47results about How to "Reduce the amount of feedback information" patented technology

The invention discloses a weighted broadcast retransmission method based on network coding. The weighted broadcast retransmission method based on network coding comprises the following steps of (1) building a weighted packet distribution matrix according to packet state information and link state information fed back by packets of receiving nodes, (2) selecting coded packets on a source node on the basis of the weighted packet distribution matrix, (3) sending the selected coded packets to the receiving nodes in a broadcast retransmission mode after the source node carries out XOR coding on the selected coded packets, (4) carrying out decoding on the coded packets through the receiving nodes after the receiving nodes receive the coded packets, feeding the state information of the decoded packets and the link state information of the decoded packets back to the source node, and (5) using the source node to judge whether transmission is accomplished or not according to the feedback information of the receiving nodes, and carrying out coding retransmission again when the transmission is not accomplished. According to the weighted broadcast retransmission method based on network coding, high transmission efficiency can be achieved through the low computation complexity in the network environment with an unbalanced link state, and the weighted broadcast retransmission method based on network coding can be widely applicable to various broadcast networks with the feedback function.

The invention discloses a pre-coding method based on a channel information covariance. The pre-coding method comprises the following steps that global channel information in a multi-cell multi-antenna interference communication system is measured to obtain a channel covariance matrix and system noise power, and the channel covariance matrix and the system noise power are transmitted to a first base station to process uniformly; the first base station carries out setting of a current iteration index and initializes the channel covariance matrixes of all users according to the number of the users of the system; the optimal transmitting power and maximum signal to interference and noise ratios of all users are computed one by one according to the covariance matrix corresponding to each user; and the covariance matrixes corresponding to all users and the current iteration index are updated until the SINR (Signal to Interference and Noise Ratio) converges, the first base station obtains an optimal covariance matrix set of the system and a corresponding optimal pre-coding matrix set, and transmits all pre-coding matrixes to other corresponding base stations. The invention also provides a cooperative communication method, which has the advantages of small amount of information and less system expenditure.

The invention discloses a method and a system for realizing antenna selection in an MIMO relay system. Antenna selection is respectively carried out on a relay node and a source node according to a signal to noise ratio; the corresponded relay node and the corresponded source node are respectively notified by the selected transmitting antennas; and then the relay node and the source node respectively use the notified transmitting antennas to carry out signal emission. According to the method and the system provided in the invention, it can be ensured that a relay node and a source node only select one antenna among a plurality of antennas for signal emission, wherein the relay node and the source node are utilized as emission nodes; and a feedback information contents, which is obviously less than that in a traditional technical scheme, can be generated, so that a design complexity can be effectively reduced.

The invention discloses a method for sensing an interception node selected by a radio system. The radio system comprises a plurality of base stations and a plurality of user station equipment, wherein the base stations are used as judgment nodes, and the user station equipment is used as interception nodes. The method comprises the following steps: the judgment notes count the N time interception information of the whole interception node and groups the interception nodes according to the relativity of the interception information; detection judgment is made according to the interception information of the whole interception node at the stage; at least one interception node is selected from each group of the interception nodes as a group header interception node to feed back P time information to the judgment notes; and the judgment notes carry out data fusion processing and make detection judgment according to the selected feedback interception information of the group header interception node. The invention provides an interception mechanism based on the relativity of CPEs historic interception information statistics so as to reduce the number of CPEs participated in feedback, thereby effectively reducing the feedback information amount when the detection probability is not reduced, and the error alarming probability is not improved.

The invention relates to the technical field of multi-cell mmWave large-scale MIMO beamforming, and discloses a multi-cell mmWave large-scale MIMO beam selection method based on a beam discovery signal. A user performs related operations on a BDS and user local ZC sequence set constructed by a ZC sequence in each beam selection period one by one so as to acquire a ZC sequence root index corresponding to a user optimal beam; and then a received signal is descrambled to acquire a service mmWave cell ID corresponding to the optimal beam; finally, the user feeds back a determined beam index to a service cell; and meanwhile, the invention designs a multi-cell partial beam interference inhibition algorithm aiming at a beam interference problem between mmWave cells, the user processes the received signal so as to acquire the optimal beam and a strong interference beam of the user, the complexity of a beam selection algorithm is greatly lowered, and the inter-cell beam interference is effectively lowered.

The embodiment of the invention provides a text sequence generation method and system. A adversarial network model comprises a generator and a discriminator; the generator comprises a feature feedbacknetwork module and is used for calling high-order features from the discriminator so as to generate a guidance feature vector; the generator generates a next word of the pre-output text sequence based on the guidance feature vector; the discriminator updates the high-order features according to the newly constructed pre-output text sequence and the initial text sequence; the above steps are iteratively executed until the total number of words of the newly constructed pre-generated text sequence reaches a set threshold, the total number of words is input into a discriminator for discrimination, and an update gradient is calculated according to a classification discrimination result; parameters of the generator are updated according to the updating gradient; and the above steps are iteratively executed until the generator and the discriminator converge, and a target text sequence is output. According to the embodiment of the invention, the feature conversion feedback module is added, and the feature vector extracted by the discriminator is transmitted to the generator as a guide signal, so that the generation quality of the generator is improved.

The invention discloses a large-scale MIMO collaborative user scheduling method for mainly solving the problem that in the prior art a spectrum efficiency of a system is reduced as state information of a channel is compressed or quantified. The implementation steps of the method are as follow: 1, a base station sends training sequences to users, and the users estimate channel information according to received signals and the sent training sequences; 2, the users in the system are grouped, and a timer is set for the users; 3, the user with the maximal channel norm is found by the timer, and the channel state information of the user is broadcasted to the other users in the same group; 4, correlation coefficients between the other users in the same group and the user with the maximal channel norm are computed, and the users of which the correlation coefficients are less than a given threshold of the system feeds back the channel state information of the users to the base station; and 5, the base station computes a set of users to send, and sending signals after being subjected to beam forming are sent to the users. According to the method, under the premise of small influence on the system and speed, the feedback quantity of the system is effectively reduced, and downlink resources of the system are effectively improved.

The present disclosure relates to a HARQ feedback method and device, a method and device for determining data to be retransmitted, a data receiving device, a data transmitting device, and a computer readable storage medium. The HARQ feedback method includes: grouping a carrier unit CC according to a binding rule, wherein the binding rule includes the number of resource units included in a CC thatcan be bound into a group; calculating a parameter value in each CC group; generating a HARQ codebook with the same number as the CC groups, wherein the length of the HARQ codebook is determined by the parameter value in the corresponding CC group and the maximum resource unit number included in a single CC in the corresponding CC group; and feeding the HARQ codebook back to a data transmitting end. In the embodiment of the present disclosure, by grouping the carrier units CC according to the binding rule, calculating the parameter value in each CC group, and generating the HARQ codebook having the same number of CC groups, the amount of feedback information is optimized, and the amount of feedback information is reduced.

The invention relates to a channel information feedback method of a beam shaping system with a Mt*2MIMO (Multiple Input Multiple Output) characteristic, which only needs to feed back characteristic values lambda 1and lambda2 and a characteristic vector v1. The method particularly comprises the steps of: estimating a channel matrix H of a descending wireless channel by a user equipment terminal MS (Memory System); carrying out SVD (singular value decomposition) on the H by the user equipment terminal MS to obtain the lambda 1, the lambda 2 and the v1; quantizing and feeding back the lambda 1, the lambda2 and the v 1 by the user equipment terminal MS; solving a v 2 through v2=eigvec2[HHH] by using a formula HHH=lambda 1v1v1H+lambda2(E-v1v1H) by a base station BS on the basis of the feedback information quantity of the user equipment terminal MS; carrying out beam forming by adopting V=[v1H, v2H]H by the base station BS; and completing the channel information feedback method of the beam shaping system with the Mt*2MIMO characteristic. The invention ensures that the MIMO beam shaping system obtains greatest system capacity and space diversity gain and simultaneously reduces channel feedback information capacity, improves the transfer effectiveness of the MIMO beam shaping system, and enhances the practicability of the MIMO beam shaping system.

The invention discloses an interference alignment method and system. The interference alignment method comprises that each base station determines an interference matrix according to the channel interference condition of each user equipment (UE) at present, and works out an emission precoding matrix correlated to strong interference according to the interference matrix, and emits a signal to the UE by means of the emission precoding matrix; after receiving the signal, each UE estimates a CSI and obtains a CSI estimation result of another UE; an emission precoding matrix in interference alignment is estimated by means of all channel state information estimation results obtained at present through a maximum SINR method; a codebook which is most strongly correlated to the estimated emission precoding matrix is determined, and an index of the codebook is reported to the base station; the base station reconstructs the precoding matrix according to the received index, and takes the reconstruction result as a new emission precoding matrix of the base station; the UE is triggered to estimate the CSI; and each UE works out a reception processing matrix of the UE through the maximum SINR method according to the CSIs estimated by the UE and the other UE. According to the invention, advantages of interference alignment can be effectively achieved, and the interference alignment method andsystem are easy to implement.

The invention discloses a multi-channel circulating oil supply valve. The oil supply valve comprises a rotating core and a stator arranged on the rotating core in a sleeved mode, the outer wall of the rotating core is in liquid sealing fit with the stator, a metering cavity is formed in the stator, a piston is arranged in the metering cavity and divides the metering cavity into a first cavity and a second cavity, 4N liquid passing openings are formed in the peripheral side of the inner wall of a rotating core cavity at equal intervals, 4N liquid outlets are formed in the peripheral side, corresponding to the section where the liquid outlets of a liquid-outlet pipe are located, of the inner wall of the rotating core cavity at equal intervals, 2N outlets are formed in the stator, every two opposite liquid outlets communicate with one outlet through a pipeline, and a rotating control device controls the rotating core to rotate 360/4N degrees each time. According to the oil supply valve, an oil-outlet annular groove in the outer diameter of an original rotating core is omitted, then multiple sets of oil-outlet pore channels are formed in the position of the original annular groove, namely the stator (a valve body) matched with the original annular groove, and one-valve multi-point quantitative oil supply can be achieved through the multiple sets of oil-outlet pore channels.