Qualcomm Incorporated

Iwen Yao

Jeremy H Lin

Arunava Chaudhuri

Raghu Narayan Challa

Hemanth Sampath

Jia Tang

Da-shan Shiu

Remi Jonathan Gurski

Atul Arvind Salvekar

Young Geun Cho

Shantanu Khare

Ali RostamPisheh

Ming-Chieh Kuo

Qiang Huang

Mrinal Mahesh Nath

Zhiqiang Cui

Tamer Adel Kadous

Jibing Wang

Yong Li

Joseph Victor Zanotelli

Kevin W Yen

Weihong Jing

Min Wu

Davie J Santos

Megan Wu

A method for wireless communication is disclosed that includes selecting a plurality of probabilities for a symbol based on a bit-to-symbol mapping; calculating a conditional mean of the symbol based on the plurality of probabilities; and, generating a signal representative of the symbol based on the conditional mean of the symbol. An apparatus for performing the method is also disclosed.

Method and apparatus for soft symbol determination

A modem (for example, a modem within a cellular telephone) includes a plurality of Wireless Communication System Modem Sub-Circuits (WCSMSCs). Each WCSMSC receives a control signal generated by a corresponding one of a plurality of programmable timers. Each timer receives the same sequence of count values from a wall clock counter. A processor that controls overall modem operation can program a timer to generate a control pulse at a particular count time of the wall clock counter. The processor can also program a timer to generate a periodic control signal. The control signals output from the timers orchestrate when the various WCSMSCs start operating in the processing of a frame. By virtue of the programmability of the timers, the wall clock timer system is programmable to generate customized control signals such that frames of new and different protocols having arbitrary frame structures can be processed by the same modem/timer system.

Wall clock timer and system for generic modem

An improved processing engine for performing Fourier transforms includes an instruction processor configured to process sequential instruction software commands and a Fourier transform engine coupled to the instruction processor. The Fourier transform engine is configured to perform Fourier transforms on a serial stream of data. The Fourier transform engine is configured to receive configuration information and operational data from the instruction processor via a set of software tasks.

Reuse engine with task list for fast fourier transform and method of using the same

A method for accessing extrinsic information in a turbo decoder is disclosed. Operation phases for Forward State Metric Calculators (FSMCs) and Reverse State Metric Calculators (RSMCs) in multiple maximum a posteriori probability (MAP) decoders are misaligned differently based on whether a current half iteration is even or odd. First extrinsic information is read from a memory into the FSMCs and RSMCs using the misaligned operation phases. Second extrinsic information is determined using the MAP decoders. Each row of the second extrinsic information is stored to a different bank in the memory using the misaligned operation phases.

Accessing memory during parallel turbo decoding

Systems, methods, devices, and computer program products are described for Turbo decoding in a wireless communication system. Turbo encoded wireless signals may be received and demodulated, and forwarded to a branch metric calculator. The branch metric calculator may calculate a set of branch metrics for the demodulated signal. A state metric unit may receive the set of branch metrics and a previously calculated set of state metrics. The state metric unit may perform various comparisons of the set of state metrics before the received set of branch metrics is added to a portion of the state metrics identified through the comparisons.

SMU architecture for turbo decoder

A de-interleaver generates a plurality of De-interleaved Reorder Physical (DRP) addresses to simultaneously write a corresponding plurality of LLR values into a multi-banked memory such that not more than one LLR value is written into each bank of the multi-banked memory at a time. A sequence of such parallel writes results in the LLR values of a transmission of a sub-packet being stored in the memory. Address translation performed during generation of the DRP addresses causes the LLR values to be stored within the banks such that a decoder can read LLR values out of the memory in a de-interleaved sequence. Each memory location of a bank is a word-location for storing multiple related LLR values, where one LLR value is stored along with its parity values. The ability to simultaneously write to multiple LLR values is used to clear locations in a fast and efficient manner.

De-interleaving mechanism involving a multi-banked LLR buffer

A flexible and reconfigurable digital system (for example, a wireless modem) includes a set of sub-circuits. Each sub-circuit includes a task manager and an amount of configurable hardware circuitry for performing a type of operation on a data stream. The task manager of a sub-circuit can configure and control the configurable hardware of the sub-circuit. A central processor configures and orchestrates operation of the sub-circuits by maintaining a set of task lists in a tightly coupled memory. Each task list includes task instructions for a corresponding sub-circuit. The task manager of a sub-circuit reads task instructions from its task list and controls its associated hardware circuitry as directed by the instructions. A timestamp task instruction and a push task instruction and the task list architecture allow modem sub-circuits to be easily reconfigured to operate in accordance with either a first air interface standard or a second air interface standard.

Reconfigurable wireless modem sub-circuits to implement multiple air interface standards

Off-line task list architecture utilizing tightly coupled memory system

A technique for performing AGC and DC compensation in a receiver. The receiver comprises an energy estimator for generating an estimate of the level of a received signal; an RF device to apply gain to the received signal; an AGC for controlling the RF device gain based on the energy estimation; a first DC compensation loop for finely adjusting the DC component of the received signal in fast or slow tracking mode (FTM or STM); and a second DC compensation loop for coarsely adjusting the DC component of the received signal. Three modes of AGC operations: In Acquisition, iterations of FTM fine DC adjustment, short energy estimation, and RF device gain adjustment are performed during signal timing detection. In Connected, long energy estimation, RF device gain adjustment, and STM fine and coarse DC adjustments are performed during superframe preamble. In Sleep, FTM fine DC adjustment, short energy estimation, and RF device gain adjustment are performed during superframe preamble.

Method and system for DC compensation and AGC

Techniques to efficiently generate memory addresses for a Turbo code interleaver using a number of look-up tables. An interleaver includes a storage unit, sets of tables, and an address generator. The storage unit stores K elements for a data packet at locations representative of an R&#xd7;C array, with the elements being stored in a first (e.g., linear) order and provided in a second (e.g., interleaved) order. A first set of table(s) stores sequences (e.g., inter-row permutation sequences P, P, P, and P) used to perform row permutation of the array to map from the first order to the second order. A second set of table(s) stores sequences (e.g., intra-row base sequences and prime number sequences) used to perform column permutation. The address generator receives a first address for the first order and generates a corresponding second address for the second order based on sequences stored in the tables.

Interleaver for turbo decoder

A buffer structure for storing intermediate results (i.e., APP data) for a Turbo decoder to increase access throughput, the buffer structure is designed to support concurrent access of APP data for two or more bits for each access cycle. This is achieved by partitioning the buffer into a number of banks, with each bank being independently accessible. To avoid access contentions, the banks are assigned to the rows and columns of a 2-dimensional array used for code interleaving such that APP data for consecutive bits are accessed from different banks. To support &ldquo;linear&rdquo; addressing, the banks can be arranged into two sets, which are assigned to even-numbered and odd-number columns of the array. To support &ldquo;interleaved&rdquo; addressing, the banks can be assigned to groups of rows of the array such that adjacent rows in the interleaved array are assigned to different groups.

Growing community of inventors

San Diego, CA, United States of America

Iwen Yao