Patents by Inventor Wen-Jeng Chang
Wen-Jeng Chang has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 11942543Abstract: A high-voltage semiconductor device structure is provided. The high-voltage semiconductor device structure includes a semiconductor substrate, a source ring in the semiconductor substrate, and a drain region in the semiconductor substrate. The high-voltage semiconductor device structure also includes a doped ring surrounding sides and a bottom of the source ring and a well region surrounding sides and bottoms of the drain region and the doped ring. The well region has a conductivity type opposite to that of the doped ring. The high-voltage semiconductor device structure further includes a conductor electrically connected to the drain region and extending over and across a periphery of the well region. In addition, the high-voltage semiconductor device structure includes a shielding element ring between the conductor and the semiconductor substrate. The shielding element ring extends over and across the periphery of the well region.Type: GrantFiled: June 29, 2022Date of Patent: March 26, 2024Assignee: Taiwan Semiconductor Manufacturing Company, Ltd.Inventors: Hung-Chou Lin, Yi-Cheng Chiu, Karthick Murukesan, Yi-Min Chen, Shiuan-Jeng Lin, Wen-Chih Chiang, Chen-Chien Chang, Chih-Yuan Chan, Kuo-Ming Wu, Chun-Lin Tsai
-
Patent number: 11928416Abstract: A method of process technology assessment is provided. The method includes: defining a scope of the process technology assessment, the scope comprising an original process technology and a first process technology; modeling a first object in an integrated circuit into a resistance domain and a capacitance domain; generating a first resistance scaling factor and a first capacitance scaling factor based on the modeling, the original process technology, and the first process technology; and utilizing, by an electronic design automation (EDA) tool, the first resistance scaling factor and the first capacitance scaling factor for simulation of the integrated circuit.Type: GrantFiled: March 1, 2023Date of Patent: March 12, 2024Assignee: TAIWAN SEMICONDUCTOR MANUFACTURING COMPANY, LTD.Inventors: Hung-Chih Ou, Kuo-Fu Lee, Wen-Hao Chen, Keh-Jeng Chang, Hsiang-Ho Chang
-
Patent number: 7555697Abstract: A data error-detecting method for detecting errors before C1 decoding procedure is provided. First, a bit modulation is performed for modulating data channel bits obtained from an optical disk into 8-bit data. When the data channel bits is determined to introduce a legal mapping, an erasure bit having a first value is derived and tagged at the modulated 8-bit data, otherwise an erasure bit having a second value is derived and tagged at the modulated 8-bit data. After the 8-bit data being arranged as a codeword composed of 32 erasure bits respectively tagged with 32 8-bit data, counting the number of the modulated 8-bit data tagged with the erasure bit having the second value. When the counted number is less than 4, the codeword composed modulated 8-bits data tagged with the erasure bits having the second value can be corrected.Type: GrantFiled: July 8, 2005Date of Patent: June 30, 2009Inventors: Wen-Jeng Chang, Pei-Jei Hu, Yu-Fu Yeh
-
Patent number: 7512764Abstract: A method is proposed for utilizing a memory of a de-interleaving unit comprising the steps of determining a size of the memory according to a maximal symbol number of a data frame, a delay value, an adjustable parameter, and a half of a total symbol number of the data frame; accessing the memory according to a write memory address determined by a first algorithm during data buffering; and accessing the memory according to a read memory address determined by a second algorithm during data retrieving. The conventional access method has the drawback of ineffective use of the memory space of the de-interleaving unit, while the proposed method uses only about half the memory space for implementation.Type: GrantFiled: May 10, 2004Date of Patent: March 31, 2009Assignee: Tian Holdings, LLCInventor: Wen-Jeng Chang
-
Patent number: 7225385Abstract: A scrambling method for generating recording data onto a disc includes vertically scrambling main data, which is stored in DRAM, to generate PO, wherein the scrambled data due to generate PO is not stored in DRAM; then scrambling data again to generate PI; the scrambled data due to generating PI adds with ID, IED, RSV, EDC, PI, and PO, arranging in orders and recording onto a disc.Type: GrantFiled: February 19, 2004Date of Patent: May 29, 2007Assignee: Via Technologies, Inc.Inventors: Kun-Long Lin, Wen-Jeng Chang, Chih-Yung Chen, Chiung-Ying Peng
-
Publication number: 20060083639Abstract: A PDMS valve-less micro pump structure includes a PDMS body having a contour surface, a membrane covering the contour surface of the PDMS body, and a PZT actuator located on the membrane. The contour surface of the PDMS body to be sealed by the membrane and the PZT actuator has in series a lead-in cavity, a lead-in nozzle structure, a main cavity, a lead-out nozzle structure, and a lead-out cavity. The PZT actuator is located right above the main cavity. By providing the PDMS as a material to form the body of the micro pump, the micro pump can then be mass-produced with less cost and simpler structuring. Also, substantial elasticity and bio-compatibility for the micro pump can be achieved.Type: ApplicationFiled: January 26, 2005Publication date: April 20, 2006Inventors: Min-Sheng Liu, Chi-Chuan Wang, Bing-Chwen Yang, Yu-Ching Cheng, Wen-Jeng Chang
-
Patent number: 7023784Abstract: An encoding method for recording gap regions or sectors having repeated data is disclosed. At least one sector is entirely encoded and stored in a memory buffer of the optical recording system. If the currently encoded are gap regions or sector having repeated data embedded in the fields of user data, only those portions affected by the modified header will be encoded to derive associated P code and Q code. The error detection code is firstly generated in the mode 1 standard when a gap region is encoded. P code is then encoded for those regions affected by the sequentially modified header and error detection code, while associated Q code is then derived according the modified header, error detection code, and P code. In the mode 2 form 1 and mode form 2 standards, only those regions affected by the sequentially modified header should be re-encoded again.Type: GrantFiled: March 28, 2003Date of Patent: April 4, 2006Assignee: Via Technologies, Inc.Inventors: Wen-Jeng Chang, Kun-Long Lin
-
Patent number: 7007225Abstract: A C3 module in an optical storage apparatus and operation method thereof are disclosed. When the format of data to be read is audio, the C3 module enters a power saving mode. When the data format is non-audio, the C3 module enters an operation mode. The C3 module has an error correction module and an error detection module. When the error correction module is operating, the error detection module enters the power saving mode. When the operation of the error detection module is finished, the error correction module enters a power saving mode. When the error detection module is operating, the error correction module enters the power saving mode. When the operation of the error detection module is finished, the error detection module enters the power saving mode.Type: GrantFiled: April 3, 2002Date of Patent: February 28, 2006Assignee: VIA Technologies, Inc.Inventor: Wen-Jeng Chang
-
Publication number: 20060015794Abstract: A data error-detecting method for detecting errors before C1 decoding procedure is provided. First, a bit modulation is performed for modulating data channel bits obtained from an optical disk into 8-bit data. When the data channel bits is determined to introduce a legal mapping, an erasure bit having a first value is derived and tagged at the modulated 8-bit data, otherwise an erasure bit having a second value is derived and tagged at the modulated 8-bit data. After the 8-bit data being arranged as a codeword composed of 32 erasure bits respectively tagged with 32 8-bit data, counting the number of the modulated 8-bit data tagged with the erasure bit having the second value. When the counted number is less than 4, the codeword composed modulated 8-bits data tagged with the erasure bits having the second value can be corrected.Type: ApplicationFiled: July 8, 2005Publication date: January 19, 2006Inventors: Wen-Jeng Chang, Pei-Jei Hu, Yufu Yeh
-
Publication number: 20040260892Abstract: A method is proposed for utilizing a memory of a de-interleaving unit comprising the steps of determining a size of the memory according to a maximal symbol number of a data frame, a delay value, an adjustable parameter, and a half of a total symbol number of the data frame; accessing the memory according to a write memory address determined by a first algorithm during data buffering; and accessing the memory according to a read memory address determined by a second algorithm during data retrieving. The conventional access method has the drawback of ineffective use of the memory space of the de-interleaving unit, while the proposed method uses only about half the memory space for implementation.Type: ApplicationFiled: May 10, 2004Publication date: December 23, 2004Inventor: Wen-Jeng Chang
-
Publication number: 20040163026Abstract: A scrambling method for generating recording data onto a disc includes vertically scrambling main data, which is stored in DRAM, to generate PO, wherein the scrambled data due to generate PO is not stored in DRAM; then scrambling data again to generate PI; the scrambled data due to generating PI adds with ID, IED, RSV, EDC, PI, and PO, arranging in orders and recording onto a disc.Type: ApplicationFiled: February 19, 2004Publication date: August 19, 2004Inventors: Kun-Long Lin, Wen-Jeng Chang, Chih-Yung Chen, Chiung-Ying Peng
-
Patent number: 6772282Abstract: A continuous burn process of a storage medium, performed when the buffer is buffer under run, is disclosed. Whether the data buffer is buffer under run is firstly detected. If buffer under run is to occur to the data buffer, the currently written data block is an incomplete data block. The incomplete data block includes at least a synchronous frame and an incomplete data frame, which is the last data to be burned. The write operation is paused then. When the data in the data buffer reaches a predetermined amount, a new data is written into a superimposed data region of the incomplete data frame. The data in the data buffer is then continuously burned into the subsequent data frame adjacent to the incomplete data frame.Type: GrantFiled: October 22, 2001Date of Patent: August 3, 2004Assignee: Via Technologies, Inc.Inventors: Kuan-Yuan Hsiao, Wen-Jeng Chang
-
Patent number: 6754790Abstract: A method is proposed for accessing the memory of a de-interleaving unit. The conventional access method has the drawback of ineffective use of the memory space of the de-interleaving unit, while the proposed method uses only about half the memory space for implementation. The proposed method utilizes a specially-formulated write-operation interleaving algorithm to write data into the memory of the de-interleaving unit, and subsequently utilizes a specially-formulated read-operation de-interleaving algorithm to read data from the memory of the de-interleaving unit.Type: GrantFiled: June 14, 2002Date of Patent: June 22, 2004Assignee: VIA Optical Solution, Inc.Inventor: Wen-Jeng Chang
-
Publication number: 20030185129Abstract: An encoding method for recording gap regions or sectors having repeated data is disclosed. At least one sector is entirely encoded and stored in a memory buffer of the optical recording system. If the currently encoded are gap regions or sector having repeated data embedded in the fields of user data, only those portions affected by the modified header will be encoded to derive associated P code and Q code. The error detection code is firstly generated in the mode 1 standard when a gap region is encoded. P code is then encoded for those regions affected by the sequentially modified header and error detection code, while associated Q code is then derived according the modified header, error detection code, and P code. In the mode 2 form 1 and mode form 2 standards, only those regions affected by the sequentially modified header should be re-encoded again.Type: ApplicationFiled: March 28, 2003Publication date: October 2, 2003Inventors: Wen-Jeng Chang, Kun-Long Lin
-
Publication number: 20030041219Abstract: A method is proposed for accessing the memory of a de-interleaving unit. The conventional access method has the drawback of ineffective use of the memory space of the de-interleaving unit, while the proposed method uses only about half the memory space for implementation. The proposed method utilizes a specially-formulated write-operation interleaving algorithm to write data into the memory of the de-interleaving unit, and subsequently utilizes a specially-formulated read-operation de-interleaving algorithm to read data from the memory of the de-interleaving unit.Type: ApplicationFiled: June 14, 2002Publication date: February 27, 2003Inventor: Wen-Jeng Chang
-
Publication number: 20020184591Abstract: A C3 module in an optical storage apparatus and operation method thereof are disclosed. When the format of data to be read is audio, the C3 module enters a power saving mode. When the data format is non-audio, the C3 module enters an operation mode. The C3 module has an error correction module and an error detection module. When the error correction module is operating, the error detection module enters the power saving mode. When the operation of the error detection module is finished, the error correction module enters a power saving mode. When the error detection module is operating, the error correction module enters the power saving mode. When the operation of the error detection module is finished, the error detection module enters the power saving mode.Type: ApplicationFiled: April 3, 2002Publication date: December 5, 2002Inventor: Wen-Jeng Chang
-
Publication number: 20020166025Abstract: A continuous burn process of a storage medium, performed when the buffer is buffer under run, is disclosed. Whether the data buffer is buffer under run is firstly detected. If buffer under run is to occur to the data buffer, the currently written data block is an incomplete data block. The incomplete data block includes at least a synchronous frame and an incomplete data frame, which is the last data to be burned. The write operation is paused then. When the data in the data buffer reaches a predetermined amount, a new data is written into a superimposed data region of the incomplete data frame. The data in the data buffer is then continuously burned into the subsequent data frame adjacent to the incomplete data frame.Type: ApplicationFiled: October 22, 2001Publication date: November 7, 2002Inventors: Kuan-Yuan Hsiao, Wen-Jeng Chang
-
Publication number: 20010023495Abstract: A data error-correction method, applied in an optical storage device, for correction an original data read from an optical storage media. A bit modulation is performed on the original data. The data is thus checked to see if the original data is error. If error exists, an error bit is added to the error-contained modulated data. An error-correction procedure is carried out to correct error in the error-contained modulated data.Type: ApplicationFiled: February 27, 2001Publication date: September 20, 2001Inventors: Wen-Jeng Chang, Pei-Jei Hu
-
Patent number: RE43019Abstract: An encoding method for recording gap regions or sectors having repeated data is disclosed. At least one sector is entirely encoded and stored in a memory buffer of the optical recording system. If the currently encoded are gap regions or sector having repeated data embedded in the fields of user data, only those portions affected by the modified header will be encoded to derive associated P code and Q code. The error detection code is firstly generated in the mode 1 standard when a gap region is encoded. P code is then encoded for those regions affected by the sequentially modified header and error detection code, while associated Q code is then derived according the modified header, error detection code, and P code. In the mode 2 form 1 and mode form 2 standards, only those regions affected by the sequentially modified header should be re-encoded again.Type: GrantFiled: April 4, 2008Date of Patent: December 13, 2011Assignee: Tian Holdings, LLCInventors: Wen-Jeng Chang, Kun-Long Lin