WO2022061499A1 - 一种基于振动信号的身份验证方法和系统 - Google Patents

一种基于振动信号的身份验证方法和系统 Download PDF

Info

Publication number
WO2022061499A1
WO2022061499A1 PCT/CN2020/116763 CN2020116763W WO2022061499A1 WO 2022061499 A1 WO2022061499 A1 WO 2022061499A1 CN 2020116763 W CN2020116763 W CN 2020116763W WO 2022061499 A1 WO2022061499 A1 WO 2022061499A1
Authority
WO
WIPO (PCT)
Prior art keywords
vibration signal
vibration
effective part
signal
mel
Prior art date
Application number
PCT/CN2020/116763
Other languages
English (en)
French (fr)
Inventor
伍楷舜
关茂柠
Original Assignee
深圳大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳大学 filed Critical 深圳大学
Priority to PCT/CN2020/116763 priority Critical patent/WO2022061499A1/zh
Publication of WO2022061499A1 publication Critical patent/WO2022061499A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
    • G06F21/31—User authentication

Definitions

  • the present invention relates to the technical field of identity verification, and more particularly, to an identity verification method and system based on vibration signals.
  • the existing verification methods of smart terminals usually include fingerprint recognition and voice recognition.
  • Fingerprint recognition uses the contact sensor to extract fingerprint features for comparison.
  • the user's finger is dipped in water or oil, it is difficult to identify effectively, and the user's fingerprint belongs to the surface information of the finger and is easily forged.
  • the purpose of the present invention is to overcome the above-mentioned defects of the prior art, and to provide a vibration signal-based identity verification method and system, which is a new technology solution for triple identity verification, which can improve the security problems existing in the existing smart device identity verification. .
  • an authentication method based on a vibration signal includes the following steps:
  • the Mel cepstral coefficients are input into the trained Hidden Markov Model to obtain the classification result of the vibration signal type;
  • identity verification is performed.
  • an identity verification system based on a vibration signal includes: a self-defining unit, which is used for the user to define the association relationship between each digital key of the smart device and the type of vibration signal, and to define the PIN code; the signal acquisition unit, which is used to collect the user's input of the PIN code.
  • the generated vibration signal ; a signal processing unit, which is used to process the collected vibration signal and extract an effective part of the vibration signal; a feature extraction unit, which is used to extract the Mel cepstral coefficient of the effective part of the vibration signal; A classification and identification unit, which is used for inputting the Mel cepstral coefficients into a trained Hidden Markov Model to obtain a classification result of the vibration signal type; an identity verification unit, which is used for setting according to the classification result and in combination with The association between the number keys and vibration signal types for authentication.
  • the advantage of the present invention is that the present invention allows the user to customize the vibration signal type corresponding to each digital key, and the different vibration signals generated by inputting the PIN code are used for user authentication.
  • the three-factor authentication method makes it difficult for illegal users to forge, and significantly improves the security and effectiveness of authentication.
  • FIG. 1 is a flowchart of an authentication method based on a vibration signal according to an embodiment of the present invention
  • FIG. 2 is a schematic process diagram of an authentication method based on a vibration signal according to an embodiment of the present invention.
  • the vibration signal-based authentication method provided by the embodiment of the present invention includes:
  • Step S110 the user defines the vibration signal type corresponding to each digital key.
  • Smart devices include but are not limited to notebook computers, desktop computers, access control systems or other terminal devices with certain computing capabilities.
  • the user-defined vibration signal types corresponding to each digital key include:
  • Step S101 the smart phone can input 10 numbers (0, 1, 2, 3, .
  • vibration signal s0 10 vibration signal types are marked as vibration signal s0, vibration signal s1, vibration signal s2, . . . , vibration signal s9.
  • step S102 different types of vibration signals are distinguished according to vibration amplitude and vibration frequency.
  • vibration amplitudes or vibration frequencies are different, as long as the types can be distinguished.
  • Step S103 the user sets an association relationship between each number and the vibration signal type.
  • the user can set the number "0" to correspond to the vibration signal s0, the number “1” to the vibration signal s1, the number “2” to correspond to the vibration signal s2, ..., the number “9” to the vibration signal according to their own wishes Signal s9.
  • Step S120 the user defines a PIN code and inputs the PIN code to generate a vibration signal.
  • Users can customize a multi-digit PIN code, for example, set a 4-8-digit PIN code for subsequent authentication.
  • Step S130 collecting the vibration signal generated by inputting the PIN code.
  • a corresponding vibration signal can be generated.
  • the vibration signal sensor connected to the Raspberry Pi to collect vibration signals, including:
  • step S301 the user holds the smart phone with the left hand, and inputs the PIN code on the phone with the right hand to generate a vibration signal.
  • Step S302 the vibration signal is transmitted to the vibration sensor through the back of the user's hand.
  • Step S303 the Raspberry Pi uses a vibration sensor to collect vibration signals generated by the smartphone.
  • the vibration sensor may be a piezoelectric thin-film vibration sensor, a piezoelectric ceramic vibration sensor, or other sensors capable of detecting vibration signals.
  • step S140 an energy-based double-threshold endpoint detection method is used to detect the effective part of the vibration signal.
  • using the energy-based dual-threshold endpoint detection method to detect the effective part of the vibration signal specifically includes:
  • step S401 after the Raspberry Pi collects the vibration signal, it uses a Butterworth bandpass filter to filter it, and the cutoff frequencies are set to 10 Hz and 1000 Hz respectively.
  • Step S402 calculating the short-term energy of the vibration signal.
  • the short-term energy is calculated as follows:
  • E is the short-term energy of the frame signal
  • L is the length of the frame signal
  • S(i) is the amplitude of the vibration signal
  • t is the time index of the frame number.
  • Step S404 set two parameters according to experience: the maximum interval maxInter between the signal peaks of the same signal and the minimum length minLen of the signal;
  • the maximum interval threshold maxInter and the minimum length threshold minLen between signal peaks can be set according to experience, which is not limited in the present invention. For example, set maxInter to 50 (frames) and minLen to 30 (frames).
  • Step S405 find a frame of signal with the highest energy in the signal and the energy of the frame signal needs to be higher than the set high threshold.
  • Step S406 extend from the frame signal to the left and right respectively, until the energy of the next frame signal is lower than the set low threshold, record the frame position at this time, it can be obtained that the frame position on the left is the starting point of the signal peak, The frame position on the right is the end point of the signal peak, and the frame energy at the position of the signal peak in the signal is set to zero.
  • Step S407 repeating S405 and S406 until all signal peaks in the entire signal are found.
  • Step S408 if the interval between the two signal peaks is less than maxInter, then combine the two signal peaks.
  • Step S409 repeat S408 until the interval between all signal peaks is greater than maxInter.
  • Step S410 if the length of the signal peak is less than minLen, the signal peak is directly discarded.
  • step S411 the number of signal peaks finally obtained should be 1, and the signal peak is an effective part of the signal.
  • step S412 if the number of signal peaks obtained in S411 is greater than 1, the signal should be regarded as an invalid signal and be directly discarded.
  • Step S150 extracting the Mel cepstral coefficients of the effective part of the vibration signal.
  • extracting the Mel cepstral coefficient of the vibration signal as a feature specifically includes:
  • step S501 pre-emphasis, framing and windowing are performed on the effective part of the extracted vibration signal.
  • the pre-emphasis coefficient is 0.96
  • the frame length is 20ms
  • the frame shift is 6ms
  • the window function is Hamming window.
  • Step S502 performing Fast Fourier Transform (FFT) on each frame of signal to obtain a corresponding frequency spectrum.
  • FFT Fast Fourier Transform
  • Step S503 passing the obtained spectrum through a Mel filter bank to obtain a Mel spectrum.
  • the mel filter frequency range is 10Hz to 1000Hz, and the number of filter channels is 28.
  • Step S504 taking the logarithm of the obtained Mel spectrum, then performing discrete cosine transform (DCT), and finally taking the first 14 coefficients as Mel cepstral coefficients (MFCCs).
  • DCT discrete cosine transform
  • MFCCs Mel cepstral coefficients
  • the extracted Mel-frequency cepstral coefficients are not limited to 14, and an appropriate number of Mel-frequency cepstral coefficients can be extracted according to the requirements of the accuracy and execution speed of the training model.
  • the existing technologies such as pre-emphasis, framing, windowing, and Fourier transform are not introduced in detail in this paper.
  • Step S160 using the extracted Mel cepstral coefficients as the observation sequence to train a hidden Markov model.
  • a training sample can be constructed to represent the corresponding relationship between the Mel cepstral coefficient and the vibration signal type, that is, the input feature of the sample is the Mel cepstral coefficient, and the sample label is the vibration signal type.
  • the Baum-Welch algorithm is used to train the hidden Markov model using the training samples, and the extracted Mel cepstral coefficients are used as the observation sequence, wherein the number of states of the hidden Markov model is 3, and each state has Two mixed Gaussian probability density functions
  • the training process includes: initializing the parameters of the hidden Markov model; calculating the forward and backward probability matrices; calculating the transition probability matrix; calculating the mean and variance of each Gaussian probability density function; calculating each Gaussian The weight of the probability density function; calculates the output probabilities of all observation sequences, and accumulates them to obtain the summed output probability.
  • the number of iterations of training can be set according to computing resources and model accuracy. For example, considering that the computing resources of the intelligent terminal are limited, the training process can be iterated only once. In addition, one hidden Markov model can be trained for each vibration signal type.
  • a Hidden Markov Model is used to classify and identify the test data to assess the effectiveness of the model.
  • the classification and identification process specifically includes: using the Viterbi algorithm to calculate the output probability of the test data (test sample) for each hidden Markov model, and giving the best state path; the category corresponding to the hidden Markov model with the largest output probability That is, the classification result of the test data.
  • the hidden Markov model can be retrained by adjusting or enriching the training samples.
  • Step S170 using the trained Hidden Markov Model to identify the type of vibration signal generated by the user to be detected inputting the PIN code, and performing identity verification based on the association between the set number keys and the type of vibration signal.
  • the trained Hidden Markov Model determines its hidden parameters.
  • the vibration signal generated by the user inputting the PIN code is extracted in real time and the corresponding Mel cepstral coefficients are extracted, and then the trained hidden Markov model is input to obtain the vibration signal type, and then According to the pre-stored association between the digital keys and the vibration signal type, the PIN input by the user is identified, thereby realizing the user's triple authentication.
  • the present invention uses the vibration signal generated by the smart terminal to input the PIN code to perform identity verification, and realizes triple identity verification. Different correspondences can be set. This is the first authentication method. Different vibration signals generated by inputting the PIN code are allowed to be used for user authentication, so different users can set different PIN codes. This is the second authentication method.
  • the generated vibration signal is transmitted to the vibration sensor of the smart device through the back of the user's hand, so the vibration signal collected by the smart device contains the biometric features corresponding to the back of the user's hand, so different users even enter the PIN code to generate The same vibration signal, but the vibration signal is transmitted to the smart device through the back of the hand of different users, so the vibration signal collected by the smart device at this time is different because it contains the biometric features of the back of the hand of different users.
  • This is the third step.
  • Authentication method This triple identity authentication method improves the security and effectiveness of smart device authentication.
  • the present invention also provides an identity verification system based on a vibration signal, which is used to implement one or more aspects of the above method.
  • the system includes: a self-defining unit, which is used for the user to customize the relationship between each digital key of the smart device and the type of vibration signal, and the custom PIN code; the signal acquisition unit, which is used to collect the user's input PIN Vibration signal generated during coding; signal processing unit, which is used to process the collected vibration signal and extract the effective part of the vibration signal; feature extraction unit, which is used to extract the Mel cepstrum of the effective part of the vibration signal A classification and identification unit, which is used to input the Mel cepstral coefficients into a trained Hidden Markov Model to obtain a classification result of the vibration signal type; an identity verification unit, which is used to combine the classification results according to the classification results.
  • the association between the set number keys and the vibration signal type is used for authentication.
  • the present invention may be a system, method and/or computer program product.
  • the computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of the present invention.
  • a computer-readable storage medium may be a tangible device that can hold and store instructions for use by the instruction execution device.
  • the computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing.
  • Non-exhaustive list of computer readable storage media include: portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM) or flash memory), static random access memory (SRAM), portable compact disk read only memory (CD-ROM), digital versatile disk (DVD), memory sticks, floppy disks, mechanically coded devices, such as printers with instructions stored thereon Hole cards or raised structures in grooves, and any suitable combination of the above.
  • RAM random access memory
  • ROM read only memory
  • EPROM erasable programmable read only memory
  • flash memory static random access memory
  • SRAM static random access memory
  • CD-ROM compact disk read only memory
  • DVD digital versatile disk
  • memory sticks floppy disks
  • mechanically coded devices such as printers with instructions stored thereon Hole cards or raised structures in grooves, and any suitable combination of the above.
  • Computer-readable storage media are not to be construed as transient signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (eg, light pulses through fiber optic cables), or through electrical wires transmitted electrical signals.
  • the computer readable program instructions described herein may be downloaded to various computing/processing devices from a computer readable storage medium, or to an external computer or external storage device over a network such as the Internet, a local area network, a wide area network, and/or a wireless network.
  • the network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and/or edge servers.
  • a network adapter card or network interface in each computing/processing device receives computer-readable program instructions from a network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in each computing/processing device .
  • the computer program instructions for carrying out the operations of the present invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state setting data, or instructions in one or more programming languages.
  • Source or object code written in any combination, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as the "C" language or similar programming languages.
  • the computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server implement.
  • the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (eg, using an Internet service provider through the Internet connect).
  • LAN local area network
  • WAN wide area network
  • custom electronic circuits such as programmable logic circuits, field programmable gate arrays (FPGAs), or programmable logic arrays (PLAs)
  • FPGAs field programmable gate arrays
  • PDAs programmable logic arrays
  • Computer readable program instructions are executed to implement various aspects of the present invention.
  • These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer or other programmable data processing apparatus to produce a machine that causes the instructions when executed by the processor of the computer or other programmable data processing apparatus , resulting in means for implementing the functions/acts specified in one or more blocks of the flowchart and/or block diagrams.
  • These computer readable program instructions can also be stored in a computer readable storage medium, these instructions cause a computer, programmable data processing apparatus and/or other equipment to operate in a specific manner, so that the computer readable medium on which the instructions are stored includes An article of manufacture comprising instructions for implementing various aspects of the functions/acts specified in one or more blocks of the flowchart and/or block diagrams.
  • Computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other equipment to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other equipment to produce a computer-implemented process , thereby causing instructions executing on a computer, other programmable data processing apparatus, or other device to implement the functions/acts specified in one or more blocks of the flowcharts and/or block diagrams.
  • each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more functions for implementing the specified logical function(s) executable instructions.
  • the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.
  • each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations can be implemented in dedicated hardware-based systems that perform the specified functions or actions , or can be implemented in a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that implementation in hardware, implementation in software, and implementation in a combination of software and hardware are all equivalent.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Security & Cryptography (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

本发明公开了一种基于振动信号的身份验证方法和系统。该方法包括：用户自定义智能设备的每个数字按键和振动信号类型之间的关联关系，并自定义PIN码；采集用户输入PIN码时产生的振动信号；对所采集的振动信号进行处理并提取振动信号的有效部分；提取所述振动信号的有效部分的梅尔倒谱系数；将所述梅尔倒谱系数输入经训练的隐马尔可夫模型，获得振动信号类型的分类结果；根据所述分类结果，并结合设定的数字按键和振动信号类型之间的关联关系，进行身份验证。本发明能够提高智能设备身份验证的安全性和有效性。

Description

一种基于振动信号的身份验证方法和系统 技术领域
本发明涉及身份验证技术领域，更具体地，涉及一种基于振动信号的身份验证方法和系统。
背景技术
随着网络的普及，越来越多的智能终端进入市场。现有的智能终端的验证方式通常包括指纹识别和语音识别等。
指纹识别通过接触性传感器提取指纹特征进行比对，在用户的手指蘸有水渍或者油渍的时候，难以进行有效识别，并且用户指纹属于手指表面信息，很容易被伪造。
目前，语音识别技术在强噪声干扰的情况下，还很难达到实用化要求。在自然发音、噪声、口音等复杂条件下，语音识别的准确率明显下降，而且用户的声音可能会被非法录制，导致重放攻击的发生。
发明内容
本发明的目的是克服上述现有技术的缺陷，提供一种基于振动信号的身份验证方法和系统，其是三重身份验证的新技术方案，能够改善现有智能设备身份验证中存在的安全性问题。
根据本发明的第一方面，提供了一种基于振动信号的身份验证方法。该方法包括以下步骤：
用户自定义智能设备的每个数字按键和振动信号类型之间的关联关系，并自定义PIN码；
采集用户输入PIN码时产生的振动信号；
对所采集的振动信号进行处理并提取振动信号的有效部分；
提取所述振动信号的有效部分的梅尔倒谱系数；
将所述梅尔倒谱系数输入经训练的隐马尔可夫模型，获得振动信号类型的分类结果；
根据所述分类结果，并结合设定的数字按键和振动信号类型之间的关联关系，进行身份验证。
根据本发明的第二方面，提供一种基于振动信号的身份验证系统。该系统包括：自定义单元，其用于用户自定义智能设备的每个数字按键和振动信号类型之间的关联关系，并自定义PIN码；信号采集单元，其用于采集用户输入PIN码时产生的振动信号；信号处理单元，其用于对所采集的振动信号进行处理并提取振动信号的有效部分；特征提取单元，其用于提取所述振动信号的有效部分的梅尔倒谱系数；分类识别单元，其用于将所述梅尔倒谱系数输入经训练的隐马尔可夫模型，获得振动信号类型的分类结果；身份验证单元，其用于根据所述分类结果，并结合设定的数字按键和振动信号类型之间的关联关系，进行身份验证。
与现有技术相比，本发明的优点在于，本发明通过允许用户自定义每一个数字按键对应的振动信号类型、经输入PIN码所产生的不同振动信号用于进行用户的身份验证，这种三重身份验证方式使非法用户难以伪造，显著提高了身份验证的安全性和有效性。
通过以下参照附图对本发明的示例性实施例的详细描述，本发明的其它特征及其优点将会变得清楚。
附图说明
被结合在说明书中并构成说明书的一部分的附图示出了本发明的实施例，并且连同其说明一起用于解释本发明的原理。
图1是根据本发明一个实施例的基于振动信号的身份验证方法的流程图；
图2是根据本发明一个实施例的基于振动信号的身份验证方法的过程示意。
具体实施方式
现在将参照附图来详细描述本发明的各种示例性实施例。应注意到：除非另外具体说明，否则在这些实施例中阐述的部件和步骤的相对布置、数字表达式和数值不限制本发明的范围。
以下对至少一个示例性实施例的描述实际上仅仅是说明性的，决不作为对本发明及其应用或使用的任何限制。
对于相关领域普通技术人员已知的技术、方法和设备可能不作详细讨论，但在适当情况下，所述技术、方法和设备应当被视为说明书的一部分。
在这里示出和讨论的所有例子中，任何具体值应被解释为仅仅是示例性的，而不是作为限制。因此，示例性实施例的其它例子可以具有不同的值。
应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一旦某一项在一个附图中被定义，则在随后的附图中不需要对其进行进一步讨论。
结合图1和图2所示，本发明实施例提供的基于振动信号的身份验证方法包括：
步骤S110，用户自定义每个数字按键对应的振动信号类型。
在此步骤中，用户可定义智能终端上各数字按键对应的振动信号类型。智能设备包括但不限于笔记本电脑、台式电脑、门禁系统或其他具有一定计算能力的终端设备。
例如，以智能手机为例，用户自定义每个数字按键对应的振动信号类型包括：
步骤S101，智能手机可以通过手机键盘输入10个数字(0，1，2，3，...，9)，同时在智能手机中存储10种类型的振动信号
例如，10种振动信号类型分别标记为振动信号s0，振动信号s1，振动信号s2，...，振动信号s9。
步骤S102，对于不同类型的振动信号，根据振动幅度和振动频率进行区分。
例如，对应于不同类型的振动信号，至少振动幅度或振动频率不同，只要能够进行类型区分即可。
步骤S103，用户设置每个数字与振动信号类型之间的关联关系。
例如，用户可以根据自己的意愿设置数字“0”对应于振动信号s0，数字“1”对应于振动信号s1，数字“2”对应于振动信号s2，...，数字“9”对应于振动信号s9。
步骤S120，用户自定义PIN码并输入PIN码来产生振动信号。
用户可自定义多位PIN码，例如，设置4-8位PIN码用于后续身份验证。
步骤S130，采集输入PIN码所产生的振动信号。
用户输入PIN码时，可以产生对应的振动信号。例如，使用树莓派连接的振动信号传感器采集振动信号，具体包括：
步骤S301，用户左手拿着智能手机，右手在手机上输入PIN码来产生振动信号。
步骤S302，振动信号通过用户的手背传到振动传感器。
步骤S303，树莓派采用振动传感器采集智能手机产生的振动信号。
在本发明实施例中，振动传感器可以是压电薄膜振动传感器、压电陶瓷振动传感器或其他能检测振动信号的传感器。
步骤S140，用基于能量的双门限端点检测法来检测振动信号的有效部分。
在一个实施例中，用基于能量的双门限端点检测法来检测振动信号的有效部分具体包括：
步骤S401，树莓派采集到振动信号之后，使用巴特沃斯带通滤波器对其进行滤波，截止频率分别设置为10Hz和1000Hz。
步骤S402，计算出振动信号的短时能量。
例如，短时能量的计算公式如下：
Figure PCTCN2020116763-appb-000001
其中，E是帧信号的短时能量，L是帧信号的长度，S(i)是振动信号的幅度，t是帧号的时间索引。
步骤S403，计算噪声的平均能量，记为u，计算信号能量的标准差， 记为σ；设置切断时的低门限为TL＝u+σ，高门限为TH＝u+3σ。
步骤S404，根据经验设置两个参数：同一个信号的信号峰之间的最大间隔maxInter和信号的最小长度minLen；
对于同一个振动信号，可根据经验设置信号峰之间的最大间隔门限maxInter和最小长度门限minLen，本发明对此不进行限制。例如，将maxInter设置为50(帧)，将minLen设置为30(帧)。
步骤S405，找出信号中能量最大的一帧信号且该帧信号的能量需要高于所设置的高门限。
步骤S406，从该帧信号分别向左和向右延伸，直到下一帧信号的能量低于所设置的低门限，记录此时的帧位置，可以得到左边的帧位置为该信号峰的起点，右边的帧位置为该信号峰的终点，同时将信号中该信号峰所在位置的帧能量置为零。
步骤S407，重复S405和S406，直到找出整段信号中的所有信号峰。
步骤S408，若两个信号峰的间隔小于maxInter，则合并两个信号峰。
步骤S409，重复S408，直到所有信号峰之间的间隔都大于maxInter。
步骤S410，若信号峰的长度小于minLen，则直接舍弃该信号峰。
步骤S411，最后得到的信号峰的数量应该为1，且该信号峰即为信号的有效部分。
步骤S412，若S411得到的信号峰的数量大于1，则该信号应视为无效信号，直接舍弃。
需说明的是，本文的“左”、“右”反映的是时序方向，例如，“向左延伸”是指搜索帧信号的前序帧，而“向右延伸”指搜索帧信号的后序帧。
步骤S150，提取振动信号的有效部分的梅尔倒谱系数。
在一个实施例中，提取振动信号的梅尔倒谱系数作为特征，具体包括：
步骤S501，对提取到的振动信号的有效部分进行预加重、分帧和加窗。
例如，预加重的系数为0.96，帧长为20ms，帧移为6ms，窗函数为Hamming窗。
步骤S502，对每一帧信号进行快速傅里叶变换(FFT)得到对应的频谱。
步骤S503，将得到的频谱通过梅尔滤波器组得到梅尔频谱。
例如，梅尔滤波频率范围为10Hz到1000Hz，滤波器通道数为28。
步骤S504，对得到的梅尔频谱取对数，然后进行离散余弦变换(DCT)，最后取前14个系数作为梅尔倒谱系数(MFCCs)。
应理解的是，所提取的梅尔频率倒谱系数不限于14个，可根据训练模型的精确度和执行速度要求提取适当数量的梅尔频率倒谱系数。此外，本文对预加重、分帧、加窗、傅里叶变换等现有技术不作具体介绍。
步骤S160，以所提取的梅尔倒谱系数作为观察序列，训练隐马尔可夫模型。
根据上述步骤可构建训练样本，用于表征梅尔倒谱系数和振动信号类型之间的对应关系，即样本的输入特征是梅尔倒谱系数，样本标签是振动信号类型。
具体地，利用训练样本使用鲍姆-韦尔奇算法训练隐马尔可夫模型，以提取的梅尔倒谱系数作为观察序列，其中，隐马尔可夫模型的状态数为3，每个状态有2个混合高斯概率密度函数，训练过程包括：对隐马尔可夫模型的参数进行初始化；计算前、后向概率矩阵；计算转移概率矩阵；计算各个高斯概率密度函数的均值和方差；计算各个高斯概率密度函数的权重；计算所有观察序列的输出概率，并进行累加得到总和输出概率。
训练的迭代次数可根据计算资源和模型精度进行设置，例如考虑到智能终端的计算资源有限，所以该训练过程可仅迭代1次。此外，对应每种振动信号类型，可各训练一个隐马尔克夫模型。
优选地，使用隐马尔可夫模型来对测试数据进行分类识别，以评估模型的有效性。分类识别过程具体包括：利用维特比算法计算测试数据(测试样本)对于各个隐马尔可夫模型的输出概率，并给出最佳的状态路径；输出概率最大的隐马尔可夫模型所对应的类别即为该测试数据的分类结果。
进一步地，在分类结果不满足预定要求的情况下，可通过调整或丰富 训练样本重新训练隐马尔克夫模型。
步骤S170，利用经训练的隐马尔可夫模型识别待检测用户输入PIN码所产生的振动信号类型，并结合设定的数字按键和振动信号类型之间的关联关系，进行身份验证。
经训练的隐马尔克夫模型即确定了其中的隐含参数。在实际应用中，对于待验证身份的用户，实时提取用户输入PIN码产生的振动信号并提取对应的梅尔倒谱系数后，输入经训练的隐马尔克夫模型，即可获得振动信号类型，进而根据预先存储的数字按键和振动信号类型之间的关联关系，识别用户输入的PIN，从而实现用户三重身份验证。
综上所述，本发明利用智能终端输入PIN码所产生的振动信号进行身份验证，实现了三重身份验证，具体地，通过允许用户自定义每一个数字按键对应的振动信号类型，所以不同的用户可以设置不同的对应关系，此为第一重身份验证方式；允许输入PIN码所产生的不同振动信号用于进行用户的身份验证，所以不同的用户可以设置不同的PIN码，此为第二重身份验证方式；产生的振动信号是通过用户的手背来传输到智能设备的振动传感器的，所以智能设备收集到的振动信号包含了用户手背对应的生物识别特征，所以不同的用户即使输入PIN码产生一样的振动信号，但是该振动信号是经过不同用户的手背传输到智能设备的，所以此时智能设备收集到的振动信号因包含了不同用户的手背的生物识别特征而不同，此为第三重身份验证方式。这种三重身份认证方式，提高了智能设备身份验证的安全性和有效性。
相应地，本发明还提供一种基于振动信号的身份验证系统，用于实现上述方法的一个方面或多个方面。例如，该系统包括：自定义单元，其用于用户自定义智能设备的每个数字按键和振动信号类型之间的关联关系，并自定义PIN码；信号采集单元，其用于采集用户输入PIN码时产生的振动信号；信号处理单元，其用于对所采集的振动信号进行处理并提取振动信号的有效部分；特征提取单元，其用于提取所述振动信号的有效部分的梅尔倒谱系数；分类识别单元，其用于将所述梅尔倒谱系数输入经训练的隐马尔可夫模型，获得振动信号类型的分类结果；身份验证单元，其用于 根据所述分类结果，并结合设定的数字按键和振动信号类型之间的关联关系，进行身份验证。
本发明可以是系统、方法和/或计算机程序产品。计算机程序产品可以包括计算机可读存储介质，其上载有用于使处理器实现本发明的各个方面的计算机可读程序指令。
计算机可读存储介质可以是可以保持和存储由指令执行设备使用的指令的有形设备。计算机可读存储介质例如可以是――但不限于――电存储设备、磁存储设备、光存储设备、电磁存储设备、半导体存储设备或者上述的任意合适的组合。计算机可读存储介质的更具体的例子(非穷举的列表)包括：便携式计算机盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、静态随机存取存储器(SRAM)、便携式压缩盘只读存储器(CD-ROM)、数字多功能盘(DVD)、记忆棒、软盘、机械编码设备、例如其上存储有指令的打孔卡或凹槽内凸起结构、以及上述的任意合适的组合。这里所使用的计算机可读存储介质不被解释为瞬时信号本身，诸如无线电波或者其他自由传播的电磁波、通过波导或其他传输媒介传播的电磁波(例如，通过光纤电缆的光脉冲)、或者通过电线传输的电信号。
这里所描述的计算机可读程序指令可以从计算机可读存储介质下载到各个计算/处理设备，或者通过网络、例如因特网、局域网、广域网和/或无线网下载到外部计算机或外部存储设备。网络可以包括铜传输电缆、光纤传输、无线传输、路由器、防火墙、交换机、网关计算机和/或边缘服务器。每个计算/处理设备中的网络适配卡或者网络接口从网络接收计算机可读程序指令，并转发该计算机可读程序指令，以供存储在各个计算/处理设备中的计算机可读存储介质中。
用于执行本发明操作的计算机程序指令可以是汇编指令、指令集架构(ISA)指令、机器指令、机器相关指令、微代码、固件指令、状态设置数据、或者以一种或多种编程语言的任意组合编写的源代码或目标代码，所述编程语言包括面向对象的编程语言—诸如Smalltalk、C++等，以及常规的过程式编程语言—诸如“C”语言或类似的编程语言。计算机可读程序指令 可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络—包括局域网(LAN)或广域网(WAN)—连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。在一些实施例中，通过利用计算机可读程序指令的状态信息来个性化定制电子电路，例如可编程逻辑电路、现场可编程门阵列(FPGA)或可编程逻辑阵列(PLA)，该电子电路可以执行计算机可读程序指令，从而实现本发明的各个方面。
这里参照根据本发明实施例的方法、装置(系统)和计算机程序产品的流程图和/或框图描述了本发明的各个方面。应当理解，流程图和/或框图的每个方框以及流程图和/或框图中各方框的组合，都可以由计算机可读程序指令实现。
这些计算机可读程序指令可以提供给通用计算机、专用计算机或其它可编程数据处理装置的处理器，从而生产出一种机器，使得这些指令在通过计算机或其它可编程数据处理装置的处理器执行时，产生了实现流程图和/或框图中的一个或多个方框中规定的功能/动作的装置。也可以把这些计算机可读程序指令存储在计算机可读存储介质中，这些指令使得计算机、可编程数据处理装置和/或其他设备以特定方式工作，从而，存储有指令的计算机可读介质则包括一个制造品，其包括实现流程图和/或框图中的一个或多个方框中规定的功能/动作的各个方面的指令。
也可以把计算机可读程序指令加载到计算机、其它可编程数据处理装置、或其它设备上，使得在计算机、其它可编程数据处理装置或其它设备上执行一系列操作步骤，以产生计算机实现的过程，从而使得在计算机、其它可编程数据处理装置、或其它设备上执行的指令实现流程图和/或框图中的一个或多个方框中规定的功能/动作。
附图中的流程图和框图显示了根据本发明的多个实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段或指令的一部分，所述 模块、程序段或指令的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个连续的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或动作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。对于本领域技术人员来说公知的是，通过硬件方式实现、通过软件方式实现以及通过软件和硬件结合的方式实现都是等价的。
以上已经描述了本发明的各实施例，上述说明是示例性的，并非穷尽性的，并且也不限于所披露的各实施例。在不偏离所说明的各实施例的范围和精神的情况下，对于本技术领域的普通技术人员来说许多修改和变更都是显而易见的。本文中所用术语的选择，旨在最好地解释各实施例的原理、实际应用或对市场中的技术改进，或者使本技术领域的其它普通技术人员能理解本文披露的各实施例。本发明的范围由所附权利要求来限定。

Claims (10)

  1. 一种基于振动信号的身份验证方法，包括以下步骤：
    用户自定义智能设备的每个数字按键和振动信号类型之间的关联关系，并自定义PIN码；
    采集用户输入PIN码时产生的振动信号；
    对所采集的振动信号进行处理并提取振动信号的有效部分；
    提取所述振动信号的有效部分的梅尔倒谱系数；
    将所述梅尔倒谱系数输入经训练的隐马尔可夫模型，获得振动信号类型的分类结果；
    根据所述分类结果，并结合设定的数字按键和振动信号类型之间的关联关系，进行身份验证。
  2. 根据权利要求1所述的方法，其中，根据以下步骤设定数字按键和振动信号类型之间的关联关系：
    对于智能设备的多个数字按键，设定每个数字按键与振动信号类型之间的一一对应关系，其中对于不同类型的振动信号，至少振动幅度或振动频率不同。
  3. 根据权利要求1所述的方法，其中，所述智能设备包括笔记本电脑、台式电脑和门禁系统。
  4. 根据权利要求1所述的方法，其中，对所采集的振动信号进行处理并提取振动信号的有效部分包括：
    利用树莓派连接的振动信号传感器采集用户输入PIN码时产生的振动信号，并采用巴特沃斯带通滤波器进行滤波；
    计算振动信号的短时能量；
    用基于短时能量的双门限端点检测确定振动信号有效部分的起点和终点；
    根据获取的起点和终点对振动信号进行截取，获得振动信号的有效部分。
  5. 根据权利要求4所述的方法，其中，所述振动传感器包括压电薄膜振动传感器和压电陶瓷振动传感器。
  6. 根据权利要求1所述的方法，其中，提取所述振动信号的有效部分的梅尔倒谱系数包括：
    对获取的信号有效部分进行预加重、分帧和加窗；
    对每一个短时分析窗，通过短时傅里叶变换得到对应的频谱；
    将获得的频谱通过梅尔滤波器组得到梅尔频谱；
    对得到的梅尔频谱取对数，并进行离散余弦变换，进而选择前14个系数作为提取出的梅尔倒谱系数。
  7. 根据权利要求1所述的方法，其中，以所述梅尔倒谱系数作为观察序列，使用鲍姆-韦尔奇算法训练隐马尔可夫模型，并且对应于每种类型的振动信号各训练一个隐马尔克夫模型。
  8. 根据权利要求7所述的方法，其中，还包括根据以下步骤评估经训练的隐马尔可夫模型：
    利用维特比算法计算测试数据对于各个隐马尔可夫模型的输出概率，并给出最佳的状态路径；
    将输出概率最大的隐马尔可夫模型所对应的振动信号类型作为该测试数据的分类结果。
  9. 一种基于振动信号的身份验证系统，包括：
    自定义单元：用于用户自定义智能设备的每个数字按键和振动信号类型之间的关联关系，并自定义PIN码；
    信号采集单元：用于采集用户输入PIN码时产生的振动信号；
    信号处理单元：用于对所采集的振动信号进行处理并提取振动信号的有效部分；
    特征提取单元：用于提取所述振动信号的有效部分的梅尔倒谱系数；
    分类识别单元：用于将所述梅尔倒谱系数输入经训练的隐马尔可夫模型，获得振动信号类型的分类结果；
    身份验证单元：用于根据所述分类结果，并结合设定的数字按键和振动信号类型之间的关联关系，进行身份验证。
  10. 一种计算机可读存储介质，其上存储有计算机程序，其中，该程序被处理器执行时实现根据权利要求1至8中任一项所述方法的步骤。
PCT/CN2020/116763 2020-09-22 2020-09-22 一种基于振动信号的身份验证方法和系统 WO2022061499A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/116763 WO2022061499A1 (zh) 2020-09-22 2020-09-22 一种基于振动信号的身份验证方法和系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/116763 WO2022061499A1 (zh) 2020-09-22 2020-09-22 一种基于振动信号的身份验证方法和系统

Publications (1)

Publication Number Publication Date
WO2022061499A1 true WO2022061499A1 (zh) 2022-03-31

Family

ID=80844776

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/116763 WO2022061499A1 (zh) 2020-09-22 2020-09-22 一种基于振动信号的身份验证方法和系统

Country Status (1)

Country Link
WO (1) WO2022061499A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1547191A (zh) * 2003-12-12 2004-11-17 北京大学 结合语义和声纹信息的说话人身份确认系统
CN101478401A (zh) * 2009-01-21 2009-07-08 东北大学 一种基于击键特征识别的认证方法及系统
WO2016049983A1 (zh) * 2014-09-29 2016-04-07 同济大学 用户键盘按键行为模式建模与分析系统及其身份识别方法
CN110058689A (zh) * 2019-04-08 2019-07-26 深圳大学 一种基于脸部振动的智能设备输入方法
CN110610070A (zh) * 2019-08-08 2019-12-24 全球能源互联网研究院有限公司 一种用户身份识别方法及装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1547191A (zh) * 2003-12-12 2004-11-17 北京大学 结合语义和声纹信息的说话人身份确认系统
CN101478401A (zh) * 2009-01-21 2009-07-08 东北大学 一种基于击键特征识别的认证方法及系统
WO2016049983A1 (zh) * 2014-09-29 2016-04-07 同济大学 用户键盘按键行为模式建模与分析系统及其身份识别方法
CN110058689A (zh) * 2019-04-08 2019-07-26 深圳大学 一种基于脸部振动的智能设备输入方法
CN110610070A (zh) * 2019-08-08 2019-12-24 全球能源互联网研究院有限公司 一种用户身份识别方法及装置

Similar Documents

Publication Publication Date Title
US10943582B2 (en) Method and apparatus of training acoustic feature extracting model, device and computer storage medium
JP7152514B2 (ja) 声紋識別方法、モデルトレーニング方法、サーバ、及びコンピュータプログラム
EP3806089B1 (en) Mixed speech recognition method and apparatus, and computer readable storage medium
JP6303971B2 (ja) 話者交替検出装置、話者交替検出方法及び話者交替検出用コンピュータプログラム
WO2018166112A1 (zh) 基于声纹识别的身份验证的方法、电子装置及存储介质
US9147400B2 (en) Method and apparatus for generating speaker-specific spoken passwords
US9530417B2 (en) Methods, systems, and circuits for text independent speaker recognition with automatic learning features
Rudrapal et al. Voice recognition and authentication as a proficient biometric tool and its application in online exam for PH people
US11100932B2 (en) Robust start-end point detection algorithm using neural network
US9564134B2 (en) Method and apparatus for speaker-calibrated speaker detection
CN110459207A (zh) 唤醒语音关键短语分割
WO2018095167A1 (zh) 声纹识别方法和声纹识别系统
Deepa et al. An overview of acoustic side-channel attack
Guo et al. MultiPAD: A multivariant partition-based method for audio adversarial examples detection
WO2022061499A1 (zh) 一种基于振动信号的身份验证方法和系统
CN112131541A (zh) 一种基于振动信号的身份验证方法和系统
US11030995B2 (en) Acoustic feature extractor selected according to status flag of frame of acoustic signal
CN113555005B (zh) 模型训练、置信度确定方法及装置、电子设备、存储介质
WO2020206579A1 (zh) 一种基于脸部振动的智能设备输入方法
Kelly Cracking passwords using keyboard acoustics and language modeling
WO2022056915A1 (zh) 一种基于电容按键的人机交互方法和交互系统
CN113035230A (zh) 认证模型的训练方法、装置及电子设备
WO2022061500A1 (zh) 一种基于电容触摸屏的人机交互系统和交互方法
Çakmak et al. Audio CAPTCHA recognition using rastaplp features by svm
CN113129926A (zh) 语音情绪识别模型训练方法、语音情绪识别方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20954353

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 23.06.2023)

122 Ep: pct application non-entry in european phase

Ref document number: 20954353

Country of ref document: EP

Kind code of ref document: A1