System and method for the remote monitoring of wireless packet data networks

• US 5,799,154 A
• Filed: 06/27/1996
• Issued: 08/25/1998
• Est. Priority Date: 06/27/1996
• Status: Expired due to Term

• Current Assignee: Verizon Patent and Licensing Incorporated (Verizon Communications Inc.)
• Original Assignee: MCI Communications Corporation (Verizon Communications Inc.)
• Inventors: Kuriyan, George W.
• Primary Examiner(s): Voeltz, Emanuel Todd
• Assistant Examiner(s): DAM, TUAN QUANG
• Application Number:
  US08/672,138
• Time in Patent Office: 789 Days
• Field of Search: 392/200.53, 392/200.54, 392/200.56, 392/180, 392/182.02, 392/183.13, 392/184.01, 392/601, 392/611, 392/613, 364/551.01, 364/242.94, 364/242.5, 364/264, 364/264.4, 364/266.5, 364/284.4, 364/919, 364/919.4, 370/319-322, 370/329, 370/331, 370/332, 370/333, 370/345, 370/346, 370/252, 370/431, 370/443, 370/913, 370/915, 370/902, 370/230, 370/232, 370/235, 379/59, 379/60, 379/58, 379/63, 379/49, 379/34, 379/111-115, 379/133-134, 379/350, 455/33.1-33.4, 455/54.1, 455/53.1, 455/67.1, 455/39, 455/49.1, 455/453, 455/447, 455/466, 455/446, 455/423
• US Class Current: 709/223
• CPC Class Codes: H04W 24/00 Supervisory, monitoring or testing arrangements

Abstract

A system and method monitors a wireless network such that degradation of performance, as measured by such performance parameters as data throughput and service availability is estimated accurately on a real time basis. The system and method scan each cell in a wireless telecommunications network for cell statistics data. Using the scanned cell statistics data, the system and method generate a model of the performance of the telecommunications network by modeling each cell in the network. The model is based on a control system relationship. Using the control system relationship, the system and method forecast the performance of the cells between actual data collection times. Forecasting the state of the cells reduces the amount of data required to monitor the telecommunications network.

Claims

1. A method for remotely monitoring the performance of a telecommunications network having at least two cells, comprising the steps:

collecting performance data for each cell wherein said data is functionally related to performance parameters in each cell;

creating a unique cell data structure for each cell, wherein each cell data structure includes a unique scan interval;

forecasting, independently, a state of each cell, for future points in time during each cell'"'"'s said unique scan interval, based on data collected prior to said unique scan interval and on prior forecasts, without collecting additional data during said unique scan interval;

comparing, independently, said collected data from each cell, to predetermined thresholds; and

modifying, independently, each unique scan interval in response to said comparing step, whereby the modification of one cell'"'"'s unique scan interval does not affect the modification of another cell'"'"'s unique scan interval.

2. The method recited in claim 1, wherein said modifying step includes the steps of:

shortening said unique scan interval when said predetermined thresholds are exceeded; and

increasing said unique scan interval when said predetermined thresholds are not exceeded.

3. The method of claim 1, wherein each unique cell data structure includes:

a vector of said collected cell performance data;

a vector of predetermined thresholds;

a vector of forecasts of a state of the network at a particular time, wherein said vector of forecasts includes a history of forecasts; and

a transformation matrix that uses cell data stored in said vector of collected cell performance data to forecast the state of the network according to a control relationship.

4. The method of claim 3, wherein each unique cell data structure further includes:

a scaling matrix that scales said history of forecasts and adds them to a forecast resulting from application of said vector of said collected cell performance data to said transformation matrix.

5. A method for remotely monitoring the performance of a telecommunications network, comprising the steps of:

collecting, prior to and after a scan interval, performance data functionally related to performance parameters in the network;

generating a network control model of the performance of the network using said data collected prior to said scan interval; and

forecasting the performance of the network at a future point in time during said scan interval, using said network control model, whereby a monitored performance of the network during said scan interval is an estimate based on data collected prior to said scan interval in said collecting step and prior forecasts determined in said forecasting step.

6. The method recited in claim 5, further comprising the steps of:

comparing said collected data to predetermined thresholds; and

adjusting a duration of said interval in response to said comparing step.

7. The method recited in claim 5, further comprising the steps of:

generating said network control model in accordance with a control relationship; and

forecasting the performance of said network by forecasting a new performance as the result of application of said control relationship to said data.

8. The method as recited in claim 5, wherein the telecommunications network includes one or more cells and said collecting step comprises the steps of:

polling the one or more cells for said data;

creating a data structure for each cell;

transmitting said data to a central location in response to said polling; and

for each data structure, storing said data polled from each cell.

9. The method as recited in claim 8, further comprising the step of:

spawning a thread for each of the one or more cells, said thread to perform said polling and collecting steps.

10. The method as recited in claim 9, further comprising the steps of:

(a) generating an A-matrix;

(b) generating a B-matrix;

(c) setting a temporary vector equal to AX+BY+C, where X is a vector corresponding to said data, Y is a vector corresponding to previous estimates of the performance of the network, and C is a vector of constants operator actions.

(d) updating a forecast of the performance of the network by adding said temporary vector to a previous estimate of the performance of the network; and

(e) repeating steps (a)-(d) one or more times during said scan interval.

11. A system for remotely monitoring the performance of a telecommunications network, comprising:

a regional scanning module for collecting, prior to and after a scan interval, performance data functionally related to performance parameters in the network; and

a performance modeling and forecasting module for generating a network control model of the performance of the network using said data collected prior to said scan interval and for forecasting the performance of the network at a future point in time, using said network control model, whereby a monitored performance of the network during said scan internal is an estimate based on data collected prior to said scan interval in said collecting step and prior forecasts determined in said forecasting step.

12. The system as recited in claim 11, wherein the telecommunications network includes one or more cells for handling calls, the system further comprising a database system for storing said performance data collected from each cell in a data structure created for each cell.

13. The system as recited in claim 12, further comprising:

a workstation;

a database system for storing said collected performance data; and

means for allowing a network administrator access to said database system by using said workstation.

14. The system as recited in claim 11, wherein said performance modeling and forecasting module further comprises:

comparing means for comparing said collected data to predetermined thresholds; and

modifying means for adjusting a duration of said scan interval in response to said comparing means.

15. The system as recited in claim 11, wherein said performance modeling and forecasting module further comprises:

generating means for generating said network control model in accordance with a control relationship; and

forecasting means for forecasting the performance of said network by forecasting a new performance as the result of application of said control relationship to said data.

16. The system as recited in claim 12, wherein said regional network scanning module comprises:

polling means for polling the one or more cells for said performance data;

transmission means for transmitting said data to a central location in response to said polling means.

17. The system as recited in claim 16, wherein said regional network and scanning module comprises:

spawning means for spawning a thread for each of the one or more cells, said thread to perform said polling means and collecting means.

18. The system as recited in claim 17, wherein said spawning means comprises:

first generating means for generating an A-matrix;

second generating means for generating a B-matrix;

setting means for setting a temporary vector equal to AX+BY+C, where X is a vector corresponding to said data, Y is a vector corresponding to previous estimates of the performance of the network, and C is a vector of constants operator actions; and

updating means for updating a forecast of the performance of the network by adding said temporary vector to a previous estimate of the performance of the network.

19. The system recited in claim 18, wherein said spawning means further comprises:

repeating means for repeatedly applying said first and second generating means, said setting means, and said updating means over a history of forecasts; and

storing means for appending a new forecast, produced by said repeating means, to said history.