0043] The core safeguards include the optimized setting of the basic threshold and the adaptive adjustment of the dynamic weighting terms: the fixed coefficient of the basic threshold. Unconstrained by strong signals, it can be optimized to the lowest detection signal-to-noise ratio required by the system (e.g., SNR = -10dB), ensuring that weak signals can be detected even if they exceed the threshold; in weak signal scenarios, the convolutional correlation value... Lower, difference ≤0, dynamically weighted term With a threshold close to 0, the adaptive threshold is approximately equal to the base threshold, preventing weak signals from being missed due to the increase in weighting terms, thus ensuring a high detection probability for weak signals.

[0044] This solution adheres to the principle of low complexity throughout and is adapted to the high-speed processing requirements of FPGA: the state machine requires only a small number of registers (to store temporary peak values, peak positions, and timer values) and comparators (to compare convolutional correlation values ​​with thresholds and temporary peak values), without the addition of additional multipliers, resulting in extremely low hardware resource consumption; power calculation, moving average, first-order low-pass filtering, and dynamically weighted attenuation processing can all be implemented using shifting and addition, avoiding complex calculations and ensuring that the system can meet the real-time detection requirements in high-speed sampling scenarios.

[0045] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

[0046] The embodiments of this application have been described above with reference to the accompanying drawings. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. This application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A synchronization head detection method adaptable to dynamic power fluctuations, characterized in that: Includes the following steps: Receive baseband I / Q signals, calculate the instantaneous power at each sampling point in real time, smooth the instantaneous power to obtain the in-band average power, and generate a basic threshold based on the in-band average power; The convolutional correlation result between the received signal and the synchronization header template is calculated. Based on the convolutional correlation result and the in-band average power, a dynamic weighting term with band time decay is constructed. The base threshold and the dynamic weighting term are superimposed to obtain the real-time adaptive decision threshold. A three-state machine including noise / search state, peak tracking state, and decision output state is used to perform peak backtracking. When the convolution correlation value exceeds the adaptive decision threshold for the first time, peak tracking is started, the true peak position is locked, the validity decision is completed, and the synchronization position detected by the synchronization head is output.

2. The synchronization head detection method adapting to dynamic power fluctuations according to claim 1, characterized in that: The formula for calculating the instantaneous power is: ； in This represents the sampled value of the baseband I-channel signal. This represents the sampled value of the baseband Q-channel signal. This is the sampling point number.

3. The synchronization head detection method and system adapting to dynamic power fluctuations according to claim 1, characterized in that: The smoothing process is a first-order low-pass filter or a length of... Moving average processing; The formula for calculating the basic threshold is: ; in Based on the basic threshold, To accommodate the lowest detection signal-to-noise ratio, a fixed coefficient is used. This represents the average power within the band.

4. The synchronization head detection method and system adapting to dynamic power fluctuations according to claim 1, characterized in that: Constructing the dynamic weighting term includes: calculating the difference between the convolution correlation result and the average power after the shift. ,in For convolution-related results, This is the preset offset coefficient; Weighted increments are generated based on differences. ,when hour ,otherwise , These are weighting coefficients; The weighted increment is subjected to time decay filtering to obtain the decayed weighted term. , It is an attenuation factor less than 1.

5. The synchronization head detection method and system adapting to dynamic power fluctuations according to claim 1, characterized in that: The calculation formula for the real-time adaptive decision threshold is as follows: .

6. The synchronization head detection method and system adapting to dynamic power fluctuations according to claim 1, characterized in that: The workflow of the three-state machine is as follows: Noise / Search State: Continuously monitor convolution correlation values ​​and adaptive decision threshold, when Record temporary peak values ​​and peak positions, start the timer and switch to peak tracking mode; Peak tracking state: Compare the current convolutional correlation value with the temporary peak value, update the peak value and position or accumulate the timer, and switch to the decision output state when the timer reaches the preset hold window length; Decision output state: Verify the validity of the peak value. If the condition is met, output the peak value position as the synchronization position, and then reset to the noise / search state.

7. The synchronization head detection method and system adapting to dynamic power fluctuations according to claim 1, characterized in that: The length of the holding window is slightly greater than the length of the synchronization header or the width of the main lobe of the relevant peak, preferably 1.5 times the width of the main lobe of the relevant peak.

8. A synchronization head detection system adaptable to dynamic power fluctuations, characterized in that: include: The power and basic threshold module is used to receive baseband I / Q signals, calculate instantaneous power and smooth it to obtain in-band average power, and generate basic threshold. The adaptive threshold module is used to calculate the convolution-related results, construct a dynamic weighting term with time decay, and superimpose them to obtain the real-time adaptive decision threshold; The peak backtracking state machine module is configured with three states: noise / search state, peak tracking state, and decision output state. It performs peak tracking, validity decision, and outputs the synchronization position.

9. The synchronization head detection method and system adapting to dynamic power fluctuations according to claim 1, characterized in that: The power and basic threshold module uses a first-order low-pass filter to achieve instantaneous power smoothing. The basic threshold is generated based on a fixed coefficient that adapts to the lowest detection signal-to-noise ratio and the in-band average power.

10. The synchronization head detection method and system for adapting to dynamic power fluctuations according to claim 1, characterized in that: The peak backtracking state machine module includes a peak storage unit, a timer, and a comparator, and is implemented using an FPGA.

Images