Method for handoff in wireless communication systems using pattern recognition

• US 6,240,291 B1
• Filed: 09/09/1999
• Issued: 05/29/2001
• Est. Priority Date: 09/09/1999
• Status: Expired due to Term

• Current Assignee: Board of Trustees of the Leland Stanford Junior University (Stanford University)
• Original Assignee: Board of Trustees of the Leland Stanford Junior University (Stanford University)
• Inventors: Narasimhan, Ravi, Cox, Donald C.
• Primary Examiner(s): Hunter, Daniel
• Assistant Examiner(s): LE, LANA N
• Application Number:
  US09/392,975
• Time in Patent Office: 628 Days
• Field of Search: 455/522, 455/422, 455/442, 455/432, 455/436, 455/438
• US Class Current: 455/436
• CPC Class Codes: H04W 36/008375 based on historical data
• CPC Class Codes: H04W 36/302 due to low signal strength

Abstract

A handoff technique for wireless communication systems uses pattern recognition of signal strength data to anticipate handoffs and reduce the total number of handoffs in the system. A criterion for system performance is used in determining the necessity for handoff. A window of signal samples from nearby base stations constitutes a pattern vector which is classified using a probabilistic neural network or other learning machine. The use of averaged signals and the sequencing of classes allow for a small number of training vectors for the pattern classifier. Substantially increased performance requires only one training vector per class. Simulation results indicate that, for a given probability of failure, the pattern recognition based handoff technique yields fewer handoffs than the conventional hysteresis rule.

Claims

1. A method for determining handoff in a wireless communication system comprising a plurality of base stations and a mobile device, the method comprising:

a) collecting at nearby base stations signal power level information representative of signal strength between the mobile device and the nearby base stations;

b) associating the power level information with predetermined power level patterns to produce an output representing measures of similarity between the power level information and the predetermined power level patterns;

c) selecting a sequence of matching patterns from the set of predetermined power level patterns, wherein each of the matching patterns has a maximal degree of similarity with a portion of the power level information; and

d) analyzing a sequence of sets of allowable base stations to determine an appropriate base station for handoff, wherein the sequence of sets of allowable base stations correspond to the sequence of matching patterns.

2. The method of claim 1 wherein the predetermined power level patterns are correlated with shadow fading in a propagation environment of the mobile device.

3. The method of claim 1 wherein the signal power level information comprises a sequence of average signal power level estimates corresponding to mobile device positions separated by less than a correlation length of a log-normal shadow fading in a propagation environment of the mobile device.

4. The method of claim 3 wherein each of the average signal power level estimates is an average of signal power level samples corresponding to mobile device positions separated by less than a wavelength of a carrier signal of the communication system.

5. The method of claim 4 wherein the signal power level samples are measured at times selected such that the samples correspond to mobile device positions separated by approximately equal spatial intervals.

6. The method of claim 1 wherein associating the power level information with the predetermined power level patterns comprises applying the power level information to a probabilistic neural network whose weights are equal to training vectors derived from training power level information.

7. The method of claim 1 wherein the predetermined power level patterns have a predetermined sequence corresponding to a sequence of patterns observed in a training session.

8. The method of claim 1 further comprising determining a degree to which adjacent patterns in the sequence of matching patterns meet an adjacent pattern matching tolerance value.

9. The method of claim 1 wherein analyzing the sequence of sets of allowable base stations to determine the appropriate base station for handoff comprises selecting a base station that is an allowable base station for a subsequent pattern in a predetermined sequence of power level patterns.