ADAPTIVE BASE BAND PARASITE SUPPRESSION SYSTEM
Patent Information
- Application Number
- TR202614299
- Authority / Receiving Office
- TR · TR
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2026-08-24
- Publication Date
- 2026-09-21
Smart Images
Figure 00000021_0000
Abstract
Description
1 TARIFF ADAPTIVE BASE BAND PARASITE SUPPRESSION SYSTEM Technical Area This invention analyzes interference in communication signals reaching the base station. by determining the type and reliability of the parasite, and using the appropriate method for the identified parasite. to apply the filtering process and analyze the communication signal according to the filtering result. It relates to a system that ensures the preservation of quality. Previous Technique Today, power control is used to reduce interference generated at base stations. Inter-cell interference coordination (ICIC), Enhanced Inter-Cell Interference Coordination (EIC) Coordination (eICIC) and baseband filtering methods are used. Fifth Fifth Generation New Radio (5G NR) dynamic time-division Crosstalk that occurs in time-division duplex (TDD) transmission systems. Cross-Link Interference (CLI) improves base station receiver performance. can have negative effects. In current systems, parasites are predetermined by 20 This is reduced by filters or filter changes depending on the measured signal values. However, the spectral and temporal characteristics of the parasite cannot be determined quickly. Suitable filters and filter coefficients reliably according to the type of interference. inability to select and the filtering result being read according to changing field conditions Deficiencies arise in the form of inability to adjust. 25 Therefore, considering the studies and shortcomings in the current technique... when present, interference in the communication signal reaching the base station determining the type and reliability of the detection, and accordingly, the appropriate filtering method. by selecting and reducing interference, the applied filtering has a 30% effect on the useful signal. 2 by controlling its effect, it modifies or reverses the filtering process as needed. It is understood that the system is needed. United States Regulation US2023090727A1, which is included in the known state of the art. The patent document describes machine 5 as a source of interference in wireless communication networks. This refers to a system that identifies and classifies using learning. The invention in question relates to the Operations, Management, and Administration (OAM) data of a wireless network. This data is used by Configuration Management (CM) and Performance Management (PM). and may include network topology data. From the obtained network data, specific 10 to each of the Physical Resource Blocks (PRBs) located in the cell during time intervals Intervention data is generated. PRB-based intervention data, different cells and bands so that the data obtained from their widths can be evaluated in the same structure normalization, resizing, zero filling, and contrast enhancement, etc. It can undergo pre-processing operations. With this data, the cell manufacturer, Metadata such as channel bandwidth, channel frequency, and operator are also in machine 15 The machine learning model is provided. The machine learning model, RAN intervention, passive Intermodulation (PIM), external interference, DECT and cable TV (CATV) interference, etc. They are trained to recognize different types of initiatives. The trained model, By analyzing unlabeled PRB intervention data and their metadata It classifies the source of the interference found in abnormal cells. Classification 20 As a result, one or more types of interference sources can be identified for each cell. and for each identified type of undertaking, a confidence in the accuracy of the classification. A confidence level value is assigned. The determined confidence level is based on a predefined threshold. When the value falls below a certain threshold, the relevant type of initiative can be excluded from the results. Furthermore, the interference characteristics of abnormal cells affect other 25 cells in the surrounding area or in the same location. cells affected by the same interference source are compared by comparing their interference characteristics. The cells can be identified. 3 Brief Description of the Invention The aim of this invention is to improve base station receiver baseband in mobile communication infrastructures. spectral power density obtained from the signal, noise level, pilot signal quality, channel estimation stability, signal-to-interference-plus-noise ratio (Signal-to-5 Interference-plus-Noise Ratio – SINR), block error rate (Block Error Rate – BLER) and hybrid automatic repeat request (HARQ) By analyzing retransmission rate data, the type of interference can be identified. determining its reliability, the identified type of parasite, and the appropriate filtering method. By selecting filtering coefficients, you can suppress noise with low latency and filter 10. then evaluating communication performance and detecting degradation in the useful signal. If detected, limit or modify the applied filtering process. or to implement a system that enables recovery. Detailed Description of the Invention 15 The "Adaptive Baseband Interference" method was developed to achieve the purpose of this invention. The "Suppression System" is shown in the attached figure; Figure 1. Schematic view of the system that is the subject of the invention. 20 The parts shown in the figure are individually numbered, and the corresponding numbers correspond to these numbers. given below: 1. System 25 2. Baseband data acquisition module 3. Digital filtering module 4. Server B. Base station 4 By analyzing the interference in the communication signals reaching the base station Identifying the type and reliability of the parasite, and applying appropriate filtering to the identified parasite. to perform the operation and improve the quality of the communication signal according to the filtering result. The system in question, developed to provide protection (1); - receiver baseband signal obtained from base station (B) in short measurement intervals 5 to examine; the power distribution of the signal in the frequency domains, the noise level changes, power peaks, pilot signal quality, channel estimation stability and signal-to-interference-plus-noise ratio Spectral parameters (SINR) will be used in the analysis of the parasite by determining the change in SINR. At least one baseband data collection configured to generate temporal data 10 module (2), - according to the detected interference structure to the receiving baseband signal of the base station (B) notch filtering, multi-notch filtering, spectral shaping, or adaptive Applying at least one of the suppression processes with low latency; filtering By applying the coefficients gradually and reversibly, the interference level is 15 at least one digital signal configured to generate a reduced baseband signal Filtering module (3), - by evaluating spectral-temporal data obtained from base station (B) The parasite can be narrow band, multiple narrow band, broad band, bursty, or time-varying. which type of enterprise it belongs to and the classification made 20 Determining the confidence score, which indicates reliability; parasite type, confidence score, based on the calculation budget, latency limit, and the risk of useful signal degradation. To determine the filtering method and filtering coefficients to be applied; filtering subsequently SINR, block error rate (BLER), hybrid autorepeat Hybrid Automatic Repeat Request (HARRQ) retransmission rate, pilot 25 by evaluating changes in signal quality and channel estimation stability If distortion is detected in the useful signal, the filtering level is adjusted. reducing, changing or reverting the last applied filtering coefficients It includes a small number of servers (4) configured to decide on the acquisition. 5 The baseband data acquisition module (2) located in the system (1) which is the subject of the invention, base station (B) radio frequency front end and analog-to-digital converter (Analog-to- After the Digital Converter (ADC), it is transferred to the receiver baseband chain. digitized I / Q (In-phase / Quadrature) signal samples 5 for one or more Orthogonal Frequency Division Multiplexing (Orthogonal Frequency Division Multiplexing (OFDM) symbol, mini time zone, time a sub-section of the segment or not exceeding a predefined low latency limit It is configured to allocate data to short metering windows. Baseband data collection module (2), source block or subcarrier 10 in each short measurement window Determining the normalized spectral power density based on neighboring frequencies To identify local power deviations and power peaks occurring in different regions. and to determine the noise floor elevation occurring across the band It is configured. The baseband data acquisition module (2) is located in the base station (B). The field is based on reference signals, pilot or preamble structures, and channel 15 to obtain measurements from estimation processes; to conduct a pilot study based on these measurements. to obtain information on quality deviation and channel estimation stability. It is configured. Baseband data acquisition module (2), base station (B) SINR and BLER are obtained from performance meters and the receiver baseband chain. HARQ Retransmission Rate and Channel Quality Indicator / Modulation and Encoding 20 scheme (Channel Quality Indicator / Modulation and Coding Scheme – CQI / MCS) to obtain radio performance data in the form of stability and before filtering In order to compare the measurements after filtering, the data in question is brief. It is configured to organize based on measurement windows. Baseband data collection module (2), obtained normalized spectral power density, local 25 power peaks, noise floor rise, pilot quality deviation, channel estimation information on the stability and variability of SINR across successive measurement windows by combining the interference in the frequency domain instead of the entire raw I / Q signal. its concentration, its change over time, and its effect on the useful signal. to generate low-dimensional spectral-temporal feature data representing and 30 The generated data is used to measure radio performance, the type of interference, and detection. 6 to transfer to the server (4) for use in determining its reliability It is being structured. The digital filtering module (3) in the system (1) which is the subject of the invention, is located in the base station (B) to receive the digitized I / Q signal located in the receiver baseband chain and 5 with the filtering mode determined by the server (4) and the filtering mode in question Using the corresponding set of coefficients, narrowband interference present in the receiver baseband signal can be detected in multiple narrowband interference, wideband noise floor amplification, explosive interference, or dynamic time-division bidirectional transmission / cross-connection interference (Time Division) The evolving initiative stemming from Duplex / Cross-Link Interference (TDD / CLI) 10 It is configured to suppress low latency. Digital filtering. module (3) filters the receiver baseband signal according to the filtering mode transmitted to it. Notch to suppress narrowband interference concentrated in a specific frequency region filtering, multiple narrow frequency regions concentrated in separate frequency domains. Multi-notch filtering to suppress band interference, band-wide noise 15 spectral shaping or adaptive suppression to reduce baseline elevation and only to suppress the explosive interference that occurs at specific time intervals to apply the temporary suppression process activated in the relevant event window It is structured. The numerical filtering module (3) performs the filtering operations finitely. Finite Impulse Response (FIR), Infinite Impulse Response (Infinite 20 Impulse Response (IIR), polyphase, frequency domain masking, minimum Least Mean Squares (LMS) or recursive least squares At least one of the adaptive filter structures based on Recursive Least Squares (RLS) It can be configured to be implemented using numerical filtering. module (3) calculates or selects the filtering coefficients to be applied itself 25 instead, take the set of coefficients determined by the server (4) and base the coefficients on buffered transition to the receiver baseband chain of station (B), stepped coefficient Filtering is applied by updating or changing parameters in limited steps. sudden instability in the baseband signal during the transition between modes It is structured to prevent this. The digital filtering module (3) performs 30 As a result of the filtering process, narrowband, multiple narrowband, wideband, burst or 7 To create a reduced baseband signal with a varying level of interference over time, The signal generated is used by the base station to continue communication operations. (B) is configured to transmit to the receiver baseband chain. Digital filtering. module (3), applied filtering mode, coefficient set and filtering behavior technical status information, communication of the filtering process performed 5 transfer to the server (4) to evaluate its impact on performance It is configured as follows: The digital filtering module (3) is configured by the server (4). created as a result of determining that there is a distortion in the useful signal. Reducing filtration depth, narrowing filter bandwidth, making it safer. 10 options to switch to filtering mode or reset the last set of coefficients. to obtain control information and perform filtering based on the received control information It is configured to apply the receiver baseband signal of station (B). The server (4) in the system in question (1) is located in the known state of the art. baseband data acquisition module (2) and 15 using any communication protocol to communicate with the digital filtering module (3) and to obtain data through this communication. It is configured to carry out the exchange. Server (4), baseband data Resource generated by the collection module (2) through short measurement windows normalized spectral power density on a block or subcarrier basis, local power peaks, power deviation relative to adjacent frequency regions, band-wide noise 20 floor rise, pilot quality deviation, channel estimation stability, and SINR. to obtain spectral-temporal feature data in the form of variability and to receive the data from the receiver the distribution of noise in the baseband signal over frequency, time its continuity and its effect on the useful communication signal It is configured to evaluate in order to determine. Server (4), baseband 25 analyze spectral-temporal feature data obtained from the data collection module (2) by causing interference from narrowband interference, multiple narrowband interference, and wideband noise floor. rise, explosive interference or dynamic TDD / CLI-related time variation assigning an enterprise to one of the classes and classifying it using a decision tree, random forest, gradient boosting, logistic regression, pure Bayes, small-scale multilayer 30 sensor, temporal convolutional network (TCN), 8 statistical thresholding, probabilistic classification, or hybrid rule-based to be carried out using at least one of the classification methods It is configured. Server (4), time-varying crossover with explosive interference. in evaluating the short-term temporal behavior of connection attempts Recurrent Neural Network (RNN), gated recurrent unit 5 (Gated Recurrent Unit – GRU) or temporal attention using at least one of the methods and the parasite that is carried out classification to determine how reliable the classification is the probability of spectral-temporal features in successive short measurement windows stability, impact on pilot quality and past similar interference incidents 10 by evaluating the success scores of the applied filtering processes together It is configured to create a confidence score. The server (4) is defined interference class and confidence score available calculation budget, allowed delay with the limit and the risk of the useful communication signal being distorted as a result of filtering 15 by evaluating together the numerical filtering module (3) Specifying the filtering mode; narrowband interference notch filtering, multiple narrowband The approach involves multi-notch filtering and a band-wide noise floor. spectral shaping or adaptive suppression and explosive interference in the rise temporary activated only within the time window in which the interference event occurs It is configured to select the appropriate suppression method. Server 20 (4), depending on the classification result if the confidence score is low to prevent the direct application of a strong filtering process and to be useful Monitoring only, low suppression to limit the risk of signal degradation. the level, in the form of limited coefficient updates or reversible trials 25 to determine the appropriate safe work practices is configured. The server (4) selects the filtering mode of the base station (B) To determine which filtering values to apply to the receiver baseband signal. Selecting the set of coefficients to be used from among pre-validated coefficient candidates or spectral peak location, interference bandwidth, required suppression depth, and allowable Using the given delay budget, a coefficient of 30 with low computational complexity. to generate through calculation and the coefficient set with the specified filtering mode 9 It is configured to transmit to the digital filtering module (3). Server (4), which sets of coefficients should be used first in subsequent similar interference events contextual multi-arm selection algorithm for determining, low-dimensional reinforcement learning, Bayesian optimization, or success score-based Use at least one of the multi-target selection algorithms and the specified coefficient is 5. buffered transition to the receiver baseband chain of the set, stepwise coefficient update or control information for implementation in the form of limited step parameter changes It is configured to transmit to the digital filtering module (3). Server (4), the filtering process performed by the digital filtering module (3) then filtering obtained by baseband data acquisition module (2) 10 SINR, BLER, HARQ retransmission rate, CQI / MCS stability, pilot quality Obtaining indicator and channel estimation stability data and filtering the data. Comparing the interference of the filtering process with the previously obtained equivalents. to determine whether it preserves the useful communication signal while reducing it It is configured. The server (4) reports the success of the filtering operation performed. Not only should it be determined based on the increase in the SINR value, but also on the relationship between SINR change and BLER. change, HARQ retransmission rate, CQI / MCS stability and pilot quality jointly assessing its protection; despite the improvement in SINR value Deterioration in BLER value or HARQ retransmission rate. In this case, using the filtering mode, the coefficient set is identified as risky and 20 Improvement in SINR and BLER values, and a decrease in HARQ retransmission rate. If pilot-grade protection is observed, the relevant interference class, filtering mode, and It is configured to increase the success score for the coefficient set. Server (4), Degradation in pilot signal quality after filtering, affecting channel estimation stability. Detection of a decrease, increase in BLER value, or increase in HARQ retransmission rate. 25 If this is done, the useful communication signal will be negatively affected by the filtering process. to determine if it is affected and, in this case, to reduce the filtering depth, filter narrowing the bandwidth, switching to a more secure filtering mode, or by generating control information to reverse the last applied set of coefficients. 30 to transmit the subject control information to the digital filtering module (3) It is configured. The server (4) sets the specified interference class and spectral-temporal after filtering the feature data with the applied filtering mode and coefficient set obtained SINR, BLER, HARQ retransmission rate and pilot quality changes To correlate which filtering mode to use under specific interference and field conditions. The coefficient set results in higher interference suppression success and a lower useful signal. to determine if it causes disruption and based on the success or risk scores obtained, 5 successful filtering parameters in subsequent similar interference events It is configured to prioritize. The server (4) obtains over time the detected parasite class distributions, spectral-temporal feature patterns, and reliability Score trends and coefficient sets with the previously used filtering mode. monitoring current performance results, 10 of the previously successful coefficient sets inability to deliver expected performance under similar conditions, decreased confidence scores. or to determine distribution drift if the parasite distribution changes, In the case of distribution shift, alternative sets of coefficients are tested within a limited trial window. testing in a low-risk manner and filtering based on performance results. to allow recalibration of its behavior and the interference performed 15 classification, confidence score, selected filtering mode, coefficient set, coefficient version, Update time, rollback decision, post-filtering performance result, and Logging recalibration events for audit trail and versioning purposes. It is structured accordingly. Industrial Application of the Invention Thanks to the system (1) which is the subject of the invention, mobile communication and telecommunication 4G (Fourth Generation Mobile Communications) in the sector (Technology) and 5G (Fifth Generation Mobile Communications 25 (Technology) The interference that occurs in signals reaching base stations is real. timely detection and reduction of interference through appropriate filtering methods by reducing data transmission errors and retransmissions, network capacity is increased. efficient use and communication with high user traffic In these environments, connection quality and signal stability are improved. 30 11 Around these fundamental concepts, the invention is titled "Adaptive Baseband Interference Suppression". It is possible to develop a wide variety of applications related to the System (1)”, and the invention This cannot be limited to the examples described here, but is primarily stated in the claims. It is like that.
Claims
12 REQUESTS 1. By analyzing the interference in the communication signals reaching the base station. to determine the type and reliability of the parasite, and to select the appropriate method for the identified parasite. Apply the filtering process and communicate based on the filtering result. 5 enabling the preservation of signal quality; - short measurement of the receiver baseband signal obtained from base station (B) to examine the power distribution of the signal in the frequency domains, changes in noise level, power peaks, pilot signal its quality, channel estimation stability, and signal-to-interference-plus-noise ratio are 10. spectral analysis, which will be used to determine the change in the parasite. at least one baseband data set structured to generate temporal data collection module (2), - to the receiving baseband signal of the base station (B), to the specified interference pattern according to notch filtering, multi-notch filtering, spectral shaping or 15 At least one of the adaptive suppression processes is performed with low latency. to implement; filtering coefficients in a gradual and reversible manner by applying at least one configured numerical filtering module (3), - Spectral-temporal data obtained from base station (B) 20 by evaluating the parasite as narrow band, multiple narrow band, wide band, burst or which of the changing types of enterprises does it belong to and what was done To determine the confidence score, which indicates the reliability of the classification; parasite type, confidence score, computational budget, latency limit, and useful signal The filtering method to be applied according to the risk of corruption and filtering 25 Determining the coefficients; SINR, block error rate, hybrid after filtering. automatic repeat request retransmission rate, pilot signal quality and channel By evaluating changes in the prediction stability, a useful signal can be obtained. If degradation is detected, the filtration level will be reduced. changing or reverting the last applied filtering coefficients 30 13 with a small server (4) configured to decide on the acquisition a characterized system (1).
2. Base station (B) radio frequency front end and analog-to-digital converter The digitized I / Q signal is then transferred to the receiver baseband chain. 5 to obtain samples and the resulting receiver baseband signal in real time In order to be examined as such, one or more Orthogonal Frequency Division (RFD) sensors are required. Multiplex symbol, mini time zone, a subdivision of a time zone or Short measurement that does not exceed a predefined low latency limit. Baseband data acquisition module 10 configured to allocate to windows A system like the one in Claim 1, characterized by (2) (1).
3. In each short measurement window, on a source block or sub-carrier basis Determining the normalized spectral power density, neighboring frequencies Detecting local power deviations and power peaks according to regions 15 to increase the noise floor that occurs across the band. with the baseband data acquisition module (2) configured to determine a system like any of the above characterized claims (1).
4. From the reference signals located at base station (B), pilot or preamble taking measurements from structures and channel estimation processes; Pilot quality deviation and channel estimation stability are determined through these measurements. Baseband data acquisition module configured to obtain information (2) a 25 as in any of the above claims characterized by system (1).
5. Performance meters of base station (B) and receiver baseband SINR, BLER, HARQ retransmission rate and channel obtained from the chain 30 in the form of quality indicator / modulation and coding scheme stability Obtaining radio performance data and comparing it before and after filtering. 14 In order to compare the measurements, the data in question are short-term measurements. Baseband data configured to organize based on windows any of the above requests characterized by the collection module (2) a system like one of them (1).
6. The resulting normalized spectral power density shows the local power peaks. points, noise floor elevation, pilot quality deviation, channel estimation stability and variability of SINR in successive measurement windows by piecing together the information, instead of the entire raw I / Q signal, only the interference is present. its concentration along the frequency axis, its change over time, and 10 low-dimensional spectral representing its effect on the useful signal. to generate temporal feature data and to use the generated data to analyze radio performance metrics of interference type and detection reliability to transfer to the server (4) for use in determining 15 characterized by the configured baseband data acquisition module (2) a system like any of the above requests (1).
7. Digitized I / Q located in the receiver baseband chain of base station (B) to receive the signal and the filtering mode determined by the server (4) The subject is the receiver baseband 20 using the coefficient set belonging to the filtering mode. narrowband interference, multiple narrowband interference, wideband interference present in the signal noise floor elevation, explosive interference, or dynamic time time due to split-duty bidirectional transmission / cross-connection interference configured to suppress the changing interference with low latency 25 of the above requests characterized by the numerical filtering module (3). a system like any other (1).
8. The receiver filters a specific baseband signal according to the filtering mode transmitted to it. Notch to suppress narrowband interference concentrated in the frequency region filtering, 30 that concentrate in multiple separate frequency regions Multiple notch filtering to suppress multiple narrowband interference, band-wide Spectral shaping to reduce radiated noise floor elevation. or adaptive suppression and bursts occurring at specific time intervals only activated in the relevant event window to suppress the attempt Digital filtering configured to apply temporary suppression. 5 in any of the above requests characterized by module (3) such a system (1).
9. Filtering operations include finite pulse response, infinite pulse response, multi-phase, frequency domain masking, least mean squares, or recursive metrics 10 by using at least one of the adaptive filter structures based on little squares. with the digital filtering module (3) configured to perform a system like any of the above characterized claims (1).
10. Calculating or selecting the filtering coefficients to be applied. 15 instead, take the set of coefficients determined by the server (4) and the coefficients buffered transition from base station (B) to receiver baseband chain, incremental by updating the coefficient or changing the parameter in limited steps Sudden changes in the baseband signal during switching between filtering modes Numerical filtering structured to prevent the occurrence of instability 20 in any of the above requests characterized by module (3) such a system (1).
11. As a result of the filtering process performed, narrowband, multi-narrowband, and wideband filters were obtained. band, baseband 25 with reduced level of explosive or time-varying interference. to generate the signal and continue the communication processes with the generated signal. to transmit to the receiving baseband chain of the base station (B) for operation The above is characterized by the structured numerical filtering module (3). a system like any of the requests (1). 16 12. Numerical filtering module (3), applied filtering mode, coefficient set and Technical status information regarding the filtering behavior, the filtering performed Evaluating the impact of the process on communication performance. Digital filtering module configured to transmit to the server (4) (3) like any of the above-mentioned claims characterized by 5 system (1).
13. Determination by the server (4) of the presence of distortion in the useful signal resulting in reduced filtering depth, filter band narrowing the width, switching to a safer filtering mode, or 10 to obtain the control information regarding the retrieval of the last set of coefficients and the received According to the control information, the filtering operation is performed on the receiver baseband of the base station (B). with the digital filtering module (3) configured to apply to the signal a system like any of the above characterized claims (1). 15 14. Baseband data acquisition module using any communication protocol. (2) and to communicate with the digital filtering module (3), this communication established with the server (4) configured to exchange data through A system like any of the above-mentioned described requirements 20 (1).
15. Short measurement windows by baseband data acquisition module (2) normalization based on the source block or subcarrier created from it. Spectral power density, local power peaks, adjacent frequency 25 power deviation according to regions, band-wide noise floor elevation, pilot quality deviation, channel estimation stability, and SINR variability. Receiving spectral-temporal feature data and receiving data from the receiver baseband the distribution of interference in the signal over frequency, time its continuity within and its effect on the useful communication signal 30 server configured to evaluate in order to determine (4) 17 a system like any of the above characterized claims (1).
16. Spectral-temporal characteristics obtained from the baseband data acquisition module (2). By analyzing the data, the interference can be narrowband interference, multiple narrowband interference, 5 broadband noise floor increase, explosive interference, or dynamics. To assign to one of the evolving startup categories derived from TDD / CLI and Classification methods include decision trees, random forests, gradient boosting, and logistics. regression, pure Bayesian, small-scale multilayer perceptron, temporal convolutional network, statistical thresholding, probabilistic classification, or hybrid rule 10 to be done using at least one of the classification-based methods the above requests are characterized by the server (4) configured to be so. a system like any other (1).
17. Short-term 15-minute burst interference and time-varying cross-linking interference. Recurrent neural networks, gated networks in evaluating temporal behavior at least one of the repetitive unit or temporal attention methods to use and to what extent the parasite classification performed is reliable To determine the probability of classification, spectral-temporal the stability of the characteristics in successive short measurement windows, pilot quality 20 the impact on and filtering applied in past similar parasitic incidents by jointly evaluating the success scores of the transactions, a combined confidence characterized by the server (4) configured to generate the score a system like any of the above requests (1).
18. Determine the parasite class and confidence score within the available calculation budget and permissions. given delay limit and filtering of the useful communication signal numerical filtering by considering the risk of corruption as a result. Determine the filtering mode to be implemented by module (3); narrowband notch filtering in interference, multiple narrowband interference, multiple notch filtering, 30 Spectral shaping in noise floor elevation spread across the band. 18 or adaptive suppression and explosive interference, only the interference event from temporary suppression processes activated in the time window in which it is located characterized by the server (4) configured to select the appropriate one a system like any of the above requests (1).
19. If the confidence score is low, the classification result will vary. to prevent the direct application of a strong filtering process and Monitoring only, to limit the risk of interference with the useful signal, is low. suppression level, limited coefficient update, or retractable trial.
10. Identifying the most appropriate of the following safer work behaviors the above requests are characterized by the server (4) configured to be so. a system like any other (1).
20. The selected filtering mode applies to the receiving baseband signal of the base station (B). 15 will be used to determine which filtering values to apply. to select the coefficient set from among pre-validated coefficient candidates or spectral peak location, interference bandwidth, required suppression depth, and Low computational complexity by using the allowed delay budget. It is created through the coefficient calculation process and the specified filtering mode. 20 to transmit the coefficient set to the numerical filtering module (3) from the above requests characterized by the configured server (4) a system like any other (1).
21. Which coefficient sets are prioritized in subsequent similar interference events? To determine which contextual multi-arm selection algorithm will be used, 25 low-dimensional reinforcement learning, Bayesian optimization, or success to use at least one of the score-based multi-target selection algorithms and buffered transition of the defined set of coefficients to the receiver baseband chain, incremental in the form of coefficient updating or parameter changes in limited steps control information for implementation to the numerical filtering module (3) 30 19 The above is characterized by the server (4) configured to transmit the above. a system like any of the requests (1).
22. Filtering performed by the digital filtering module (3) 5 obtained by the baseband data acquisition module (2) after the operation SINR, BLER, HARQ retransmission rate, CQI / MCS after filtering. stability, pilot quality indicator and channel estimation stability data to obtain and filter the data and their corresponding values beforehand. By comparing, the filtering process reduces interference while facilitating useful communication. Server configured to determine whether it protects its signal (4) 10 as in any of the above claims characterized by system (1).
23. The success of the filtering process performed can only be measured by the SINR value. Not determining based on the increase, SINR change and BLER change, HARQ 15 retransmission rate, CQI / MCS stability, and pilot quality protection To evaluate together; although there was an improvement in SINR value, BLER deterioration in value or HARQ retransmission rate using the filtering mode to identify the coefficient set as risky. and improvement in SINR and BLER values, HARQ retransmission rate increased by 20 If a decrease and preservation in pilot quality is observed, the relevant class of parasites, To improve the success score for the filtering mode and coefficient set. from the above requests characterized by the configured server (4) a system like any other (1).
24. Degradation in pilot signal quality and channel estimation after filtering. decrease in stability, increase in BLER value, or HARQ retransmission If an increase in the rate is detected, the useful communication signal to determine if it is negatively affected by the filtering process and in this case Reducing the filtration depth, narrowing the filter bandwidth, more 30 switching to a safe filtering mode or the last set of coefficients applied By creating control information for its reversal, the control in question configured to transmit its information to the numerical filtering module (3) any of the above requests characterized by the server (4) such a system (1).
25. The determined parasite class and spectral-temporal characteristic data were applied. SINR obtained after filtering with filtering mode and coefficient set, BLER, HARQ retransmission rate and pilot quality changes. To correlate which filtering mode to use under specific interference and field conditions. with a coefficient set that results in higher interference suppression success and a lower value of 10. to determine if it provides beneficial signal interference and the success or failure achieved Successful filtering parameters based on risk scores for the next similar one. Server configured to prioritize in parasitic events (4) a system like any of the above characterized claims (1). 15 26. Spectral-temporal distributions of parasite classes obtained over time. trait patterns, confidence score trends, and previously used Current performance results of coefficient sets with filtering mode to monitor, previously successful coefficient sets under similar conditions 20 failure to deliver expected performance, decreased confidence scores, or To determine distribution shifts if the parasite distribution changes. In the case of distribution shift, a limited trial of alternative coefficient sets. testing it in a low-risk manner within the window and the performance results to ensure that the filtering behavior is recalibrated accordingly and 25 Parasite classification performed, confidence score, selected filtering. mode, coefficient set, coefficient version, update time, rollback decision, post-filtering performance results and recalibration events configured to log for audit trail and versioning 30 in any of the above requests characterized by the server (4). such a system (1).