Analysis on improving the network connection rate of switching devices

According to statistics, at present, China's telecom network service access rate is only about 45%. If the access rate increases by one percentage point, the income can reach 1 billion yuan, and the current switch overload in China can be significantly improved, therefore, it is very important to improve the service access rate of the telecom network.

With the development of China's telecom technology, the focus of China's telecom operation and maintenance work has shifted from one-sided pursuit of the incoming call rate to how to improve the network access rate. When the No. 7 signaling relay is used, the network connection rate can be expressed by the following formula: 100% × (ACM + SLB + STB)/(IAM + IAI). We can see from the formula that, the network connection rate is mainly used to analyze the network-side causes of call failure, that is, the device causes. Call loss is no longer caused by the busy call and long call, make the assessment results more reasonable, so as to accurately reflect the maintenance level of the switching equipment.

Reasonably configure and maintain vswitch database data

The C & C08 vswitch database has three main Relational Tables: User index table, user data table, and device table. Data parameter consistency and rationality should be maintained during data configuration, in this way, the call loss, congestion signal generation, and signal combination errors are avoided.

Impact of Table missing data records on incoming relay calls

For versions 02 or 03B, the correct relationship between data in each table is usually guaranteed by the er. Therefore, there is often a shortage of data or inconsistency between tables, these results in Table retrieval failure and call failure. Because of the storage process encapsulation, the consistency between tables is ensured by the program, avoiding data inconsistency between tables and reducing the occurrence of the above phenomenon.

Settings of relay-related data, such as crown

If this department is an end Bureau, it is necessary to ask the higher-level Bureau to make a detailed title of the Bureau. Otherwise, the wrong title will lead to blank and wrong call loss for the number entered for relay. The relay group call sources sent by the higher authorities should be separated from the call sources sent by the higher authorities as much as possible, and the calls sent by the higher authorities should be implemented to avoid the endless circulation of words. In the CEN, especially for C3.

For the setting of the Crown data, pay attention to the following principles: If this Bureau is a local bureau, the pre-receiving number length can be set to equal to the local bureau number length for incoming relay calls, reduces the number of internal message packets and number analysis times. For outgoing calls, the pre-receive number length and minimum number length should also be set as large as possible, and the circuit occupation point of the relay should also be set as large as possible. It is best to send codes in groups, the benefits of setting big data not only improve the processing efficiency of the switch, improve the circuit utilization, but also minimize the error of signal cooperation.

For outgoing TUP calls, it is best to send the caller number to reduce the number of inter-office message packets. The new national standard has made this provision. The caller display function is enabled throughout the network. The initiator must send an IAI forward message with the Caller information. In version 03I, when the software parameter "TUP operation retention parameter 4" BIT1 = 0, TUP will take the initiative to send the caller number, without the need for a supplemental signaling table, which is very convenient; For ISUP, when the software parameter "ISUP operation retention parameter 10" BIT10 = 1, it also has the same effect.

The sending method from the Board to the local board should be "send code in groups", but a few of the calls from the Board's MS3 are now "send code overlapping ", when the number is not full-time, it is easy to cause a "give up in dialing" call loss. This problem should be solved by the Bureau.

Reasonably configure network parameters for Optical Paths, relay, and networking

Network parameter configuration principles

Relay module configuration principles. In the case of multiple relay modules, multiple TSM modules are mounted under one AM32 or AM128. To ensure that a single module failure does not affect the global connection with other bureaus, the circuit allocated to the same board must be allocated to different relay modules. Similarly, the routes to the same bureau must also be distributed on the No. 7 signaling board of different relay modules, even if there is only one relay module, the links going to the same bureau should also be distributed on different 7th signaling boards to avoid a full blocking fault in a certain board due to damage to a 7th signaling board. In addition, if Link 7 is configured, the time slot should be allocated to different DT boards to avoid interruptions of multiple links due to damage to one DT board.

Link selection principles. Try to keep the route in the same module as its preferred link, so as to avoid the problem of TSM1 and TSM2 and TSM3, and reduce the message packets between modules.

For example, a management module of a certain Bureau has four relay modules, and four relay modules have 24 RSM Ⅱ, with only one S1240 egress board, the circuit 7 is deployed on four relay modules. Through the data modification process shown in tables 1, 2, and 3, users with different call source codes can occupy a certain relay module when making outgoing calls, thus reducing the number of message packets between modules, reduces the congestion rate of A vswitch. In the ST User table, the source code of the RSM Ⅱ User call carried by the four relay modules is defined as 1, 2, 3, and 4, respectively.

How to reasonably configure network parameters

Reasonably configure the Crown data. If this Bureau is an end server, for incoming relay calls (set the call source code 5 for the incoming relay group separately), in the call source table, the reservation number length corresponding to source code 5 is set to equal to the current call number length, reducing the number of internal message packets and number analysis times. For outgoing calls, especially long-distance calls, automatic paging, and mobile phones that require the title number, in the called analysis table, the "Supplemental signaling sign" corresponding to the Crown is set to "yes ". In the supplemental signaling table, set "whether to send the caller number" to "yes" (sending IAI) corresponding to each Crown to reduce the number of inter-Bureau message packets. In addition, the pre-receipt number length (call source table) and minimum number length (called analysis table) should be configured as much as possible, and the number of circuit occupation points carried by the relay should also be configured larger, it is best to configure it as a group sending code to improve the processing efficiency and circuit utilization of the switch, and minimize the matching error of the signal.

Set the relay circuit line selection method. For the No. 7 signaling circuit, the National Standard specifies two line selection methods: Size/small size, Master/non-master (LIFO/FIFO), and the master/non-master mode is preferred. Some switches, such as 5ESS, may adopt the size/small-size mode. Note that one side of the two ports is selected from the largest Circuit of CIC, the other party selects a small circuit from the CIC, but the Circuit Selection of the 08 machine is based on the relay internal code. Pay attention to the CIC ing relationship between the CIC and the relay internal code and make the correct data. Because circuit 7 is a two-way circuit, the circuit may be in the same way. Therefore, the two switches connected to each other are required to control half of the circuit, the non-primary control releases the occupation of the circuit and initiates a call on another idle circuit. Between the two switches, one party with the largest signaling point code controls the circuit where the CIC is an even number, and the other party with the smallest signaling point code controls the circuit where the CIC is an odd number, if the two sides control the circuit in the same direction, the impact will be greater, and the circuit may be suspended. Please note that. When the traffic volume is large, the same-rush has a certain impact on the access rate, so avoid it.

For inter-module relay line selection, we recommend that you configure SM and RSMII (or SMII) it is best to configure the "minimum", "maximum", or "Main Control/non-main control" for the relay line selection mode ".

Allocate traffic reasonably. When configuring data, it is sometimes difficult to predict which users belong to high-traffic users. When a user box (or RSA box) when too many "webworms" or hotline users (such as paging stations, hotline services, and stock reviews) are concentrated, the system's traffic volume will be too concentrated, resulting in call loss. Through the traffic statistics function of the C & C08 switch, you can promptly discover unreasonable components in the Data configuration, and then optimize the Data configuration. The main statistical methods include:

· User line group statistics. By analyzing the distribution of user calls in the module, User box, and RSA box, find the User box and module with the maximum or minimum traffic, and take corresponding improvement measures.

· Bureau-to-destination traffic distribution statistics. Analyze the outbound traffic distribution from a certain Bureau to each destination, find out the traffic that is large, and adjust the routing selection code of the corresponding crown in the called analysis table to balance the traffic.

· Module communication status and traffic statistics. Count the traffic distribution between modules. If traffic on some links is found to be extremely large and traffic on some links is extremely small, you must reconfigure the data to reduce inter-module calls.

· Module internal circuit group statistics. By collecting statistics on OPT, IDT, E16, and other traffic, we can determine whether there are traffic overload and long congestion periods, and take further improvement measures.

Reasonably allocate message traffic between modules. It is usually caused by unreasonable arrangement of the relay circuit and the signaling link. For example, if a module only has traffic and no link, each outgoing call must go through the optical path between modules, the message traffic between modules is too large, resulting in insufficient OPT network resources.

Reduces the occurrence of CGC signals and rationally configures relay data. Generally, when the data configuration of the route table, route analysis table, relay group table, and relay group/sub-route table is incorrect, the CGC signal may occur.

Reasons for users' hardware and status

The most common is external line faults. The user's external line faults are mainly caused by broken lines, mixed lines, single line Ground Gas, and user locks. When an external line is disconnected, the user can be called each time to hear the Back-to-bell sound, but there is always no answer, which is similar to a user who is not called for a long time. When there is a single line of ground, this will lead to false disconnection reporting, out-of-order dialing, and other phenomena, which will affect the connection rate. When the user's port is locked, it will cause a busy call, which is similar to the user's busy schedule. Because disconnection and locking are similar to "Long call should not" and "called Busy", You need to note that they are separated from "really long call should not" and "really busy" users for different processing. The above situations are often the most direct and fundamental reasons for the low access rate of C4 and C5 bureaus.

A few of them are ASL and DRV board faults. The failure cause of traffic statistics is "terminal fault ". Call this user, mostly for "beep, beep" fault sound; tracking, mostly for locking, automatic dial-up, repeated disconnects and disconnects, the two sides cannot hear the sound after the disconnects (single line); test, mostly inter-Line Insulation difference, touch other user line, single line air and so on.

In addition, there are also ONU User power outages, no inter-module time slots (CTN Control Memory suspended), user long call should not wait. When an ONU user loses power, the traffic statistics fail because of a "temporary fault ". All observed users are V5ST users, and the event alerts include V5ST user port blocking alerts, which are found to be users with ONU power outages. Solution: handle the first line of the ONU header for power outages and send congested audio; for time slots without inter-module traffic (CTN control ), for example, if the 62 time slots in the User box (56 at the far end of RSA) are used up, the reason for the failure is "no available resources ". For the user for a long time should not, use the following method to solve: to add a user "no response to voice mail" New Business: mod st: DN = PQABCB, NS = CFNRAVM-1; modify this information table, for example, if the direct index number is set to 0, the timer duration of the remote call should not be changed to 1423 in the voice mailbox; the timer duration of the remote call should not be changed to mod realtimer: PID = 13, and the timer group number = 0, group Index = 25, NVL = 30.

It is worth noting that, for a long time, the default value of the timer duration should not be 20 s, that is, it starts to be transferred to the voice mailbox about 5 times after the zhenling. To prevent calls that have been transferred when the call is received five times after the call, the timer duration for a long call should not be converted to the voice mailbox is usually set to 30 s. CFL should not be sent for a long time. It can also be improved through the call-forward function. However, if the call-forward function is still unsuccessful, the CFL should still be sent back. In addition, for ISDN users, the device is not activated due to external power outages and other reasons, and CFL is also sent back when the device is called, which can be solved according to the specific situation.

Summary

In short, there are a variety of factors that affect the switch connection rate. during routine maintenance, we have adopted corresponding solutions by analyzing data consistency, reasonably configuring network parameters, and analyzing user faults, this significantly improves the network connection rate and the quality of system operation and maintenance.

Blog: http://blog.163.com/xutong_hust/blog/static/15818372008513112757702/