KR20110115519A - Method and apparatus for user state recognition in a wireless communication system - Google Patents

Abstract

The present invention provides a method for recognizing a user state of a wireless communication system including a mobile terminal having a microphone, the method comprising: receiving audio signals using the microphone and dividing the received audio signals into a plurality of segments Calculating an audio signal level indicator for each of the divided segments; and setting a user state to a first state if at least one of the audio signal level indicators is less than a predetermined threshold value. Include.

Description

Method and apparatus for recognizing user state in wireless communication system {METHOD AND APPARATUS FOR USER STATE RECOGNITION IN A WIRELESS COMMUNICATION SYSTEM}

The present invention relates generally to a wireless communication system, but is not limited thereto, and more particularly, to a method and an apparatus for recognizing a user state by analyzing audio signals received by a microphone of a mobile terminal in a wireless communication system. This is applicable to GSM, CDMA, IEEE 802.16, IEEE 802.11n, 3GPP LTE, and other wireless communication systems.

In general, wireless communication systems include mobile devices, which provide a communication function with base stations and provide various functions. Other functions provided by the mobile terminal include other devices. functions include communications with devices and equipment, communications with a Global Positioning Service (GPS), entertainment functions, music playing, games, user alert systems, and the like.

Recognizing the state of a user in a wireless communication system plays an important and practical role in improving the functionality of a mobile terminal. The user state may refer to whether the user is indoors or outdoors and recognizes a stationary or moving state. Information about these conditions may provide a variety of different ways to extend the functionality of the mobile terminal.

 For example, when the user is indoors, communication with the GPS may be stopped to reduce battery power consumption of the portable terminal. As another example, there is a user alarm system embedded in mobile terminals, and the user alarm system informs the user that a specific action is required according to the state of the mobile terminal. Another example is a sound vlume adaptation system used depending on whether the user is walking, driving, or stationary in a quiet indoor environment. Another example is call forwarding from a mobile phone to a hands-free automatic audio playback device while driving.

Known user state recognition methods include the use of global navigation systems, such as GPS, standard positioning services, and signal specifications (described on page 46 of the 2nd edition, June 2, 1999 as a Signal Specification). For example. However, since the signals of the global navigation systems are not always available, these systems cannot always provide a user location. In addition, many errors may occur in user location determination in multipath environments, which are particularly typical when determining user location in an urban environment. In such a situation, it may be a significant advantage to use some means to identify some user conditions in the wireless communication system.

Current user state recognition techniques include methods based on user location provided by both GPS and base station signal strength (eg, “Location Determination Systems and Methods” of British Patent Application GB 2454939, published May 27, 2009). There is). The base station signal strength can vary greatly with other obstacles, so-called log-normal fading of the wireless signal strength. In this case, errors in user positioning can affect hundreds of meters. These errors may also make it impossible to detect whether the user is indoors or outdoors.

In addition, as a user state recognition method, methods based on signals of small built-in mechanical devices such as 2D accelerometers (for example, in US Patent Application 2006/0187847 A1, published by Cysco on August 24, 2006) A communication state determination technique using accelerometer data "is known. The accelerometer may be used to determine whether the user is stationary or moving. However, the method has at least two disadvantages. First, the hardware of the portable terminal (eg accelerometer) must be upgraded, which means that a new portable terminal (eg accelerometer included) is essential. Accordingly, the cost of the portable terminal will increase, and there is a problem that is not compatible with the similar portable terminals available. Secondly, small mechanical systems are generally quite sensitive to physical shocks. For example, if the user drops the mobile phone, the interior of the mechanical devices can easily be broken.

Several user state recognition methods (e.g., "apparatus and method using wireless signals" of US patent application 2009/0227271 A1, dated September 10, 2009) provide wireless signals of base stations in the wireless communication system. It is based on a statistical analysis of. The method has the disadvantage that it relies heavily on base station signal levels on a number of uncontrolled elements. This means that the reliability of the user state recognition according to the signal level of the base station in the wireless communication system is very low. This reliability can be improved by increasing the observation time, and statistics accumulated through long time observations will be able to recognize more reliable user state. However, such long observations (typically 10 to 15 minutes) delay the determination of the user state for a long time, which may result in an outdated or useless determination of the user state.

Since known methods for recognizing a plurality of user states in the wireless communication system have the above disadvantages, it would be advantageous to recognize the user states only through software upgrade without having to upgrade the hardware of the portable terminal.

The prior art on the claimed method is "Situational Awareness through Changes in GSM Signal Intensity" (authors Ian Anderson and Henk Muller, published in May 2006 at the Fourth International Conference on Pervasive Computing, ISBN number 3 -85403-207-2 and described on pages 27-31).

The gist of the method according to the prior art is as follows.

At least one base station and at least one user portable terminal are used, and user states are regularly estimated at a specific time period (once per second) every cycle, and are visible to the user portable terminal on the user portable terminal. Measure the number of base stations, measure the signal power of at least one base station, and sum the powers of the signals received from all base stations to obtain two results, the total signal power and the total number of base stations, and the obtained total signal Two results, power and total number of base stations, are transmitted to an embedded neuron network preconfigured and trained with eight hidden elements, and the neuron network gives different weights to the obtained two results. A user state estimate is constructed and sent to the user portable terminal.

The method according to the prior art has at least the following disadvantages.

The prior art method determines user states using the signal strength of the base station. As mentioned above, the signal strength of the base station is greatly affected by a number of uncontrolled factors. This is a very low reliability for the user state recognition based on the signal strength estimation of the base station in the wireless communication system. This reliability can be improved by increasing the observation time, and statistics accumulated through long time observations will be able to recognize more reliable user state. However, such long observations (typically 10 to 15 minutes) delay the determination of the user state for a long time, which may result in an outdated or useless determination of the user state.

Accordingly, the present invention seeks to ameliorate the problems of the system according to the prior art described above.

Revise after claim review

According to a first aspect of the present invention, a method for recognizing a user state of a wireless communication system including a mobile terminal having a microphone includes: receiving audio signals using the microphone, and segmenting the received audio signals into a plurality of segments. Dividing the signal into a plurality of segments; calculating an audio signal level indicator for each of the divided segments; and if at least one of the audio signal level indicators is less than a predetermined threshold value, the user state is set to a first state. It characterized in that it comprises a process of setting to.

According to an embodiment of the present invention, the first state indicates that the mobile terminal is indoors. This has the advantage that mobile terminals can be used more efficiently in wireless communication systems, particularly through more accurate estimation of whether a user is indoors or outdoors in urban environments.

  Embodiments of the present invention are based on the following concept. In general, acoustic noise in urban environments is quite high and free of substantial pauses due to many cars and other noise factors. Indoor acoustic noise is usually caused by the conversation of one or many people. Short pauses generally occur during this conversation. Accordingly, the present method for user state recognition in the wireless communication system is based on indoor and outdoor audio signal recognition algorithms, and is typically based on a concept called quiet period search or silence search to achieve a low audio signal level (audio signal level). Search for a short period of time, which can be represented by an indicator). The audio signal level need not necessarily be zero, but is a relatively low level that is below a predefined threshold. The audio signal level indicator is, for example, the calculated standard deviation of the audio signal level.

If the short period of time (ie, a low audio signal level is observed) is found, the user determines that he is indoors. If not, the user is determined to be outdoors. Therefore, the present invention according to the embodiment can recognize the user state more accurately than the methods according to the prior art, it is possible to use the mobile terminals more efficiently in wireless communication systems.

According to one embodiment, the present invention has the advantage of improving the efficiency of mobile terminals in wireless communication systems by ensuring more accurate user state recognition than prior art in wireless communication systems. This advantage can be achieved by statistically analyzing and processing the external audio signal levels measured by the mobile terminal.

Accurate user state estimates obtained in the wireless communication system stop communication with the GPS when the user is indoors, for example, to save battery power of the mobile terminal, and use the user state estimates to determine the state of the user portable terminal. Inform the user of specific actions to be taken, adjust the phone sound according to whether the user is walking, driving, or stationary in a quiet indoor environment using the user state estimates, and driving using the user state estimates. It enables a variety of operations that extend the functionality of a mobile terminal, such as a call transfer from a mobile phone to a hands-free automatic audio playback device.

Setting of the user state according to an embodiment of the present invention,

Determining a minimum value of the audio signal level indicator for each set of segments to compare the minimum value with the predetermined threshold value; and setting the user state to a first state if the minimum value is less than the predetermined threshold value. It is done through the process.

This has the advantage that the user state setting can be efficiently implemented.

According to one embodiment of the invention, the segments are adjacent segments. This has the advantage that a quiet period between segments is not missed.

According to an embodiment of the present invention, the first state indicates that the user terminal is located indoors.

According to an embodiment of the present invention, the method may include setting the user state to a second state when the minimum value is greater than or equal to the predetermined threshold value, wherein the second state is located outdoors of the user terminal. It shows.

According to an embodiment of the present invention, the method includes recording the received audio signals and determining the audio signal level indicator for each segment in the recorded received audio signals. This has the advantage that the received audio signals do not have to be processed in real time.

According to one embodiment of the invention, the minimum value of the voice signal level indicator corresponds to the minimum value of the standard deviation of audio signal levels in each segment of the received audio signals. This conveniently derives the audio signal level indicator. Provide a way to do it.

According to a second aspect of the present invention, a portable terminal having a microphone in a wireless communication system receives audio signals using the microphone, divides the received audio signals into segments, and audio signal level for each segment. And calculating the indicators and setting the user state to the first state when the value of the at least one audio signal level indicator is less than a predetermined threshold.

According to a third aspect of the invention, a non-volatile recording medium readable by a computer storing one or more instructions executable by a computer device, the instructions being executed by the computer device such that the computer device communicates wirelessly. A system for performing a method of recognizing a user's state, wherein the wireless communication system includes a portable terminal having a microphone, the method for recognizing a user's state, the process of receiving audio signals using the microphone And dividing the received audio signals into segments, calculating an audio signal level indicator for each segment, and when the value of the at least one audio signal level indicator is less than a predetermined threshold value. Characterized in that it comprises the step of setting to the first state do..

Features and advantages of the present invention will become apparent from the preferred embodiments of the present invention, which are presented by way of example.

As noted above, the present invention has the advantage that mobile terminals can be used more efficiently in wireless communication systems, enabling more accurate estimation of whether a user is indoors or outdoors, especially in urban environments. .

1 is a flowchart illustrating an algorithm for user state recognition in wireless communication systems according to an embodiment of the present invention.
2 is a graph showing typical amplitudes of audio signals measured near a road in a city.
3 is a graph showing typical amplitudes of audio signals measured in an office.
4 is a graph illustrating typical autocorrelation functions of audio signals measured in an office and near a road.
5 is a graph showing typical power spectral densities of audio signals in decibels measured in an office and near a road.
FIG. 6 is a graph of indicators of audio signal levels measured in the office and near the road for 4.5 seconds, specifically the standard deviation of the audio signals.
FIG. 7 is a graph showing the minimum moving value of consecutive 100 standard deviations of audio signals measured in the office and near the road for 4.5 seconds.

Hereinafter, an embodiment according to the present invention will be described taking the situation of a user terminal (for example, a mobile terminal) such as a mobile handset in a mobile wireless system as an example. However, this is merely an example, and other embodiments may include other types of wireless networks or wireless communication systems and other types of terminals such as, for example, portable computers.

One embodiment according to the present invention can be implemented on a mobile terminal that is a user terminal in the order shown in FIG. According to one embodiment, it is assumed that the user portable terminal includes at least one built-in microphone and a calculation module (processor module). In step S101, the microphone may operate after a specific time period through software or the like to receive and record external audio signals. Such reception may for example be monitoring external audio signals. In step S102, using the arithmetic module, the external audio signals may be recorded (or recorded or stored) in the form of a standard "wav" file for a specific time period. In steps S103-S104, in general, when the recording is completed, the "wav" file is read and converted into a set of external audio signal level values, the set of values being segments having a certain length, generally adjacent segments. It can be divided into

In step S106, the audio signal level indicator (eg, the standard deviation of the audio signal level in the segment) is calculated for each segment, and for a specified duration corresponding to a set of segments. A row may be selected, wherein a particular time-phased column consists of successive values of the calculated audio signal level indicators. For example, in steps S107-S108, a standard deviation of audio signal levels may be selected, which is compared with a preset threshold after the minimum audio signal level indicator is selected in relation to the row described above. The state of the user's portable terminal can be determined. In step S109, if the minimum audio signal level indicator value is less than the preset threshold, the user portable terminal is determined to be indoors and the user status indicator is set to the first state. In operation S110, when the minimum indicator value is greater than or equal to the threshold value, the user portable terminal determines that the user is outdoors and sets the user status indicator to the second state.

Embodiments according to the present invention will be described in detail with reference to FIGS. 2 to 7.

Acoustic noise in an office can have statistically significantly different characteristics from acoustic noise outdoors (eg, on a roadside in an urban environment). In general, acoustic noise closer to a more complex urban road than in an office is much louder. In addition, the spectral content of noise indoors and outdoors may also be different. Thus, in the embodiment of the present invention, two kinds of user states, i.e., indoor state (eg, in an office) and outdoor state (eg, complicated city) based on audio signals measured by the user's portable terminal, are illustrated. Rules for differentiating roads and neighbors) may be applied.

Since many portable terminals have a built-in microphone, acoustic noise can be measured without a hardware upgrade. The measured audio signals can be recorded in wav files via the calculation module. The audio signals from the wav files may be decoded in the form of realizing an initial audio signal strength. At this time, the received audio signals are not necessarily recorded, and the signals are divided into segments in real time, and the audio level indicator may be calculated for each segment or for each sliding time window.

To perform statistical analysis on the audio signals, a field test is performed near the road and office for the selected sampling frequency and time interval between two neighboring measured values as shown in FIGS. 2 and 3 respectively. Signal acquisition (eg, realization). Since the acoustic noise in the office is inherently a non-stationary stochastic process, a preliminary classification was performed by identifying each of the typical parts of the implementations.

2 and 3 show two typical measured implementations of audio signals when the portable terminal is outdoors (2) near a road (shown in FIG. 2) and indoors (4) in an office (shown in FIG. 3). 2 and 4 are shown. The sample size is 10000 samples, corresponding to 0.45 seconds for each implementation.

The above realizations show that the average amplitude of audio signals indoors (eg in an office) may be lower or higher than the average amplitudes of outdoor (eg roadside) signals at different times.

4 shows typical normalized autocorrelation functions of the respective audio signals 6, 8. 4 shows that the autocorrelation function of an audio signal measured indoors 6 and outdoors 8 may not be essentially different.

범위 내에서 측정된 오디오 신호들의 파워 스펙트럼 밀도 추정들(power spectral density estimates)을 보여주고 있다. 여기서 Fs는 샘플링 주파수를 나타낸다.The general method of acoustic noise analysis is based on spectral studies rather than the correlation characteristics of audio signals. Figure 5 shows the maximum Nyquist frequency at 0 Hz

Power spectral density estimates of audio signals measured in range are shown. Where Fs represents the sampling frequency.

Figure 5 clearly shows the main intensity of the signal focuses measured in the 0-1 kHz range. In addition, FIG. 5 shows that, due to limited analysis data, it is difficult to define patterns of such differences, but the spectral components of the audio signals of the office 10 and the roadside 12 are slightly different.

2 to 5, the user state discrimination algorithm based on the correlation or spectral characteristics of indoor and outdoor acoustic noise may be somewhat complicated because of similar correlation function shapes and spectral densities corresponding to the states. It can be seen that there is.

Therefore, the user state recognition method according to an embodiment of the present invention can be implemented in the wireless communication system based on the difference of fluctuation values of wireless signals between indoors such as an office and outdoors such as near a road.

Embodiments of the present invention may be based on the following concept. In urban environments, acoustic noise is quite high due to many cars and other noise factors, and there is generally no practical pause. Room acoustic noise is typically caused by conversations of one or several people, and typically short pauses occur during the conversation. According to an embodiment of the present invention, user state recognition in the wireless communication system is a concept of "silence search" or "quiet search", and it is not necessary to have a relatively low acoustic signal level (eg, a level of 0). None is based on indoor and outdoor audio signal recognition algorithms that search for the short time interval observed. When the above-mentioned short time intervals are found, it may be determined that the user is indoors, or it may be determined that the user is outdoors.

The acoustic noise calculated in the set time period, that is, the sampling standard deviation of the received audio signals may be used as a measurement function of the acoustic noise level.

x _t represents acoustic noise realization, ie received audio signals. Then, the sample estimation of the audio signal level indicator, i.e., the standard deviation of the audio signal at time interval L , is as follows.

Where μ is the average audio signal value at time interval L. The algorithm for user state recognition may be performed as follows.

Step 1 . In this step, the audio signal standard deviation values K in adjacent time periods are calculated according to the following equation.

,

,

,

,

.......................

.

.

As a result, the audio signal standard deviation values K in adjacent time intervals become as follows.

Specifically, the K value in this case is 100 may be selected. Therefore, the total time period required for the determination of the user state is K · L samples corresponding to 4.5 seconds.

FIG. 6 shows typical standard deviation estimates of audio signals measured at office 16 and outdoors 14 near a road. FIG. 6 shows that the standard deviations of audio signals measured in an office can often be higher than those measured outdoors in the vicinity of a road. Thus, the following steps may be used to reduce possible errors in recognizing user status.

Step 2 . The minimum value of the standard deviation values K of the audio signal levels is calculated as follows.

FIG. 7 shows typical minimum indicator values for audio signals measured in office 22 and near road 18. 7 also shows that the minimum values of audio signal levels measured in the office are generally not higher than those measured near the road.

Step 3 . In this step, the minimum value of the audio signal level indicators is compared with a minimum threshold value of 20, i.e., a predetermined threshold value h , to determine the user state according to the comparison result. The threshold can be obtained by experiment. The decision is made as follows.

When the minimum audio signal level indicator value SD _min is less than the threshold value h , it may be determined that the user portable terminal is indoors, for example, in an office.

When the minimum audio signal level indicator SD _min is greater than or equal to the threshold h, it may be determined that the user portable terminal is outdoors, for example, near a road. The outdoor location need not be near the road and will have similar characteristics in the acoustic environment at other outdoor locations. For example, there may be a relatively constant noise level due to fewer pauses or silent periods.

FIG. 7 shows typical minimum indicators of audio signal levels measured near the office 22 and the road 18, as well as the threshold 20. 7 shows that the minimum indicators of the audio signal level measured in the office are below the predetermined threshold, and the minimum indicators of the audio signal level measured outdoors, such as near a road, exceed the predetermined threshold. . Thus, in this example, the proposed algorithm for the audio signal implementations according to the method of the present invention can accurately determine the user state without error.

Embodiments of the present invention can be implemented in a user's portable computer according to the above algorithm. In this case, the portable computer should have a microphone and an arithmetic module (processor module) available in almost all computers.

In the above, the present invention has been described with reference to the above embodiments, but the embodiments of the present invention are merely examples. Any features described in connection with an embodiment of the invention may be used alone or in combination with other features, and may be used with one or more features of other embodiments or in any combination of other embodiments. have. Also, even if not mentioned above, equivalents and modifications according to the present invention may be applied without departing from the scope of the present invention, which is the following claims.

Claims (15)

In the user state recognition method of a wireless communication system comprising a portable terminal having a microphone,
Receiving audio signals using the microphone;
Dividing the received audio signals into a plurality of segments;
Calculating an audio signal level indicator for each of the divided segments;
And setting a user state to a first state if at least one of the audio signal level indicators is less than a predetermined threshold value.
According to claim 1, The setting process,
Determining a minimum value of the audio signal level indicator for each set of segments and comparing the minimum value with the predetermined threshold value;
And setting the user state to a first state if the minimum value is less than the predetermined threshold value.
The method according to claim 1 or 2,
And the segments are adjacent to each other.
The method of claim 1,
And the first state indicates that the mobile terminal is located indoors.
The method of claim 2,
If the minimum value is greater than or equal to the predetermined threshold value, setting the user state to a second state.
The method of claim 5,
The second state indicates that the mobile terminal is located outdoors.
The method of claim 1, wherein the voice signal level indicator,
And after the received audio signals are recorded, are calculated from the recorded audio signals.
The method of claim 2, wherein the minimum value of the voice signal level indicator,
Corresponds to a minimum value of a standard deviation of audio signal levels within each segment of the received audio signals,
The standard deviation SD _t is calculated through Equation 1 below, wherein x _t is a received audio signal, and μ corresponds to an average audio signal value in a time interval L. State awareness method.

... Equation 1
In a wireless terminal having a microphone in a wireless communication system,
Receive audio signals using the microphone,
Divide the received audio signals into segments,
Calculate audio signal level indicators for each segment,
And when the value of the at least one audio signal level indicator is less than a predetermined threshold, setting the user state to the first state.
10. The method of claim 9,
The first state is set when the minimum value of the audio signal level indicator is less than the threshold, and the first state indicates that the mobile terminal is located indoors.
10. The method of claim 9,
And if the value of the at least one audio signal level indicator is equal to or greater than the threshold, set a user state to a second state.
The method of claim 11,
The second state is set when the minimum value of the audio signal level indicator is greater than or equal to the threshold value, and the second state indicates that the mobile terminal is located outdoors.
A computer-readable nonvolatile recording medium storing one or more instructions executable by a computer device, the method comprising:
The instructions are executed by a computer device to cause the computer device to perform a method of recognizing a user's state in a wireless communication system,
The wireless communication system includes a mobile terminal having a microphone, the method for recognizing the state of the user,
Receiving audio signals using the microphone;
Dividing the received audio signals into segments;
Calculating an audio signal level indicator for each segment;
And setting a user state to a first state when the value of the at least one audio signal level indicator is less than a predetermined threshold.
The method of claim 13,
And wherein the first state is set when the minimum value of the audio signal level indicator is less than the threshold value, and the first state indicates that the portable terminal is located indoors.
The method of claim 13,
If the value of the at least one audio signal level indicator is greater than or equal to a predetermined threshold, the user state is set to a second state, wherein the second state indicates that the mobile terminal is located outdoors. media.

Images