83results about How to "Significant performance gain" patented technology

A processor generation system includes the ability to describe processors with three instruction sizes. In one example implementation, instructions can be 16-, 24- and 64-bits. This enables a new range of architectures that can exploit parallelism in architectures. In particular, this enables the generation of VLIW architectures. According to another aspect, the processor generator allows a designer to add a configurable number of load/store units to the processor. In order to accommodate multiple load/store units, local memories connected to the processor can have multiple read and write ports (one for each load/store unit). This further allows the local memories to be connected in any arbitrary connection topology. Connection box hardware is automatically generated that provides an interface between the load/store units and the local memories based on the configuration.

An interference suppression scheme is provided for a radio receiver. According to the provided interference suppression scheme, the bandwidth of a received pilot signal and a data signal is divided into a plurality of frequency sub-bands. The pilot signal and the data signal have been transmitted according to single carrier data transmission technology. Interference parameters are calculated for each frequency sub-band separately. Interference suppression may be carried out jointly or separately for each frequency sub-band. After the interference suppression, the frequency sub-bands are combined. A filter bank may be used for dividing the total frequency band into sub-bands.

The invention discloses a channel information quantized feedback method and device and a data precoding method and device. The channel information quantized feedback method comprises the following steps: receiving a channel measuring pilot signal from a base station; performing channel measurement according to the channel measuring pilot signal to obtain channel information; calculating CMatrix1 and CMatrix2 by adopting the channel information, wherein the CMatrix1 and the CMatrix2 are used for quantifying the channel information or instructing a base station to perform precoding; and feeding back first indicating information of the CMatrix1 and/or second indicating information of the CMatrix2 to the base station. According to a technical scheme provided by the invention, the influence of polarization leakage is considered in feedback design, so that a remarkable performance gain exists under the practical situation of polarization leakage.

A system, terminal, receiver and method are provided for receiving and processing a signal received from a spread-spectrum downlink channel. The method includes receiving the signal from the downlink channel, and generating an estimate of a transmitted chip sequence in accordance with an iterative process. In this regard, the iterative process includes calculating statistical information, such as a mean and covariance, for a previous iteration of a transmitted chip sequence, the statistical information being selectively calculated based upon one of soft bits and soft symbols associated with the previous iteration of the transmitted chip sequence. The estimate for a current iteration of the chip sequence is then generated based upon the statistical information.

The invention discloses a data transmission method and a base station in a multi-base station coordinated system, comprising that: when performing user scheduling for a plurality of times, with respect to the terminal of a current transmission retransmission data package in the district, the base station selects the channel information which has large response relativity with the local district terminal channel, small response relativity with the adjacent district terminal channel, and causes the terminal to obtain highest scheduling priority as the channel information corresponding to the terminal; then the base station selects the terminal with the highest scheduling priority to perform data transmission according to the channel information of each terminal in the local district. In the invention, each base station independently performs user scheduling, and realizes partial avoiding of multi-base station interference through updating the channel information.

The invention provides a chrominance component prediction method in a hybrid video coding standard, and belongs to the field of video encoding. The purpose of the invention is to effectively eliminate the correlation between a chrominance component, a luminance component and another chrominance component in videos, and provides the chrominance component prediction method in the hybrid video coding standard, so as to further improve the performance of video coding. The chrominance component prediction method comprises the following steps: when performing chrominance component prediction through a linear model, using pixel points reconstructed around a current coding block to calculate parameters of the linear model; when performing parameter calculation, increasing the surrounding effective pixel points and removing the surrounding unusable pixel points; predicting a Cb chrominance component according to the obtained parameters through a reconstructed luminance component Y; and predicting a Cr chrominance component through the reconstructed luminance component Y and the reconstructed chrominance component Cb. According to the chrominance component prediction method in the hybrid video coding standard provided by the invention, model parameters are calculated through the effective pixel points surrounding the block, and the chrominance component Cr is predicted adaptively through the luminance component Y and the chrominance component Cb, so that the code efficiency is further improved.

A method and system of allocating a bandwidth and a power for a wireless user. The method includes for an allocated bandwidth determining a transmission power and adjusting the transmission power based on whether a bit rate and an error rate associated with the transmission power and the allocated bandwidth satisfy one or more predetermined criteria. The method further processing information associated with the adjusted transmission power and determining a power level indicator. Additionally, the method includes determining whether the power level indicator satisfies one or more predetermined third criteria. If the power level indicator is determined not to satisfy the one or more predetermined third criteria, the method includes adjusting the allocated bandwidth from the first bandwidth to a second bandwidth for the wireless user. Otherwise, the method includes keeping the allocated bandwidth.

A method for scheduling messages transmitted on a forward common channel of a CDMA system. The method includes determining the ratio between the slotted and unslotted messages transmitted over the forward common channel over a period of time. For each time slot S(i) of the forward common channel, s % time is allocated from the duration of the time slot to slotted messages and u % time is allocated from the duration of the time slot to unslotted messages. The s % time and u % time are determined based on the ratio between the slotted and unslotted messages transmitted over the forward common channel over a period of time.

A method for cooperative relay beamforming based on limited feedback is applicable to a communication system. The communication system comprises a transmitter, a plurality of relays and a receiver. The transmitter is provided with a plurality of antennae, each relay is provided with a single antenna, the receiver is provided with a plurality of antennae, and the plurality of the relays cooperate and adopt an amplification and forwarding protocol; the whole communication process is finished in two time slots, wherein in the first time sot, after performing transmit beamforming on a data flow, the transmitter broadcasts the data flow to all relays; and in the second time slot, all relays determine distributed beamforming code words according to limited feedback information containing code word indicators and power amplification factors from the receiver and forward receipt signals subjected to respective code word component treatment to a information sink; and the design of a code book is completed by code book design and cord word updating algorithm, the relays select code words which can maximize the instant signal to noise ratio of the receiver from the code book according to each channel condition as a beamforming vector.

The invention discloses a data model dual-drive joint MIMO channel estimation and signal detection method. Channel estimation and signal detection are carried out through a channel estimator and a signal detector. The signal detector takes the channel estimation result and the received data signal as the input of each layer of network, and the tth layer of network is combined with the output of the (t-1) th layer of network to calculate an error variance vector; the tth layer of network calculates and obtain external information according to the input parameters to be trained, the error variance estimation vector and the linear estimator, obtains a signal estimation value by adopting an unscattered estimation function according to the external information, and outputs and transmits the signal estimation value to the next layer of network; and the estimated value of the final sending symbol is output by the Tth layer of network. Meanwhile, the signal detector feeds back the estimated sending symbol to the channel estimator, so that the channel estimation result is further improved. According to the invention, the network performance can be greatly improved, dynamic updating and network self-adaption are realized, the performance of the receiver can be improved, and remarkable performance gain is obtained on the basis of a traditional iterative receiver.

The invention provides a time division duplex (TDD) transmission method and a base station and a user device which both support the method. The method comprises the following steps: channel estimation is carried out by a base station on a down link from the base station to a user device; the base station establishes a precoder aiming at the user device according to an estimated down link channel; the base station sends a channel quality indication reference signaling (CQI-RS) which has been subjected to precoding by the precoder to the user device, and sends indication information in first physical down link control channel (PDCCH) and/or an enhanced physical down link control channel (ePDCCH); the indication information indicates information needed by the user device in detecting the pre-coded CQI-RS; the user device detects the pre-coded CQI-RS, and the corresponding CQI is exported; and the user device gives a feedback of the exported CQI to the base station on a PUCCH. The beneficial effects of the technical scheme of the present invention are that: advanced multi-user multiple-input-multiple-output (MU-MIMO) and/or coordinated multi-point (CoMP) operation can be supported, and the feedback of the CQI can be accurately and timely given, so that compared with existing TDD transmission methods, the TDD transmission method of the invention has a considerable performance gain; and the TDD transmission method of the invention is low in signaling cost and simple to realize.

A method for dividing a display into zones at system initialization for tracking which zones have any pixels revised so that, when the time comes to update the display, only the zones requiring revision (that is, those zones in which any pixel has been revised) are copied from shadow memory to the frame buffer for display on the display device. The memory for tracking these zones can be allocated at initialization and held since it is relatively small. Consequently, a significant performance gain may be achieved by avoiding the shortcomings of the existing methods in the art notwithstanding the fact that some “clean” pixels in each zone having even a single changed pixel are also rewritten to the frame buffer.

The invention relates to a method for cooperating relays to form beams based on limited feedback, which is applicable to a communication system. The communication system comprises an transmitter, a plurality of relays and a receiver; the transmitter is configured with a plurality of antennas; each relay is configured with single antenna; the receiver is configured with a plurality of antennas; a plurality of relays cooperate mutually and adopt amplifying and retransmission protocol; the whole communication process is finished in two time slot; in a first time slot, the transmitter broadcasts data flow to all the relays after the transmitting beams are formed; in a second time slot, all the relays determine the code words formed by distributed beams according to limited feedback information containing code word labels and power amplified factors from the receiver; then the plurality of relays transmit the received signals to information sink after component handling of respective code word; the design of a codebook is finished by a codebook design and code word updating algorithm, the relays select code words which can lead the instantaneous signal to noise ratio of the receiver to be maximum as beams according to each channel condition so as to form vectors.

The invention discloses a multi-class leukocyte automatic identification method based on a deep convolutional neural network, and belongs to automatic identification of a medical microscope image by using a deep learning machine vision scheme. The method comprises the following steps: firstly, carrying out arrangement and data enhancement operation on a data set of multiple types of leukocytes; then, using an inception module to perform cascading to form a multi-scale feature fusion convolutional neural network model; during cascade combination, establishing the mutual dependence relationshipbetween the characteristic channels through continuous squeeze operation and excitation operation, so that the network performance is improved, and the accuracy of leukocyte classification is improved.

According to the image classification processing method of the neural network based on frequency domain wavelet basis processing, frequency domain analysis is combined into image classification processing, operation is simple, classification precision is high, accuracy is good, and the method is closer to image processing of human vision. The method comprises the following steps: step 1, decomposing information in a natural image into three groups according to frequency band distribution by utilizing multistage discrete wavelet transform; and step 2, gradually injecting the obtained three groups of information into the neural network from low frequency to high frequency. An input image is decomposed through given wavelet transform, and coefficients of wavelet transform are gradually fed into different depth map layers of a deep neural network according to decomposition levels. Furthermore, while feed-in is carried out, an attention module is used for promoting fusion of neural networkfeatures and injection information, so that remarkable performance gain is generated. Therefore, how the information of different frequencies affects the accuracy of the deep neural network is obtained.

Propagation channels are reconstructed from measurements in disjoint subbands of a wideband channel of interest. By using high-resolution estimation of multipath parameters, and suitable soft combining of the results, a channel estimate and subseqeuntly channel models can be extracted that accurately interpolate between the measured subbands.

The embodiment of the invention discloses a method for issuing assignment information, and comprises the step of generating assignment information carrying with a distinguishing mark according to the feedback channel quality information (CQI) sent by a terminal, wherein the distinguishing mark is used for distinguishing whether modulation coding mode (MCS) is corresponding to the last feedback CQI and sending the assignment information. The embodiment of the invention also discloses a device and a communication system for issuing assignment information. The embodiment of the invention utilizes the MCS domain in the original assignment information or the whole information content to transmit new information, thus saving expense for resource assignment information, and having obvious gain in performance compared with issuing no or less resource assignment information as the MCS is adjustable.

The invention discloses a high-precision indoor wireless positioning method. The method comprises the following steps: A, obtaining distances from all N anchor nodes to a measured node and coordinates of the anchor nodes at a moment t; B, mining spatial correlation of adjacent moments through Kalman filtering, defining Kalman state equation coefficients on three coordinates of x, y and z at the moment t, and summarizing vectors and matrixes at all moments into a specified set form; C, performing factor decomposition on the global probability distribution by utilizing a total probability formula and a Kalman state equation; D, modeling by utilizing the factor decomposition obtained in the step C, and dividing the factor graph model into a lattice network part, a Kalman network part and a Kalman parameter estimation part; and E, carrying out message calculation on the lattice network part, the Kalman network part and the Kalman parameter estimation part through a message transmission algorithm. According to the method, the high-precision position information of the indoor to-be-tested node can be obtained efficiently and accurately.

The invention discloses a concatenated code iterative decoding method for correcting synchronization and substitution errors. The method comprises the steps of generating an expanded grid diagram capable of describing a synchronization error, decoding each frame of received data through utilization of a bidirectional viterbi decoding algorithm based on the expanded grid diagram, and carrying out RS code decoding; and reinitializing convolutional codes through utilization of accurately decoded RS code symbols, taking the expanded grid diagram after the synchronization error occurs into consideration, and further carrying out viterbi decoding and RS code decoding. According to the method, the expanded grid diagram is effectively clipped through adoption of the accurately decoded RS code symbols, so the complexity of the grid diagram is reduced, the integrated iterative decoding complexity is reduced, and the superior error correction capability is achieved.

The invention relates to the technical field of communication, in particular to adaptive data retransmission method and device. The method includes: a data receiver decodes received data and transmits a retransmission request to a base station in case of decoding errors; and after receiving the retransmission request, the base station selects an optimal retransmission coding matrix from a space-time coding matrix set according to optimal retransmission selection rules and respectively transmits the retransmitted data, newly transmitted data and an optimal preprocessing matrix number to relays, and the relays load the retransmitted data and the newly transmitted data transmitted by the base station into the optimal retransmission coding matrix in a distributed manner and forward the data to the data receiver. The method guarantees full diversity, high rate and linear decoding complexity of a retransmitted data stream and also guarantees real-time transmission of another data stream to be insusceptible to retransmission of the data stream, and resource utilization rate is high.

The invention relates to an adaptive collaborative transmission method for backhaul links of a wireless relay. When a plurality of users access a base station through the wireless relay, if the resource occupation of the backhaul links exceeds a preset threshold, the relay carries out network coding processing on partial direct information of users and partial forwarded information of the users, then, sends the information to the base station by adopting single wireless resources, so that the resource occupation of the backhaul links is reduced at the same that the number of target users of relay service is increased, and the transmission performance of a wireless relay system is improved; and the base station further carries out centralized scheduling on users who adopt the network coding processing, and detects and recovers received network coding mixed information and direct information from the users, and the diversity gain can be obtained without increasing additional system overhead. With the adoption of the method, the problem that the resources of the backhaul links of a new-generation broadband mobile communication system are limited is solved, the wireless relay is ensured to be capable of expanding coverage and reducing the decrease of the spectrum efficiency caused by the introduction of the relay.

The invention discloses a precoding method and device cooperating with a multipoint transmission system. Signal receiving quality of cooperation UE and cooperated UE is guaranteed. The method includes the steps that a cooperation base station collects signal precoding information of the cooperation UE and interference precoding information of the cooperated UE; the signal precoding information is adopted by the cooperation base station to find precoding matrixes in a codebook, and the precoding matrixes are used as signal vectors; the cooperation base station determines comprehensive interference null space by means of the interference precoding information; the cooperation base station determines the latest precoding matrix of the cooperation UE according to the spatial relationship between the signal vectors and the comprehensive interference null space. The device is located on the cooperation base station and comprises a collecting module, a signal vector obtaining module, an interference null space determining module and a precoding matrix determining module. By the adoption of the method and device, the signal receiving quality of the cooperation UE is guaranteed, interference of the cooperated UE is also reduced, and the signal receiving quality of the cooperated UE is ensured.

The invention discloses a double-sided power generation double-glass assembly. The assembly is provided with front glass, a front hot melting adhesive film, a solar battery string, a back hot melting adhesive film and punching glass in sequence from top to bottom; a plurality of double-faced battery pieces are serially arranged on the solar battery string; thin grid lines are arranged on the surfaces of the double-faced battery pieces; thick grid lines are arranged at two ends of the front face and the back face of each double-faced battery piece; the heads and tails of the adjacent double-faced battery pieces are overlapped mutually to form a cascaded structure; and the thick grid lines are connected with a junction box on the punching glass through bus bars. According to the double-glass assembly structure disclosed by the invention, an N-type PERT double-faced power generation battery is adopted, and the adjacent batteries are connected together in a manner that the heads and tails are overlapped in virtue of a conducting material, so that the output power and conversion efficiency of the assembly can be obviously improved.

The invention discloses a multi-user sending precoding design method in visible light communication, and belongs to the field of wireless optical communication. According to the method, the nonlinearity and illumination requirements of the LED are comprehensively considered, and a multi-user sending precoding algorithm under optical constraint is established, so that multi-user interference can besuppressed while the expected indoor illumination level is kept, and the method is more suitable for being applied to a multi-user visible light communication system. According to the method, optimalsending precoding conforming to actual constraints is formulated, the purpose of maximizing the minimum signal-to-interference-plus-noise ratio between users is achieved, and compared with a traditional zero-forcing algorithm based on pseudo-inverse, the method has obvious performance gain; furthermore, due to the fact that the multi-user precoding sending design problem in visible light communication is quasi-convex, the optimal solution of the precoding sending matrix can be effectively obtained through the dichotomy, meanwhile, the direct current bias voltage needed by the expected dimminglevel is obtained, and the dimming control function is achieved.