Search

CN-122002322-A - Wi-Fi router signal interference identification and coverage optimization method

CN122002322ACN 122002322 ACN122002322 ACN 122002322ACN-122002322-A

Abstract

The invention relates to the technical field of signal interference identification and discloses a Wi-Fi router signal interference identification and coverage optimization method, which comprises the steps of dividing a transmission time slot into a service transmission time slot and a micro-sampling time slot by adopting a time slot slicing asynchronous sampling mechanism, and completing interference preliminary classification and unknown interference feature extraction by combining feature similarity calculation; the method comprises the steps of finishing Do-calculus causal operation through frame gap lightweight intervention actions, finishing network health quantification by adopting multidimensional indexes, accurately positioning the degradation root cause of synchronous finishing performance, finishing optimal optimization action combination screening through multi-objective ratio operation, finishing uninterrupted optimization execution by adopting an MLO main and standby link dual-active frame, finishing automatic identification of an application scene through feature matching, finishing interference intensity prediction, finishing prepositive preventive optimization by matching a scene exclusive optimization strategy, and finishing optimization rollback and long-term self-adaptive closed loop through real-time index comparison.

Inventors

  • SONG JIAYI

Assignees

  • 上海鑫笙垚科技有限公司

Dates

Publication Date
20260508
Application Date
20260317

Claims (10)

  1. 1. A Wi-Fi router signal interference identification and coverage optimization method is characterized by comprising the following steps: Dividing a transmission time slot into a service transmission time slot and a micro-sampling time slot by adopting a time slot slicing asynchronous sampling mechanism, completing full-frequency-band, uplink and downlink and airspace full-dimension data acquisition by uplink reverse probe acquisition, airspace beam scanning acquisition and lightweight federal acquisition, and completing interference preliminary classification and unknown interference feature extraction by combining feature similarity calculation; Combining the collected full-dimensional signals with terminal data to construct a Wi-Fi network full-element causal directed acyclic graph, completing Do-calculus causal operation through frame gap lightweight intervention actions, completing network health quantification by adopting multi-dimensional indexes, and accurately positioning performance degradation root causes synchronously; Combining the output root cause tracing conclusion with the health degree quantification result, constructing a locally mapped network digital twin body, completing optimal optimization action combination screening through multi-objective ratio operation, completing uninterrupted optimization execution by adopting an MLO main-standby link dual-active framework, and synchronously realizing PHY and MAC cross-layer joint interference suppression; combining the full-link acquisition data with the optimization action execution data, completing automatic identification of an application scene through feature matching, completing interference intensity prediction through combining historical time sequence data, completing pre-preventive optimization through matching a scene exclusive optimization strategy, and completing optimization rollback and long-term self-adaptive closed loop through real-time index comparison.
  2. 2. The Wi-Fi router signal interference identification and coverage optimization method of claim 1, wherein the method for time slot slicing asynchronous sampling is: After the router is started, wi-Fi transmission time slots are automatically divided into service transmission time slots and micro-sampling time slots, 1 micro-sampling time slot is inserted into each 1 service frame every interval, a time slot slicing asynchronous sampling mode is adopted to conduct slicing acquisition on all frequency bands of 2.4GHz, 5GHz and 6GHz, the micro-sampling time slot interval is shortened to 0.5 service frames in an industrial scene, a default interval is kept in a household scene, and dynamic intervals are adopted in enterprise and stadium scenes.
  3. 3. The Wi-Fi router signal interference identification and coverage optimization method of claim 1, wherein the method for uplink reverse probe acquisition is: The router sends a reverse lightweight probe in a silent time slot, triggers directional time slot response of all terminals in a coverage area, collects collision characteristics, time slot offset and receiving power of response signals, and does not need the terminals to actively report data in the collection process.
  4. 4. The Wi-Fi router signal interference identification and coverage optimization method of claim 1, wherein the method for scanning and acquiring the airspace beam is as follows: and utilizing a multi-antenna array of the router to perform microsecond directional beam scanning in a micro-sampling time slot, collecting the intensity and the arrival angle of interference signals in different directions, capturing the direction and the distance characteristics of an interference source, and distinguishing near field/far field interference and co-frequency router interference/non Wi-Fi interference.
  5. 5. The Wi-Fi router signal interference identification and coverage optimization method of claim 1, wherein the lightweight federal acquisition method is: the terminal locally preprocesses self-health data, wherein the self-health data comprises a terminal type, a transmission rate, a packet loss condition and a service type, the terminal only uploads the encrypted characteristic value, and the router gathers the characteristic values uploaded by all the terminals to complete full-coverage detection of the health state of the whole network terminal.
  6. 6. The Wi-Fi router signal interference identification and coverage optimization method of claim 1, wherein the feature similarity calculation method is as follows: Denoising the acquired full-band signal data, uplink response data, airspace data and terminal characteristic data, extracting 3 core characteristics of time domain pulse width, frequency domain bandwidth and duty ratio of interference by adopting short-time Fourier transform of a sliding window, and primarily distinguishing Wi-Fi interference from non-Wi-Fi interference through lightweight characteristic matching; adopting an interference feature similarity calculation method to realize interference preliminary classification and unknown interference marking, and calculating feature similarity of the ith acquisition signal and known interference features The formula is as follows: Wherein, the For the time-domain pulse width of the ith acquisition signal, Is a standard time-domain pulse width for known interference, For the frequency domain bandwidth of the ith acquisition signal, Is the standard frequency domain bandwidth for known interference, For the duty cycle of the ith acquisition signal, And if the feature similarity is less than 0.7, the method judges the unknown interference, automatically extracts the features of the unknown interference and stores the features in a local feature library.
  7. 7. The Wi-Fi router signal interference identification and coverage optimization method of claim 1, wherein the causal root cause positioning method is as follows: The router locally builds a Wi-Fi network all-element causal directed acyclic graph, and the core nodes comprise interference source characteristics, channel states, MAC layer collision states, transmission layer performances, service experience and non-interference confusion variables, and are subjected to node association based on collected data; And inserting a microsecond-level directional channel probe, time slot detection and RTS/CTS directional triggering lightweight intervention action into a frame gap without interrupting service, removing the influence of a confusion variable through Do-calculus causal operation, and distinguishing specific causes of performance degradation.
  8. 8. The Wi-Fi router signal interference identification and coverage optimization method of claim 1, wherein the network health quantification method is as follows: constructing a health degree scoring system with 5 dimensions including coverage effectiveness, interference cleanliness, transmission reliability, service experience and equipment stability, quantifying the coverage effectiveness by adopting an effective signal-to-noise ratio, and calculating an effective signal-to-noise ratio ESNR, wherein the formula is as follows: Wherein, the For the total signal-to-noise ratio collected by the router, In order to determine the number of disturbances to be detected, For the similarity of the ith interference, The intensity of the ith interference; and calculating the influence of the quantized interference of the interference cleanliness on the channel, and distinguishing uplink/downlink and continuous/burst interference, wherein the influence is calculated by the formula: obtaining interference cleanliness , wherein, For the total strength of the uplink interference, For the total strength of the downlink interference, Maximum interference strength that can be tolerated by the router; Uplink and downlink bidirectional independent quantization covers packet loss rate, RTT and jitter, and the method comprises the following steps of: obtaining transmission reliability , wherein, For the average packet loss rate, For the maximum allowed round trip delay of the corresponding traffic, For the actual detected round trip delay, For the maximum allowable jitter of the corresponding traffic, Jitter is the actual detected jitter; the service experience degree is mapped according to core indexes of different service types, and the equipment stability is mapped to 0 minutes and 100 minutes according to the CPU occupancy rate, the radio frequency temperature and the terminal access stability.
  9. 9. The Wi-Fi router signal interference identification and coverage optimization method of claim 1, wherein the method for interruption-free optimization is: The router locally builds a 1:1 mapped virtual twin network based on the collected full-dimension data, health evaluation and root cause tracing result, and restores the positions, characteristics, channel states and service attributes of all routers, terminals and interference sources; By the calculation formula: obtaining a globally optimized action combination , wherein, In order to optimize the total intensity of the post-interference, In order to optimize the total strength of the pre-interference, In order to optimize the effective coverage area before, In order to optimize the effective coverage after the optimization, In order to optimize the post-average time delay, To optimize the front average time delay; when the optimization action is executed, all traffic flows are switched to the standby links, the main link enters an optimization mode, the channel switching, the power adjustment and the beam optimization action are executed, after the optimization is completed, three core indexes of ESNR, IC, TR are verified to be stable, and then the flows are switched back to the main link; Aiming at terminals which do not support MLO, the transmitting power and the beam forming weight are gradually and finely adjusted in the gap time slot of a service frame, the amplitude of each fine adjustment is less than or equal to 2dB, ESNR and IC changes are detected in real time after fine adjustment, physical layer and MAC layer parameters are adjusted by combining root cause tracing results, exclusive interference avoidance time slots are divided according to the time slot characteristics of interference, a directional RTS/CTS mechanism is adopted to send CTS signals to the terminals in the direction of hidden nodes, and the special micro time slots are allocated to each terminal by overlapping distributed time slot scheduling aiming at a high-density scene.
  10. 10. The Wi-Fi router signal interference identification and coverage optimization method of claim 1, wherein the pre-preventative optimization method is: the router automatically identifies core scenes based on the collected terminal number, terminal type, service model, interference rule and space characteristic data through lightweight characteristic matching, and each scene is preconfigured with a special optimization strategy; Through light time sequence feature extraction, the time rules of interference change, terminal access and service flow are learned, and the following formula is adopted: obtaining the predicted total interference intensity of the future 1 hour , wherein, For the average total interference intensity over the same period of 7 days, Total interference intensity for the same period of day k of the past 7 days; Acquiring ESNR, IC, TR core indexes in real time according to a prediction result, comparing the three core indexes with indexes before optimization, and if the optimized indexes meet the requirements that ESNR (equivalent series noise ratio) is more than or equal to 3dB, IC (integrated circuit) is more than or equal to 10%, TR (total power ratio) is more than or equal to 5%, reserving the optimization action; Combining feedback data, on-line adjusting a special strategy parameter of a scene and an optimization decision threshold value, and continuously updating an unknown interference feature library and a time sequence prediction model; and for periodic interference, according to a time sequence rule, solidifying and optimizing strategies.

Description

Wi-Fi router signal interference identification and coverage optimization method Technical Field The invention relates to the technical field of signal interference identification, in particular to a Wi-Fi router signal interference identification and coverage optimization method. Background With the large-scale deployment of Wi-Fi6 and Wi-Fi7 technologies, the requirements of low-delay, high-reliability and wide coverage of Wi-Fi networks in household high-density houses, enterprise offices, industrial sites, large venues and other scenes are continuously improved, signal interference identification and coverage optimization become the core direction of router performance improvement, but the prior art has the core defects of systematicness and scenerization, and cannot adapt to the rigidity requirement of complex actual scenes; In the prior art, the interference identification link has the dual defects of sampling blind areas and root cause misjudgment, namely, the interference identification link only depends on downlink passive sampling, cannot cover detection blind areas such as uplink hidden nodes, low-power consumption internet of things (IoT) equipment, airspace interference and the like, the inherent contradiction between full-band high-precision sampling and embedded platform low-calculation-power requirements cannot be solved, meanwhile, the inherent contradiction between the requirements cannot be solved, the relevance and causality are mixed, the performance degradation caused by non-interference factors such as shielding fading, terminal hardware bottleneck and the like is misjudged as interference, the coverage is evaluated by a single physical layer index such as RSSI and the like, the signal deficiency caused by interference superposition cannot be distinguished, the technical index and the real experience of a user are completely disjointed, the inherent conflict between coverage promotion and interference suppression cannot be balanced by adopting single-target logic in optimization, secondary interference is easy to be caused, the service interruption is trapped into invalid optimization circulation, the radio frequency parameter adjustment cannot be adapted to zero-interruption requirements such as industrial control and real-time communication, the differential requirements of different scenes cannot be adapted, the pre-preventive optimization capability is not provided, and the self-adaptive capability of a long-term environment is lost; In view of this, it is desirable to provide a Wi-Fi router signal interference identification and coverage optimization method. Disclosure of Invention The invention aims to provide a Wi-Fi router signal interference identification and coverage optimization method, which aims to solve the problems in the prior art, and is realized by the following technical scheme: In a first aspect, the method for identifying Wi-Fi router signal interference and optimizing coverage provided by the embodiment of the present invention specifically includes the following steps: Dividing a transmission time slot into a service transmission time slot and a micro-sampling time slot by adopting a time slot slicing asynchronous sampling mechanism, completing full-frequency band, uplink and downlink and airspace full-dimension data acquisition by uplink reverse probe, airspace beam scanning and light federal acquisition, and completing interference preliminary classification and unknown interference feature extraction by combining feature similarity calculation; Combining the collected full-dimensional signals with terminal data, constructing a Wi-Fi network full-element causal directed acyclic graph, completing Do-calculus causal operation through a frame gap lightweight intervention action, completing network health quantification by adopting a multi-dimensional index, and accurately positioning a performance degradation root cause synchronously; Combining the output root cause tracing conclusion with the health degree quantification result, constructing a local mapped network digital twin body, completing optimal optimization action combination screening through multi-objective ratio operation, completing uninterrupted optimization execution by adopting an MLO main-standby link dual-active framework, and synchronously realizing PHY and MAC cross-layer joint interference suppression; And step four, combining all-link acquired data with optimized action execution data, completing automatic identification of an application scene through feature matching, completing interference intensity prediction through historical time sequence data, completing pre-preventive optimization through matching a scene exclusive optimization strategy, and completing optimization rollback and long-term self-adaptive closed loop through real-time index comparison. In a second aspect, the system for identifying signal interference and optimizing coverage of a Wi-Fi router provided by the embodiment