Packet network monitoring device

• US 7,362,714 B2
• Filed: 03/11/2004
• Issued: 04/22/2008
• Est. Priority Date: 01/29/1996
• Status: Expired due to Fees

• Current Assignee: CommScope Technologies, LLC (CommScope Holding Company, Inc.)
• Original Assignee: ADC Telecommunications Incorporated (TE Connectivity Limited)
• Inventors: Swanson, Leslie H., Kovummal, Ashwin, Richardson, William M., Mukherjee, Sandeepan
• Primary Examiner(s): NGUYEN, STEVEN H D
• Application Number:
  US10/797,984
• Publication Number:
  US 20050058076A1
• Time in Patent Office: 1,503 Days
• Field of Search: 370/241, 370/245, 370/247, 370/248, 370252-253, 324/522, 324/533, 324/534, 324/628
• US Class Current: 370/242
• CPC Class Codes: H04L 43/022 by sampling
• CPC Class Codes: H04L 43/0847 Transmission error
• CPC Class Codes: H04L 43/12 Network monitoring probes
• CPC Class Codes: H04L 43/16 Threshold monitoring
• CPC Class Codes: H04L 43/18 Protocol analysers
• CPC Class Codes: H04L 43/50 Testing arrangements

Abstract

A network diagnostic device is disclosed that digitally samples the voltages on the cabling of the network, but does so at a much higher rate and with greater resolution then is required to minimally detect digital transmissions on the cabling. This sampling provides information on the analog characteristics of digital, noise, and interference signals on the network. Thus, network problems can be precisely diagnosed. The device includes a fast digitizer with a long memory and a system processor that statistically analyzes the signal events captured by the digitizer. The invention is also capable of performing time domain reflectometry (TDR) analysis of a functioning network. This is accomplished by placing a TDR signal on the network surrounded by a transmission that the network devices will interpret as a broadcast diagnostic packet. This will cause the network nodes to ignore the transmission. The digitizer, however, is able to detect the networks response to the TDR signal. Methods for identifying unknown network sources and Manchester decoding are also disclosed.

Claims

1. A network analysis device for a digital data computer network, comprising:

a digitizer which digitally samples analog characteristics of digital communication events between network devices connected to the network independently of packet contents associated with any of the digital communication events, the analog characteristics including a component of at least one of voltage and frequency, the samples taken at least at one of a rate and a resolution greater than is required to minimally detect digital transitions, wherein the samples are taken at a sampling frequency of at least approximately 500 megahertz;

a system processor which downloads data of the sampled signal events from the digitizer, which analyzes the analog characteristics, and which decodes the signal events, which are digital communications between the devices, based on the data, and wherein the system processor classifies the signal events as digital communications, noise, interference and/or crosstalk, wherein the system processor classifies the signal events as digital communications, noise, interference and/or crosstalk to selectively predict a failure mode.

7. A method as described in claim 1, further comprising classifying the signal events as digital communications, noise, interference and/or crosstalk based using parametric analysis of each event.

8. A method as described in claim 1, further comprising simultaneously connecting to multiple links of the network.

9. A method as described in claim 1, further comprising simultaneously connecting to multiple links of a star topology network.

10. A method as described in claim 9, further comprising tagging sampled signal events to identify the link from which the event originated.

11. A method as described in claim 9, further comprising determining whether the network communications are within frequency and voltage specifications for the network.

12. A method as described in claim 9, further comprising analyzing transmission characteristics of the network analysis by driving a predetermined signal out onto the network and detecting the response of the network.

2. A method for monitoring the operation of a computer network, comprising:

digitally sampling analog characteristics of signal events on the network with a digitizer independently of packet contents associated with any of the signal events, the analog characteristics including a component of at least one of voltage and frequency, the sampling taking place at least at one of a rate and a resolution greater than is required to minimally detect digital transitions, wherein the samples are taken at a sampling frequency of at least approximately 500 megahertz;

downloading data arrays of the signal events to a system processor;

analyzing the data arrays in the system processor to identify the signal events;

determining analog characteristics of the signal events; and

decoding the signal events, which are digital communications between network devices, based on the data, to selectively predict a failure mode.

3. A method as described in claim 2, further comprising

classifying the events as collisions between network devices;

determining transmission start and stop times for colliders in collision signal events.

4. A method as described in claim 2, further comprising locating network devices that improperly react to collisions with other network devices by reference to the start and stop times.

5. A method as described in claim 2, further comprising identifying sources of transmissions on a network by calculating parameters for transmissions from known sources, calculating the parameters for a transmission from an unknown source, and identifying the unknown source based upon the degree to which the parameters match parameters from the known sources.

6. A method as described in claim 2, further comprising classifying the signal events as digital communications, noise, interference and/or crosstalk based on the analog characteristics.