Data processing: artificial intelligence – Neural network – Learning task
Reexamination Certificate
1998-10-27
2001-10-16
Davis, George B. (Department: 2122)
Data processing: artificial intelligence
Neural network
Learning task
C704S232000, C704S259000
Reexamination Certificate
active
06304865
ABSTRACT:
RESERVATION OF COPYRIGHT
A portion of the disclosure of this patent document contains material to which a claim of copyright protection is made. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure as it appears in the Patent and Trademark Office patent file or records, but reserves all other rights whatsoever.
BACKGROUND
1. Field
This field relates to diagnostic tests for audio cards used in computers.
2. Description of the Related Art
Computer systems in general and International Business Machines (IBM) compatible personal computer systems in particular have attained widespread use for providing computer power to many segments of today's modem society. A personal computer system can usually be defined as a desk top, floor standing, or portable microcomputer that includes a system unit having a system processor and associated volatile and non-volatile memory, a display monitor, a keyboard, one or more diskette drives, a fixed disk storage device and an optional printer. One of the distinguishing characteristics of these systems is the use of a system board to electrically connect these components together. These personal computer systems are information handling systems which are designed primarily to give independent computing power to a single user (or a relatively small group of users in the case of personal computers which serve as computer server systems) and are inexpensively priced for purchase by individuals or small businesses. A personal computer system may also include one or a plurality of I/O devices (i.e. peripheral devices) which are coupled to the system processor and which perform specialized functions. Examples of I/O devices include modems, sound and video devices or specialized communication devices. Mass storage devices such as hard disks, CD-ROM drives and magneto-optical drives are also considered to be peripheral devices. Computers producing multimedia effects (e.g., sound coupled with visual images) are in increased demand as computers are used for artistic endeavors, for entertainment, and for education. Sound makes game playing more realistic and helps reinforce knowledge and make educational programs more enjoyable to use. Digital effects and music can also be created on the computer and played through attached speakers without the need for additional musical instruments or components.
Multimedia systems today often include audio devices (e.g., sound cards) connected to the computer to which speakers can be attached for playing music and other sound effects. Testing and diagnosis of these audio devices poses challenges in modem computer manufacturing and repair facilities. Currently, the technician attaches speakers to the audio device and invokes a test procedure that plays music or a test pattern on the speakers and listening to the result.
A challenge of testing and diagnosing audio devices by listening to the result is that the area is often noisy and difficult to distinguish one system being tested from another. In addition, human error, which may be caused by repetitively listening to numerous systems may cause the technician to pass an audio device which would otherwise fail. Finally, to be done well, the technician must listen to a variety of sounds to ensure that the audio device is working properly which is time consuming and can effect the throughput of the manufacturing facility.
Automated methods have been used to test audio devices such as sound cards. Audio devices have also been tested using a loopback connector connecting the input and output ports of the audio device. These prior methods involved recording simple sine waves across the audio device and checking for a sine wave match (i.e., zero crossing at the same point for the input and output sine waves) or by using a frequency domain, such as a Fast Fourier Transform (“FFT”), analysis to check the sine wave. Other methods of testing an audio device have recorded a square wave and charting the periodic changes in the recorded data. The challenges posed by these prior methods is that they did not allow for much, if any, variance in the results obtained leading to many false failures. If, for example, the zero crossing did not occur at precisely the right position, the audio device did not pass the given test even though the audio device may be within specifications and qualitatively sound in operation. False positives may lead to increased diagnostic time using a human technician to listen to output from the audio device only to eventually determine that the audio device was suitable for shipment to a customer.
An improvement to the testing and diagnosis of audio devices is needed which removes the need for technicians to manually listen to sounds generated from such devices and which also allows for variances to improve the fault tolerance of automated test systems.
SUMMARY
The audio testing and diagnostic method of the present invention trains a neural network based upon an actual working audio device. The audio device is configured so that a line from the output port loops back into an input port. A test signal is transmitted through the audio port and received at the input port. The test signal is then converted into a frequency spectrum for analysis. The frequency spectrum is provided as input to a trained neural network, the neural network being previously trained to recognize the frequency spectrum pattern created by a properly working, or ideal, audio device. The neural network is trained by connecting the input port to the output port of an audio device from which the training is to occur, i.e., a properly working audio device. A signal is transmitted through the output port and received from the input port. The signal is converted to a frequency spectrum which is provided as input to train the neural network. Training the neural network results in weighted values associated with neurons, the neurons being arranged into a neural network topology. The weighted values and topology are stored in the trained neural network so that the trained neural network can test other audio cards. The audio device may include both sound cards, modems, or other sound generating devices. Converting signals to a frequency spectrum may be computed using a fourier transform. Prior to converting signals to a frequency spectrum, the waveform characteristics of the signal may be iteratively evaluated and recording levels adjusted so that the signal received has characteristics that can be tested by the neural network. In addition, the signal that is analyzed may be a portion of the signal so that analog to digital converters in the audio device have stabilized before the portion of the signal is taken. The neural network generates a confidence level based on a comparison of the pattern of the tested audio device's frequency spectrum to the frequency spectrum of a working audio device. A pass value may be predetermined so that the tested audio device is reported as passing or failing the test by comparing the confidence level value generated by the system with the predetermined pass value.
The computer system of the present invention gains the same advantages as the method by including a processor, memory, and an audio device with an input and output port into a computer system. A connector is provided to connect the input and output ports. A signal generator generates a signal from the output port which, in turn, is received at the input port by a signal receiver. A frequency converter, which may use a fourier transform, converts the received signal to a frequency spectrum. A trained neural network analyzes the frequency spectrum to generate a confidence level.
REFERENCES:
patent: 4271513 (1981-06-01), Maejima et al.
patent: 4430530 (1984-02-01), Kandell et al.
patent: 4953165 (1990-08-01), Jackson
patent: 5040215 (1991-08-01), Amano et al.
patent: 5093899 (1992-03-01), Hiraiwa
patent: 5119469 (1992-06-01), Alkon et al.
patent: 5150323 (1992-09-01), Castelaz
patent: 5163111 (1992-11-01), Baji et al.
patent: 5175
Broadbent Christopher F.
Christensen Alan K.
Davis George B.
Dell U.S.A. L.P.
Skjerven Morrill & MacPherson LLP
Terrile Stephen A.
LandOfFree
Audio diagnostic system and method using frequency spectrum... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Audio diagnostic system and method using frequency spectrum..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Audio diagnostic system and method using frequency spectrum... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2580307