Primax Electronics Ltd.

Skymizer Taiwan Inc.

Sampo Technology Corp.

Shu-Ming Liu

Wen Li Tang

Kai-Chiang Wu

A-Ming Chang

Hong-Che Yen

Chen-Ming Chang

Heng-Chun Ho

Der-Yu Tsai

Po-Sheng Chang

Jen-ho Kuo

Chien-Fa Chen

Han Che Wang

A memory apparatus and a data rearrangement method for computing in memory (CIM) are provided. The method includes determining whether first sequence data has two target bits that are both of a first value, inserting a non-target bit of a second value between the two target bits that are both of the first value and adjacent to each other to generate second sequence data, and receiving the second sequence data through memory cells in a memory to perform a multiply-accumulate (MAC) operation on the second sequence data. Each bit in the first sequence data is the first value or the second value. One of the two target bits is located adjacent to the other one of the two target bits in the first sequence data. The two target bits and the non-target bit are located in the first sequence data. Accordingly, the error rate is decreased.

Memory apparatus and data rearrangement method for computing in memory

A data processing circuit and a fault mitigating method are provided. The method is adapted for a memory having at least one fault bit. The memory provides a block for data storage. A difference between an output of a value of a plurality of bits input to at least one computing layer in a neural network and a correct value is determined. The bits are respectively considered the at least one fault bit. A repair condition is determined based on the difference. The repair condition includes a correspondence between a position where the fault bit is located in the block and at least one non-fault bit in the memory. A value of at least one non-fault bit of the memory replaces a value of the fault bit based on the repair condition.

Data processing circuit and fault mitigating method

A data processing circuit and a fault-mitigating method are provided. In the method, multiple sub-sequences are divided from sequence data. A first sub-sequence of the sub-sequences is accessed from a memory for a multiply-accumulate (MAC) operation to obtain a first computed result. The MAC operation is performed on a second sub-sequence of the sub-sequences in the memory to obtain a second computed result. The first and the second computed results are combined, where the combined result of the first and the second computed results is related to the result of the MAC operation on the sequence data directly. Accordingly, the error rate could be reduced, so as to mitigate fault.

Data processing circuit and fault-mitigating method

An error calibration apparatus and method are provided. The method is adapted for calibrating a machine learning (ML) accelerator. The ML accelerator achieves computation by using an analog circuit. An error between an output value of one or more computing layers of a neural network and a corresponding corrected value is determined. The computation of the computing layers is achieved by the analog circuit. A calibration node is generated according to the error. The calibration node is located at the next layer of the computing layers. The calibration node is used to minimize the error. The calibration node is achieved by a digital circuit. Accordingly, error and distortion of the analog circuit could be reduced.

Error calibration apparatus and method

A data processing circuit and a fault-mitigating method, which are adapted for a memory having a faulty bit, are provided. The memory is configured to store data related to an image, a weight for a multiply-accumulate (MAC) operation of image feature extraction, and/or a value for an activation operation. Sequence data is written into the memory. The bit number of the sequence data equals to the bit number used for storing data in a sequence block of the memory. The sequence data is accessed from the memory, wherein the access of the faulty bit in the memory is ignored. The value of the faulty bit is replaced by the value of a non-faulty bit in the memory to form new sequence data. The new sequence data is used for MAC. Accordingly, the accuracy of image recognition can be improved for the faulty memory.

A remote control toy car set comprising a toy car with a housing having a driving device, a camera system, and a first radio module. The camera system is fixed on the housing for taking images of the toy car in motion and generating corresponding image signals, and the first radio module is fixed on the housing and electrically connects with both the driving device and the camera system. The first radio module transfers the image signals from the camera system and receives radio control signals in order to control the operations of the driving device. The remote control toy car set also includes a remote control device for controlling the toy car. The remote control device includes a display for displaying images, a remote control interface for generating control signals, and a second radio module for transferring both the image signals from the toy car to the display and the control signals from the remote control interface to the toy car.

Toy car camera system and rear vision mirrors

A method of simulating two-way communications over a one-way communications system, enabling an indicator on the transmitter to reflect the manner in which signals from a function key on the transmitter side will be interpreted by an application on the receiver side of the communications system, is carried out by causing a list of current states of the function on the receiver side of the communications system rather than transmitting a signal concerning the current state of the list to the transmitter side. The transmitter side of the communication system is part of a wireless keyboard that does not include a receiver to properly indicate which of two possible states a function of the computer will be in even though the computer has no way of communicating that state to the receiver.

Initialization method for one-way communications and apparatus for use therewith

The present invention discloses a pointing device for scrolling a screen image. The pointing device comprises a chassis, a housing having an opening on its top mounted on the chassis, a vertical sensor installed on the chassis below the opening, a plurality of horizontal sensors installed on the chassis beside the vertical sensor, and a control stick vertically installed inside the opening of the housing comprising a top end and a bottom end. The bottom end of the control stick is in contact with the vertical sensor whereas the top end of the control stick protrudes from the opening of the housing and can be horizontally moved within a predetermined angle with respect to the bottom end of the control stick serving as its center. The pointing device further comprises a plate horizontally installed at the bottom end of the control stick with its bottom side positioned above the horizontal sensors. Horizontal movement of the top end of the control stick will cause corresponding movement of the plate which then causes the bottom surface of the plate to trigger at least one horizontal sensor. The pointing device also comprises a control device electrically connected to the vertical and horizontal sensors for detecting vertical movements of the control stick using the vertical sensor and generating correspondent vertical signals, and detecting horizontal movements of the control stick using the horizontal sensors and generating correspondent horizontal signals.

Controlling scrolls of a screen image

The present invention provides a computer joystick for controlling cursor or object movements in three dimensions. The computer joystick comprises a chassis, a control stick vertically installed on the chassis in a rotatable manner, a control stick sensor installed inside the chassis for detecting rotations of the control stick and generating corresponding stick position signals, a joystick handle comprising a vertical slot, and a rotation detector. The joystick handle is rotatably installed on the control stick through the vertical slot. The rotation detector comprises an optical panel fixed on the control stick, and a light sensor fixed inside the vertical slot of the joystick handle for detecting rotations of the joystick handle on the control stick and generating corresponding rotation signals.

Computer joystick for moving an object in three dimensions

The present invention discloses an input device for transmitting an input signal to a computer through a game port of the computer. The game port comprises a capacitor and a detecting circuit for detecting a time period needed for charging the capacitor to a threshold voltage, discharging the capacitor when the capacitor is charged to the threshold voltage, and transmitting the detected time period as an input to the computer. The input device comprises a first charging circuit wired to the capacitor for slowly charging the capacitor, a second charging circuit wired to the capacitor for quickly charging the capacitor, and a control circuit wired to the capacitor used for activating the second charging circuit according to the input signal. The control circuit activates the second charging circuit according to the voltage of the capacitor and the input signal so that the capacitor can be quickly charged to the threshold voltage.

Input device for transmitting an input signal to a computer game port

The invention discloses a computer joystick having two optical sensors to generate vector signals for indicating the current position of the joystick's handle. The computer joystick comprises a housing having an opening at its top, two mutually perpendicular shafts rotatably installed inside the housing, a joystick handle rotatably installed in the opening of the housing, an engaging means mounted at a bottom end of the joystick handle for rotatably engaging the two shafts, two optical sensors installed in the housing close to the two shafts for detecting rotations of the two shafts and generating corresponding displacement signals, and a control circuit having a memory for storing a coordinate as a position of the joystick handle, and a processor wired to the two optical sensors for updating the coordinate of the joystick handle according to the displacement signals generated by the two optical sensors and generating vector signals according to the coordinate of the joystick handle to indicate the current position of the joystick handle.

Computer joystick having two optical sensors for generating vector

The present invention discloses an optical sensing system which uses a plurality of detectors installed at various positions next to an optical panel to generate a position signal for indicating the position of the optical panel. The optical sensing system has an optical panel having a plurality of sensing areas each having a plurality of encoding holes arranged in a predetermined direction, a sensing device having a plurality of detectors for detecting movements of the optical panel corresponding to the sensing device, each of the detectors being corresponded to one of the sensing areas of the optical panel for detecting movements of the encoding holes along the predetermined direction and generating correspondent sensing signals, and a control circuit electrically connected to the sensing device for collecting the sensing signals generated by the detectors of the sensing device to form a position signal so as to indicate the optical panel's position.

Optical sensing system using multiple detectors at various positions for

A quadrature encoding device with slope-triggered digitizing circuit is provided for use in a mechanical mouse. The quadrature encoding device is capable of producing two square-wave signals indicative of the current mouse movement for position control of the mouse cursor on the computer screen. The slope-triggered digitizing circuit includes a delay circuit and an analog comparator, which is capable of generating a first voltage output when the slope of the opto-electrical signal from the photo detector is positive, and a second voltage output when the slope is negative. It is characteristic feature of the quadrature encoding device that the output characteristic of the slope-triggered digitizing circuit would not be affected by any deviations in the output characteristic of the associated photo detectors, and also would not be affected by the ambient light. Moreover, the quadrature encoding device can operate without having to make adjustments on threshold voltage, which makes the quadrature encoding device more convenient to use, simpler in circuit complexity, and lower in manufacturing cost as compared to the prior art.

Growing community of inventors

Taipei, Taiwan

Shu-Ming Liu