Dynamic subcarrier distribution method and system for time-frequency multiplexing passive optical network

Abstract

The invention provides a dynamic subcarrier allocation method and system for a time-frequency multiplexing passive optical network, and the method comprises the steps: constructing a dynamic scheduling model of a time-frequency two-dimensional resource pool in each scheduling period according to the bandwidth demand condition of an ONU (Optical Network Unit) and the requirements of business demands for throughput, fairness and power balance, and carrying out the dynamic scheduling of the time-frequency two-dimensional resource pool; coding each sub-carrier allocation scheme as a particle position, optimizing targets such as throughput, fairness and power balance of the system at the same time by using a multi-target particle swarm, and reserving a diversified compromise optimal solution set through Pareto non-dominated sorting and congestion degree distance maintenance in iteration; and finally, selecting a feasible scheme from the Pareto frontier solution set according to network strategy preference, and issuing the feasible scheme to the OLT for execution, thereby realizing real-time self-adaptive distribution of the DSC. According to the method, time-frequency resources can be fully utilized in a mixed service scene, the overall throughput of a system is improved, the service quality of a weak link or a low-priority ONU is considered, and the basic physical feasibility requirement is met.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of fiber access network, in particular to a dynamic subcarrier allocation method and system of time-frequency multiplexing passive optical network. BACKGROUND

[0002] With the rapid rise of bandwidth and latency requirements of cloud computing, big data, 5G / 6G wireless front-haul, ultra-high-definition video (4K / 8K), and online games, access networks are facing unprecedented performance pressure and flexibility challenges. Among various access schemes, passive optical network (PON) is widely deployed and becomes the most important access method in fiber-to-the-home (FTTP / FTTR) and other fields due to its point-to-multipoint passive distributed architecture and centralized OLT and distributed ONU management mode. Early EPON / GPON systems achieved 1.25-2.5 Gbps user downlink bandwidth sharing through millisecond-level time division multiplexing (TDM), but it is difficult to provide sufficient bandwidth and flexibility in the face of explosive growth of high-definition video and large-scale data services; the subsequent 10G-PON (such as IEEE802.3av, ITU-T G.987) is increased to 10 Gbps on a single wavelength, but its fixed time slot allocation method often makes the resource utilization rate lower than expected; the hybrid time and wavelength division optical access network (TWDM-PON) based on ITU-T G.989 increases the capacity to 40-80 Gbps by superimposing multiple wavelengths (4-8 wavelengths), which significantly improves the downlink bandwidth, but the dependence on tunable lasers and arrayed waveguide grating (AWG) and other color optical devices significantly increases the cost and operation complexity of ONUs. Under this background, how to further improve the bandwidth utilization efficiency while maintaining or reducing the cost of terminals has become a problem to be solved in the development of PON technology.

[0003] Coherent TDM-PON based on coherent optical communication can achieve high-speed burst transmission on a single wavelength, but it is easily affected by coverage range and service diversity, resulting in scheduling delay, and coherent WDM- / TWDM-PON increases bandwidth through multiple wavelengths, but brings the significant disadvantage of a large increase in cost and complexity. Compared with the above, time and frequency division multiplexing (TFDM) coherent PON uses digital subcarrier multiplexing technology to divide a single optical carrier into multiple GHz-level subcarriers (DSC) in the frequency domain, while supporting burst / continuous transmission in the time domain, and constructs a time-frequency two-dimensional resource pool. Under the premise of using single-wavelength transmission / reception hardware, TFDM-PON can flexibly allocate resources on the microsecond-level time slot and GHz-level subcarrier, significantly improve the spectrum utilization rate, reduce the blocking rate, and reduce the demand for color optical devices, thereby balancing the cost and performance, and is one of the solutions widely concerned in the development and evolution of PON technology.

[0004] TFDM-PON can realize highly flexible resource multiplexing in time and frequency domain by dividing GHz-level digital subcarriers and combining microsecond-level time slots on a single-wavelength platform. However, the existing public experiments (all DSCs are uniformly configured as 6.25GBd / 16QAM) still adopt static allocation, which cannot dynamically adjust according to the link loss or service delay requirement of different ONUs, and it is also difficult to balance the low-order modulation delay constraint of 5G front transmission and the high-order wideband demand of 8K video, resulting in that the actual spectrum utilization is often less than 50%. At the same time, these methods lack a systematic trade-off strategy for multiple targets such as throughput, fairness and delay, and also fail to provide real-time response mechanism when ONUs are offline or services burst. Therefore, it is urgent to propose an intelligent optimization method for dynamic allocation of digital subcarrier resources in TFDM-PON in the time-frequency two-dimensional space, so as to fully release the performance advantages of TFDM-PON.

[0005] Dynamic subcarrier allocation in the TFDM-PON scenario is a high-dimensional, multi-objective combinatorial optimization problem, which needs to cooperatively optimize the conflicting targets such as system throughput, resource allocation fairness and link robustness under the premise of meeting the physical layer hard constraints (such as ONU power balance range, carrier frequency offset tolerance, dispersion tolerance, etc.) and system real-time requirements (such as transmission delay meeting microsecond-level requirement in industrial scenarios). Traditional heuristic scheduling methods (such as polling, fixed priority allocation) are often inefficient and have significant delay fluctuations in the decision space of 128 subcarriers x multiple modulation formats; while single-objective optimization algorithms (such as particle swarm optimization PSO which only maximizes throughput) can improve utilization in some scenarios, but lack multi-objective trade-off ability and are difficult to meet the multiple needs in mixed service scenarios. SUMMARY

[0006] In view of the shortcomings of static configuration of TFDM-PON subcarriers, the present application proposes a dynamic subcarrier allocation method and system for time-frequency multiplexing passive optical network, which simultaneously optimizes system throughput, fairness and power balance by using multi-objective particle swarm optimization, and realizes adaptive allocation of subcarriers through Pareto non-dominated sorting and crowding distance maintenance in iteration, to solve the problems proposed in the above background art. The technical scheme provided by the present application is as follows:

[0007] In the first aspect, a dynamic subcarrier allocation method for time-frequency multiplexing passive optical network is provided, and the following steps are implemented at the optical line terminal OLT side, which are repeated every time slot:

[0008] Step 1, read the concurrent capacity of N optical network units ONUs and bandwidth demand ;

[0009] Step 2, generate M particles, each particle encodes an allocation vector X (m), initialize velocity vector V (m) , the Pareto Archive is used to record the non-dominated solution set in the current iteration, and the initialization method is determined according to the strategy;

[0010] Step 3, multi-objective particle swarm iterative search;

[0011] Step 4, select the final scheme X* from the Pareto Archive according to the network strategy preference;

[0012] Step 5, map X* to OLT subcarrier control instructions, instructing subcarrier k to be transmitted or idle, and the ONU side starts / silences the corresponding subcarrier k according to the instructions;

[0013] Step 6, record X*, if the iteration is not completed or there is no legal solution, use the last time slot scheme or static scheme to fall back;

[0014] Step 7, if there is a subsequent time slot, jump back to step 1 to start again; if the overall scheduling is completed or the system is closed, terminate the process.

[0015] Preferably, the initialization strategy of step 2 is: when it is desired to improve the consistency and convergence speed of the scheduling solution between consecutive time slots, part of the Pareto solution of the last time slot is introduced into the current Archive as the initial reference; when it is desired to maintain the diversity of the solution set or avoid historical bias, the Archive is set to an empty set, and the current particle swarm explores a new solution set.

[0016] Preferably, the specific steps of step 3 are:

[0017] Step 3.1, fitness evaluation: for each particle X (m) , construct the allocation matrix , calculate the bandwidth of each ONU , calculate and , where K is the total number of allocable subcarriers, is the fixed subcarrier rate, is the total system throughput, is the fairness index;

[0018] Step 3.2, update the Pareto Archive: merge the current particle with the Archive scheme, perform non-dominated sorting, and extract the first layer solution; if the capacity is exceeded, remove the densest solution by congestion;

[0019] Step 3.3, update the particle optimal position and the particle swarm optimal position : update according to the non-dominated relationship, select from the Pareto Archive according to the strategy;

[0020] Step 3.4, Discrete PSO update position: adjust X in a discrete way based on inertia / individual / global influence, so that ;

[0021] Step 3.5, Conflict correction: correct the subcarrier mutual exclusion and ONU concurrency restriction, keep one or discard the redundant mapping to 0 when conflict;

[0022] Step 3.6, Convergence discrimination: if the Pareto front changes slightly or reaches the iteration upper limit or the running time is close to the time slot cutoff, jump out of the loop; otherwise, continue iteration.

[0023] Preferably, the strategy for selecting the final scheme is: when "throughput priority" is selected, the scheme with the maximum is selected; when "fairness priority" is selected, the scheme with the maximum is selected; when "balanced mode" is selected, the scheme with both and are high and the compromise is good is selected.

[0024] The second aspect is a dynamic subcarrier allocation system of a time-frequency multiplexing passive optical network, comprising an optical line terminal OLT and a plurality of optical network units ONUs connected through a passive optical splitter, wherein the OLT side comprises a dynamic subcarrier allocation control module, which is responsible for issuing subcarrier allocation instructions according to time slots through the method of any one of claims 1-4; and the ONU side comprises a fixed modulation format transceiver module, which is turned on or silenced according to the instructions.

[0025] Compared with the prior art, the present application has the following beneficial effects:

[0026] 1. The K digital subcarriers and T time slots of a time-frequency multiplexing passive optical network TFDM-PON are regarded as a time-frequency two-dimensional resource pool, so as to realize joint dynamic scheduling of the DSC in the time domain and the frequency domain;

[0027] 2. A multi-objective optimization framework is constructed, the global search capability of a particle swarm PSO is combined with a Pareto front compromise decision, and the best compromise among conflicting objectives such as system throughput, fairness and power balance is systematically obtained;

[0028] 3. The method is simple and easy to implement: the particle coding, PSO parameters and Pareto selection strategy are universally expandable and can be adapted to different sizes of TFDM-PON systems;

[0029] 4. The deployment cost is low: only the algorithm needs to be implemented in the OLT controller, without the need to add expensive optical components, and it is easy to integrate into the existing TFDM-PON platform.

[0030] The method is suitable for various time-frequency multiplexing passive optical networks (TFDM-PON), and in particular in a mixed service (high-bandwidth video, low-latency forwarding, a large number of small-bandwidth disturbances, etc.) scene, the method significantly optimizes resource utilization and service quality through multi-target dynamic allocation, and is simple, real-time and easy to deploy in an existing system. BRIEF DESCRIPTION OF DRAWINGS

[0031] The accompanying drawings are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and are included to provide a further understanding of the application, and are made a part of the description. In the drawings:

[0032] Figure 1 is a flow chart of the method of the application;

[0033] Figure 2 is a schematic diagram of the system structure of the application. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings accompanying the embodiments of the application. Obviously, the described embodiments are only a part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments of the application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the scope of protection of the application.

[0035] In order to make the above-mentioned purposes, features and effects of the application more obvious and easy to understand, the application will be further described in detail below with reference to the drawings and specific embodiments.

[0036] Embodiment 1: This embodiment will be described in detail in combination with the drawings to explain a dynamic subcarrier allocation method of a time-frequency multiplexing passive optical network (TFDM-PON) based on multi-target PSO and Pareto front analysis. The method involves the following premises and settings.

[0037] Fixed transmission rate: all subcarriers use the same modulation format, and the rate is denoted as (constant), in the scheduling algorithm, as a fixed parameter, which does not need to be updated in real time.

[0038] ONU concurrent capability: each The maximum number of subcarriers that can be concurrently used in a single time slot is determined by its hardware or service policy, denoted as The OLT reads the of each ONU when scheduling; if the service demand provides a bandwidth reference value , the value can be used as a priority reference factor in allocation initialization or conflict correction, but it is not a mandatory limit condition that must be strictly met in the algorithm.

[0039] Slot length: Set the length of the scheduling slot , it is required to ensure that the dynamic scheduling algorithm completes and issues instructions within this time. The specific value is determined by the OLT controller computing capacity and real-time service requirements (such as 100 μs, 500 μs, 1 ms, etc.).

[0040] Number of subcarriers: The system has K digital subcarriers, all of which are considered as allocable resources. When dynamically allocating, no channel feasibility filtering is performed in each scheduling, simplifying the calculation.

[0041] This method regards K subcarriers and one time slot as K resource units, forming a time-frequency two-dimensional resource pool. If T consecutive time slots are considered (where T represents the total number of time slots in the scheduling period), the entire resource pool can be represented as a KxT resource grid. However, this embodiment independently schedules each time slot, so each scheduling only focuses on the allocation of K resource units in the current time slot. The allocation scheme is encoded using a vector: define a vector of length K , where . If , it means that subcarrier k is allocated to ; if , it means that subcarrier k is idle. Thus, the allocation matrix is obtained, which is when and only when ; otherwise .

[0042] The objectives of multi-objective optimization include:

[0043] System total throughput: , which is equivalent to multiplying the number of allocated subcarriers by the rate.

[0044] Fairness index: To evaluate the resource allocation balance between different ONUs in the system, the Jain fairness index is defined as follows: Let there be N ONUs in the system, and the actual bandwidth obtained by each ONU is , the fairness index is defined as: , where , the closer the value is to 1, the more fair the allocation is; if all ONUs are allocated bandwidth completely equally, , if it is extremely uneven (such as one ONU monopolizing all bandwidth), then tends to 1 / N.

[0045] Constraint conditions include:

[0046] Subcarrier exclusivity: Each subcarrier is allocated to at most one ONU in the same time slot, .

[0047] ONU concurrency limit: Each allocating at most subcarriers per slot .

[0048] A dynamic subcarrier allocation method for time-frequency multiplexed passive optical networks, at OLT side, the following steps are implemented, repeating every slot:

[0049] Step 1, read the concurrent capability of each optical network unit ONU and bandwidth demand .

[0050] Step 2, generate M particles, each particle encodes an allocation vector X (m) , initialize velocity vector V (m) , and the Pareto Archive is used to record the non-dominated solution set in the current iteration, and the initialization method is determined by the strategy: when it is desired to improve the consistency and convergence speed of the scheduling solution between consecutive slots, part of the Pareto solution of the previous slot is introduced into the current Archive as the initial reference; when it is desired to maintain the diversity of the solution set or avoid historical bias, the Archive is set to an empty set, and the current particle swarm explores a new solution set.

[0051] Step 3, multi-objective particle swarm optimization (MOPSO) iterative search:

[0052] Step 3.1, fitness evaluation: for each particle X (m) , construct the allocation matrix , calculate the bandwidth of each ONU , calculate and ;

[0053] Step 3.2, update the Pareto Archive: merge the current particle with the Archive scheme, perform non-dominated sorting, and extract the first layer solution; if the capacity is exceeded, remove the densest solution by congestion;

[0054] Step 3.3, update the particle optimal position and the particle swarm optimal position : update according to the non-dominated relationship, select from the Pareto Archive according to the strategy;

[0055] Step 3.4, discrete PSO update position: based on inertia / individual / global influence, adjust X in a discrete manner so that ;

[0056] Step 3.5, conflict correction: correct the subcarrier mutual exclusion and ONU concurrency limit, and when there is a conflict, keep one or discard the redundant mapping and set it to 0;

[0057] Step 3.6, convergence criterion: if the Pareto frontier changes little or reaches the iteration upper limit I or the running time is close to the time slot cutoff , break the loop; otherwise, continue the iteration.

[0058] Step 4, select the final scheme X* from the Archive (Pareto frontier solutions) according to the network policy preference: when "throughput priority", select the scheme with the maximum in the frontier solution; when "fairness priority", select the scheme with the maximum ; when "balanced mode", the scheme with both and are high and the compromise is good (such as and are close to the solution at the inflection point of the frontier).

[0059] Step 5, map X* to OLT DSP control instructions, instructing the subcarrier k to be transmitted or idle, and the ONU side to turn on / off the corresponding subcarrier according to the instructions.

[0060] Step 6, record X* and the indicators, and if the iteration is not completed or there is no legal solution, use the last time slot scheme or the static scheme to fall back.

[0061] Step 7, if there is a subsequent time slot, jump back to step 1 to start again; if the overall scheduling is completed or the system is turned off, terminate the process.

[0062] Embodiment 2: a dynamic subcarrier allocation system for a time-frequency multiplexing passive optical network, as shown in Figure 2 , comprising an optical line terminal OLT and a plurality of optical network units ONUs connected through a passive optical splitter, the OLT side containing a dynamic subcarrier allocation control module responsible for issuing subcarrier allocation instructions by time slot; the ONU side maintains a fixed modulation format transmission module and only needs to turn on or mute the corresponding subcarrier according to the instructions.

[0063] Embodiment 3: a computer readable storage medium of the present embodiment, which stores a computer program, the program being executed by a processor to realize the steps of the dynamic subcarrier allocation method for a time-frequency multiplexing passive optical network of embodiment 1.

[0064] The computer readable storage medium of the present embodiment can be an internal storage unit of the terminal, such as the hard disk or memory of the terminal; the computer readable storage medium of the present embodiment can also be an external storage device of the terminal, such as a plug-in hard disk, a smart memory card, a secure digital card, a flash memory card, etc. equipped on the terminal; further, the computer readable storage medium can include both the internal storage unit and the external storage device of the terminal.

[0065] The computer readable storage medium of the embodiment is used to store a computer program and other programs and data required by the terminal, and can also be used to temporarily store data that has been output or will be output.

[0066] Embodiment 4: The computer device of the embodiment comprises a processor, a memory, and a computer program stored on the memory and executable on the processor, and the processor implements the steps in the dynamic subcarrier allocation method of the time-frequency multiplexing passive optical network of embodiment 1 when executing the program.

[0067] In the embodiment, the processor can be a central processing unit, and can also be other general-purpose processors, digital signal processors, application-specific integrated circuits, ready programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. The memory can include read-only memory and random access memory, and provide instructions and data for the processor. Part of the memory can also include non-volatile random access memory, for example, the memory can also store device type information.

[0068] Those skilled in the art can clearly understand that each embodiment can be realized by means of software and necessary general hardware platforms, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the method described in each embodiment or some parts of the embodiment.

[0069] The above only describes the preferred embodiments of the present application and is not used to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can modify the technical solutions described in the foregoing embodiments or make equivalent replacements to some technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A dynamic subcarrier allocation method for a time-frequency multiplexing passive optical network, characterized in that: On the optical line terminal (OLT), perform the following steps, repeating them for each time slot: Step 1: Read the concurrent capabilities of N optical network units (ONUs) and bandwidth requirements ; Step 2: Generate M particles, each particle encodes the distribution vector X (m) , initialize the velocity vector V (m) ,Pareto archive Archive is used to record the non-dominated solution set in the current iteration, and the initialization method is set according to the strategy; Step 3, multi-target particle swarm iterative search; Step 4: Select the final solution X* from the Pareto archive based on the network policy preference; Step 5: Map X* to the subcarrier control instruction of OLT, indicating that subcarrier k is Transmitting or receiving or idle, the ONU side turns on / off the corresponding subcarrier k according to the instruction; Step 6: Record X*. If the iteration is not completed or there is no legal solution, use the previous time slot solution or the static solution to fall back. Step 7: If there is a subsequent time slot, jump back to step 1 and start again; if the overall scheduling is completed or the system is shut down, terminate the process.

2. The method for dynamic subcarrier allocation in a time-frequency multiplexing passive optical network according to claim 1, wherein: The initialization strategy of step 2 is: when it is desired to improve the consistency and convergence speed of the scheduling solutions between consecutive time slots, the partial Pareto solution of the previous time slot is imported into the current Archive as the initial reference; when it is desired to maintain the diversity of the solution set or avoid historical bias, the Archive is set to an empty set, and the current particle swarm explores the new solution set.

3. The method for dynamic subcarrier allocation in a time-frequency multiplexing passive optical network according to claim 2, wherein: The specific steps of step 3 are: Step 3.1, fitness evaluation: for each particle X (m) , construct the allocation matrix , calculate the bandwidth of each ONU ,calculate and , where K is the total number of allocable subcarriers, is a fixed subcarrier rate, is the total system throughput, is a fairness indicator; Step 3.2, update the Pareto archive: merge the current particle and the Archive solution, perform non-dominated sorting, and extract the first-level solution; if it exceeds the capacity, eliminate the most dense solution based on the congestion degree; Step 3.3, update the optimal position of the particle and particle swarm optimal position : Update by non-dominant relationship, Select by strategy from the Pareto archive; Step 3.4, discrete PSO update position: Based on inertia / individual / global influence, adjust X in a discrete manner so that ; Step 3.5, conflict correction: Correct the subcarrier mutual exclusion and ONU concurrency limit, retain one item or discard the redundant mapping and set it to 0 when there is a conflict; Step 3.6, convergence judgment: If the Pareto front changes slightly or reaches the iteration limit or the running time is close to the time slot end, jump out of the loop; otherwise, continue to iterate.

4. The method for dynamic subcarrier allocation in a time-frequency multiplexing passive optical network according to claim 3, wherein: The strategy for selecting the final solution is: when "throughput priority", select the one with the maximum When "fairness first", choose the one with the largest In "Balanced Mode", select and Both are relatively high and are a good compromise solution.

5. A dynamic subcarrier allocation system for a time-frequency multiplexing passive optical network, characterized in that: The invention comprises an optical line terminal (OLT) and multiple optical network units (ONUs), which are connected via a passive optical splitter. The OLT side comprises a dynamic subcarrier allocation control module, which is responsible for sending subcarrier allocation instructions according to time slots by using the method described in any one of claims 1 to 4. The ONU side maintains a fixed modulation format transceiver module, which turns on or silences the corresponding subcarrier according to the instruction.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the dynamic subcarrier allocation method for a time-frequency multiplexing passive optical network according to any one of claims 1 to 4 are implemented.

7. A computer device comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the dynamic subcarrier allocation method for a time-frequency multiplexing passive optical network according to any one of claims 1 to 4 are implemented.

Images