A method for optimizing signal transmission paths in an integrated RF terminal chip

Abstract

This invention discloses a signal transmission path optimization method for an integrated RF terminal chip, relating to the field of RF technology. The invention includes: S100 real-time parameter acquisition, calculation of mismatch loss factor, impedance deviation factor, and interference rise factor; S200 weighted fusion of the three factors to obtain a disabling coefficient, constructing a disabling coefficient curve, performing curve feature analysis to generate a reconstruction index, and comparing it with a threshold to determine whether reconstruction is necessary; S300 when reconstruction is required, querying the target control word of the matching network and filter based on a multi-dimensional lookup table, and performing priority correction based on interference and power occupancy; S400 executing control word writing, verifying the effect, and if it does not meet the standard, performing local search and fine-tuning until performance is restored. This invention achieves adaptive optimization of signal transmission paths in multiple frequency bands and scenarios through a closed-loop dynamic reconstruction process of perception, decision-making, execution, and verification, effectively reducing mismatch loss and interference impact, and improving the overall performance of the RF terminal chip.

Description

Technical Field

[0001] This invention relates to the field of radio frequency technology, and more specifically, to a method for optimizing the signal transmission path of an integrated radio frequency terminal chip. Background Technology

[0002] In an integrated RF terminal chip, signal transmission refers to the entire process of RF signals being generated inside the chip, passing through the chip's internal circuitry, package interconnection structures, and printed circuit board traces, and finally reaching the antenna, or the signals received from the antenna being transmitted back to the chip for processing. This path involves multiple physical links such as pads, leads, bumps, substrate wiring, package pins, and microstrip lines on the PCB. The signal propagates in the form of electromagnetic waves in these heterogeneous media and structures, and its quality directly determines the performance of the communication system.

[0003] As terminal devices increasingly demand support for multi-band and multi-mode communication, if a dedicated, optimized RF path is designed for each frequency band, key components such as filters, switches, and matching networks must be provided as a set. This leads to an explosive increase in the number of RF front-end devices, a significant increase in package size, and a substantial rise in power consumption and manufacturing and testing costs. It also exacerbates the complexity of packaging and antenna co-design and mass production consistency issues. Conversely, while using a single broadband or compromise RF path can reduce the number of devices and simplify implementation, it is difficult to maintain low reflection and low insertion loss at all operating frequencies due to the frequency dependence of matching networks and filters, as well as the temperature drift and environmental sensitivity of devices. This results in decreased receiver sensitivity, loss of transmit efficiency, and difficulty in intermodulation control, thereby affecting system performance and spectrum compliance. Summary of the Invention

[0004] The main objective of this invention is to provide a signal transmission path optimization method for an integrated radio frequency terminal chip, so as to overcome the problems mentioned in the background art.

[0005] To achieve the above objectives, a signal transmission path optimization method for an integrated RF terminal chip is provided, comprising:

[0006] S100 collects chip operating status parameters in real time and calculates key performance indicators of the current signal transmission path based on the collected parameters. The key performance indicators include mismatch loss factor, impedance deviation factor and interference rise factor. The operating status parameters include the current operating frequency, forward and reverse power of the antenna port, transmit power, chip temperature, complex reflection coefficient, reflection coefficient amplitude and interference intensity.

[0007] S200: The mismatch loss factor, impedance deviation factor and interference rise factor are weighted and fused to obtain the disabling coefficient, and a reconstruction index is generated based on the time series curve of the disabling coefficient. The reconstruction index is compared with a preset reconstruction threshold. If it is greater than or equal to the threshold, dynamic reconstruction is triggered.

[0008] S300: When dynamic reconfiguration is triggered, the target control word of the adjustable matching network and the adjustable filter is determined by querying a multidimensional lookup table based on the current working state parameters.

[0009] S400: Write the target control word into the corresponding device to complete the reconstruction and verify the disabling coefficient after reconstruction. If the requirements are not met, perform a local search adjustment until the performance is restored or the search is exited.

[0010] Furthermore, the calculation process for the key performance indicators includes:

[0011] The forward and reverse power of the antenna port at the current operating frequency are collected. The reverse power is divided by the forward power and the result is resized to obtain the reflection coefficient amplitude, denoted as r(t), where t represents the acquisition time index. The reflection coefficient amplitude is then calculated using the formula... The mismatch loss M(t) is calculated; a maximum mismatch loss threshold is set, and the mismatch loss is normalized using the maximum mismatch loss threshold to obtain the mismatch loss factor: when the mismatch loss is less than the maximum mismatch loss threshold, the mismatch loss factor = mismatch loss divided by the maximum mismatch loss threshold; when the mismatch loss is greater than or equal to the maximum mismatch loss threshold, the mismatch loss factor = 1; the range of the mismatch loss factor is [0,1], which represents the degree of power loss caused by the current antenna mismatch;

[0012] The complex reflection coefficients at various times are obtained using a vector coupler or a six-port network. The equivalent antenna impedance is then calculated using transmission line theory. The calculation formula is as follows: Z0 is the system characteristic impedance; the impedance plate deviation is obtained by subtracting the system characteristic impedance from the equivalent antenna impedance and taking the absolute value; a maximum allowable impedance deviation threshold is set, and the impedance deviation is normalized using the maximum allowable impedance deviation threshold to obtain the impedance deviation factor: when the impedance deviation is less than the maximum allowable impedance deviation threshold, the impedance deviation factor = impedance deviation divided by the maximum allowable impedance deviation threshold; when the impedance deviation is greater than or equal to, the impedance deviation factor = 1; the range of the impedance deviation factor is [0,1], which represents the degree to which the current antenna impedance deviates from the ideal matching value;

[0013] The theoretical thermal noise floor N0 under the current channel bandwidth is calculated based on the interference intensity I(t). The calculation formula is: N0 = -174 + 10 × log 10(B), where B is the bandwidth of the current channel in Hertz; the interference rise ΔI(t) is defined as the interference intensity I(t) minus the theoretical thermal noise floor N0, representing the increment of the actual interference power relative to the theoretical noise floor; if the interference intensity is less than the theoretical thermal noise floor, then ΔI(t) is set to zero; a maximum allowable interference rise threshold is set, and the interference rise is normalized using the maximum allowable interference rise threshold: when the interference rise is less than the maximum allowable interference rise threshold, then the interference rise factor = interference rise divided by the maximum allowable interference rise threshold; when the interference rise is greater than or equal to the maximum allowable interference rise threshold, then the interference rise factor = 1; the range of the interference rise factor is [0,1], which characterizes the degree of degradation of the receiving link performance by the current interference.

[0014] Furthermore, the step of generating the remodeling index based on the time series curve of the disability coefficient includes:

[0015] The arithmetic mean of the disability coefficients at each time point is calculated to obtain the mean disability value. The maximum value of the disability coefficients at each time point is taken as the maximum disability value. The slope of the disability coefficient curve is further fitted by the least squares method to obtain the performance change rate. The non-negative part of the performance change rate is taken and normalized. The non-negative part of the normalized performance change rate, the mean disability value, and the maximum disability value are then linearly weighted and fused to obtain the reconstruction index.

[0016] Furthermore, the target control word for determining the tunable matching network and the tunable filter includes:

[0017] The system acquires the current operating frequency, chip temperature, equivalent antenna impedance, interference rise factor, and transmit power. When the interference rise factor exceeds the preset interference threshold, significant interference is determined to exist. The baseband processor performs a fast Fourier transform analysis on the received signal and extracts the frequency corresponding to the maximum power peak in the spectrum as the interference frequency.

[0018] A pre-saved multidimensional lookup table is configured. Using the current operating frequency, temperature, and antenna impedance as indices, the system searches the multidimensional lookup table and outputs an adjustable matching network control word that minimizes mismatch loss and maximizes transmission efficiency. Then, using the current operating frequency and interference frequency as indexes, a search is performed in a multidimensional lookup table to output the adjustable filter control word that maximizes interference suppression in the receiving channel while minimizing insertion loss of the useful signal. ;

[0019] The corresponding notch mode control word is retrieved from a multidimensional lookup table based on the interference frequency. And with its covered adjustable filter control word ,Right now = The power utilization rate is obtained by dividing the transmit power by the maximum output power corresponding to the linear region of the power amplifier. A preset utilization threshold is set. When the power utilization rate is greater than or equal to the utilization threshold, the fine-tuning amount of the matching network is read from the multidimensional lookup table. And based on this, the basic control word After correction, the final matching network control word is: = + When the power utilization rate is less than the utilization threshold, no linearity correction is required, and the final matching network control word directly adopts the basic control word. = ;

[0020] Therefore, the final adjustable matching network control word can be obtained. and adjustable filter control word , as the target control word.

[0021] Furthermore, the target control word is written to the corresponding device to complete the reconstruction, and the disabling factor after reconstruction is verified:

[0022] The target control word is transmitted via the MIPI RFFE bus. , Write the corresponding adjustable matching network register and adjustable filter register respectively; wait for the device to switch to its complete state and stabilize before completing the reconstruction process;

[0023] After the reconstruction process is completed, return to steps S100 and S200 to obtain the latest reconstruction index. Compare the latest reconstruction index with the preset reconstruction threshold. If the latest reconstruction index is less than the reconstruction threshold, the reconstruction is successful and the current configuration is recorded for quick subsequent calls. If the latest reconstruction index is greater than or equal to the reconstruction threshold, perform a local search adjustment.

[0024] Furthermore, the local search adjustment includes:

[0025] With the target control word currently being written ( , Centered on a target, a search neighborhood is defined. The control word is a digital value, and the search step size is the minimum resolution, which is to add or subtract one least significant bit for each adjustable dimension. For the matching network control word... Search for each K before and after it. m There are steps, totaling (2×K) m +1) candidate values; for the filter control word Search its preceding and following K. f There are steps, totaling (2×K) f+1) candidate values; traverse all candidate control word combinations to perform search space judgment to determine the search method, and output the optimal control word combination according to the search method. The search methods include matching network priority search, filter priority search and two-dimensional joint search.

[0026] The obtained optimal control word combination ( , Write the data to the corresponding device, wait for stabilization, and repeat steps S100 and S200 to calculate the final reconstruction index. If the final reconstruction index is less than the reconstruction threshold, the reconstruction is considered successful, and the process returns to step S100. If the final reconstruction index is greater than or equal to the reconstruction threshold, and less than the reconstruction index after the reconstruction process is completed (i.e., less than the reconstruction index before the search), the process returns to steps S100 to S400 to reconstruct the data. To avoid infinite loops, a maximum number of local searches, Nmax, is set. In this embodiment, Nmax = 3. If the reconstruction index still cannot be reduced below the reconstruction threshold after Nmax consecutive local searches, the search stops, the current optimal configuration is maintained, and an abnormal status is reported to the baseband processor. If the final reconstruction index is greater than or equal to the reconstruction threshold, and greater than or equal to the reconstruction index after the reconstruction process is completed, the process reverts to the control word combination before the current local search, i.e., (…). , It also reports the abnormal status to the baseband processor.

[0027] Furthermore, the search space determination includes:

[0028] Extract the mismatch loss factor and interference rise factor corresponding to the current reconstruction index. If the ratio of the mismatch loss factor to the interference rise factor is greater than 1.5, then perform a matching network priority search. If the ratio of the interference rise factor to the mismatch loss is greater than 1.5, then perform a filter priority search. In other cases, it means that the contributions of the mismatch loss factor and the interference rise factor are equal, so the matching network and the filter are simultaneously adjusted, and a two-dimensional joint search is adopted.

[0029] Furthermore, the matching network priority search includes:

[0030] With the current value fixed, the filter control word is traversed (2×K). m +1) candidate control words for matching networks; sequentially combine each candidate matching network control word with the fixed filter control word to form a temporary combination, write the temporary combination to the corresponding device via the MIPI RFFE bus and stabilize it, repeat steps S100 and S200, calculate the disabling coefficient under the current combination, record the matching network control word that minimizes the disabling coefficient among all candidate control word combinations as the optimal matching network control word, and combine it with the current value of the current fixed filter control word to form the optimal control word combination for this round of search, denoted as ( , ).

[0031] Furthermore, the filter priority search includes:

[0032] The fixed matching network control word is the current value, and the process iterates through (2×K) f +1) candidate filter control words are selected. Each candidate filter control word is combined with a fixed matching network control word to form a temporary combination. The temporary combination is written to the corresponding device via the MIPI RFFE bus. After stabilization, steps S100 and S200 are repeated to calculate the disabling coefficient under the current combination. The filter control word that minimizes the disabling coefficient among all candidate control word combinations is recorded as the optimal filter control word. This optimal control word combination is formed by combining the optimal filter control word with the current fixed matching network control word in this round of search and denoted as ( ). , ).

[0033] Furthermore, the two-dimensional joint search includes:

[0034] With the currently written control sub-combination ( , Centered on , the search step size is defined as the least significant bit, and the search range is the K bits before and after it. m and K f Each step, that is, matching the network control word (2×K) m +1) candidate values ​​and the (2×K) filter control word f +1) candidate values ​​are permuted and combined to obtain (2×K) m +1)×(2×K) f +1) candidate control word combinations, and simultaneously traverse the candidate control word combinations of the matching network and filter: write the candidate control word to the corresponding device through the MIPI RFFE bus, wait for stabilization, repeat steps S100 and S200, calculate the disabling coefficient under the current candidate control word combination, and record the candidate control word combination that minimizes the disabling coefficient among all candidate combinations as the optimal control word combination, denoted as ( , ).

[0035] The beneficial effects of this invention are:

[0036] (1) By collecting working status parameters in real time, the key performance indicators of the current signal transmission path are calculated accordingly. The key performance indicators include mismatch loss factor, impedance deviation factor and interference rise factor. The complex performance degradation of radio frequency signal transmission is transformed into a normalized quantitative indicator, eliminating the influence of different physical dimensions and providing a unified basis for subsequent comprehensive evaluation. At the same time, by setting reasonable engineering thresholds, the sensitivity and accuracy of performance degradation are ensured, providing a reliable input for dynamic reconstruction.

[0037] (2) By weighted fusion of mismatch loss factor, impedance deviation factor and interference rise factor, the disabling coefficient is obtained, the disabling coefficient curve is constructed and the curve characteristics are analyzed to generate the reconstruction index, thereby realizing the comprehensive quantification of the health status of the signal transmission path, which can provide early warning of performance degradation and avoid false triggering caused by instantaneous fluctuations.

[0038] (3) Based on a multidimensional lookup table, the state is quickly mapped to the control word. Combined with multidimensional linear interpolation, the accurate control word can be obtained even between discrete sampling points. Through interference detection and filter priority adjustment, the receiving sensitivity is prioritized. Through power utilization monitoring and linearity correction, the target control word is obtained to avoid nonlinear distortion caused by power amplifier saturation. The factory calibration data is fully utilized to transform the complex radio frequency nonlinear relationship into a simple lookup table operation. The decision is completed in microseconds, which not only ensures real-time performance but also improves the accuracy of reconstruction, providing a clear target for subsequent execution.

[0039] (4) The target control word is quickly written through the MIPI RFFE bus and the reconstruction effect is verified to form a closed loop. When the reconstruction effect is not up to standard, a priority-based multi-dimensional local search is adopted to finely optimize in the neighborhood with the minimum effective step size, avoiding the high overhead of global search. At the same time, the maximum number of times limit and back-off mechanism are used to prevent the system from falling into instability. Two-level optimization from coarse tuning to fine tuning is realized, which can compensate for model error and environmental changes, ensure that the signal transmission path always approaches the optimal state, and significantly improve the adaptive capability of the RF front end in multiple frequency bands and multiple scenarios. Attached Figure Description

[0040] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0041] Figure 1 This is a schematic diagram of the principle of the present invention. Detailed Implementation

[0042] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0043] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0044] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of the invention described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0045] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0046] Application Scenarios: As terminal devices increasingly demand support for multi-band and multi-mode communication, designing independent and optimized dedicated RF paths for each frequency band necessitates the integration of key components such as filters, switches, and matching networks. This leads to an explosive increase in the number of RF front-end devices, a significant increase in package size, and a substantial rise in power consumption and manufacturing / testing costs. It also exacerbates the complexity of packaging and antenna co-design and mass production consistency issues. Conversely, while using a single broadband or compromise RF path can reduce components and simplify implementation, the frequency dependence of matching networks and filters, as well as the temperature drift and environmental sensitivity of components, make it difficult to maintain low reflection and low insertion loss at all operating frequencies. This results in decreased receiver sensitivity, loss of transmit efficiency, and difficulty in intermodulation harmonic control, thus affecting system performance and spectrum compliance. This invention achieves adaptive optimization of an integrated RF terminal chip across multiple frequency bands and scenarios by dynamically sensing signal transmission path mismatch, impedance deviation, and interference status, making closed-loop reconstruction of the adjustable matching network and filter. This significantly improves signal transmission quality and system stability.

[0047] Please see Figure 1 As shown, the present invention is a signal transmission path optimization method for an integrated radio frequency terminal chip, comprising:

[0048] S100 periodically acquires data or performs data acquisition based on event triggers during chip operation, processes the acquired parameters, and calculates the key performance indicators of the current signal transmission path, specifically including:

[0049] The forward and reverse power of the antenna port at the current operating frequency are collected. It should be noted that under ideal lossless and perfectly matched conditions, the transmit power should equal the forward power. In signal path transmission, the forward power is the transmit power minus link losses such as coupler insertion loss and switch insertion loss. The reflection coefficient amplitude, denoted as r(t), is obtained by dividing the reverse power by the forward power and opening a new sign. Here, t represents the acquisition time index, and the reflection coefficient amplitude represents the proportion of energy reflected back from the antenna to the incident energy. Its value range is between 0 and 1. r(t) = 0 indicates an ideal match with no energy reflection; this is the ideal case. r(t) = 1 indicates total internal reflection, usually due to an open or short-circuited antenna. The reflection coefficient amplitude is then calculated using the formula... The mismatch loss M(t) is calculated. Mismatch loss represents the power loss that is not actually transmitted to the antenna due to impedance mismatch; the larger the value, the more severe the antenna mismatch. A maximum mismatch loss threshold is set, which is 0.5 dB in this embodiment. In RF front-end design, mismatch loss directly leads to ineffective dissipation of transmit power and deterioration of the signal-to-noise ratio of the received signal. When the mismatch loss is 0.5 dB, the corresponding voltage standing wave ratio (VSWR) is approximately 1.4, and the reflection coefficient amplitude is approximately 0.2. This value is considered a critical point for optimization: below 0.5 dB, the impact of mismatch on system performance is generally tolerable; above 0.5 dB, mismatch... The resulting power reflections and signal distortions will significantly degrade communication quality, requiring dynamic reconfiguration for compensation. Furthermore, 0.5 dB is a common matching tolerance indicator in most communication standards and RF device specifications. The mismatch loss factor is obtained by normalizing the mismatch loss M(t) using the maximum mismatch loss threshold: when the mismatch loss M(t) is less than the maximum mismatch loss threshold, the mismatch loss factor = mismatch loss M(t) divided by the maximum mismatch loss threshold; when the mismatch loss M(t) is greater than or equal to the maximum mismatch loss threshold, the mismatch loss factor = 1. The range of the mismatch loss factor is [0,1], representing the degree of power loss caused by the current antenna mismatch.

[0050] The complex reflection coefficients at various times are obtained using a vector coupler or a six-port network. The equivalent antenna impedance is then calculated using transmission line theory. The calculation formula is as follows: Where Z0 is the system characteristic impedance, which is taken as 50 ohms in this embodiment, and the equivalent antenna impedance Z(t) reflects the actual load impedance of the antenna port in the current environment; the impedance plate deviation is obtained by subtracting the system characteristic impedance Z0 from the equivalent antenna impedance Z(t) and taking the absolute value; it should be noted that the complex reflection coefficient contains amplitude and phase information, and the amplitude of the reflection coefficient is the magnitude of the complex reflection coefficient, that is The maximum allowable impedance deviation threshold is set, which is set to 20 ohms in this embodiment. When the impedance deviation is 20 ohms, the corresponding voltage standing wave ratio (VSWR) is approximately 1.5, which is a commonly accepted matching boundary in engineering. The impedance deviation is normalized using the maximum allowable impedance deviation threshold to obtain the impedance deviation factor: when the impedance deviation is less than the maximum allowable impedance deviation threshold, the impedance deviation factor = impedance deviation divided by the maximum allowable impedance deviation threshold; when the impedance deviation is greater than or equal to the maximum allowable impedance deviation threshold, the impedance deviation factor = 1; the range of the impedance deviation factor is [0,1], representing the degree to which the current antenna impedance deviates from the ideal matching value.

[0051] During periods when no useful signal is transmitted or when a known useful signal is idle, the total interference power within the current receiving channel is measured using spectrum information provided by a broadband energy detector or baseband, and this is expressed as the interference intensity I(t), measured in decibels. This value includes various types of noise interference signals within the channel. The theoretical thermal noise floor N0 under the current channel bandwidth is calculated using the formula: N0 = -174 + 10 × log 10 (B), where B is the bandwidth of the current channel in Hertz; the interference rise ΔI(t) is defined as the interference intensity I(t) minus the theoretical thermal noise floor N0, representing the increment of the actual interference power relative to the theoretical noise floor; if the interference intensity I(t) is less than the theoretical thermal noise floor N0, then ΔI(t) is taken as zero, and is considered as no significant interference; a maximum allowable interference rise threshold is set, which is set to 3 dB in this embodiment; a 3 dB interference rise will reduce the received signal-to-noise ratio by 3 dB, resulting in a direct deterioration of the receiving sensitivity by 3 dB. B is the upper limit that is usually tolerated in communication system design. The interference rise ΔI(t) is normalized using the maximum allowable interference rise threshold: when the interference rise ΔI(t) is less than the maximum allowable interference rise threshold, the interference rise factor = interference rise ΔI(t) divided by the maximum allowable interference rise threshold; when the interference rise ΔI(t) is greater than or equal to the maximum allowable interference rise threshold, the interference rise factor = 1; the range of the interference rise factor is [0,1], which represents the degree of degradation of the receiving link performance by the current interference.

[0052] S200 is a real-time receiver of key performance indicators of the current signal transmission path, specifically including the mismatch loss factor, impedance deviation factor, and interference rise factor. A weighting factor is assigned to each of these factors, and the sum of these weighting factors is one. The mismatch loss factor, impedance deviation factor, and interference rise factor are then weighted and fused according to the assigned weighting factors to obtain the disabling coefficient S(t). It should be noted that the weighting factor allocation depends on the application scenario. If the terminal is in a transmit-dominant scenario, such as a scenario with high uplink traffic, the weight of the mismatch loss factor should be higher, because mismatch directly affects transmit efficiency and power consumption. If the terminal is in a receive-dominant scenario... For applications in industries with weak signals, the interference factor weight should be higher, as interference directly affects receiver sensitivity. The specific weighting should be adjusted according to the current communication mode to better suit current performance requirements. The disabling coefficients at each acquisition time t are plotted in chronological order on a two-dimensional Cartesian coordinate system, with time on the horizontal axis and disabling coefficient on the vertical axis. A smooth curve connects each point to obtain a disabling coefficient curve, which visually reflects the dynamic changes in signal transmission path performance over time. Curve feature analysis is performed on the disabling coefficient curve to obtain the reconstruction index. A preset reconstruction threshold is established, and the disabling coefficient S(t) represents the normalized degradation degree across three dimensions: mismatch, impedance deviation, and interference. When S(t) reaches 0.6, it means that at least one dimension's degradation exceeds 60% of its threshold, or multiple dimensions combined lead to a significant decrease in overall performance. Taking a typical weighted allocation as an example, a transmit efficiency loss of 0.6 may exceed 0.3dB, and a receive sensitivity degradation may exceed 1.8dB, which exceeds the conventional link budget margin of the communication system and requires intervention. In practical applications, this threshold can be dynamically adjusted according to the terminal's operating mode. For example, the threshold can be appropriately lowered in high-speed mobile or weak field environments to optimize in advance; the threshold can be increased in standby or low-power modes to reduce the number of reconfiguration attempts. This embodiment uses a fixed threshold of 0.6 as the base. The accuracy can be extended to configurable parameters in specific implementations. When the reconstruction index is greater than or equal to the reconstruction threshold, it indicates that the performance of the current signal transmission path has significantly deteriorated and dynamic reconstruction needs to be initiated. In this case, step S300 is executed to make an adjustment decision. If the reconstruction index is less than the reconstruction threshold, it indicates that the current performance meets the standard requirements and no reconstruction is needed. The process returns to step S100 to continue monitoring the disability index at the next moment and updating the curve. In order to extract quantified performance degradation characteristics from the disability coefficient curve, this embodiment adopts a multi-dimensional curve feature analysis method to comprehensively evaluate the current state and trend of the signal transmission path and generate a reconstruction index. The specific process includes:

[0053] The arithmetic mean of the disability coefficients S(t) is calculated, and the maximum value of S(t) is taken as the maximum disability value. The performance change rate is obtained by fitting the slope of the disability coefficient curve using the least squares method. A performance change rate greater than zero indicates that performance is continuously deteriorating, and the steeper the slope, the faster the deterioration. A performance change rate less than zero indicates that performance is recovering, and a performance change rate of zero indicates that performance is stable. The non-negative part of the performance change rate is taken and normalized. The non-negative part means that if the slope is negative, indicating performance recovery, this term is set to 0 to avoid negative features lowering the reconstruction index. Finally, the non-negative part of the normalized performance change rate, the mean disability value, and the maximum disability value are linearly weighted and fused. The reconstruction index is calculated. It should be noted that in this embodiment, the weighting factors for the non-negative part of the normalized performance change rate, the mean disability value, and the maximum disability value are set to 0.5, 0.3, and 0.2, respectively. These specific settings are adjusted according to the specific application mode to ensure that the reconstruction index always matches the current performance requirements. For example, in the uplink high traffic volume mode, the weighting factors for the non-negative part of the normalized performance change rate, the mean disability value, and the maximum disability value are set to 0.6, 0.2, and 0.2, respectively; in the downlink high traffic volume and weak signal mode, and in the uplink high traffic volume mode, the weighting factors for the non-negative part of the normalized performance change rate, the mean disability value, and the maximum disability value are set to 0.7, 0.2, and 0.1, respectively.

[0054] S300, based on the current real-time state, determines the optimal adjustment strategy for the tunable matching network and the tunable filter, specifically including:

[0055] The system acquires the current operating frequency, chip temperature, equivalent antenna impedance, interference rise factor, and transmit power. When significant interference is detected in the receiving channel via a broadband energy detector or baseband spectrum information, i.e., the interference rise factor exceeds a preset interference threshold (in this embodiment, the interference threshold is set to 0.5), a rise factor greater than 0.5 indicates that the actual interference rise has reached or exceeded 1.5 dB, significantly degrading the receiver sensitivity and indicating significant interference. The baseband processor performs Fast Fourier Transform (FFT) analysis on the received signal to extract the frequency corresponding to the maximum power peak in the spectrum, which is then used as the interference frequency. A pre-saved multidimensional lookup table (LUT) is used; the LUT is generated during the chip's factory calibration phase and covers the optimal configuration under all possible frequency, temperature, and antenna load combinations. Antenna load refers to antenna impedance. A multidimensional lookup table (LUT) enables rapid mapping from environmental conditions to the optimal control word. Three independent LUTs are constructed to meet different optimization objectives: the matching network LUT takes frequency, temperature, and antenna impedance as input, and outputs a matching network control word that minimizes mismatch loss and maximizes transmission efficiency; the filter LUT takes frequency and interference frequency as input, and outputs a filter control word that maximizes the suppression of specified interference and minimizes insertion loss of the useful signal in the receiving channel, corresponding to the default bandpass mode when there is no interference; the predistortion correction LUT takes transmit power and frequency as input, and outputs a matching network fine-tuning amount used to optimize linearity when the power amplifier is near saturation. All three LUTs are generated through full-state traversal during factory calibration, and high-precision lookup is achieved during runtime using multidimensional linear interpolation, providing an accurate data foundation for dynamic reconstruction.

[0056] Using the current operating frequency, temperature, and antenna impedance as indices, a multidimensional lookup table is searched to output an adjustable matching network control word that minimizes mismatch loss and maximizes transmission efficiency. Then, using the current operating frequency and interference frequency as indexes, a search is performed in a multidimensional lookup table to output the adjustable filter control word that maximizes interference suppression in the receiving channel while minimizing insertion loss of the useful signal. It should be noted that if there is no interference, the default bandpass mode control word will be output. Since the actual state may not fall completely on the discrete sampling points, multidimensional linear interpolation is used to calculate the control word corresponding to the current state. Specifically, for the equivalent antenna impedance, interpolation is performed between the real and imaginary parts of the sampled points. For temperature and frequency, interpolation is also performed between the nearby temperature and frequency points.

[0057] The corresponding notch mode control word is retrieved from a multidimensional lookup table based on the interference frequency. And with its covered adjustable filter control word ,Right now = Interference often has a more direct impact on receiver sensitivity than mismatch, and filter adjustment and matching network adjustment are relatively independent. Prioritizing interference suppression can quickly restore receiver performance. The power utilization rate is obtained by dividing the transmit power by the maximum output power corresponding to the linear region of the power amplifier. The maximum output power corresponding to the linear region of the power amplifier is the maximum output power that the power amplifier can maintain linear amplification at a specified operating frequency and temperature. A preset utilization threshold is set to 80% in this embodiment. This threshold is set based on engineering experience. Specifically: when the power utilization rate is less than 80%, the power amplifier operates in the linear region, and changes in the matching network have little impact on linearity; when the power utilization rate is greater than or equal to 80%, it indicates that the current transmit power is close to the upper limit of the linear region, and the power amplifier's operating point is close to the saturation region. At this time, small changes in the matching network will significantly affect the amplifier's load-pulling characteristics, which may lead to linearity deterioration. Therefore, linearity correction needs to be initiated; when the power utilization rate is greater than or equal to the utilization threshold, the fine-tuning amount of the matching network is read from the multidimensional lookup table. And based on this, the basic control word The following modifications were made to optimize linearity while maintaining high efficiency. The final matching network control word is: = + When the power utilization rate is less than the utilization threshold, the power amplifier operates in the linear region. Changes in the matching network have little impact on linearity, so no linearity correction is needed. The final matching network control word directly adopts the basic control word. = By monitoring the power utilization rate, the system can dynamically sense the power amplifier's operating status and actively fine-tune the counterweight when it is close to saturation, avoiding nonlinear distortion caused by matching adjustments and ensuring the quality of the transmitted signal.

[0058] Therefore, the final adjustable matching network control word can be obtained. and adjustable filter control word And pass it to step S400; it should be noted that if the system supports independent adjustment of multiple channels, then each affected channel generates a corresponding control word;

[0059] S400 performs dynamic reconstruction based on the target control word of an adjustable matched network and an adjustable filter, verifies the reconstruction effect, and performs fine-tuning as necessary to ensure that the signal transmission rate path is restored to the optimal state; specifically including:

[0060] Adjustable matching network control word via MIPI RFFE bus and adjustable filter control word Write to the corresponding adjustable matching network register and adjustable filter register respectively. The write operation adopts atomic operation to ensure that it takes effect simultaneously; after the device completes the state switch and stabilizes, the reconstruction process is completed. It should be noted that the adjustable device usually refers to the SOI-based variable capacitor and adjustable filter, whose stabilization time is usually less than 10 microseconds. Therefore, in this embodiment, after the switch is complete, wait 10 microseconds before completing the reconstruction process.

[0061] After the reconstruction process is completed, the process returns to steps S100 and S200 to obtain the latest disabling coefficient. This coefficient is then compared to a preset disabling threshold. If the disabling coefficient is less than the preset disabling threshold, the reconstruction is successful, and the signal transmission path has returned to normal. The current configuration is recorded for quick subsequent recall. If the disabling coefficient is greater than or equal to the preset disabling threshold, the signal transmission path performance still does not meet requirements. A local search adjustment is then performed to further optimize the control word. This search adjustment includes defining a search neighborhood centered on the currently written adjustable matching network control word and adjustable filter control word. The control word is a digital quantity, such as an N-bit binary code, corresponding to 2... N The search step size is the minimum resolution, which means adding or subtracting one least significant bit from each adjustable dimension. The least significant bit refers to the bit with the lowest weight in a binary number, representing the finest adjustment granularity that the hardware can achieve. Assuming the control word is a 4-bit binary number, i.e., N=4, with a value range of 0000-1111, and the weight of the most significant bit is 2. 3 =8, the weight of the least significant bit is 2. 0 =1; The specific search scope is as follows: for matching network control words... Search for each K before and after it. m There are steps, totaling (2×K) m +1) candidate values, in this embodiment, K is taken. m =2, that is, to examine ( -2), ( -1) 、( +1) and ( +2), a total of 5;

[0062] For filter control word Search its preceding and following K. f There are steps, totaling (2×K) f +1) candidate values; similarly, in this embodiment, K is taken. f =2, that is, to examine ( -2), ( -1) 、( +1) and ( +2), a total of 5 candidate values; traversing all candidate control word combinations to perform search space judgment to determine the optimal control word combination, specifically including:

[0063] Extract the mismatch loss factor and interference rise factor corresponding to the current disability coefficient. If the mismatch loss factor is significantly greater than the interference rise factor (i.e., the ratio of the mismatch loss factor to the interference rise factor is greater than 1.5), then perform a matching network priority search. If the mismatch loss factor is significantly less than the interference rise factor (i.e., the ratio of the interference rise factor to the mismatch loss factor is greater than 1.5), then perform a filter priority search. By quantifying and comparing the contribution of each factor, the dimension with the greatest influence can be adjusted first, reducing unnecessary combination traversal and improving search efficiency. In other cases, it indicates that the contributions of the mismatch loss factor and the interference rise factor are equal, so the matching network and the filter are adjusted simultaneously, and a two-dimensional joint search is adopted.

[0064] Among them, the matching network prioritizes the search: the filter control word is fixed at its current value, and the candidate control words of the matching network are traversed ( -2), ( -1) 、( +1) and ( +2), a total of 5 candidate combinations; each candidate matching network control word is sequentially combined with the fixed filter control word to form a temporary combination, and the temporary combination is written to the corresponding device through the MIPI RFFE bus, and waits for 10 microseconds to stabilize; steps S100 and S200 are quickly repeated, the disabling coefficient under the current combination is calculated, and the matching network control word that minimizes the disabling coefficient among all candidate control word combinations is recorded as the optimal matching network control word, and it is combined with the current value of the current fixed filter control word to form the optimal control word combination for this round of search ( , ), and then combine the optimal control words ( , The reconstruction index is used as the optimization reconstruction index;

[0065] Among them, the filter priority search is as follows: the current value of the fixed matching network control word is used to traverse the candidate control words of the filter ( -2), ( -1) 、( +1) and ( +2), a total of 5 candidate combinations; each candidate filter control word is sequentially combined with the fixed matching network control word to form a temporary combination, and the temporary combination is written to the corresponding device through the MIPI RFFE bus, and waits for 10 microseconds to stabilize; steps S100 and S200 are quickly repeated, the disabling coefficient under the current combination is calculated, and the filter control word that minimizes the disabling coefficient among all candidate control word combinations is recorded as the optimal filter control word, and it is combined with the current fixed matching network control word to form the optimal control word combination for this round of search ( , ), and then combine the optimal control words ( , The reconstruction index is used as the optimization reconstruction index;

[0066] Among them, the two-dimensional joint search: based on the currently written control sub-combination ( , Centered on , the search step size is defined as the least significant bit, and the search range is the K bits before and after it. m and K f Each step, that is, matching the network control word (2×K) m +1) candidate values ​​and the (2×K) filter control word f +1) candidate values ​​are permuted and combined to obtain (2×K) m +1)×(2×K) f +1) candidate control word combinations, resulting in a total of 25 candidate control word combinations in this embodiment; specifically K m =K f Simultaneously, iterate through the candidate control word combinations of the matching network and filter: write the candidate control word to the corresponding device via the MIPI RFFE bus, wait for the device to stabilize (i.e., wait 10 microseconds); quickly repeat steps S100 and S200, calculate the disabling coefficient under the current candidate control word combination, and record the candidate control word combination that minimizes the disabling coefficient among all candidate combinations as the optimal control word combination. , );

[0067] The optimal control word combination will be obtained through the above search. , Write the corresponding device, wait for stabilization, repeat steps S100 and S200, calculate the final reconstruction index. If the final reconstruction index is less than the reconstruction threshold, the reconstruction is successful, the optimal configuration is recorded for subsequent quick recall, and the process returns to step S100. If the final reconstruction index is greater than or equal to the reconstruction threshold, and the final reconstruction index is less than the reconstruction index after the reconstruction process is completed (i.e., less than the reconstruction index before the search), it indicates an improvement compared to before the search. Then, return to steps S100 to S400 to reconstruct. To avoid infinite loops, a maximum number of local searches Nmax is set. In this embodiment, Nmax = 3. If the disability coefficient still cannot be reduced below the disability threshold after Nmax consecutive local searches, the search is stopped, the current optimal configuration is maintained, and the abnormal status is reported to the baseband processor. If the final reconstruction index is greater than or equal to the reconstruction threshold, and the optimized reconstruction index is greater than or equal to the reconstruction index after the reconstruction process is completed, it indicates no improvement or even deterioration. Then, the process reverts to the control word combination before this local search, i.e., ( , It reports abnormal states to the baseband processor to avoid system instability caused by continuous attempts; and records fault information for subsequent analysis; local search can further refine the adjustment between discrete sampling points in the lookup table, compensate for model errors and environmental changes, and ensure that the performance after reconstruction is close to the true optimal value; at the same time, by limiting the search range and number of searches, it balances the optimization effect with the real-time requirements.

[0068] The above are merely embodiments of the present invention and are not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A method for optimizing the signal transmission path of an integrated radio frequency terminal chip, comprising: S100 collects chip operating status parameters in real time and calculates key performance indicators of the current signal transmission path based on the collected parameters. The key performance indicators include mismatch loss factor, impedance deviation factor, and interference rise factor. The operating status parameters include the current operating frequency, forward and reverse power of the antenna port, transmit power, chip temperature, complex reflection coefficient, reflection coefficient amplitude, and interference intensity. Its characteristic is that it further includes: S200: The mismatch loss factor, impedance deviation factor and interference rise factor are weighted and fused to obtain the disabling coefficient, and a reconstruction index is generated based on the time series curve of the disabling coefficient. The reconstruction index is compared with a preset reconstruction threshold. If it is greater than or equal to the threshold, dynamic reconstruction is triggered. S300: When dynamic reconfiguration is triggered, the target control word of the adjustable matching network and the adjustable filter is determined by querying a multidimensional lookup table based on the current working state parameters. S400: Write the target control word into the corresponding device to complete the reconstruction and verify the disabling coefficient after reconstruction. If the requirements are not met, perform a local search adjustment until the performance is restored or the search is exited.

2. The signal transmission path optimization method for an integrated radio frequency terminal chip according to claim 1, characterized in that, The calculation process for the key performance indicators includes: The forward and reverse power of the antenna port at the current operating frequency are collected. The reverse power is divided by the forward power and the square root is taken to obtain the reflection coefficient amplitude, denoted as r(t), where t represents the acquisition time index. The reflection coefficient amplitude is then calculated using the formula... The mismatch loss M(t) is calculated; a maximum mismatch loss threshold is set, and the mismatch loss is normalized using the maximum mismatch loss threshold to obtain the mismatch loss factor: when the mismatch loss is less than the maximum mismatch loss threshold, the mismatch loss factor = mismatch loss divided by the maximum mismatch loss threshold; when the mismatch loss is greater than or equal to the maximum mismatch loss threshold, the mismatch loss factor = 1; the range of the mismatch loss factor is [0,1], which represents the degree of power loss caused by the current antenna mismatch; The complex reflection coefficients at each moment are obtained through a vector coupler or a six-port network. The equivalent antenna impedance is then calculated based on transmission line theory. The impedance patch deviation is obtained by subtracting the system characteristic impedance from the equivalent antenna impedance and taking the absolute value. A maximum allowable impedance deviation threshold is set, and the impedance deviation is normalized using this threshold to obtain the impedance deviation factor: when the impedance deviation is less than the maximum allowable impedance deviation threshold, the impedance deviation factor = impedance deviation divided by the maximum allowable impedance deviation threshold; when the impedance deviation is greater than or equal to the maximum allowable impedance deviation threshold, the impedance deviation factor = 1. The range of the impedance deviation factor is [0,1], representing the degree to which the current antenna impedance deviates from the ideal matching value. The theoretical thermal noise floor N0 under the current channel bandwidth is calculated based on the interference intensity I(t). The calculation formula is: N0 = -174 + 10 × log 10 (B), where B is the bandwidth of the current channel; the interference rise ΔI(t) is defined as equal to the interference intensity I(t) minus the theoretical thermal noise floor N0, representing the increment of the actual interference power relative to the theoretical noise floor; if the interference intensity is less than the theoretical thermal noise floor, then ΔI(t) is set to zero; a maximum allowable interference rise threshold is set, and the interference rise is normalized using the maximum allowable interference rise threshold: when the interference rise is less than the maximum allowable interference rise threshold, then the interference rise factor = interference rise divided by the maximum allowable interference rise threshold; when the interference rise is greater than or equal to the maximum allowable interference rise threshold, then the interference rise factor = 1; the range of the interference rise factor is [0,1], which characterizes the degree of degradation of the receiving link performance by the current interference.

3. The signal transmission path optimization method for an integrated radio frequency terminal chip according to claim 2, characterized in that, The process of generating the remodeling index based on the time series curve of the disability coefficient includes: The arithmetic mean of the disability coefficients at each time point is calculated to obtain the mean disability value. The maximum value of the disability coefficients at each time point is taken as the maximum disability value. The slope of the disability coefficient curve is further fitted by the least squares method to obtain the performance change rate. The non-negative part of the performance change rate is taken and normalized. The non-negative part of the normalized performance change rate, the mean disability value, and the maximum disability value are then linearly weighted and fused to obtain the reconstruction index.

4. The signal transmission path optimization method for an integrated radio frequency terminal chip according to claim 3, characterized in that, The target control word for determining the tunable matched network and the tunable filter includes: The system acquires the current operating frequency, chip temperature, equivalent antenna impedance, interference rise factor, and transmit power. When the interference rise factor exceeds the preset interference threshold, significant interference is determined to exist. The baseband processor performs a fast Fourier transform analysis on the received signal and extracts the frequency corresponding to the maximum power peak in the spectrum as the interference frequency. A pre-saved multidimensional lookup table is configured. Using the current operating frequency, temperature, and antenna impedance as indices, the system searches the multidimensional lookup table and outputs an adjustable matching network control word that minimizes mismatch loss and maximizes transmission efficiency. Then, using the current operating frequency and interference frequency as indexes, a search is performed in a multidimensional lookup table to output the adjustable filter control word that maximizes interference suppression in the receiving channel while minimizing insertion loss of the useful signal. ; The corresponding notch mode control word is retrieved from a multidimensional lookup table based on the interference frequency. And with its covered adjustable filter control word ,Right now = The power utilization rate is obtained by dividing the transmit power by the maximum output power corresponding to the linear region of the power amplifier. A preset utilization threshold is set. When the power utilization rate is greater than or equal to the utilization threshold, the fine-tuning amount of the matching network is read from the multidimensional lookup table. And based on this, the basic control word After correction, the final matching network control word is: = + When the power utilization rate is less than the utilization threshold, no linearity correction is required, and the final matching network control word directly adopts the basic control word. = ; Therefore, the final adjustable matching network control word can be obtained. and adjustable filter control word , as the target control word.

5. The signal transmission path optimization method for an integrated radio frequency terminal chip according to claim 4, characterized in that, The target control word is written to the corresponding device to complete the reconstruction, and the disabling factor after reconstruction is verified: The target control word is transmitted via the MIPI RFFE bus. , Write the corresponding adjustable matching network register and adjustable filter register respectively; wait for the device to switch to its complete state and stabilize before completing the reconstruction process; After the reconstruction process is completed, return to steps S100 and S200 to obtain the latest reconstruction index. Compare the latest reconstruction index with the preset reconstruction threshold. If the latest reconstruction index is less than the reconstruction threshold, the reconstruction is successful and the current configuration is recorded for quick subsequent calls. If the latest reconstruction index is greater than or equal to the reconstruction threshold, perform a local search adjustment.

6. The signal transmission path optimization method for an integrated radio frequency terminal chip according to claim 5, characterized in that, The local search adjustment includes: With the target control word currently being written ( , Centered on a target, a search neighborhood is defined. The control word is a digital value, and the search step size is the minimum resolution, which is to add or subtract one least significant bit for each adjustable dimension. For the matching network control word... Search for each K before and after it. m There are steps, totaling (2×K) m +1) candidate values; for the filter control word Search its preceding and following K. f There are steps, totaling (2×K) f +1) candidate values; traverse all candidate control word combinations to perform search space judgment to determine the search method, and output the optimal control word combination according to the search method. The search methods include matching network priority search, filter priority search and two-dimensional joint search. The obtained optimal control word combination ( , Write the data to the corresponding device, wait for stabilization, and repeat steps S100 and S200 to calculate the final reconstruction index. If the final reconstruction index is less than the reconstruction threshold, the reconstruction is considered successful, and the process returns to step S100. If the final reconstruction index is greater than or equal to the reconstruction threshold, and less than the reconstruction index after the reconstruction process is completed (i.e., less than the reconstruction index before the search), the process returns to steps S100 to S400 to reconstruct the data. To avoid infinite loops, a maximum number of local searches, Nmax, is set. In this embodiment, Nmax = 3. If the reconstruction index still cannot be reduced below the reconstruction threshold after Nmax consecutive local searches, the search stops, the current optimal configuration is maintained, and an abnormal status is reported to the baseband processor. If the final reconstruction index is greater than or equal to the reconstruction threshold, and greater than or equal to the reconstruction index after the reconstruction process is completed, the process reverts to the control word combination before the current local search, i.e., (…). , It also reports the abnormal status to the baseband processor.

7. The signal transmission path optimization method for an integrated radio frequency terminal chip according to claim 6, characterized in that, The search space determination includes: Extract the mismatch loss factor and interference rise factor corresponding to the current reconstruction index. If the ratio of the mismatch loss factor to the interference rise factor is greater than 1.5, then perform a matching network priority search. If the ratio of the interference rise factor to the mismatch loss is greater than 1.5, then perform a filter priority search. In other cases, it means that the contributions of the mismatch loss factor and the interference rise factor are equal, so the matching network and the filter are simultaneously adjusted, and a two-dimensional joint search is adopted.

8. The signal transmission path optimization method for an integrated radio frequency terminal chip according to claim 7, characterized in that, The matching network priority search includes: With the current value fixed, the filter control word is traversed (2×K). m +1) candidate control words for matching networks; sequentially combine each candidate matching network control word with the fixed filter control word to form a temporary combination, write the temporary combination to the corresponding device via the MIPI RFFE bus and stabilize it, repeat steps S100 and S200, calculate the disabling coefficient under the current combination, record the matching network control word that minimizes the disabling coefficient among all candidate control word combinations as the optimal matching network control word, and combine it with the current value of the current fixed filter control word to form the optimal control word combination for this round of search, denoted as ( , ).

9. The signal transmission path optimization method for an integrated radio frequency terminal chip according to claim 8, characterized in that, The filter priority search includes: With the fixed matching network control word as the current value, iterate through (2×K) f +1) candidate filter control words are selected. Each candidate filter control word is combined with a fixed matching network control word to form a temporary combination. The temporary combination is written to the corresponding device via the MIPI RFFE bus. After stabilization, steps S100 and S200 are repeated to calculate the disabling coefficient under the current combination. The filter control word that minimizes the disabling coefficient among all candidate control word combinations is recorded as the optimal filter control word. This optimal control word combination is formed by combining the optimal filter control word with the current fixed matching network control word in this round of search and denoted as ( ). , ).

10. A signal transmission path optimization method for an integrated radio frequency terminal chip according to claim 9, characterized in that, The two-dimensional joint search includes: With the currently written control sub-combination ( , Centered on , the search step size is defined as the least significant bit, and the search range is the K bits before and after it. m and K f Each step, that is, matching the network control word (2×K) m +1) candidate values ​​and the (2×K) filter control word f +1) candidate values ​​are permuted and combined to obtain (2×K) m +1)×(2×K) f +1) candidate control word combinations, and simultaneously traverse the candidate control word combinations of the matching network and filter: write the candidate control word to the corresponding device through the MIPI RFFE bus, wait for stabilization, repeat steps S100 and S200, calculate the disabling coefficient under the current candidate control word combination, and record the candidate control word combination that minimizes the disabling coefficient among all candidate combinations as the optimal control word combination, denoted as ( , ).

Images