KR101481533B1 - Sensor Signal Filtering Method and Apparatus using Multi Block Filter - Google Patents

Abstract

A method and an apparatus for filtering a sensor signal using a multi-block filter are provided. In the filtering method according to the embodiment of the present invention, the selected filters are connected to generate one multi-block filter, and the input multi-block filter is used to filter the input signal. As a result, the multi-filter can be designed more easily by the developer, so that it is possible to provide universality and scalability.

Description

[0001] The present invention relates to a sensor signal filtering method and apparatus using a multi-block filter,

The present invention relates to a method and apparatus for filtering a sensor signal, and more particularly to a method and apparatus for filtering a sensor signal received from a sensor installed in a garment.

In addition to smart phones, smart devices such as smart apparel and sportswear have been equipped with ultra-small devices equipped with multiple sensors with more than six axes. Axis sensor, 3-axis gyroscope (3-axis), and temperature sensor. The 9-axis sensor adds one of the compass sensor, magnetic field sensor, gravity sensor and pressure sensor to the 6-axis sensor function. Axis sensor is formed.

A separate physical filter for ensuring reliability and efficiency is provided in the composite sensor to a plurality of raw data generated in such a composite sensor. Physical filters are mostly composed of digital filters. Physical filters can control not only the noise and ripple but also the cycle of information gathering according to the behavior inference service and the physical error of the sensor. However, for various behavior inference services, digital filter technology is required to process various raw data for individual user services. However, since the physical filter or the controller is fixed in the existing complex sensor, it is difficult to provide various behavior inference services. To this end, a separate low-speed, low-power microprocessor is installed and software is used to create a digital filter. New data for the service demanded by the user can be added to the sensor such as physical sensitivity, data collection distribution, Provide information by processing the information according to the service.

In order to overcome this problem, a number of physical filters are included in the composite sensor and proposed by a combination of users. In addition to the noise and ripple, we propose the role of providing the raw data collected by the sensor as data suitable for various services desired by the user.

SUMMARY OF THE INVENTION The present invention has been conceived to solve the problems as described above, and it is an object of the present invention to provide a method and apparatus for enhancing vulnerability of individual physical digital filters, The present invention provides a method and apparatus for filtering a sensor signal using a multi-block filter as a method for providing raw data processing information optimized for a service to secure data reliability and designing a filter more conveniently.

According to an aspect of the present invention, there is provided a filtering method comprising: receiving filters; Generating one multi-block filter by connecting the filters selected in the selecting step; And filtering the input signal using the multi-block filter generated in the generating step.

In the selection step, the type and characteristics of the filter can be selected.

In addition, the generation step may generate the multi-block filter by serially connecting the filters selected in the selection step.

The filtering method according to an embodiment of the present invention may further include changing the order of the filters included in the multi-block filter.

Also, the changing step may be performed when the filtering performance of the multi-block filter whose order has been changed is superior to the filtering performance of the multi-block filter whose order is not changed.

The filtering method according to an embodiment of the present invention may further include replacing at least one of the filters included in the multi-block filter with another filter.

Also, in the changing step, at least one of the filters included in the multi-block filter may be replaced with another type of filter performing the same function.

The changing step may be performed when the performance of the multi-block filter in which at least one filter is replaced with another filter is superior to the performance of the multi-block filter in which at least one filter is not replaced with another filter.

In addition, the generation step may generate the multi-block filter by connecting the filters selected in the selection step in a serial / parallel manner.

According to another aspect of the present invention, there is provided a filtering apparatus comprising: a storage unit storing a plurality of filters; And a processor for coupling the selected filters among the filters stored in the storage unit to generate one multi-block filter and filtering the input signal using the generated multi-block filter.

As described above, according to the embodiments of the present invention, it is possible to improve the reliability of the raw data of the sensor (minimize ripple and noise according to the external environment). In addition, the multi-filter can be designed more conveniently by the developer, thereby providing universality and scalability.

In addition, the data for deducing the behavior pattern of the wearer is accurate, and the inference error can be minimized.

1 is a block diagram of a sensor signal processing apparatus according to an embodiment of the present invention;
Fig. 2 is a view provided to explain a change of order of filters constituting a multi-block filter, Fig.
FIG. 3 is a view provided to explain a replacement of a part of filters constituting a multi-block filter,
4 is a view illustrating another connection method of filters constituting the multi-block filter.

Hereinafter, the present invention will be described in detail with reference to the drawings.

1 is a block diagram of a sensor signal processing apparatus according to an embodiment of the present invention. The sensor signal processing apparatus according to the present embodiment enhances the reliability of data by performing filtering for eliminating noise, etc., on raw data received from a sensor installed in a garment.

As shown in FIG. 1, the sensor signal processing apparatus according to the present embodiment, which performs such a function, includes input units 110-1, 110-2, and 110-3, ADCs 120-1 and 120-2, 2, and 120-3, a processor 130, an output unit 180, a communication interface 190, and an operation unit 194. [

The input units 110-1, 110-2, and 110-3 are interface units that receive sensor signals from the composite sensors installed in the clothes. The input units 110-1, 110-2, and 110-3 receive the sensor signals from the ADCs 120-1, 120-2, -3).

The ADCs 120-1, 120-2, and 120-3 convert the sensor signals received from the input units 110-1, 110-2, and 110-3 from analog to digital. The ADCs 120-1, 120-2 and 120-3 may be omitted if the digital sensor signals are input through the input units 110-1, 110-2 and 110-3.

The processor 130 performs filtering on the digital sensor signals received from the ADCs 120-1, 120-2, and 120-3, and transmits the filtered signals to the external system through the output unit 180 .

The processor 130 includes a buffer 140, a multi-block filter 150, a memory 160, and a factory calibrator 170 for this purpose.

The buffer 140 is a space in which the digital sensor signals received from the ADCs 120-1, 120-2, and 120-3 are temporarily stored.

The multi-block filter 150 is a software module that is executed by the processor 130. For digital sensor signals, the multi-block filter 150 performs digital filtering. The multi-block filter 150 has a plurality of filters 150-l, 150-m and 150-n connected in series as shown in FIG. The filters 150-1, 150-m and 150-n constituting the multi-block filter 150 are filters selected by the developer.

Specifically, the developer selects desired filters among the filters 160-1, 160-2, 160-3, ... stored in the memory 160 through the communication interface 190 or the operation unit 195 in order The processor 130 configures the filters of the multi-block filter 150 according to the developer's selection.

The memory 160 is constituted by the selection of the physically configured filters and the input conditions. The selection condition has an individual selection value for the block filter 150 configuration, and the input condition defines the required value and range of filters for the selected filters. This information is stored and remembered. Further, there is no limitation on the filter to be selected and constitute the multi-block filter 150. [ A Kalman filter, a dual Kalman filter, a moving average filter, a median filter, a smoothing filter, a maximum likelihood filter, and a maximum likelihood filter. (Maximum Likelihood) filter can be applied.

The developer's choice includes the characteristics of the filter in addition to the type of filter. In other words, the developer must select a desired filter among the filters 160-1, 160-2, 160-3, ... stored in the memory 160, Should be determined and input. Even if the same kind of filter has different filtering characteristics, different filtering results are obtained for the same input signal.

The filters 150-1, 150-m and 150-n constituting the multi-block filter 150 may be replaced with other filters. For example, if a sensor that transmits an additional sensor signal is changed / added through the input unit 110, the physically configured filters constituting the multi-block filter 150 may be replaced to select a multi-block filter 150 < / RTI >

Since the multi-block filter 150 is a physical internal structure module that is executed by the physically configured controller 130, the developer can control the filter of the filter stored in the memory 160 via the communication interface 190 or the operating unit 195 160-1, 160-2, 160-3, ...), it is possible to change the filter configuration of the multi-block filter 150 by sequentially selecting necessary filters among the individually operable blocks.

The filter configuration change of the multi-block filter 150 may be performed automatically, not by a command of the developer, and will be described in detail below.

The processor 130 generates a virtual sensor signal, inputs a noise signal to the generated virtual sensor signal to the multi-block filter 150, compares the filtered signal with a virtual sensor signal generated in the beginning, The filtering performance of the multi-block filter 150 can be measured (measurement result 1).

The processor 130 further includes a filter-1 150-l and a filter-m 150-m among the filters constituting the multi-block filter 150, as shown in "a) The filtering performance of the multi-block filter 150 can be measured (measurement result 2).

The processor 130 then selects the filter-m 150-m and the filter-n 150-n among the filters constituting the multi-block filter 150, as shown in "b) The filtering performance of the multi-block filter 150 can be measured (measurement result 3).

If the measurement result (2) shows the best filtering performance, the processor 130 changes the order of the filters constituting the multi-block filter 150 as shown in "a)" If the measurement result ③ shows the best filtering performance, the processor 130 changes the order of the filters constituting the multi-block filter 150 as shown in "b)"

On the other hand, if the measurement result (1) shows the best filtering performance, the processor (130) maintains the order of the filters constituting the multi-block filter (150) unchanged.

On the other hand, changing the order of the filters constituting the multi-block filter 150 is also possible for non-adjacent filters. That is, it is also possible to replace the order of the non-adjacent filter-1 (150-l) and the filter-n (150-n).

Hereinafter, a method of changing the filter configuration of the multi-block filter 150 by replacing a part of the filters constituting the multi-block filter 150 will be described in detail with reference to FIG.

The method includes the steps of:

1) The filter-1 (150-l) of the multi-block filter 150 is replaced with the filter-x 150-x as shown in "a)" of FIG. 3,

2) By replacing the filter-m 150-m of the multi-block filter 150 with the filter-y 150-y as shown in "b)" of FIG. 3,

3) By replacing the filter-n 150-n of the multi-block filter 150 with the filter-z 150-z as shown in FIG. 3C, filtering performance is measured,

And replacing the configuration of the multi-block filter 150 according to the replacement method showing the best filtering performance. If the original method exhibits the best filtering performance, the processor 130 maintains the order of the filters that make up the multi-block filter 150 unchanged.

Here, the filter replacement may be replaced with another type of filter performing the same function as the filter to be replaced, or may be performed by changing only the characteristics of the filter.

It is also possible to change at least two of the filters constituting the multi-block filter 150 at a time. That is, as shown in Fig. 3, it is not possible to change only one filter.

Up to now, a method and an apparatus for filtering a sensor signal using a multi-block filter have been described in detail with reference to preferred embodiments.

In the above embodiment, it is assumed that the filters constituting the multi-block filter 150 are connected in series. However, this is merely an example for convenience of explanation. The technical idea of the present invention is also applicable to the case where the filters constituting the multi-block filter 150 are connected in parallel or in a series / parallel combination. FIG. 4 illustrates a case where filters constituting the multi-block filter 150 are connected in a direct / parallel mix.

In addition, one filter constituting the multi-block filter 150 may be implemented by a front end filter and a back end filter. This is for design needs and convenience.

In addition, the technical idea of the present invention can be applied to filtering of signals other than the sensor signals, and it is applicable to the field of image signal processing as well as processing of wired / wireless communication signals as well as sensor signal processing.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention.

110-1, 110-2, and 110-3: input units 120-1, 120-2, and 120-3: ADC
130: Processor 140: Buffer
150: Multi-block filter 160: Memory
160: Memory 170: Factory Calibratoion
180: output unit 190: communication interface
195:

Claims (10)

The filtering device selecting the filters;
The filtering apparatus may further include a step of connecting the filters selected in the selecting step to generate a multi-block filter.
Wherein the filtering device comprises: changing the order of the filters included in the multi-block filter;
Wherein the filtering device comprises: replacing at least one of the filters included in the multi-block filter with another filter; And
And filtering the input signal by the filtering device using the multi-block filter.
The method according to claim 1,
Wherein the selecting step comprises:
Wherein a type and a characteristic of the filter are selected.
The method according to claim 1,
Wherein the generating comprises:
Wherein the multi-block filter is generated by serially connecting the filters selected in the selecting step.
delete The method according to claim 1,
Wherein the modifying comprises:
Wherein the filtering performance of the multi-block filter whose order is changed is superior to the filtering performance of the multi-block filter whose order has not been changed.
delete The method according to claim 1,
Wherein the replacing comprises:
Wherein at least one of the filters included in the multi-block filter is replaced with another type of filter performing the same function.
The method according to claim 1,
Wherein the replacing comprises:
Wherein the performance of the multi-block filter in which at least one filter is replaced with another filter is performed when the at least one filter is superior to the performance of the multi-block filter not replaced with another filter.
The method according to claim 1,
Wherein the generating comprises:
Wherein the multi-block filter is generated by connecting selected filters in a serial / parallel manner in the selecting step.
A storage for storing a plurality of filters; And
And a processor for coupling the selected filters among the filters stored in the storage unit to generate one multi-block filter and filtering the input signal using the generated multi-block filter,
The processor comprising:
Wherein the order of the filters included in the multi-block filter is changed, and at least one of the filters included in the multi-block filter is replaced with another filter.

Images