KR100634530B1 - Method and apparatus for character input and function selection by combining user's key input and motion - Google Patents

Abstract

The present invention relates to a method and apparatus for inputting a character input or performing a specific function using a combination of a user's key button input and a movement in a terminal such as a cellular phone. Receiving a key input from a user; Detecting a user's movement using a sensor; Recognizing a pattern of the sensed movement; And performing a function corresponding to a combination of a key input and a recognized movement pattern.

According to the present invention, when inputting a character or performing a specific function such as a wireless network connection in a mobile phone, the character input speed is improved by enabling a function corresponding to a combination of a user's key input and a user's movement. In addition, since a character or a function more than a limited number of key input buttons can be input by a key input, it may provide convenience to a user.

Description

Character and function input method combining user's key input and movement, and terminal using the same {Method and apparatus for character input and function selection by combining user's key input and motion}

1 is a diagram showing the configuration of a text input button of a cellular phone.

2 is a block diagram illustrating a configuration of a function input device combining a key input and a movement of a user according to the present invention.

3 is a block diagram illustrating a configuration of a setting device that allows a user to set a specific function by combining a key input and a movement.

FIG. 4 is a diagram illustrating a first embodiment of a character input method combining a user's key input and movement.

FIG. 5 is a diagram illustrating a second embodiment of a character input method combining a user's key input and movement.

FIG. 6 is a diagram illustrating a third embodiment of a character input method combining a user's key input and movement.

7 is a table illustrating an embodiment of a method of performing a mobile phone function combining a user's key input and movement.

8 is a graph illustrating output signals of an inertial sensor according to left and right movements and a stop state of a user.

9 is a diagram illustrating an embodiment of a user's motion trajectory.

FIG. 10 is a graph illustrating output signals of an inertial sensor according to the user motion trajectory of FIG. 9.

11 is a block diagram illustrating a configuration of a pattern recognition unit that recognizes a motion pattern from a motion signal.

12 is a flowchart illustrating a function input method combining a user's key input and movement.

FIG. 13 is a flowchart illustrating a first embodiment of a method for setting a user to perform a specific function by combining a key input and a movement.

FIG. 14 is a flowchart illustrating a second embodiment of a method for setting a user to perform a specific function by combining a key input and a movement.

The present invention relates to a method and an apparatus for inputting a character or selecting a specific function in a terminal such as a mobile phone, and more particularly, by using a combination of a user's key button input and a movement pattern to input a character or a specific function corresponding thereto. A method and apparatus are provided.

In general, the user inputs Korean, English, and numbers using a key keyboard installed on one side of the mobile phone. However, to input Korean and English characters through the limited keys of the keyboard, several Korean vowels / consonants and English are assigned to one key. In order to repeatedly input any one of a plurality of characters assigned to one key, the same key must be repeatedly pressed at predetermined time intervals or a character may be repeatedly input using another special key.

Particularly, when one wants to repeatedly input any one of several consonants assigned to the same key in the Korean input method by Cheonjiin, one consonant is input, and after a while, the consonant is input again. Alternatively, enter one consonant, press the arrow keys, and then enter the consonant again. If the consonant key is pressed again within a certain amount of time, the consonant key is changed to another consonant assigned to the consonant key. Therefore, a problem arises in that the input time is slowed down by a large number of key input times and a delay time until the letter is determined.

FIG. 1 illustrates a configuration of a text input button of a mobile phone, and a general text input method will be described with reference to FIG. 1.

Figure 1 (a) shows a mobile phone English letter input button, the user can enter three alphabets with one button. For example, if the user wants to input CLEAR with the input button, the user presses two buttons 100 three times in succession, five buttons 110 consecutively three times, and then presses three buttons 120 two times consecutively. After pressing the 2 button 100 once, the 7 button 140 is pressed twice in succession to enter.

FIG. 1 (b) shows a Korean character input button which is a heaven and earth of a mobile phone, and a user can input two consonants or one vowel with one button. For example, if the user wants to enter a patent using the input button, the user presses the c button 150 twice in succession, presses the-button 150 once, presses the a button 160 once, and the S button ( After pressing 170 two times in succession, press button 180 once and then press button 190 once to enter.

In addition, recently, due to the multi-functionality of mobile phones, various functions such as mobile phone access to wireless Internet to obtain information, listen to music, take pictures, etc. are simply added by means of communication. Compared with the addition of the functions of the mobile phone as described above, the number of keys that can be input to the mobile phone is limited, so that each time a new function is added, the number of key inputs for performing the function increases.

For example, in order to download the latest ringtones by accessing the wireless Internet from a mobile phone, press the wireless internet connection button, select the My Bell menu after connecting to the wireless Internet, and the ringtone menu, which is a submenu of the My Bell menu. After selecting, a total of four key inputs are required, such as selecting the latest song menu, which is a submenu of the ringtone menu.

As described above, when inputting a character or performing a specific function in a mobile phone using a conventional input method, it is inconvenient because there are a large number of buttons to be pressed, and in particular, when a character is input, one button is continuously input in a short time. There is a problem that there is a lot of error of character input and time delay occurs until character input is completed.

The technical problem to be achieved by the present invention, in order to solve the above problems in inputting text in a mobile phone, such as using a combination of the user's key button input and the movement of a small number of key input to perform a character or a specific function It is to provide a method and apparatus for inputting.

The input method according to the present invention for solving the above technical problem, the step of receiving a key input from the user; Detecting a user's movement using a sensor; Recognizing the pattern of the detected movement; And performing a function corresponding to a combination of the key input and the recognized movement pattern.

Preferably, the performing of the function may further include generating a character corresponding to the combination of the key input and the recognized movement pattern, and displaying the generated character.

Recognizing the movement pattern may be any one of artificial neural network, template matching, hidden Markov model, SVM (Support Vector Machine), it is preferable to recognize the detected pattern of the user's movement.

Preferably, the input method receives one of predetermined movement patterns, one of predetermined keys, and a function to perform from a user, and corresponds to the inputted function in combination of the input movement pattern and key input. It further comprises the step of.

The input method may further include receiving a user's movement, one of predetermined keys, and a function to be performed by a user, and corresponding the input function to a combination of the user's movement pattern and key input. desirable.

Preferably, the detecting of the movement may include detecting at least one of a user's movement using at least one of an angular velocity sensor and an acceleration sensor.

The detecting of the motion may include detecting a motion during a key input time of the user by using a sensor, or detecting a motion for a predetermined time after the user's key input.

Preferably, the step of recognizing the movement pattern is to recognize the pattern of the detected movement trajectory of the user, it is preferable to recognize the movement of the detected user as one of the predetermined number of movement patterns. .

Recognizing the movement pattern may include extracting a feature of the detected movement; And recognizing a motion pattern of a predetermined number of motion patterns by using the extracted motion feature.

Advantageously, said movement pattern comprises left movement, right movement, and a stationary state.

An input device according to the present invention for solving the above technical problem. A key input unit for generating and outputting a key input signal corresponding to a user's key input; A sensor unit which detects a user's movement and generates and outputs a motion signal corresponding to the detected movement; A pattern recognition unit recognizing a movement pattern of the user from the movement signal; Memory for storing information on a performance function corresponding to a combination of a key input and a movement pattern; And a signal generation unit configured to read information about a function corresponding to the combination of the key input signal and the recognized movement pattern from the memory to generate and output a signal corresponding to the function to be performed.

Preferably, the signal generator generates and outputs a signal corresponding to a character corresponding to a combination of the key input signal and the recognized movement pattern.

The input device may include a pattern selection unit receiving one of predetermined movement patterns from a user; A function input unit for receiving a function to be performed from a user; And a first setting unit for storing a combination of a movement pattern input from the pattern selection unit and a key input of a user input from the key input unit in the memory in correspondence with a function input from the function input unit.

Preferably, the input device comprises a function input unit for receiving a function to be performed from a user; And a second setting unit configured to store a combination of a user's movement input from the sensor unit and a user's key input input from the key input unit in the memory in correspondence with a function input from the function input unit.

Preferably, the sensor unit includes at least one of an angular velocity sensor and an acceleration sensor, and detects a user's movement during a key input time of the user, generates and outputs a motion signal corresponding to the detected movement, or the user's key. It is preferable to generate and output a motion signal corresponding to the detected movement by detecting a user's movement for a predetermined time after the input.

Preferably, the pattern recognition unit recognizes the user's movement as one of a predetermined number of movement patterns using the user's motion signal, and recognizes the pattern of the user's motion trajectory using the motion signal. It is desirable to.

The pattern recognition unit extracts a feature of the user's movement using the motion signal; And a pattern selector which selects one motion pattern from a predetermined number of motion patterns by using the extracted motion feature.

Preferably, the pattern recognition unit uses any one of artificial neural networks, template matching, hidden Markov models, and SVM (Support Vector Machine), one of a predetermined number of motion patterns from the user's motion signal. Recognizes movement patterns.

The artificial neural network, template matching, hidden Markov model, and SVM may be learned by a user.

Advantageously, said movement pattern comprises left movement, right movement, and a stationary state.

The character or function input method may be embodied as a computer-readable recording medium recording a program for execution on a computer.

Hereinafter, a method and apparatus for inputting a character and a function using a key input and a movement of a user according to the present invention will be described in detail with reference to the accompanying drawings.

2 is a block diagram illustrating a configuration of a function input device 200 using a user's key input and movement according to the present invention. The illustrated input device 200 includes a key input unit 205, a sensor unit 210, a pattern recognition unit 220, a signal generator 230, and a memory 240. An operation of the illustrated input device 200 will be described with reference to a flowchart illustrating the input method illustrated in FIG. 12.

The key input unit 205 receives a key input for character input or a function performance selection from a user, generates and outputs a key input signal corresponding to the key input (step 1200). The key input unit 205 is preferably made of alphabets or Hangul consonants, buttons displaying vowels, wireless Internet connection buttons or menu buttons.

The sensor unit 210 detects a user's movement and generates and outputs a sensor output signal corresponding to the detected movement (step 1210). The sensor unit 210 includes an angular velocity sensor that detects an angular velocity of a user's movement, an acceleration sensor that detects an acceleration of a user's movement, or includes both an angular velocity sensor and an acceleration sensor. It is desirable to simultaneously detect acceleration and acceleration. The sensor unit 210 may detect a user's movement by using a geomagnetic sensor.

The angular velocity and acceleration of the portable device change according to the movement of the user, for example, the movement of the user's hand holding a portable device such as a mobile phone for text input. Therefore, the angular velocity sensor attached to the portable device detects the angular velocity of the portable device, that is, whether the portable device rotates left / right or up / down or rotates clockwise / counterclockwise and corresponds to the detected angular velocity. To generate a sensor output signal. An acceleration sensor detects an acceleration of the portable device, that is, a change in the movement speed of the portable device, and generates a sensor output signal corresponding to the sensed acceleration.

The sensor unit 210 detects the user's movement while the user presses a key of the key input unit 205 while the user presses a key through the key input unit 205, for example, the time. It is desirable to generate and output a sensor output signal for the user's movement during the period. In addition, the sensor unit 210 is a key for a predetermined time after the user inputs the key through the key input unit 205, for example, for 1 second after the user presses and releases the button of the key input unit 205. By detecting the user's movement of the sensor may generate and output a sensor output signal for the user's movement during the time.

The pattern recognition unit 220 receives a motion signal, which is a sensor output signal that detects a user's movement, from the sensor unit 210 and recognizes a pattern of the user's movement in operation 1220. In operation 1220 of recognizing the motion pattern, one feature of a predetermined number of motion patterns stored in the memory 240 is extracted by using a feature of the motion signal and stored in the memory 240. As such, it is preferable to generate and output a signal corresponding to the recognized movement pattern by recognizing the movement signal.

The method for recognizing a movement pattern from the user's movement signals, the recognition of the movement pattern using a pre-trained artificial neural network, pattern recognition such as template matching, hidden Markov model, SVM (Support Vector Machine) It is desirable to recognize the movement pattern using the method.

 In order to perform a character input or a specific function by combining a key input and a movement pattern, the memory 240 stores a specific character or a specific function in correspondence with a combination of a specific key input and a specific movement pattern. For example, the memory 240 may store a ring tone changing function corresponding to a combination of a menu button input and a rightward movement pattern.

The signal generation unit 230 receives a key input signal received from the key input unit 205 and a movement pattern of the user received from the pattern recognition unit 220 as input, and specifies a specific correspondence to the combination of the key input and the movement pattern. After reading a character or function from the memory 240 (step 1230), a signal corresponding to the specific character or performing function is generated and output (step 1230).

The function performing unit 250 included in the device such as a mobile phone receives the generated signal from the signal generator 230 to generate a character corresponding to the input signal or perform a function (step 1250). When the character is generated in step 1250, the display unit 260 may display the generated character on the screen so that the user can see the input character.

The specific characters or specific functions corresponding to the key input, the movement pattern, and the combination thereof stored in the memory 240 are stored in the memory 240 in advance by the seller's setting in the production of the device such as a mobile phone, and then the device is It is preferable that the user who purchases the information about the character or performing function input according to the combination be provided. Alternatively, after a user purchases a device, the user may store a desired character or performance function in the memory 240 in response to a combination of a desired key input and a movement pattern, and then store the specific character or function according to the combination of the key input and the movement pattern. It is desirable to set performance arbitrarily.

3 is a block diagram illustrating a configuration of a setting device that allows a user to set a specific function by performing combination of a specific key input and a movement. The illustrated setting device includes a key input unit 205, a sensor unit 210, The pattern selector 300 includes a function input unit 310, a setting unit 320, and a memory 240. The operation of the setting apparatus illustrated in FIG. 3 will be described with reference to a flowchart illustrating a method of setting a specific function to be performed on the combination of the key input and the movement pattern illustrated in FIG. 13.

The key input unit 205 receives a key input for performing a specific character or function from the user (step 1300), and the pattern selector 300 selects one movement pattern from a predetermined number of movement patterns. (Step 1310). In operation 1310, after the display unit 260 displays the movement patterns selectable to the user, the pattern selection unit 300 preferably receives one of the displayed movement patterns from the user. In operation 1310, the setting apparatus does not include a separate pattern selector 300, and preferably selects a movement pattern from a user through the keys including the key input unit 205. For example, the movement pattern may be selected by the user using number buttons such as 1, 2, 3, 4, and 5 included in the key button of the mobile phone.

The function input unit 310 receives a function to be performed by the user according to the combination of the key input input in step 1300 and the movement pattern input in step 1310 (step 1320). In operation 1320, after the display unit 260 displays the functions selectable by the user, the function input unit 310 may receive one of the displayed functions from the user. In operation 1320, the setting apparatus does not include a separate function input unit 310, and receives a function to be performed by a user through keys included in the key input unit 205.

The setting unit 320 associates the input function with the combination of the input key input and the movement pattern and stores the input function in the memory 240 (step 1330).

Referring to another method of setting a user to perform a specific function by combining a key input and a movement with reference to the flowchart shown in FIG. 14, the key input unit 205 may be configured to perform a specific character or function from the user. Receive a key input (step 1400). The sensor unit 210 detects an arbitrary movement of the user for performing a specific character or function, and outputs a motion signal corresponding to the detected movement in operation 1410. It is preferable that the arbitrary movement of the user is not limited in the trajectory, the direction, or the size of the movement.

The function input unit 310 receives a function to be performed by the user according to a combination of the key input input in step 1400 and the movement detected in step 1410 (step 1420). The setting unit 320 associates the input function with a combination of the input key input and the detected movement and stores the input function in the memory 240 (step 1430). In step 1420, it is preferable to repeat the movement to be input to the user two or more times, and store the plurality of detected motion signals or overlapping features of the plurality of motion signals in the memory 240.

When a function corresponding to a combination of a key input and an arbitrary movement is set by a user using the method illustrated in FIG. 14, it is preferable to use a template matching method for pattern recognition.

FIG. 4 illustrates a first embodiment of a method of inputting a character by mapping a specific character to a combination of a key input and a movement of the user, wherein the user's movement pattern includes three kinds of movements: leftward movement, stop, and rightward movement. Preset with a pattern.

If the user grabs the portable device with the sensor with his hand and presses two buttons 400 among the illustrated character input buttons and moves to the left as shown in FIG. 4 (b), A of the displayed alphabets is input. . Further, if the user stops the portable device as shown in Fig. 4 (c) while pressing the 2 button 400, B of the displayed alphabets is input, and the portable button is pressed while the 2 button 400 is pressed. C is input when the device is moved to the right as shown in Fig. 4 (d). As described above, it is preferable to recognize the movement pattern in consideration of the movement of the user while the user presses the button.

FIG. 5 illustrates a second embodiment of a character input method using a user's key input and movement. FIG. 5 (a) shows a user's key input and FIG. 5 (b) shows a user's movement pattern. It is shown. When the user wants to input the SUM, the signal generator 230 moves the hand holding the portable device to the right while pressing the 7 button 500 and combines the key input and the movement pattern to signal S. And stop the portable device for a predetermined time while pressing the 8 button 510 to generate a signal corresponding to U by the combination, and the hand holding the portable device while pressing the 6 button 520. Move left to generate a signal corresponding to M, so that the user can enter the character SUM.

FIG. 6 illustrates an embodiment of a method of inputting a character through a key input and a movement of a user by using a celestial key input unit, in which case a character of Korea is to be input.

When the user moves the portable device to the right while pressing the button A, the input ㅎ is input, and when the user presses the button 610 to move the portable device to the right, a letter is displayed on the display unit 260. If the user stops the portable device for more than a predetermined time while pressing the button 620, the letter Han is displayed on the display unit 250.

Then, when the user stops the portable device for more than a predetermined time while pressing the button a 630, the letter a is input, and when the user moves the portable device to the right while pressing the button 640, the phrase “pure” is displayed on the display unit ( 260, and when the user stops the portable device for more than a predetermined time while pressing the a button 630, the letter “station” is displayed on the display unit 260. By the above-described sequence of key input and movement, the user can input the character of Korea in a portable device such as a mobile phone.

FIG. 7 is a table showing an embodiment of a method of mapping a combination of a user's key input and a movement pattern to a specific function of a mobile phone. When the network button and the user's movement B pattern are input, the wireless corresponding to the combination is input. It connects to the network and connects to the ring setting service. In addition, when a network button and a user's movement M pattern are input, a function of connecting to a wireless network corresponding to the combination is connected to a mail service.

When the menu button and the user's motion B pattern are input, the ring tone setting function corresponding to the combination is performed. When the menu button and the user's motion M pattern are input, the mobile phone text message input function corresponding to the combination is performed.

8 is a graph illustrating output signals of an inertial sensor according to left and right movements and a stop state of the user. FIG. 8 (a) shows left movement of the user, and FIG. 8 (b) shows a stationary state of the user. 8 (c) shows the output signals of the angular velocity sensor and the acceleration sensor according to the user's rightward movement. In each graph, three angular velocity sensor output signals and three acceleration sensor output signals are displayed, and the sensor output signals for two of the movements corresponding to the above-described movement pattern are displayed. As shown, the leftward movement, the stationary state and the rightward movement may be distinguished from each other according to the characteristics of the motion signal which is the sensor output.

FIG. 9 illustrates an embodiment of a user's motion trajectory, and output signals of the inertial sensor for the motion trajectories 1 to 5 of the numerical user's motion trajectories shown in FIG. 9 are shown in FIG. 10. .

Hereinafter, a method of recognizing the movement pattern from the detected movement signal of the user by the pattern recognition unit 220 will be described in detail.

The general procedure of applying the pattern recognition method is as follows.

First, we collect a lot of data about {input X, class C} from the user. Second, the collected data is divided into training data and test data. Third, the learning data is presented to the pattern recognition system to perform the learning. In this case, the model parameters of the pattern recognition system are changed to match the data. Finally, only the input X is presented to the system, which causes the system to export the class as output.

FIG. 11 is a block diagram illustrating a configuration of a pattern recognition unit 220 that recognizes a motion pattern from a motion signal using an artificial neural network. The illustrated pattern recognition unit preferably recognizes a user's movement pattern among predetermined movement patterns using the artificial neural network 1100. The artificial neural network is a model that simplifies and mathematically interprets a neurotransmission process of a living organism, and solves a problem through a kind of learning process that controls the connection weight between neurons while passing complicated intertwined nerve cells. In the analysis, the above process is similar to the process of learning and remembering a person, through which reasoning, classification, and prediction can be performed. In the artificial neural network, nerve cells correspond to nodes, and connection strength between nerve cells corresponds to weights connected to arcs between nodes. Preferably, the artificial neural network is a multilayer perceptron neural network composed of a plurality of single layer perceptrons, and the neural network is preferably learned by a back propagation learning method.

The backward propagation learning method is a model which extends the Weidrow-Hoff learning rule, which is one of neural network models, to a multilayer network and a nonlinear differential transition function, and is widely used for character recognition and nonlinear prediction. Each node of the neural network uses one of several differentiable transition functions to generate their output, most often using a log- sigmoid function such as Equation 1 below.

The function produces an output between 0 and 1 depending on the input value coming from the negative infinity to the positive infinity range. When the desired function is learned, the learning is performed by reducing the deviation between the desired output value and the actual output value using a backpropagation algorithm.

When a signal output from each sensor is processed and input to an input layer node of the artificial neural network, the signal is converted at each node and transmitted to an intermediate layer, and finally outputs a movement pattern at the output layer. By comparing the activation values output from the neural network with the activation values defined for each pattern during learning, the connection strength between nodes is adjusted in the direction of reducing the difference, and the back layer is propagated from the upper layer according to the delta learning rule. On the basis of this, the connection strength of the magnetic layer is adjusted again to minimize the error. The delta learning rule adjusts the connection strength so that the input / output function minimizes the sum of the squares of the error between the output obtained from all input patterns and the target output for a network of nonlinear neurons.

When the artificial neural network 1100 learning for all the movement patterns preset by the learning process is completed, the artificial neural network 1100 receives a user's motion signal detected from the sensor unit 210 and is preset. The pattern may be recognized as one of predetermined movement patterns.

The artificial neural network 1100 may be re-learned by the user, if desired by the user. For example, when the user selects a movement pattern to relearn the artificial neural network and performs a plurality of movements corresponding to the movement pattern, the re-learning of the artificial neural network 1100 to reflect the performed movement is performed. It can be done.

As a second embodiment of the method for recognizing the pattern, a movement pattern of a user may be recognized by using a support vector machine (SVM). According to this method, after forming the N-dimensional space from the characteristics of the N motion signals and finding an appropriate hyperplane by the training data, the patterns can be classified by the hyperplane, and each of the patterns It can be defined by Equation 2.

In Equation 2, W is a weight matrix, X is an input vector, and b is an offset.

As a third embodiment of the method for recognizing the pattern, a movement pattern may be recognized using template matching. According to this method, after generating the template data in which the pattern is classified from the training data, the pattern data can be classified by finding the template data closest in distance to the currently input input. In other words, for the input data X = P (x ₁ , ..., x _n ) and the i-th data Y _i = P (y ₁ , .. y _n ) of the training data, Y ^* It can be defined together.

In Equation 3, Distance (X, Y) can be calculated by calculating as shown in Equation 4.

According to Equations 3 and 4, the input X is classified into a pattern to which the data Y ^* belongs.

As a fourth embodiment of the method for recognizing the pattern, a hidden Markov model can be used. A hidden Markov model is a collection of states connected by transitions, with each transition representing a set of two probabilities. One is the transition probability required to make the transition, and the other is the output probability which represents the conditional probability of generating each output symbol belonging to a finite alphabet each time a transition occurs. The hidden Markov model evaluates the similarity between the input pattern and the reference pattern, so it can generate several candidates. Since the spatiotemporal variation is stochastic in each state and transition, No further consideration is needed.

In addition to the pattern recognition algorithm described above, various pattern recognition algorithms can be applied to the present invention.

The invention can also be embodied as computer readable code on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disks, optical data storage devices, and the like, which are also implemented in the form of carrier waves (for example, transmission over the Internet). Include. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion. And functional programs, codes and code segments for implementing the present invention can be easily inferred by programmers in the art to which the present invention belongs.

Although the preferred embodiments of the present invention have been described in detail above, those skilled in the art to which the present invention pertains should make the present invention without departing from the spirit and scope of the present invention as defined in the appended claims. It will be appreciated that various modifications or changes can be made. Accordingly, modifications to future embodiments of the present invention will not depart from the technology of the present invention.

According to the character input method and apparatus using the user key input and the movement according to the present invention as described above, by inputting the character by combining the user's key input and the user's movement, the character input speed can be improved By using a limited character input button, more than a number of characters or functions that can be combined may be input, thereby providing convenience to the user.

Claims (27)

In the method for performing a function performed in a terminal having one or more keys, Receiving a key value by a user operating the key; Sensing the position of the terminal integrated with the user and movable using a sensor provided in the terminal for a predetermined time; Recognizing the temporal trajectory of the detected position as a movement pattern of the terminal; And And performing a function corresponding to the input key value and the recognized movement pattern among one or more functions that can be performed in the terminal. The method of claim 1, wherein the performing of the step And receiving a character corresponding to the input key value and the recognized movement pattern. The method of claim 2, wherein the function performing method is And displaying the input characters. The method of claim 1, wherein the recognizing step Method of performing a function characterized in that it is performed using any one of artificial neural network, template matching, hidden Markov model, SVM (Support Vector Machine). The method of claim 1, wherein the performing of the step Reading function information matched to the input key value and the recognized movement pattern, from among the function information matched and stored for each possible combination of possible key values and possible movement patterns; And And performing a function shown in the read function information. delete The method of claim 1, wherein the detecting step Method for performing a function, characterized in that performed using at least one of the angular velocity sensor and the acceleration sensor. The method of claim 1, wherein the sensing is performed while the key is operated. The method of claim 1, wherein the detecting of the key is performed during the predetermined time when the manipulation of the key is started. delete The method of claim 1, wherein the recognizing step Reading a trajectory closest to a trajectory in time of the detected position among a plurality of previously stored trajectories; And And recognizing the read trajectory as a movement pattern of the terminal. The method of claim 1, wherein the recognizing step Learning characteristics common to a given plurality of temporal trajectories, and generating and storing a model using the learned results; Calculating a similarity between each of the plurality of stored models and the temporal trajectory of the sensed position; Reading a model having a maximum similarity among the calculated similarities among a plurality of the stored models; And And recognizing the read model as a movement pattern of the terminal. The method of claim 1, wherein the movement pattern is A method for performing a function comprising a left movement, a right movement and a stop state. In the function performing device for a terminal, A key input unit having one or more keys and generating a key input signal corresponding to a key value of the operated key; A sensor unit for sensing a position of the terminal for a predetermined time and generating a motion signal corresponding to a time trajectory of the detected position; A pattern recognizing unit recognizing a temporal trajectory of the detected position as a movement pattern of the terminal; And a function execution unit configured to perform a function corresponding to the key input signal and the recognized movement pattern among one or more functions that can be performed by the terminal. The method of claim 14, wherein the function performing unit And inputting characters corresponding to the key input signal and the recognized movement pattern into the terminal. The apparatus of claim 14, wherein the function performing apparatus is A memory for matching and storing function information for each possible combination of possible key values and possible movement patterns, And the function execution unit reads the function information matched to the key input signal and the recognized movement pattern, and performs a function indicated in the read function information. delete The method of claim 14, wherein the sensor unit And at least one of an angular velocity sensor and an acceleration sensor. The apparatus of claim 14, wherein the sensor unit is operated while the key input unit is in operation. The apparatus of claim 14, wherein the sensor unit operates after the operation of the key input unit is completed. The method of claim 14, wherein the pattern recognition unit A function performing apparatus comprising: reading a trajectory closest to a time trajectory of the detected position among a plurality of previously stored trajectories, and recognizing the read trajectory as a movement pattern of the terminal. delete The method of claim 14, wherein the pattern recognition unit A feature extractor configured to extract a feature of the temporal trajectory of the detected position by comparing the temporal trajectory of the detected position with a preset trajectory; And And a pattern selector configured to read a trace matched to the extracted feature among a plurality of traces matched for each feature and to recognize the read trace as a movement pattern of the terminal. The method of claim 14, wherein the pattern recognition unit Recognizing a movement pattern of a predetermined number of movement patterns from the movement signal of the user using any one of artificial neural network, template matching, hidden Markov model, and support vector machine (SVM) Function performing device. The method of claim 24, The artificial neural network, template matching, hidden Markov model, SVM can be learned by a user. The method of claim 14, wherein the movement pattern is Character function performing apparatus comprising a left movement, a right movement and a stop state. A computer-readable recording medium having recorded thereon a program for executing the method according to any one of claims 1 to 13.

Images