Device and method for testing of multi-branch optical network

• US 6,512,610 B1
• Filed: 03/03/1999
• Issued: 01/28/2003
• Est. Priority Date: 03/11/1998
• Status: Expired due to Fees

• Current Assignee: Kansai Electric Power Company Incorporated, Nobuaki Takeuci, Takao Minami, Yokogawa Electric Corporation
• Original Assignee: Kansai Electric Power Company Incorporated, Ando Electric Co., Ltd. (Yokogawa Electric Corporation)
• Inventors: Shimizu, Keiichi, Minami, Takao, Shinozaki, Koichi, Genji, Takamu, Takeuchi, Nobuaki
• Primary Examiner(s): Chan, Jason
• Assistant Examiner(s): HUYNH, TUANSON H
• Application Number:
  US09/261,835
• Time in Patent Office: 1,427 Days
• Field of Search: 359/110
• US Class Current: 398/21
• CPC Class Codes: G01M 11/3136 for testing of multiple fibers

Abstract

A multi-branch optical network testing method (or device) is provided to perform a fault isolation test on an optical network that branches off at a branch point by a number of optical lines having terminal ends respectively. Herein, optical pulses are input to the optical network, from which they are returned as reflection beams. Then, response beams corresponding to mixture of the reflection beams are converted to OTDR waveform data representing a waveform whose optical power gradually decreases in accordance with a distance from an OTDR measurement device and which has a number of reflection peaks. The OTDR waveform data are subjected to logarithmic conversion to produce logarithmic waveform data representing a logarithmic waveform. An approximation method of least squares is effected on the logarithmic waveform data to produce an approximation line, which crosses the logarithmic waveform at points of intersection corresponding to Fresnel reflection points. Using the Fresnel reflection points as split points to split the OTDR waveform data into a number of ranges. Attenuation constants are repeatedly calculated with respect to each of the ranges every measurement time and are stored in a storage device. Thereafter, fault determination is automatically performed based on the attenuation constants stored in the storage device with respect to the fault occurrence time, fault occurrence line and fault occurrence distance.

Claims

1. A multi-branch optical network testing method comprising the steps of:

inputting optical pulses to a branch point at which a multi-branch optical network branches off by a plurality of optical lines;

receiving response beams corresponding to mixture of reflection beams which are produced by reflecting the optical beams at selected portions of the optical lines respectively;

performing logarithmic conversion on waveform data corresponding to electric signals, which are produced by converting the response beams;

performing calculations on the logarithmically-converted waveform data to produce an approximation line;

comparing the logarithmically-converted waveform data with the approximation line to detect Fresnel reflection points on the waveform data;

using the Fresnel reflection points as split points to split the waveform data to a number of ranges;

performing separative analysis on each of the ranges to calculate attenuation constants with respect to each of the optical lines;

storing the attenuation constants in a storage device in connection with each of measurement times; and

determining a fault occurrence time, a fault occurrence line and a fault occurrence distance with respect to a fault that occurs in the multi-branch optical network having the optical lines on the basis of the attenuation constants which are repeatedly calculated every prescribed time and stored in the storage device.

2. A multi-branch optical network testing method according to claim 1, wherein the logarithmic conversion is performed in accordance with an equation of

3. A multi-branch optical network testing method according to claim 1 or 2, wherein the Fresnel reflection points are detected based on points of intersection formed between a waveform corresponding to the logarithmically-converted waveform data and the approximation line.

4. A multi-branch optical network testing method according to claims 1 or 2, wherein comparison is performed on the attenuation constants sequentially calculated at consecutive measurement times so that fault occurrence is determined in response to a change that is detected between the attenuation constants, and

wherein the measurement time corresponding to the change of the attenuation constants is determined as the fault occurrence time, distance of a reflection peak is detected from the attenuation constants so that an optical line corresponding a change that occurs on the distance of the reflection peak is determined as the fault occurrence line, and the distance of the reflection peak after the change is determined as the fault occurrence distance.

5. A multi-branch optical network testing device comprising:

light emitting means for emitting optical pulses input to a branch point at which a multi-branch optical network branches off by a plurality of optical lines;

light receiving means for receiving response beams corresponding to mixture of reflection beams which are produced by reflecting the optical beams at selected portions of the optical lines respectively;

conversion means for performing logarithmic conversion on waveform data corresponding to electric signals, which are produced by converting the response beams;

approximation means for performing calculations on the logarithmically-converted waveform data to produce an approximation line;

comparison means for comparing the logarithmically-converted waveform data with the approximation line to detect Fresnel reflection points on the waveform data;

split means for using the Fresnel reflection points as split points to split the waveform data to a number of ranges;

analysis means for performing separative analysis on each of the ranges to calculate attenuation constants with respect to each of the optical lines;

write means for storing the attenuation constants in a storage device in connection with each of measurement times; and

determination means for determining a fault occurrence time, a fault occurrence line and a fault occurrence distance with respect to a fault that occurs in the multi-branch optical network having the optical lines on the basis of the attenuation constants which are repeatedly calculated every prescribed time and stored in the storage device.

6. A multi-branch optical network testing device according to claim 5, wherein the conversion means performs the logarithmic conversion in accordance with an equation of

7. A multi-branch optical network testing device according to claim 5 or 6, wherein the Fresnel reflection points are detected based on points of intersection formed between a waveform corresponding to the logarithmically-converted waveform data and the approximation line.

8. A multi-branch optical network testing device according to claims 5 or 6, wherein the determination means performs comparison on the attenuation constants sequentially calculated at consecutive measurement times so that fault occurrence is determined in response to a change that is detected between the attenuation constants, and

wherein the measurement time corresponding to the change of the attenuation constants is determined as the fault occurrence time, distance of a reflection peak is detected from the attenuation constants so that an optical line corresponding a change that occurs on the distance of the reflection peak is determined as the fault occurrence line, and the distance of the reflection peak after the change is determined as the fault occurrence distance.

9. A multi-branch optical network testing method comprising the steps of:

inputting optical pulses to an optical network that branches off using an optical coupler by a plurality of optical lines having terminal ends respectively, wherein the optical pulses are reflected by the optical coupler as well as in the plurality of optical lines respectively so that reflection beams are produced;

receiving response beams corresponding to mixture of the reflection beams from the optical coupler;

converting the response beams to electric signals to produce OTDR waveform data representing a waveform whose optical power gradually decreases in accordance with a distance and which has reflection peaks in response to the optical coupler and the terminal ends of the plurality of optical lines respectively;

performing logarithmic conversion on the OTDR waveform data to produce logarithmic waveform data representing a logarithmic waveform having peaks corresponding to the reflection peaks of the waveform of the OTDR waveform data;

effecting an approximation method of least squares on the logarithmic waveform data to produce an approximation line which crosses the logarithmic waveform at points of intersection that correspond to Fresnel reflection points;

using the Fresnel reflection points as split points to split the logarithmic waveform data into a plurality of ranges, a number of which is determined in connection with the plurality of optical lines;

calculating at least one attenuation constant with respect to each of the plurality of ranges, wherein calculation of the attenuation constant is repeated every measurement time so that attenuation constants are sequentially calculated with respect to each of the plurality of ranges;

storing the attenuation constants in a storage device; and

performing fault determination with regard to the optical network on the basis of the attenuation constants stored in the storage device.

10. A multi-branch optical network testing method according to claim 9, wherein the logarithmic conversion is performed in accordance with an equation of

11. A multi-branch optical network testing method according to claim 9 or 10, wherein occurrence of a fault is determined in response to a change that occurs between the attenuation constants sequentially calculated at consecutive measurement times with respect to each of the plurality of ranges.

12. A multi-branch optical network testing method according to claims 9 or 10, wherein the fault determination is performed with regard to a fault occurrence time, a fault occurrence line and a fault occurrence distance.