Classification of wireless LAN signals
AssigneeNational Instruments Corporation
InventorsKunal Agarwal, Baijayanta Ray, Abhay Samant
Life of this patent
Estimating the term…
Abstract
A system and method for classifying an input signal suspected of being a wireless LAN signal. Bursts are isolated based on an analysis of an average power signal that is derived from the input signal. A correlation-based test is performed to determine if the input signal contains a Barker code typical of DSSS transmissions. Another correlation-based test is performed to determine of the input signal contains a long training sequence typical of OFDM transmissions. The results of the two tests are used to classify the input signal as being DSSS, OFDM or neither.
Claims24
A method comprising:
- a computer system receiving an input signal;
- the computer system detecting a start location and stop location of a burst within the input signal based on an average power signal derived from the input signal;
- the computer system computing a correlation function based on (a) samples of the input signal corresponding to at least a portion of the burst and (b) a known Barker code of a direct sequence spread spectrum (DSSS) standard, wherein the correlation function is a function of shift index, wherein a value of the correlation function at a given value of the shift index is determined by computing inner products between respective windows of the samples and the Barker code and by computing a sum of magnitudes of the inner products, wherein the respective windows are shifted relative to an initial series of windows by an amount equal to the shift index value;
- the computer system determining whether or not the correlation function includes two peaks with amplitude greater than a first threshold and separated by a sample distance equal to a known sample distance between successive DSSS symbols;
- the computer system outputting a decision indicating whether or not the burst corresponds to a DSSS signal based on a result of said determining.
A method comprising:
- a computer system receiving an input signal;
- the computer system detecting a start location and stop location of a burst within the input signal based on an average power signal derived from the input signal;
- the computer system computing a correlation function between a template and samples of the input signal corresponding to the burst, wherein the template represents a training sequence of a preamble conforming to an OFDM standard, wherein the training sequence includes two repetitions of a base pattern;
- the computer system determining whether or not the correlation function includes three peaks with:
- (a) amplitude greater than a threshold and (b) interpeak spacing corresponding to a known symbol distance between the two repetitions of the base pattern;
- in response to determining that the correlation function includes said three peaks, the computer system performing an autocorrelation analysis on a selected portion of the burst to determine whether or not the selected portion corresponds to an instance of the training sequence;
- the computer system outputting a decision indicating whether or not the burst corresponds to an OFDM signal based on a result of said determining whether or not the selected portion corresponds to an instance of the training sequence.
A method comprising:
- a computer system receiving an input signal;
- the computer system detecting a start location and stop location of a burst within the input signal based on an average power signal derived from the input signal;
- the computer system computing a first correlation function based on (a) samples of the input signal corresponding to at least a portion of the burst and (b) a known Barker code of a direct sequence spread spectrum (DSSS) standard;
- the computer system determining whether or not the first correlation function includes two peaks with amplitude greater than a first threshold and separated by a sample distance equal to a known sample distance between successive DSSS symbols;
- the computer system determining whether or not the burst contains a known training sequence conforming to an OFDM standard, wherein the known training sequence includes two repetitions of a base pattern;
- the computer system classifying the burst as corresponding to a DSSS signal, an OFDM signal or neither based on whether or not the first correlation function includes the two peaks and based on whether or not the burst contains the known training sequence.
A non-transitory computer-readable memory medium storing program instructions, wherein the program instructions, when executed by a computer system, cause the computer system to execute a method comprising:
- receiving an input signal;
- detecting a start location and stop location of a burst within the input signal based on an average power signal derived from the input signal;
- computing a correlation function based on (a) samples of the input signal corresponding to at least a portion of the burst and (b) a known Barker code of a direct sequence spread spectrum (DSSS) standard, wherein the correlation function is a function of shift index, wherein a value of the correlation function at a given value of the shift index is determined by computing inner products between respective windows of the samples and the Barker code and by computing a sum of magnitudes of the inner products, wherein the respective windows are shifted relative to an initial series of windows by an amount equal to the shift index value;
- determining whether or not the correlation function includes two peaks with amplitude greater than a first threshold and separated by a sample distance equal to a known sample distance between successive DSSS symbols;
- outputting a decision indicating whether or not the burst corresponds to a DSSS signal based on a result of said determining.
A non-transitory computer-readable memory medium storing program instructions, wherein the program instructions, when executed by a computer system, cause the computer system to execute a method comprising:
- receiving an input signal;
- detecting a start location and stop location of a burst within the input signal based on an average power signal derived from the input signal;
- computing a correlation function between a template and samples of the input signal corresponding to the burst, wherein the template represents a training sequence of a preamble conforming to an OFDM standard, wherein the training sequence includes two repetitions of a base pattern;
- determining whether or not the correlation function includes three peaks with:
- (a) amplitude greater than a threshold and (b) interpeak spacing corresponding to a known symbol distance between the two repetitions of the base pattern;
- in response to determining that the correlation function includes said three peaks, performing an autocorrelation analysis on a selected portion of the burst to determine whether or not the selected portion corresponds to an instance of the training sequence;
- outputting a decision indicating whether or not the burst corresponds to an OFDM signal based on a result of said determining whether or not the selected portion corresponds to an instance of the training sequence.
Dependencies are inferred from the claim text.