CN101826919B - Mixed type passive optical network structure and method for positioning and restoring faults thereof - Google Patents

Abstract

The invention discloses a hybrid passive optical network and its fault location and recovery method. The passive optical network includes an optical line terminal, optical network units divided into several optical network unit groups, and a remote node. The optical line terminal and the remote node are connected through two feeder optical fibers, the remote node and each optical network unit group are connected in a dual-fiber star topology, and the optical network unit groups are connected in a dual-fiber tree or cross bus type The optical network unit that needs to be protected in the structural connection group; the present invention adopts hybrid TDM/WDM technology to effectively improve the access expansion capability of the network, and provides convenience for the gradual increase of network bandwidth and the number of users; the network provided by the present invention can be analyzed through statistics The method performs fault location, and can realize flexible network fault recovery through the action of the optical line terminal, and improve the survival rate of the network.

Description

A Hybrid Passive Optical Network and Its Fault Location and Restoration Method

technical field

The invention relates to a passive optical network, and in particular designs a passive optical network with a protection function and a fault location and recovery method thereof.

Background technique

Optical fiber access technology is one of the main technologies to realize the next-generation broadband access network. It has the characteristics of high bandwidth, large capacity, high reliability, and good service quality, and it is convenient to realize multi-network integration of voice, data, and video. In such a special occasion as the application of the access network, with the increase of the service type and bandwidth, especially the possible incorporation of the voice service, the requirements for its reliability and survivability are getting higher and higher, and the network is required to have Redundant protection and rapid recovery capabilities should keep the childlike innocence to a minimum even in the event of a catastrophic accident. From a cost point of view, protection should also be achieved at a small cost. In addition, the access network is also required to have good scalability to meet the ever-increasing requirements of bandwidth, number of access users, access service types, etc., so as to meet the needs of network upgrades.

In the passive optical network (hereinafter referred to as PON), there are currently two technologies: wavelength division multiplexing passive optical network (hereinafter referred to as WDM-PON) and time division multiplexing passive optical network (hereinafter referred to as TDM-PON). Among them, TDM-PON technology is relatively mature, with high bandwidth utilization rate and low device cost, and has been widely used; while WDM-PON can provide higher bandwidth for each access terminal and has good scalability, which can be used in the future. The upgrade of network applications can be guaranteed, but the cost of components is relatively high, and the utilization rate of optical broadband resources is low. In addition, whether it is TDM-PON technology or WDM-PON technology, the maximum number of users that can be accessed is limited by power and bandwidth, which still has limitations for some high-density access applications. In contrast, the mixed use of WDM-PON and TDM-PON technologies can achieve as many as hundreds or thousands of user accesses under the premise of guaranteed bandwidth, with higher bandwidth resource utilization, and can be used for TDM-PON The upgrade process to WDM-PON provides a smooth transition.

As far as the network topology is concerned, the structures currently in use are mainly tree (including star, bus) and ring. The tree structure can disperse the risk of distributing optical fibers, and has good scalability, but the failure of pig liver and each branch feeder optical fiber will affect all the ONUs connected to it, and the protection of its optical fibers is generally realized by double optical fibers. The redundancy is high, the backup efficiency is low, and the backup and working fibers have the same route, so there is a risk of sharing. The ring structure can achieve redundant protection with fewer optical fibers, and there is no shared risk problem between the working and backup optical fibers on the ring. The number of ONU nodes, and its scalability is poor.

For TDM-PON and WDM-PON, there are currently various schemes that use double redundancy of network components to achieve protection, such as using double backbone optical fibers or all optical fibers, and double optical line terminals (hereinafter referred to as OLTs). ) or the transceiver in the OLT, the double ONU or the transmitter and receiver in the ONU, etc., combined with the optical splitter or optical switch of the splitter, the gating of the electrical control unit, etc. The protection of the components in the ONU or the transceiver in the ONU, and the comprehensive protection of each component. In addition, there is a method of realizing mutual backup and protection of nodes by connecting two adjacent OLT or ONU nodes. For the hybrid passive optical network (hereinafter referred to as WDM/TDM-PON) structure, there is currently no corresponding network protection method.

For the detection of network faults, the optical time domain reflectometer (OTDR detection below) can be used for fault monitoring and location, or by detecting the power of the new uplink/downlink model to calculate the line loss for fault monitoring and location, or by detecting the uplink Signal to detect faults, or use a dedicated detection link for detection, etc. Among these methods, the use of OTDR is not only expensive to implement (OTDR is expensive), but also not suitable for branch fault location of TDM-PON. When used for WDM-PON, it is necessary to have a multi-wavelength OTDR light source, further increasing cost of testing equipment. The power detection unit is respectively set in the ONU and the OLT by detecting the uplink/downlink loss. Only when the normal communication between the ONU and the OLT is guaranteed can the interaction of the optical power information measured by the two be realized to calculate the loss. Faults in the optical path under test may result in the inability to interact with test information. However, when a dedicated detection network is used to transmit network performance monitoring and management information, misjudgment may occur when the monitoring network fails and the data transmission network is not faulty, resulting in normal communication being affected. Fault judgment only by detecting the information of the uplink signal is applicable to the case where the uplink and downlink signals are transmitted bidirectionally over a single fiber, and can quickly realize fault detection.

According to the location of protection control and recovery actions, network protection and recovery can be divided into distributed (ONU-side control and switching) and centralized (OLT-side control and switching) protection and recovery. Multi-point control restoration type and centralized single-point control restoration type. Centralized control and recovery can reduce the cost of the ONU, which is conducive to the reduction of the terminal price of the optical access network and the promotion of applications. In the OLT, when a single switch is used for single-point control, fewer control components are required, and the total network cost is relatively low. The nitrogen protection action often restores the business of the ONU affected by the fault, and often causes some ONUs that are working normally to be affected. In addition, it can generally only recover a single fault; while using multiple optical switches for control recovery at the cost of more components can achieve partial recovery of local faults, and may recover multiple faults.

Contents of the invention

Purpose of the invention: In view of the above problems, the present invention provides a hybrid passive optical network with a centralized control optical fiber protection function, and a method for fault location and fast recovery based on the network, which can effectively improve network access expansion It provides convenience for the gradual increase of network bandwidth and the number of users, and can realize flexible network protection. It can realize partial restoration of partial faults in the distribution fiber, and can realize the maximum degree of network in the case of multiple faults in the network. recover.

Technical solution: In order to achieve the above object, the technical solution provided by the invention is:

A hybrid passive optical network, including an OLT, a remote node (hereinafter referred to as RN) and m ONUs, where:

There are two ports outside the OLT that can back up each other and work at the same time, called the main port and the backup port. The main port is connected to the remote node through the main feeder fiber, and the backup port is connected to the remote node through the backup feeder fiber. , the uplink/downlink signals in the network are transmitted by co-fiber bidirectional transmission.

All ONUs are divided into n ONU groups (1≤n≤m) according to the bandwidth required by each ONU, the location of the ONU, etc., and each ONU group adopts dual-fiber tree, dual-fiber star, cross bus or Other tree structures connect the ONUs that need to be protected in the group, and the upstream signals of each ONU are sent to the optical paths of the two distribution fibers connected to it at the same time. For other ONUs that do not need protection, various existing access structures can be used to realize their connection with RN;

The RN consists of a wavelength division multiplexer (p≥n) with 2×2p ports (g×h means having g input ports and h output ports, or having h input ports and g output ports), to The dual-fiber star topology is connected to each ONU group.

A simple implementation of the OLT is to use two same wavelength division multiplexing optical transceiver modules (or optical transceiver arrays, etc.) inside the OLT, which are called the main optical transceiver module and the backup optical transceiver The port and the backup port, the main and backup optical transceiver modules can work independently of each other under the control of the control circuit and can work in wavelength division or/and time division coordination. Both can transmit and receive n wavelength signals corresponding to n groups.

Another implementation of the OLT is to use a set of wavelength division multiplexing optical transceiver modules (or optical transceiver arrays, etc.) inside the OLT to connect to the main port and the backup port through a 1×2 optical path switching module. The 1×2 optical path switching module controls the optical signal of the optical transceiver module to input or output from the active and standby ports through the control circuit.

The 1×2 optical path switching module includes a 1×2 optical coupler, a wavelength gating unit, and an optical switch unit with at least 2×2 ports; the single-port port (with one port) of the optical path switching module The access end of the optical coupler is called a single-port end, and the access end with two ports is called a dual-port end) connected to the single-port end of the optical coupler, and one port of the dual-port end of the optical coupler is connected to the input port of the wavelength gating unit, so The other port of the dual-port port of the optical coupler and the output port of the wavelength gating unit are respectively connected to two ports of the input end of the optical switch unit, and the output ports of the wavelength gating unit of the optical switch unit are respectively connected to the The two ports of the dual-port port of the above-mentioned optical path switching module. The wavelength gating unit can be composed of a wavelength tunable filter. At this time, the optical switch unit needs to use an optical switch with a port number of at least 2×2. The wavelength gating unit can also be composed of a 1×1 optical switch and a wavelength The tunable filter is formed in series; the tunable filter can be a single-wavelength filter or a multi-wavelength filter, that is, an optical filter with multiple adjustable passbands, such as an acousto-optic tunable filter.

When the cross bus type connection is used inside the ONU group, each ONU group has two buses corresponding to the same wavelength or a pair of wavelengths or a group of wavelengths extending from the output end of the RN (called the main optical fiber bus and the backup optical fiber). bus) to connect all the ONUs in the group in opposite directions, so as to avoid the risk sharing problem of using the same route for the main optical fiber and the backup optical fiber. The connection method is as follows: the main optical fiber bus and the standby optical fiber bus downstream from the RN connect the ONUs sequentially in the opposite direction and order to form a bus structure that crosses each other. For example: for an ONU group with k ONUs, its The order (from the RN end) on the active optical fiber bus is the i+1th ONU, and its order on the standby optical fiber is the k-1th. The two buses can be connected to each ONU through an optical splitter. The optical fiber at the end of the bus is directly connected to the end ONU without splitting, or the other end is connected to an attenuator after being connected to the end ONU through an optical splitter to facilitate further expansion of the network in the future.

The ONU group can also be connected in a star structure. For an ONU group with k ONUs, the two optical fibers extending from the RN need to use 1:k optical splitters to connect to all k ONUs in the group.

The hybrid passive optical network adopts the following bandwidth utilization mode: ONU groups are connected in WDM mode, and ONUs in each group are connected in TDM or TDM/WDM hybrid mode. Each ONU group uses different wavelengths, which correspond to the wavelengths corresponding to different ports of the connected RN. Each ONU group can use the wavelength at the corresponding position in one or more free spectrum intervals (hereinafter referred to as FSR) of the RN. If multiple FSRs are used, the wavelength interval between ONU upstream/downstream signals can be a non-zero integer multiple of FSR; the same or different FSR can be used for signals on two different optical paths in the same ONU group; unused ONUs within the ONU group can use The same wavelength can also use different wavelengths whose wavelength interval is a non-zero integer multiple of FSR.

The present invention also provides a method for realizing fault location and recovery when an optical fiber fault occurs in this network by statistically analyzing the communication status of each ONU. The fault location method is as follows:

1) Detect the uplink signal of the main feeder fiber, and judge whether it can communicate normally by detecting the communication status of each optical network unit;

2) If all optical network units cannot communicate normally, it is judged that the main feeder fiber may be faulty;

3) If only some optical network units cannot communicate normally, it is judged that the main distribution fiber is faulty, and:

a) For a network that adopts a crossover bus type connection within the optical network unit group, if the i-th optical network unit on the main optical fiber bus and all subsequent optical network units cannot communicate normally, it is judged that the i-th optical network unit on the bus If the i and the i+1th optical network unit are not all unable to communicate normally, it is judged that the i-th optical network unit may be faulty;

b) For a network using a dual-fiber star connection inside the ONU group, if the i-th ONU cannot communicate normally, it is judged that the main distribution fiber connected to the i-th ONU may be faulty;

4) Perform uplink signal detection on the standby feeder fiber, judge whether it can communicate normally by detecting the communication status of each optical network unit, and perform the same judgment as steps 2) and 3) on the detection structure.

5) Comparing the judgment structure of the main feeder fiber and the backup feeder fiber, if the two optical paths corresponding to the same optical network unit fail only on one hop optical path, it is judged as a single fiber failure; if both optical paths fail If the fault occurs on the same link, it is judged as a link dual-fiber fault; if the fault occurs on both optical paths but the fault is located on a different link, it is judged as a dual-fiber interleaved fault.

For the various faults that the PON described in the present invention may encounter in use and work, the network recovery can be carried out by several controls and actions of the OLT, and its recovery method is as follows:

When a single-fiber failure occurs in the main feeder fiber being used, the fault is recovered through the backup feeder fiber by using the backup port alone;

When a single-fiber failure occurs in the main distribution fiber connected to the main port of the optical line terminal, the failure is recovered by using the backup port alone through the backup feeder fiber;

When there is a double-optical path interleaving fault in the same optical network unit grouping in the distributed optical fiber, use the simultaneous coordination of each port corresponding to the wavelength of the optical line terminal to recover from the fault;

For the network connected by the cross bus structure inside the optical network unit group, when the dual-fiber fault of the link occurs in the distribution fiber, the fault is recovered by using the simultaneous work of the corresponding wavelengths of the two ports of the optical line terminal.

Beneficial effects: Compared with the prior art, the present invention has the following advantages: 1. The network adopts hybrid TDM/WDM technology, so that the entire network has better expansion capability than TDM-PON and WDM-PON, and can be accessed The total number of users is more difficult and the total capacity of the network is larger. The method of expanding the number of users of this PON can not only add new ONUs inside the ONU group, but also increase the number of ONUs by adding new wavelength groups (some ports can be reserved when selecting and designing RNs). If TDM is used alone - The maximum number of users accessed by PON and WDM-PON are respectively m1 and m2, then the maximum number of users that can be accessed by this TDM/WDM-PON can reach m1×m2. There are two ways to increase network capacity, which can add new wavelengths or increase the speed of existing OLT and ONU terminals. At the same time, the upgrade and expansion of the network may not affect the communication services of existing users.

2. Using centralized control, protection and recovery technology, and keeping ONU low cost, it can realize partial recovery for the partial failure of the distributed optical fiber without affecting the normal communication of other ONU groups. In the case of dual-fiber failure, all normal communication The ONU will not be affected by the recovery action, and at the same time, the network recovery can be realized to the greatest extent in the case of multiple concurrent faults in the network, which improves the survival rate of the network.

3. The cross-bus structure used in the ONU group has better scalability than the ring structure. It can not only reduce the impact on the normal communication ONU in the process of network expansion, such as adding a new ONU at the end of the bus, but also because the signal does not need Loopback, so under the condition of the same maximum transmission distance, it has more accessible users than the passive ring network, and has a longer transmission distance under the same condition of the same number of users; compared with the dual-fiber star structure, it is lower The degree of risk correlation between the working and backup paths is improved, and at the same time, the amount of required optical fibers is effectively reduced, and the amount of required optical fibers is the least among various known protection structures. When the dual-fiber star structure is used in the ONU group, the network can have better scalability than the cross-bus structure in the group, and the number of intervening users under the condition of power limitation can be further increased (due to the insertion of devices on the optical path) The cumulative value of the loss is relatively small), and at the same time, the access of the new ONU from any position of the RN will not affect the original ONU.

4. Fault location for optical fiber and ONU faults by detecting upstream signals, avoiding the use of expensive OTDR or additional optical transceivers specially used for fault detection, reducing the cost of fault detection and location, and judging results based on fault location A simple and fast fault location method is given.

5. When the bus structure is adopted in the group, the end of the bus is directly terminated at the ONU or link attenuator at the end, which avoids the interference problem of the end reflection on the signal that usually exists in the bus type/ring structure.

Description of drawings

Fig. 1 is a structural representation of the present invention;

Figure 2 is a schematic diagram of two structures of RN used in the present invention;

Fig. 3 is two kinds of structural representations of the ONU grouping internal link used among the present invention;

Fig. 4 is two kinds of structural representations of the OLT used in the present invention;

FIG. 5 is a schematic diagram of two structures of the optical path switching module in FIG. 4;

Fig. 6 is a schematic diagram of recovery situations of the network provided by the present invention under several failure situations;

Figure 7 is a calculation result diagram of the survival rate of access networks with various topological structures changing with the number of faults.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings.

Description of text and symbols in graphics:

As shown in FIG. 1 , it is a schematic structural diagram of a hybrid passive optical network with a centralized control redundancy function proposed by the present invention. The network includes an OLT with two ports that can work independently of each other, back up each other or work at the same time (referred to as the main port and the standby port respectively), m ONUs divided into n ONU groups, and an OLT with 2×2p RN of ports, where p≥n. The main port and backup port of the OLT are connected to the RN through two independent trunk/feeder fibers (respectively called the main feeder fiber and the backup feeder fiber), and the signals in the two trunk/feeder fibers are input from the RN input port Afterwards, output from at least n output ports at the output end respectively; each output port of RN is connected to each corresponding ONU group through the main distribution optical fiber and the spare distribution optical fiber. The solid line in the figure indicates the main optical path, and the dotted line indicates the standby optical path.

As shown in FIG. 2, there are two structures of the RN in the structure in FIG. 1, wherein the solid line indicates the main optical path, and the dotted line indicates the backup optical path, and the main and backup optical paths can be replaced with each other. The RN shown in Figure 2(a) is composed of an N×N wavelength division multiplexer, where N≥n+1. In the figure, an N×N arrayed waveguide grating (hereinafter referred to as AWG) is taken as an example. Its completely symmetrical There are a pair of symmetrical ports at both ends connected to the main/standby port in the OLT through the main/standby feeder fiber, and the other ports are connected to all ONU groups through the main/standby distribution fiber, corresponding to the two ports of the same ONU group The positions of the leading ends are symmetrical. Figure 2(b) consists of two 1×n wavelength division multiplexers. The multiplexer ends of the two wavelength division multiplexers are respectively connected to the main/standby port of the OLT through the main/standby feeder fiber. The wavelength ports are respectively connected to all ONU groups through the main/standby distribution fiber, and the ports corresponding to the same wavelength are connected to the same ONU group.

As shown in Figure 3, the internal connection structure of the ONU group in the structure of Figure 1 contains k ONUs. Different ONUs in the group use TDM technology or TDM/WDM hybrid technology to achieve bandwidth allocation, where the solid line represents the main optical path, and the dotted line represents Alternate light path. Figure 3 (a) uses a cross-bus structure, and Figure 3 (b) uses a dual-fiber star structure.

Figure 3 (a) The specific implementation of the crossover bus structure is as follows: the independent main and backup optical fiber buses downstream from the RN use a series of optical splitters to connect the ONUs in the group to the bus, and the main optical fiber bus At the end of the first section of distribution fiber, use the first optical splitter to connect an ONU (ONU _{i, k} in the figure) at one end of the group to the bus, and the other end of the first optical splitter is connected to the main The second section of distribution fiber on the optical fiber bus, and the end of which uses an optical splitter to connect the next ONU (ONU _{i, k-1} in the figure), that is, the second ONU on the main optical fiber bus to the bus. Sequentially connect all ONUs in the group to the main optical fiber bus; at the same time, the standby optical fiber bus starts from the other end of the ONU group (ONU _{i, 1} ) and connects each ONU in the group in reverse order in the opposite direction. For the ONU whose order is i+1 on the active optical fiber bus, its order on the standby optical fiber is ki. After the active and standby optical fiber buses connect all the ONUs in the ONU group, each ONU is located on the active and standby optical fiber buses at the same time. At this time, each ONU can only be connected to the active on the fiber optic bus. The optical fiber at the end of the bus is directly connected to the end ONU without splitting. It can also be connected to the end ONU through an optical splitter and then connected to an attenuator at another point to facilitate further expansion of the network.

Figure 4 shows the two structures of the OLT in Figure 1. The OLT shown in Figure 4 (a) uses two sets of wavelength division multiplexing optical transceiver modules (or optical transceiver arrays), respectively called The main optical transceiver module and the backup optical transceiver module are respectively connected to the main port and the backup port. Both can send and receive n wavelength signals corresponding to n ONU groups. The OLT shown in Figure 4(b) uses a set of wavelength division multiplexing optical transceiver modules (or optical transceiver arrays) to connect to the active port and the standby port through a 1×2 optical path switching module. The 1×2 optical path The switching module controls the optical signal of the optical transceiver module to input or output from the active and standby ports.

Figure 5 is a schematic structural diagram of the 1×2 optical path switching module in Figure 4(b), which includes a 1×2 optical coupler, a wavelength gating unit, and an optical switch unit with at least 2×2 ports ; The single-port end of the optical path switching module is connected to the single-port end of the optical coupler, one port of the dual-port end of the optical coupler is connected to the input port of the wavelength gating unit, and the other port of the dual-port end of the optical coupler is connected to the The output ports of the wavelength gating unit are respectively connected to two ports of the input end of the optical switch unit, and the two ports of the output end of the optical switch unit are respectively connected to two ports of the dual-port end of the optical path switching module. The wavelength gating unit can be composed of a wavelength tunable filter. At this time, the optical switch unit uses an optical switch with a port number of at least 2×2. The wavelength gating unit can also be composed of a 1×1 optical switch and a wavelength tunable The tunable filter is formed in series; the tunable filter can use a single-wavelength filter or a multi-wavelength filter, that is, an optical filter with multiple adjustable passbands, such as an acousto-optic tunable filter.

As shown in Figure 6, the recovery situation of the network provided by the present invention in the event of a failure, where an N×N AWG (a multi-port thin-film filter, etc. can also be used) is used as the wavelength division multiplexer in the RN, the intra-group Use a cross-bus structure connection (the recovery situation is similar when using a two-wire star structure), the two optical paths use the same wavelength, and Meizu has k ONUs as an example to illustrate. First, determine the type of network fault according to the following methods:

1) Detect the uplink signal of the main feeder fiber, and judge whether it can communicate normally by detecting the communication status of each optical network unit;

2) If all optical network units cannot communicate normally, it is judged that the main feeder fiber may be faulty;

3) If only some optical network units cannot communicate normally, it is judged that the main distribution fiber is faulty, and:

a) For a network that adopts a cross bus type connection within the optical network unit group, if the i-th optical network unit on the main optical fiber bus and all subsequent optical network units cannot communicate normally, it is judged that it is on the bus There may be a fiber failure between the i-th ONU and the i+1-th ONU; if the i-th ONU fails to communicate normally, but not all of its subsequent ONUs fail to communicate normally, it is judged that the i-th ONU The network unit may have failed;

b) For a network that adopts a dual-fiber star connection within the ONU group, if the i-th ONU cannot communicate normally, it is judged that the main distribution fiber connected to the i-th ONU may be faulty;

4) Detect the uplink signal of the spare feeder optical fiber, judge whether it can communicate normally by detecting the communication status of each optical network unit, and perform the same judgment as steps 2) and 3) on the detection results;

5) Comparing the judgment results of the main feeder fiber and the backup feeder fiber, if the two optical paths corresponding to the same optical network unit fail only on one hop optical path, it is judged as a single fiber failure; If there is a fault and the fault is located on the same link, it is judged as a dual-fiber fault on Lu'an Road; if a fault occurs on both optical paths but the fault is located on a different link, it is judged as a dual-optical path interleaved fault.

According to the above method, the judgment result is as follows: Figure 6 (a) single-fiber failure of the main optical fiber, after the repair of the main feeder fiber, it can be further judged that the main distribution fiber is located in the internal link RN and ONU ₁ of the first group of ONU _{, the optical fiber of k} has a single fiber failure, and at the same time, the optical fiber between ONU _{n, 2} and ONU _{n, 3} in the nth group of the main distribution fiber has a single fiber failure; Figure 6 (b) is the main distribution fiber The optical fiber connected to RN and ONU _{1, k} in the first group of ONU has a single-fiber fault. At the same time, the OLT can judge the occurrence of TRx1 through direct detection of the light source signal (such as the detection of the photodiode on the back of the laser). Fault. The solid line in the figure indicates the optical path for signal transmission, and the dotted line indicates the optical path for no signal transmission.

As shown in Figure 6(a), when the main feeder fiber connected to the main port of the OLT in use in the network fails, or the main feeder fiber and the distribution fiber connected to it fail at the same time (in the figure, the first and second n ONU groups corresponding to the distribution fiber failure, for example), and the backup feeder fiber connected to the OLT backup port and the distribution fiber connected to it are not faulty, by enabling the backup port of the OLT to replace the active port in use, it can be realized through the backup optical path Restoration of network communications.

Figure 6(b) shows the recovery situation when faults occur on both the distribution fibers of the active and standby optical paths. Take the distribution fiber failure of the 1st and nth ONU groups as an example, the main optical fiber in use between the RN of the local 1 ONU group fails, and the ONU _{n, 2} and ONU _n of the nth ONU group _{, when there is a double-fiber failure in the link between 3} , the failure of the first ONU group can be solved by starting the OLT standby port corresponding to a pair of working wavelengths λ ₁ , λ ₁ , λ _n , λ _n of the first ONU group, Work at the same time to achieve recovery. All other ONUs in the group except ONU _{n, 1} and ONU _{n, 2} still communicate with the active port of the OLT through the active optical path, while ONU _{n, 1} and ONU _{n, 2} communicate through the standby The communication between the optical path and the standby port of the OLT is restored. It can be seen from the recovery process that the fault-free ONU group, that is, the normal communication of the ONU group from the 2nd to n-1 will not be affected during the network recovery process. At the same time, the nth ONU group except ONU _{n, 1} and ONU _n, The communication of all other ONUs other than ₂ will not be affected during the network recovery process.

As shown in Figure 6(c), the OLT uses two sets of independent main and backup optical transceiver modules (the optical transceiver array is taken as an example in the figure) to send and receive signals respectively. When an optical transceiver in the main optical transceiver module and/or a distribution fiber connected to it fails, the corresponding transceiver in the backup optical transceiver module can be used to restore network communication. Likewise, this restoration process will not affect the communications of other ONU groups.

In order to further illustrate the survivability of the network of the present invention in the case of multiple faults, Fig. 7 shows the variation of the survival rate with the number of faulty optical fibers when the number of accessible nodes is 128 for various topology access networks with protection functions The calculation structure, where the survival rate refers to the expected value of the percentage of network accessible nodes (that is, ONU) in the case of a given number of faults in the network, is the percentage of the number of accessible nodes in the network to the total number of nodes in various Statistical mean in case of failure distribution. The double-feeder double-star structure and the double-feeder double-star-crossover bus structure (the crossover bus type structure is used inside the ONU group) are the topology structures provided by the present invention.

It can be seen from the figure that the topology adopted in the present invention has better fault tolerance, and can make the network survival rate decrease more slowly with the increase of the number of faults. Among them, the dual-feeder dual-star structure has the best survivability in most cases, not only can achieve complete recovery when a single fault occurs, but also its survival rate decreases linearly with the increase in the number of faults, and the rate of decline is the slowest ( The slope is the smallest); the double-feeder-double-star-cross bus structure network has higher survivability when the number of ONU groups is large (the survival rate when n=16 in the figure is better than the survival rate when n=8), but compared to double The feeder double-star structure has a slightly lower survival rate, but the amount of fiber required is significantly less than that of the double-feeder double-star structure. In comparison, although the existing ring structure network can achieve complete recovery when the number of failures is extremely small, for the bidirectional double-ring structure and the bidirectional single-ring structure, when the number of failures exceeds 3 and 1, respectively, the survival rate will decrease rapidly. It is lower than the survival rate of several other networks; and the survival rate of the existing double-feeder single-star structure network is also significantly lower than the double-feeder double-star-crossover bus structure with 16 ONU groups within the range of the calculated number of faults The network is close to the double-feeder double-star-cross bus structure network with grouping number of 8 only when the number of faults is greater than 12, and it is close to the double-feeder double-star-crossover bus-type network with grouping number of 8 only when the number of faults is greater than 12.

The above is only a preferred embodiment of the present invention, it should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, some improvements and modifications can also be made, and these improvements and modifications are also possible. It should be regarded as the protection scope of the present invention.

Claims (4)

1. A hybrid passive optical network, characterized in that: The network includes the following three nodes: 1) An optical line terminal, with two external ports that can back up each other and work at the same time, are called the main port and the backup port, and output the main and backup downlink signals respectively; 2) m optical network units are divided into n optical network unit groups, where 1≤n≤m; 3) A remote node with 2x2p ports, where p≥n; The main port of the optical line terminal is connected to the main input port of the remote node through the main feeder fiber, the backup port is connected to the backup input port of the remote node through the backup feeder fiber, and the remote node uses the main input port The signal is distributed to a group of output ports according to the wavelength. This group of output ports is called the main output port, and the signal of the backup input port is distributed to the backup output port; The remote node and each optical network unit group are connected in a dual-fiber star topology, and each optical network unit group uses a dual-fiber tree or cross bus structure to connect the optical network unit that needs to be protected in the group, and the remote node and optical network unit The part between the network units is called the distribution network, the fiber connecting the main output port of the remote node and the optical network unit is called the main distribution fiber, and the fiber connecting the backup output port of the remote node and the optical network unit is called the backup distribution fiber: network The upstream/downstream signals are transmitted in the same fiber bidirectional transmission mode, and the upstream signals of each optical network unit are sent to the optical path where the main and standby distribution optical fibers connected to it are simultaneously sent through the splitter; The optical line terminal includes a control module, a wavelength division multiplexing optical transceiver module, and may also include an optical path switching module. The division multiplexing optical transceiver module can send and receive the uplink and downlink wavelength signals of n channels corresponding to n optical network units, and the control module controls the output of the main and backup ports by controlling the transceiver of the wavelength division multiplexing optical transceiver module or the direction of the optical path of the optical path switching module The signal and the optical path of the uplink signal in the optical line terminal; The passive optical network structure adopts the following bandwidth utilization method: all optical network unit groups are accessed in a wavelength division multiplexing manner, and different optical network units in each optical network unit group adopt a time division multiplexing method or time division multiplexing and WDM hybrid access. the 2. The hybrid passive optical network according to claim 1, characterized in that: when each optical network unit group uses a cross bus structure to connect the optical network units in the group to be protected, each optical network unit group There are two buses corresponding to the same wavelength or a pair of wavelengths or a group of wavelengths extending from the output end of the remote node, that is, the main optical fiber bus and the backup optical fiber bus, and all optical network units in the group are in opposite directions. To be connected, each optical network unit is connected to the bus through an optical splitter, and the end of the bus is connected to an attenuator or directly connected to the last optical network unit without an optical splitter. the 3. a fault location method of hybrid passive optical network as claimed in claim 1, is characterized in that, this location method is as follows: 1) Detect the uplink signal of the main feeder optical fiber, and judge whether it can communicate normally by detecting the communication status of each optical network unit; 2) If all optical network units cannot communicate normally, it is judged that the main feeder fiber may be faulty; 3) If only some optical network units cannot communicate normally, it is judged that the main distribution optical fiber is faulty, and: a) For a network that adopts a crossover bus type connection within the optical network unit group, if the i-th optical network unit on the main optical fiber bus and all subsequent optical network units cannot communicate normally, it is judged that the i-th optical network unit on the bus There may be a fiber failure between the i-th ONU and the i+1th ONU; if the i-th ONU cannot communicate normally, but not all of its subsequent ONUs cannot communicate normally, then judge the i-th ONU a malfunction may have occurred; b) For a network using a dual-fiber star connection inside the ONU group, if the i-th ONU cannot communicate normally, it is judged that the main distribution fiber connected to the i-th ONU may be faulty: 4) Detect the uplink signal of the spare feeder optical fiber, judge whether it can communicate normally by detecting the communication status of each optical network unit, and perform the same judgment on the detection results as steps 2) and 3); 5) Comparing the judging results of the main feeder fiber and the backup feeder fiber, if one of the main optical paths and backup optical paths corresponding to the same optical network unit has a fault, it is judged as a single fiber failure: if two If a fault occurs on both optical paths and the fault is located on the same link, it is judged as a dual-fiber fault of the link; if a fault occurs on both the active optical path and the backup optical path but the fault is located on a different link, it is judged as a dual-optical path interleaved fault; here corresponds to The active optical path of an optical network unit refers to the optical channel that transmits the main uplink and downlink signals from the corresponding optical transceiver in the optical line terminal to the optical network unit, and the standby optical path refers to the optical channel from the corresponding optical transceiver in the optical line terminal to the optical network unit. The optical channel through which the standby uplink and downlink signals of the optical network unit pass. the 4. A method of failure recovery using the hybrid passive optical network as claimed in claim 1, characterized in that, the method is realized by the action of the optical line terminal, and its specific implementation method is as follows: When a single fiber failure occurs in the main distribution fiber, use the control module in the optical line terminal to make the downlink wavelength corresponding to the faulty group output from the backup port of the optical line terminal, and the downlink wavelength corresponding to the faulty group passes through the backup optical path where the grouping backup distribution fiber is located. Transmission to achieve failure recovery; When a single fiber failure occurs in the main feeder fiber being used, use the control module in the optical line terminal to make the backup port replace the main port to output all downlink signals, and the downlink signals of each optical network unit group are transmitted through their respective backup optical paths, Implement failure recovery; When a double optical path interleaving fault occurs in a group of optical network units in the distribution network, the control module in the optical line terminal is used to make the active port and the standby port output the downlink signals corresponding to different optical network units in the group in time-sharing, where the main The unaffected ONUs on the active optical path still use the main optical path to receive downlink signals, and the affected ONUs on the main optical path receive downlink signals from the standby port of the optical line terminal through the standby optical path, so as to realize the service of the affected ONUs. recovery; For a network connected by a cross-bus structure within the grouping of optical network units, when a dual-fiber fault occurs in the link in the distribution network, the fault is recovered by using the simultaneous work of the two ports of the optical line terminal corresponding to the wavelength, in which the main optical line The unaffected ONUs still use the main optical path to receive downlink signals, and the affected ONUs on the main optical path receive downlink signals from the backup port of the OLT through the backup optical path, so as to restore the services of the affected ONUs. the

Images