207 results about "Address generation unit" patented technology

A memory may include a plurality of word lines, one or more redundancy word lines for replacing one or more word lines among the plurality of word lines, a target address generation unit suitable for generating one or more target addresses using a stored address, and a control unit suitable for sequentially refreshing the plurality of word lines in response to a refresh command which is periodically inputted, refreshing a word line selected based on the target address when the refresh command is inputted M times, and refreshing the one or more redundancy word lines whenever the refresh command is inputted N times, wherein the M and N are natural numbers.

A parallel decoder, which is simpler and more flexible than conventional devices, is provided in decoding device for a LDPC code. The present invention includes a plurality of memory units for storing a received value and a message generated during a Message-Passing decoding, a plurality of variable node function units, a plurality of check node function units, a plurality of address generation units for generating an address of each of memory units, and a plurality of shuffle network units for determining a connection between variable node function units and check node function units. An address generation unit generates an address on the basis of a plurality of permutations. Each shuffle network unit is connected to some of the variable node function units. This connection is determined on the basis of a plurality of permutations. A change of the permutations in the address generation units and a change of the permutations in the shuffle network units are performed in the same cycle in a decoding process.

In a line by line image forming apparatus, line switching information which depends on scan line curve and an overlap data length to be read that overlaps across a plurality of lines before and after a switching position instructed by line switching information are set in a register. If switching to the line above or below is instructed by the line switching information, an address generating unit, when reading image data from an image memory, generates the read address and read data length of the image memory in accordance with the line switching information and the overlap data length, and reads image data corresponding to a current line and the line above or below the current line in accordance with the generated data.

A memory may include a plurality of word lines coupled to one or more memory cells; a target address generation unit suitable for generating one or more target addresses using a stored address; and a control unit suitable for sequentially refreshing the plurality of word lines in response to a refresh command which is periodically inputted, and refreshing a word line corresponding to the target address in response to the refresh command at a random time.

The invention provides an SOC-based data reuse convolutional neural network accelerator. Input data of image input, weight parameters, offset parameters and the like of a convolutional neural networkis grouped; a large amount of the input data is classified into reusable block data; and reusable data blocks are read through a control state machine. The convolutional neural network has large parameter quantity and requires strong calculation capability, so that the convolutional neural network accelerator needs to provide very large data bandwidth and calculation capability. A large load is subjected to reusable segmentation, and data reuse is realized through a control unit and an address generation unit, so that the calculation delay and the required bandwidth of the convolutional neuralnetwork are shortened and reduced, and the calculation efficiency is improved.

An image processing apparatus converts an input data indicative of a fish-eye image into an output data indicative of a plane image. The apparatus includes an input data conversion unit that converts the fish-eye image into the input data including an input data pixel information and an input data color information; a fish-eye image corrected address generating unit that generates, based on the input data pixel information, an output data pixel information; an input data color information acquiring unit that acquires the input data color information based on a pattern of a color filter array; an output data color information calculating unit that calculates, based on the input data color information, an output data color information; an output data generating unit that generates the output data including the output data pixel information and the output data color information; and a storage unit that stores the input and output data.

Various embodiments are disclosed of a multiprocessor system with processing elements optimized for high performance and low power dissipation and an associated method of programming the processing elements. Each processing element may comprise a fetch unit and a plurality of address generator units and a plurality of pipelined datapaths. The fetch unit may be configured to receive a multi-part instruction, wherein the multi-part instruction includes a plurality of fields. A first address generator unit may be configured to perform an arithmetic operation dependent upon a first field of the plurality of fields. A second address generator unit may be configured to generate at least one address of a plurality of addresses, wherein each address is dependent upon a respective field of the plurality of fields. A parallel assembly language may be used to control the plurality of address generator units and the plurality of pipelined datapaths.

An apparatus and method for scalable qubit addressing. For example, one embodiment of a processor comprises: a decoder comprising quantum instruction decode circuitry to decode quantum instructions to generate quantum microoperations (uops) and non-quantum decode circuitry to decode non-quantum instructions to generate non-quantum uops; execution circuitry comprising: an address generation unit (AGU) to generate a system memory address responsive to execution of one or more of the non-quantum uops; and quantum index generation circuitry to generate quantum index values responsive to execution of one or more of the quantum uops, each quantum index value uniquely identifying a quantum bit (qubit) in a quantum processor; wherein to generate a first quantum index value for a first quantum uop, the quantum index generation circuitry is to read the first quantum index value from a first architectural register identified by the first quantum uop.

A processor includes a memory port for accessing a physical memory under control of an address. A processing unit executing instructions stored in the memory and/or operates on data stored in the memory. An address generation unit (“AGU”) generates address for controlling access to the memory; the AGU being associated with a plurality of N registers enabling the AGU to generate the address under control of an address generation mechanism. A memory unit is operative to save/load k of the N registers, where 2<=k<=N, triggered by one operation. To this end, the memory unit includes a concatenator for concatenating the k registers to one memory word to be written to the memory through the memory port and a splitter for separating a word read from the memory through the memory port into the k registers.

This invention discloses one digital signal processor with bit expend and bit compression unit, which comprises digital signal processor nuclear, order memory and data memory, wherein, the nuclear comprises program control unit, order coding unit, address generation unit and data computation unit. This invention is characterized by the following: the computation unit is integrated with one bit expend and bit compression unit; the unit is composed of timer, displacement device and two source operation number and one aim register device.

A pointer processing apparatus in an SDH transmission system used to serially conducting a pointer process on inputted multiplex data has an address generating unit for allocating an address to each channel of the multiplex data, a RAM for holding an information group obtained by a pointer extracting process and a pointer process, and RAM controlling unit for controlling a sequence of an operation to write-in/read-out the RAM to serially conduct the pointer process on the received multiplex data, thereby largely decreasing the circuit scale, the power consumption, the number of distributions and the like. A POH terminating operation process is conducted in a POH terminating operation process unit, and an obtained result of the POH terminating operation is stored in a storage area for a corresponding channel of a storage unit, whereby the POH terminating operation process can be conducted without separating a multiplex signal into channels. By conducting the POH terminating operation process on the multiplex signal without separating the multiplex signal into channels, a scale and a power consumption of an apparatus can be largely decreased.

In a block interleaving apparatus, a block deinterleaving apparatus, a block interleaving method, and a block deinterleaving method for performing block interleaving and block deinterleaving by using a single plane of a storage unit having a storage area of one block, in order to realized further reductions in circuit scale and power consumption, a comparison reference value of a comparator 123 included in an address generation unit 103 for generating addresses of a storage unit 104 is set at a minimum value which appears in the output of a multiplier 111 and is larger than L×M−1, thereby reducing the scale of the comparator.

Methods and apparatus are provided for efficiently executing extended custom instructions on a programmable chip. Components of a processor core such as arithmetic logic units, program sequencer units, and address generation units are integrated with customizable logic blocks. Various customizable logic blocks can be invoked in a pipelined manner using an available customized instruction set while allowing a processor to continue simultaneous operation. Program counter snooping is also provided to add custom instruction functionality to a processor with no additional provisions for adding custom instructions.

An encoder includes a precoder for encoding an input information object according to a preset encoding scheme and storing the encoded information object in a precoder buffer, a sample number/address generation unit for generating a sample number of each sample and an address, which corresponds to each bit of each sample and the address of the precoder buffer, a multiplexer for selecting a bit of the precoder buffer corresponding to the address generated by the sample number/address generation module, a sampling buffer for storing a bit of each sample output from the multiplexer, a control packet generation module for generating a control packet including information on the sample number generated by the sample number/address generation module, a packet assembling unit for assembling the sample stored in the sampling buffer with the control packet generated by the control data generation module, and a modulation module for modulating the packet output from the packet assembling unit into a sound signal according to a preset scheme.

Provided is directed to a negative word line driver, including: a block select address generation unit for generating first and second block select addresses having a block information according to an active signal; a row decoder controller for generating a control signal to disable a word line; a main word line driver for accessing a main word line by being driven in response to a signal coding the first block select address and the control signal; and a phi X driver for accessing a sub word line by being driven in response to a signal coding the second block select address and the control signal wloff.

An apparatus and method to increase memory bandwidth is presented. In one embodiment, the apparatus comprises a load array having: a first array to store a plurality of load operation entries and a second array to store a second plurality of load operation entries. The apparatus further comprises: a store array having a plurality of store operation entries; a first address generation unit coupled to send a linear address of a first load operation to the first array and to send a linear address of a first store operation to the store array; and a second address generation unit coupled to send a linear address of a second load operation to the second array and to send a linear address of a second store operation to the store array.

A Fourier transform architecture and system for FFT and IFFT processing within multicarrier transceiver is disclosed that includes a programmable butterfly component, a memory and a programmable address generation unit. The architecture includes a butterfly component configured to perform a plurality of radix butterfly calculations, and a four bank memory configured to operate on sample data. The architecture further includes a programmable address generation unit coupled to the pipeline to enable the architecture to perform calculations independent of Fourier-based algorithms. A method for addressing memory banks for an FFT pipeline includes expressing an index in radix notation, computing a bank address for a bank memory, converting the bank address to a reduced size by ignoring one or more bits, and calculating the bank address within the reduced memory bank.

The invention discloses a high efficiency storing method for the coding digit of an LDPC coder based on an FPGA. In the method, the coding digit and external information (or channel information) share one storage block, thereby effectively reducing the requirement of the coder to the quantity of the storage resources; and the coding digit and the external information can be withdrawn when reading the storage block. Therefore, a check equation computational unit PCU and a check node update unit CNU can share one group of address information, and the PCU does not need additional address generation units; and finally, the design method of a step thinning production line is adopted to realize process units VNU, CNU and PCU, thereby effectively reducing the key path delay of the coder and providing a necessary safeguard to improve the throughput of the LDPC coder. When realizing the method with the FPGA, the invention not only can save a large number of the resources for respectively storing the digit and the logic resources for generating addresses which are needed by the PCU, but also can improve the throughput of the coders.