Patents by Inventor Kaichien Tsai
Kaichien Tsai 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: 10560909Abstract: A network includes a plurality of parent nodes to communicate in a wireless network via a wireless network protocol. A network node establishes a network connection to one of the parent nodes of the plurality of parent nodes in response to received beacons from the parent nodes. A signal analyzer in the network node processes a received signal strength of the beacons and the number of beacons received over a given time period from each of the parent nodes. The signal analyzer selects the parent node by analyzing the received signal strength for a number of beacons received from each parent node with respect to a number of expected beacons transmitted from each parent node of the plurality of parent nodes over the given time period.Type: GrantFiled: September 13, 2017Date of Patent: February 11, 2020Assignee: TEXAS INSTRUMETNS INCORPORATEDInventors: Ariton E. Xhafa, Jianwei Zhou, Il Han Kim, Kaichien Tsai, Xiaolin Lu
-
Patent number: 10545182Abstract: A system such as an electric meter includes a single-ended analog-to-digital converter which is susceptible to crosstalk. Methods and apparatus are disclosed to calibrate and cancel crosstalk effects. The system determines whether voltage-induced cross-talk is in-phase or out-of-phase with respect to voltage. The system determines a first calibration factor based on minimum and maximum measured current values and voltage. If the cross-talk is in-phase, the system sets a second calibration factor to 0. If the cross-talk is out-of-phase, the system computes the second calibration factor based on a measured current when a power factor angle is set to 90 degrees. Calibration factors may be stored in the multi-channel system. In use, the system measures current and voltage and computes the actual current, voltage and power based on the measured current and voltage by employing a crosstalk cancellation technique using the calibration factors.Type: GrantFiled: March 17, 2016Date of Patent: January 28, 2020Assignee: TEXAS INSTRUMENTS INCORPORATEDInventors: Minghua Fu, Kaichien Tsai, Prashanth Saidu
-
Patent number: 10447142Abstract: A switch-mode power supply includes a DC-DC converter and metering circuitry that is coupled to the DC-DC converter. The metering circuitry includes scaling circuitry, a current source, a capacitor, switching circuitry, and a comparator. The scaling circuitry is configured to generate a reference current scaled to be a predetermined fraction of a peak current flowing in an inductor of the DC-DC converter. The current source is configured to output a first current that is one-half of the reference current. The capacitor is coupled to the current source. The switching circuitry is configured to switchably connect the current source to the capacitor. The comparator is coupled to the capacitor. The comparator is configured to generate a signal indicating that a voltage across the capacitor exceeds a threshold voltage.Type: GrantFiled: December 20, 2018Date of Patent: October 15, 2019Assignee: TEXAS INSTRUMENTS INCORPORATEDInventors: Vinod Joseph Menezes, Manikandan RR, Rajat Chauhan, Vipul Kumar Singhal, Mahesh Madhukar Mehendale, Kaichien Tsai
-
Patent number: 10291520Abstract: A BLE-Mesh device includes a controller, an RF driver for driving the transceiver adapted to be coupled to an antenna, and a counter. The controller implements an applications layer including BLE and Mesh Applications, and a BLE stack and a mesh stack. A redundant traffic suppression relaying algorithm is for waiting for a random time within a selected time window from W1 to a later W2 before attempting to transmit a first packet that contains a unique source (SRC) address and a packet sequence (SEQ) number of a device that is the source of the first packet. If during the random time a packet with both the source device's SRC address and the SEQ number is received, the counter is incremented from an initial count to a current count. After the random time elapses, the current count is compared to a Cthreshold value, and the first packet is transmitted only if the current count<the Cthreshold value.Type: GrantFiled: April 24, 2017Date of Patent: May 14, 2019Assignee: TEXAS INSTRUMENTS INCORPORATEDInventors: Arvind Kandhalu Raghu, Kaichien Tsai, Ramanuja Vedantham
-
Publication number: 20180242266Abstract: A network includes a plurality of parent nodes to communicate in a wireless network via a wireless network protocol. A network node establishes a network connection to one of the parent nodes of the plurality of parent nodes in response to received beacons from the parent nodes. A signal analyzer in the network node processes a received signal strength of the beacons and the number of beacons received over a given time period from each of the parent nodes. The signal analyzer selects the parent node by analyzing the received signal strength for a number of beacons received from each parent node with respect to a number of expected beacons transmitted from each parent node of the plurality of parent nodes over the given time period.Type: ApplicationFiled: September 13, 2017Publication date: August 23, 2018Inventors: ARITON E. XHAFA, JIANWEI ZHOU, IL HAN KIM, KAICHIEN TSAI, XIAOLIN LU
-
Publication number: 20180234893Abstract: A network includes a mobile network node (MNN) that includes a mobile node communications manager (MNCM) to facilitate wireless communications to a plurality of stationary network nodes (SNNs) in a wireless network via a wireless network protocol. The MNCM utilizes a multicast address received over the wireless network. The multicast address is assigned to a predetermined network time slot to communicate uplink data from the MNN to the SNNs. The MNN receives downlink data via a separate predetermined network address and time slot assigned to a given SNN.Type: ApplicationFiled: September 13, 2017Publication date: August 16, 2018Inventors: ARITON XHAFA, JIANWEI ZHOU, XIAOLIN LU, IL HAN KIM, KAICHIEN TSAI
-
Patent number: 9894512Abstract: A BLE-Mesh device includes a controller coupled to a memory and to a transceiver adapted to be coupled to an antenna, wherein the controller implements an applications layer including BLE and Mesh applications, and a BLE stack and a mesh stack. The BLE-Mesh device has a switchable high-speed and low-speed mode and a speed switching algorithm for implementing a method of communications in BLE-mesh network. A broadcast ping is sent to neighborhood devices with a time to live (TTL)=1. A manufacturer's ID is analyzed to identify in a manufacturer's ID field in pongs received to determine a higher-speed capable device or a lower-speed device. A higher-speed data rate is utilized for mesh communications if a percentage of higher-speed capable devices is?a threshold percentage or a lower-speed data rate is utilized for the mesh communications if the percentage of higher-speed capable devices is<the threshold percentage.Type: GrantFiled: April 21, 2017Date of Patent: February 13, 2018Assignee: TEXAS INSTRUMENTS INCORPORATEDInventors: Arvind Kandhalu Raghu, Kaichien Tsai, Ramanuja Vedantham
-
Publication number: 20170325127Abstract: A BLE-Mesh device includes a controller, an RF driver for driving the transceiver adapted to be coupled to an antenna, and a counter. The controller implements an applications layer including BLE and Mesh Applications, and a BLE stack and a mesh stack. A redundant traffic suppression relaying algorithm is for waiting for a random time within a selected time window from W1 to a later W2 before attempting to transmit a first packet that has the first relay device's sender's source (SRC) address and a sequence (SEQ) number. If during the random time a packet with both the device's SRC address and the SEQ number is received, the counter is incremented from an initial count to a current count. After the random time elapses, the current count is compared to a Cthreshold value, and the first packet is transmitted only if the current count <the Cthreshold value.Type: ApplicationFiled: April 24, 2017Publication date: November 9, 2017Inventors: ARVIND KANDHALU RAGHU, KAICHIEN TSAI, RAMANUJA VEDANTHAM
-
Publication number: 20170317713Abstract: A method for pseudo channel hopping in a node of a wireless mesh network is provided that includes scanning each channel of a plurality of channels used for packet transmission by the node, wherein each channel is scanned for a scan dwell time associated with the channel, updating statistics for each channel based on packets received by the node during the scanning of the channel, and changing scan dwell times of the plurality of channels periodically based on the statistics.Type: ApplicationFiled: February 3, 2017Publication date: November 2, 2017Inventors: Kaichien Tsai, Arvind Kandhalu Raghu, Ramanuja Vedantham
-
Publication number: 20170318453Abstract: A BLE-Mesh device includes a controller coupled to a memory and to a transceiver adapted to be coupled to an antenna, wherein the controller implements an applications layer including BLE and Mesh applications, and a BLE stack and a mesh stack. The BLE-Mesh device has a switchable high-speed and low-speed mode and a speed switching algorithm for implementing a method of communications in BLE-mesh network. A broadcast ping is sent to neighborhood devices with a time to live (TTL)=1. A manufacturer's ID is analyzed to identify in a manufacturer's ID field in pongs received to determine a higher-speed capable device or a lower-speed device. A higher-speed data rate is utilized for mesh communications if a percentage of higher-speed capable devices is?a threshold percentage or a lower-speed data rate is utilized for the mesh communications if the percentage of higher-speed capable devices is<the threshold percentage.Type: ApplicationFiled: April 21, 2017Publication date: November 2, 2017Inventors: ARVIND KANDHALU RAGHU, KAICHIEN TSAI, RAMANUJA VEDANTHAM
-
Publication number: 20170317906Abstract: A method for pseudo channel hopping in a node of a wireless mesh network is provided that includes scanning each channel of a plurality of channels used for packet transmission by the node, wherein each channel is scanned for a scan dwell time associated with the channel, updating statistics for each channel based on packets received by the node during the scanning of the channel, and selecting a channel of the plurality of channels for scanning based on the statistics when the scan dwell time of a currently scanned channel ends.Type: ApplicationFiled: February 3, 2017Publication date: November 2, 2017Inventors: Kaichien Tsai, Arvind Kandhalu Raghu, Ramanuja Vedantham
-
Publication number: 20170269134Abstract: A system such as an electric meter includes a single-ended analog-to-digital converter which is susceptible to crosstalk. Methods and apparatus are disclosed to calibrate and cancel crosstalk effects. The system determines whether voltage-induced cross-talk is in-phase or out-of-phase with respect to voltage. The system determines a first calibration factor based on minimum and maximum measured current values and voltage. If the cross-talk is in-phase, the system sets a second calibration factor to 0. If the cross-talk is out-of-phase, the system computes the second calibration factor based on a measured current when a power factor angle is set to 90 degrees. Calibration factors may be stored in the multi-channel system. In use, the system measures current and voltage and computes the actual current, voltage and power based on the measured current and voltage by employing a crosstalk cancellation technique using the calibration factors.Type: ApplicationFiled: March 17, 2016Publication date: September 21, 2017Inventors: Minghua FU, Kaichien TSAI, Prashanth SAIDU
-
Patent number: 9274201Abstract: A system is provided for calibrating a device. The system includes a reference component, a sampling component, a calibration component, a comparing component and a proportional integral component. The reference component provides a reference power signal based on a voltage instruction and a current instruction. The sampling component samples a voltage signal to obtain a sampled voltage value and samples a current signal to obtain a sampled current value. The calibration component generates a calibrated power signal based on the sampled voltage value and the sampled current. The comparing component generates an error signal based on the reference power signal and the calibrated power signal. The proportional integral component and the calibration component are a feedback system that is operable to calibrate the gain of the sampled voltage and the sample current based on the error signal.Type: GrantFiled: June 27, 2014Date of Patent: March 1, 2016Assignee: TEXAS INSTRUMENTS INCORPORATEDInventors: Kaichien Tsai, Minghua Fu, Anand Dabak
-
Publication number: 20150061636Abstract: A system is provided for calibrating a device. The system includes a reference component, a sampling component, a calibration component, a comparing component and a proportional integral component. The reference component provides a reference power signal based on a voltage instruction and a current instruction. The sampling component samples a voltage signal to obtain a sampled voltage value and samples a current signal to obtain a sampled current value. The calibration component generates a calibrated power signal based on the sampled voltage value and the sampled current. The comparing component generates an error signal based on the reference power signal and the calibrated power signal. The proportional integral component and the calibration component are a feedback system that is operable to calibrate the gain of the sampled voltage and the sample current based on the error signal.Type: ApplicationFiled: June 27, 2014Publication date: March 5, 2015Inventors: Kaichien Tsai, Minghua Fu, Anand Dabak