Intelligent home network quality optimization system and method based on FTTR

Abstract

The invention belongs to the technical field of home network quality optimization, and particularly relates to an intelligent home network quality optimization system and method based on FTTR, and the system comprises a data collection module, a data processing module, an optical fiber topology reconstruction module, a resource distribution module and a management control module. The method comprises the following steps: acquiring optical fiber micro-strain of each node in real time to identify wall vibration or wiring abnormal data, constructing a 3D network health map, fusing multi-node data, predicting service flow fluctuation data in a specified time period in the future, and adjusting an optical fiber topological structure between a master optical gateway and a slave optical gateway in real time according to a user behavior mode and network load change. Optical path time delay and energy consumption efficiency are optimized, optical domain signal processing is realized by adopting a photon integrated circuit, optical wavelengths and wireless channels are dynamically distributed, microsecond-level optical path protection switching and equipment abnormity isolation are realized through a built-in optical link fault prediction model and an edge inference engine, and the reliability of the system is improved. The problems of coverage and rate contradiction, multi-device interference and switching existing in a traditional home network are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of home network quality optimization, and in particular relates to an intelligent home network quality optimization system and method based on FTTR. Background Art

[0002] The traditional FTTH home network refers to connecting optical fiber directly to the user's home and providing broadband services through an optical modem. Users need to handle the wired and wireless coverage after the optical modem by themselves. FTTH (Fiber To The Home) is a transmission method for optical fiber communications. It installs the optical network unit (ONU) at residential or corporate users. It is the optical access network application type closest to the user in the optical access series except for FTTD (Fiber to the Desktop). In a traditional FTTH home network, the operator will pull the optical fiber into the user's home and connect it to the optical modem, but the connection after the optical modem is the responsibility of the user. This usually includes using RJ45 network cables for wired deployment, or using Wi-Fi coverage to achieve network coverage throughout the house.

[0003] Traditional home networks have the following technical bottlenecks:

[0004] Conflict between coverage and speed: Although traditional FTTH (fiber to the home) can provide gigabit access to homes, a single router cannot cover large apartments (e.g., over 150 square meters). This results in severe attenuation of the WiFi signal in the room, and the actual speed is less than 100 megabits, which cannot meet the needs of high-bandwidth services such as 8K / VR.

[0005] Multi-device interference and switching issues: Multi-AP (access point) networking has problems such as high switching delay (>50ms) and significant packet loss rate (>10%), which affects the real-time service experience of online education, remote office, etc.

[0006] Therefore, the coverage and speed contradictions, multi-device interference and switching problems existing in traditional home networks are technical issues that need to be solved urgently. Summary of the Invention

[0007] The purpose of the present invention is to provide a FTTR-based smart home network quality optimization system and method to solve the coverage and rate contradiction, multi-device interference and switching problems existing in traditional home networks.

[0008] In order to solve the above technical problems, the technical solutions adopted by the present invention are as follows:

[0009] In a first aspect, a FTTR-based smart home network quality optimization system is provided, comprising a data acquisition module, a data processing module, a fiber topology reconstruction module, a resource allocation module, and a management control module, wherein the data acquisition module is connected to the data processing module, the data processing module is connected to the fiber topology reconstruction module, and the fiber topology reconstruction module and the management control module are both connected to the resource allocation module;

[0010] The data acquisition module is used to collect optical layer parameters, optical attenuation and IP layer parameters in real time, and upload the data to the management and control module through an encrypted channel;

[0011] The data processing module is used to build a global network quality prediction model based on a distributed learning framework by integrating local data of multiple home nodes, and output dynamic resource allocation strategies and optical-wireless collaborative optimization instructions. The local data includes channel state information and device load;

[0012] The fiber topology reconstruction module adjusts the fiber topology between the master and slave optical gateways in real time according to user behavior patterns and network load changes, optimizing optical path delay and energy efficiency;

[0013] The resource allocation module is used to implement optical domain signal processing using photonic integrated circuits and dynamically allocate optical wavelengths and wireless channels;

[0014] The management and control module implements microsecond-level optical path protection switching and equipment anomaly isolation through a built-in optical link fault prediction model and an edge inference engine.

[0015] Preferably, the data acquisition module includes a fiber optic acoustic wave sensing unit and a quantum noise encryption unit. The fiber optic acoustic wave sensing unit identifies wall vibration or wiring abnormality by detecting optical fiber microstrain and locates the fault point in combination with TDR technology. The quantum noise encryption unit encrypts the monitoring data based on quantum encryption of a quantum random number generator.

[0016] Preferably, the optical fiber acoustic wave sensing unit includes a sensing optical fiber module, a demodulation module, an acoustic wave signal reconstruction module, a feature extraction and classification module and a communication module, wherein the sensing optical fiber module is connected to the demodulation module, the demodulation module is connected to the acoustic wave signal reconstruction module, the acoustic wave signal reconstruction module is connected to the feature extraction and classification module, and the feature extraction and classification module is connected to the communication module;

[0017] The sensing fiber module is based on weak reflection grating array technology, which uses femtosecond laser to write thousands of grating points on a single optical fiber;

[0018] The demodulation module converts the optical signal into an electrical signal and demodulates the phase change of the acoustic wave through phase-sensitive optical time domain reflectometry;

[0019] The acoustic signal reconstruction module uses the intrinsic mode decomposition algorithm to denoise the demodulated signal and reconstructs the effective acoustic wave components through correlation screening;

[0020] The feature extraction and classification module extracts the time-frequency features of the sound wave based on wavelet transform, constructs a multi-dimensional feature matrix, and classifies the sound wave signal based on the multi-dimensional feature matrix;

[0021] The communication module transmits the collected data to the data processing module.

[0022] Preferably, the data processing module includes a privacy protection aggregation unit and a cross-domain knowledge migration unit. The privacy protection aggregation unit uses differential privacy to add noise to the local model gradient to ensure that user data does not go out of the domain. The cross-domain knowledge migration unit models the topological relationship of different apartment types through graph neural networks, and migrates the large-sized apartment training model to small and medium-sized apartments.

[0023] Preferably, the optical fiber topology reconstruction module performs dynamic optical fiber topology reconstruction through a reinforcement learning algorithm, and the reward function is expressed as follows:

[0024]

[0025] Among them, α = 0.6, β = 0.3, γ = 0.1, and the optimal topology strategy is output through the preset network;

[0026] During the dynamic fiber topology reconstruction process, fiber topology reconstruction is achieved based on dynamic bifurcation and merging of optical paths.

[0027] Preferably, the resource allocation module includes an optical domain service scheduling unit and an intelligent slicing unit. The optical domain service scheduling unit realizes hybrid scheduling of time division multiplexing and wavelength division multiplexing through silicon-based optical switches, and reserves dedicated optical channels for real-time services; the intelligent slicing unit divides the wireless spectrum into physical resource blocks, dynamically allocates them according to service needs, and enhances mobile broadband and reliable low-latency communication.

[0028] In a second aspect, a method for optimizing the quality of an FTTR-based smart home network is provided, which is implemented based on the FTTR-based smart home network quality optimization system, and includes the following steps:

[0029] S1: The data acquisition module acquires optical fiber micro-strain data from each node in real time to identify wall vibration or wiring anomalies and build a 3D network health map.

[0030] S2: The data processing module integrates multi-node data and predicts business traffic fluctuation data within a specified time period in the future;

[0031] S3: Fiber topology reconstruction module, which adjusts the fiber topology between master and slave optical gateways in real time based on user behavior patterns and network load changes, optimizing optical path latency and energy efficiency;

[0032] S4: Use photonic integrated circuits to implement optical domain signal processing and dynamically allocate optical wavelengths and wireless channels;

[0033] S5: Achieve microsecond-level optical path protection switching and device anomaly isolation through the built-in optical link fault prediction model and edge inference engine.

[0034] The beneficial effects of the present invention include:

[0035] The FTTR-based smart home network quality optimization system and method provided by the present invention includes a data acquisition module, a data processing module, a fiber topology reconstruction module, a resource allocation module, and a management and control module. This system acquires real-time fiber microstrain data from each node to identify wall vibration or wiring anomalies, constructs a 3D network health map, integrates multi-node data, and predicts traffic fluctuations within a specified future time period. Based on user behavior patterns and network load changes, it adjusts the fiber topology between master and slave optical gateways in real time, optimizing optical path latency and energy efficiency. It uses photonic integrated circuits for optical domain signal processing, dynamically allocates optical wavelengths and wireless channels, and implements microsecond-level optical path protection switching and device anomaly isolation through a built-in optical link fault prediction model and edge inference engine. This solves the coverage-rate conflict, multi-device interference, and switching issues that plague traditional home networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a structural diagram of the FTTR-based smart home network quality optimization system of the present invention.

[0037] Figure 2 Schematic diagram of the process of the FTTR-based smart home network quality optimization method of the present invention. DETAILED DESCRIPTION

[0038] The following is combined with Figures 1 and 2 The present invention is described in further detail:

[0039] Example 1

[0040] See attached Figure 1 As shown, a smart home network quality optimization system based on FTTR includes a data acquisition module, a data processing module, a fiber topology reconstruction module, a resource allocation module and a management control module. The data acquisition module is connected to the data processing module, the data processing module is connected to the fiber topology reconstruction module, and the fiber topology reconstruction module and the management control module are both connected to the resource allocation module.

[0041] The data acquisition module is used to collect optical layer parameters, optical attenuation and IP layer parameters in real time, and upload the data to the management and control module through an encrypted channel. The data processing module is used to build a global network quality prediction model based on a distributed learning framework by integrating local data of multiple home nodes, and output dynamic resource allocation strategies and optical-wireless collaborative optimization instructions. The local data includes channel status information and equipment load. The optical fiber topology reconstruction module adjusts the optical fiber topology between the master and slave optical gateways in real time according to user behavior patterns and network load changes, and optimizes optical path delay and energy efficiency. The resource allocation module is used to implement optical domain signal processing using photonic integrated circuits and dynamically allocate optical wavelengths and wireless channels. The management and control module realizes microsecond optical path protection switching and equipment abnormality isolation through a built-in optical link fault prediction model and an edge inference engine.

[0042] Example 2

[0043] Based on Example 1, the data acquisition module includes a fiber optic acoustic wave sensing unit and a quantum noise encryption unit. The fiber optic acoustic wave sensing unit identifies wall vibration or wiring abnormalities by detecting optical fiber microstrain and locates the fault point in combination with TDR technology. The quantum noise encryption unit encrypts the monitoring data based on quantum encryption of a quantum random number generator.

[0044] The fiber-optic acoustic wave sensing unit is deployed along the home's fiber optic cabling pathways (such as baseboards and ceilings). It utilizes a weak-reflection grating array (GRA) technology, inscribing a grating dot every 10 cm (reflectivity 0.05%) on the single-mode fiber. By detecting microstrain in the fiber (sensitivity 1με / √Hz), it can identify wall vibrations, such as those caused by drilling, or wiring anomalies, such as fiber bends. Incorporating time-domain reflectometry (TDR) technology, when strain anomalies are detected, it triggers the emission of a 1550nm wavelength, 10ns pulse width optical pulse. The reflected signal's time delay difference is used to locate the fault point with an accuracy of ±0.1m. The quantum noise encryption unit utilizes a true random key generation rate of 10Mbps based on a quantum random number generator (QRNG). It encrypts the collected acoustic wave data and network parameters using AES-256-GCM encryption, with a key update cycle of 1 minute. The encrypted data is uploaded to the management and control module via a dedicated management channel (independent of the service channel).

[0045] In this embodiment, the fiber optic acoustic wave sensing unit includes a sensing fiber module, a demodulation module, an acoustic wave signal reconstruction module, a feature extraction and classification module, and a communication module. The sensing fiber module is connected to the demodulation module, the demodulation module is connected to the acoustic wave signal reconstruction module, the acoustic wave signal reconstruction module is connected to the feature extraction and classification module, and the feature extraction and classification module is connected to the communication module. The sensing fiber module is based on weak reflection grating array technology, and thousands of grating points are engraved on a single optical fiber by a femtosecond laser. The demodulation module converts the optical signal into an electrical signal and demodulates the phase change of the acoustic wave through phase-sensitive optical time domain reflection. The acoustic wave signal reconstruction module uses the intrinsic mode decomposition algorithm to denoise the demodulated signal and reconstructs the effective acoustic wave component through correlation screening. The feature extraction and classification module extracts the time-frequency characteristics of the acoustic wave based on wavelet transform, constructs a multidimensional feature matrix, and classifies the acoustic wave signal based on the multidimensional feature matrix. The communication module transmits the collected data to the data processing module.

[0046] Example 3

[0047] Based on Example 1 or Example 2, the data processing module includes a privacy protection aggregation unit and a cross-domain knowledge migration unit. The privacy protection aggregation unit uses differential privacy to add noise to the local model gradient to ensure that user data does not go out of the domain. The cross-domain knowledge migration unit models the topological relationship of different apartment types through a graph neural network, and migrates the large-sized apartment training model to small and medium-sized apartments.

[0048] The privacy-preserving aggregation unit trains a 128-node LSTM network locally on each home node to learn the correlation between device load and channel state. Laplace noise (noise variance δ = 0.1) is added before uploading the model gradients, satisfying (ε = 0.5, δ = 0.01) differential privacy. The cross-domain knowledge transfer unit constructs a household topology graph (nodes are optical gateways, edges are fiber connections), extracts topological features using a graph attention network (GAT), and migrates the model trained on large households to apartment-sized units. After migration, the model's prediction error is reduced to less than 5%.

[0049] The fiber topology reconstruction module performs dynamic fiber topology reconstruction through a reinforcement learning algorithm. The reward function is expressed as follows:

[0050]

[0051] Among them, α = 0.6, β = 0.3, γ = 0.1, and the optimal topology strategy is output through the preset network;

[0052] During the dynamic fiber topology reconstruction process, fiber topology reconstruction is achieved based on dynamic bifurcation and merging of optical paths. A 1×4 silicon-based optical switch is used to bifurcate a single optical path into three sub-paths during the evening peak period, carrying video, gaming, and IoT services respectively. Phase conflicts are eliminated through optical couplers during merging.

[0053] The resource allocation module includes an optical domain service scheduling unit and an intelligent slicing unit. The optical domain service scheduling unit implements hybrid scheduling of time division multiplexing and wavelength division multiplexing through silicon-based optical switches, reserving dedicated optical channels for real-time services. The intelligent slicing unit divides the wireless spectrum into physical resource blocks and dynamically allocates them according to service needs, enhancing mobile broadband and reliable low-latency communications.

[0054] A Mach-Zehnder modulator with a bandwidth of 40 GHz enables hybrid TDM / WDM scheduling, allocating exclusive wavelengths (λ3 = 1551.72 nm) for real-time services such as cloud gaming, with latency less than 1 ms. Dynamic demultiplexing uses microring resonators to separate eight wavelengths, spaced 200 GHz apart, for a total single-fiber capacity of 800 Gbps.

[0055] The intelligent slicing unit divides wireless channels: the 6GHz band is divided into four 160MHz channels, and the optimal channel is dynamically bound based on the terminal's location. For example, CH3 is assigned to a remote terminal in a bedroom to mitigate wall attenuation. Air interface resource block allocation: The radio frame is divided into 100 physical resource blocks (PRBs), which are dynamically allocated based on service priority: eMBB: 60% PRB, URLLC: 30% PRB, and mMTC: 10% PRB.

[0056] The management and control module can predict optical link faults by collecting data such as laser bias current and temperature drift, training an LSTM model to predict lifespan (mean absolute error < 3 days), and providing early warning of optical module failures.

[0057] Microsecond-level self-healing control is achieved after optical link fault prediction: Optical path protection switching: When the primary wavelength λ1 fails, it switches to the backup wavelength λ2 (pre-configured in the wavelength routing table) within 10μs, with service interruption time less than 1ms. Device abnormality isolation: If a slave optical gateway CPU is overloaded (>90%), load migration is triggered (migration rate 10Gbps / s) and the maximum number of concurrent connections is limited to 50.

[0058] A method for optimizing the quality of an intelligent home network based on FTTR is implemented based on the system for optimizing the quality of an intelligent home network based on FTTR. Figure 2 , including the following steps:

[0059] S1: The data acquisition module acquires optical fiber micro-strain data from each node in real time to identify wall vibration or wiring anomalies and build a 3D network health map.

[0060] S2: The data processing module integrates multi-node data and predicts business traffic fluctuation data within a specified time period in the future;

[0061] S3: Fiber topology reconstruction module, which adjusts the fiber topology between master and slave optical gateways in real time based on user behavior patterns and network load changes, optimizing optical path latency and energy efficiency;

[0062] S4: Use photonic integrated circuits to implement optical domain signal processing and dynamically allocate optical wavelengths and wireless channels;

[0063] S5: Achieve microsecond-level optical path protection switching and device anomaly isolation through the built-in optical link fault prediction model and edge inference engine.

[0064] In another implementation of this embodiment, a 3D network health map is first constructed. The data acquisition module obtains the optical fiber microstrain data of each node in real time, and combines the TDR positioning results to generate a 3D map containing wall vibration intensity, optical fiber bending loss and equipment status (XYZ axes are physical position, signal strength, and time dimension respectively). Then, business traffic prediction is performed. The data processing module integrates multi-node historical data (time window of 1 hour) and predicts business traffic fluctuations in the next 30 minutes through a learning model, such as a video traffic peak of 8Gbps±5%. Dynamic topology reconstruction is then performed. According to user behavior, such as mobile phone positioning showing that the user moves from the study to the living room, a tree topology is switched to, shortening the number of optical path hops, reducing the delay from 15ms to 8ms, and reducing power consumption by 25%. Photonic-wireless collaborative scheduling is then performed, and the optical domain allocates λ3-λ5 wavelengths to carry high-priority services. The wireless side reserves 30% PRB for URLLC slices to ensure that the upload delay of smart door lock alarms is less than 50ms. Finally, fault isolation and recovery are performed. When it is predicted that the living room optical module has 2 days left in its life, the backup module is automatically isolated and started, and a maintenance work order is simultaneously pushed to the user app.

[0065] In summary, the present invention provides a system and method for optimizing the quality of an intelligent home network based on FTTR.

Claims

1. A smart home network quality optimization system based on FTTR, characterized in that: It includes a data acquisition module, a data processing module, a fiber topology reconstruction module, a resource allocation module and a management control module. The data acquisition module is connected to the data processing module, the data processing module is connected to the fiber topology reconstruction module, and the fiber topology reconstruction module and the management control module are both connected to the resource allocation module. The data acquisition module is used to collect optical layer parameters, optical attenuation and IP layer parameters in real time, and upload the data to the management and control module through an encrypted channel; The data processing module is used to build a global network quality prediction model based on a distributed learning framework by integrating local data of multiple home nodes, and output dynamic resource allocation strategies and optical-wireless collaborative optimization instructions; The fiber topology reconstruction module adjusts the fiber topology between the master and slave optical gateways in real time according to user behavior patterns and network load changes, optimizing optical path delay and energy efficiency; The resource allocation module is used to implement optical domain signal processing using photonic integrated circuits and dynamically allocate optical wavelengths and wireless channels; The management and control module realizes microsecond-level optical path protection switching and equipment abnormality isolation through a built-in optical link fault prediction model and an edge inference engine.

2. The FTTR-based smart home network quality optimization system according to claim 1, characterized in that: The data acquisition module includes a fiber optic acoustic wave sensing unit and a quantum noise encryption unit. The fiber optic acoustic wave sensing unit identifies wall vibration or wiring anomalies by detecting optical fiber microstrain and locates the fault point in combination with TDR technology. The quantum noise encryption unit encrypts the monitoring data based on quantum encryption using a quantum random number generator.

3. The FTTR-based smart home network quality optimization system according to claim 2, characterized in that: The optical fiber acoustic wave sensing unit includes a sensing optical fiber module, a demodulation module, an acoustic wave signal reconstruction module, a feature extraction and classification module and a communication module, wherein the sensing optical fiber module is connected to the demodulation module, the demodulation module is connected to the acoustic wave signal reconstruction module, the acoustic wave signal reconstruction module is connected to the feature extraction and classification module, and the feature extraction and classification module is connected to the communication module; The sensing fiber module is based on weak reflection grating array technology, which uses femtosecond laser to write thousands of grating points on a single optical fiber; The demodulation module converts the optical signal into an electrical signal and demodulates the phase change of the acoustic wave through phase-sensitive optical time domain reflectometry; The acoustic signal reconstruction module uses the intrinsic mode decomposition algorithm to denoise the demodulated signal and reconstructs the effective acoustic wave components through correlation screening; The feature extraction and classification module extracts the time-frequency features of the sound wave based on wavelet transform, constructs a multi-dimensional feature matrix, and classifies the sound wave signal based on the multi-dimensional feature matrix; The communication module transmits the collected data to the data processing module.

4. The FTTR-based smart home network quality optimization system according to claim 1, characterized in that: The data processing module includes a privacy protection aggregation unit and a cross-domain knowledge transfer unit. The privacy protection aggregation unit uses differential privacy to add noise to the local model gradient to ensure that user data does not go out of the domain. The cross-domain knowledge transfer unit uses a graph neural network to model the topological relationship of different apartment types and migrate the large-sized apartment training model to small and medium-sized apartments.

5. The FTTR-based smart home network quality optimization system according to claim 1, characterized in that: The fiber topology reconstruction module performs dynamic fiber topology reconstruction through a reinforcement learning algorithm. The reward function is expressed as follows: Among them, α = 0.6, β = 0.3, γ = 0.1, and the optimal topology strategy is output through the preset network; During the dynamic fiber topology reconstruction process, fiber topology reconstruction is achieved based on dynamic bifurcation and merging of optical paths.

6. The FTTR-based smart home network quality optimization system according to claim 1, characterized in that: The resource allocation module includes an optical domain service scheduling unit and an intelligent slicing unit. The optical domain service scheduling unit implements hybrid scheduling of time division multiplexing and wavelength division multiplexing through silicon-based optical switches, reserving dedicated optical channels for real-time services. The intelligent slicing unit divides the wireless spectrum into physical resource blocks and dynamically allocates them according to service needs, enhancing mobile broadband and reliable low-latency communications.

7. A method for optimizing the quality of an intelligent home network based on FTTR, implemented based on the system for optimizing the quality of an intelligent home network based on FTTR according to any one of claims 1 to 6, characterized in that: The following steps are involved: S1: The data acquisition module acquires optical fiber micro-strain data from each node in real time to identify wall vibration or wiring anomalies and build a 3D network health map. S2: The data processing module integrates multi-node data and predicts business traffic fluctuation data within a specified time period in the future; S3: Fiber topology reconstruction module, which adjusts the fiber topology between master and slave optical gateways in real time based on user behavior patterns and network load changes, optimizing optical path latency and energy efficiency; S4: Use photonic integrated circuits to implement optical domain signal processing and dynamically allocate optical wavelengths and wireless channels; S5: Achieve microsecond-level optical path protection switching and device anomaly isolation through the built-in optical link fault prediction model and edge inference engine.

Images