30results about How to "Overcoming costly problems" patented technology

The invention discloses a method and a device for feeding back channel state information reports. The method includes the following steps that: a base station configures parameters of a plurality of groups of channel state information (CSI) reports; and the base station informs UE about the parameters of the plurality of groups of channel state information (CSI) reports and instructs the UE to feed back the channel state information (CSI) reports of a predetermined group. The parameters of each group of channel state information (CSI) reports in the plurality of groups of channel state information (CSI) reports comprise a resource allocation parameter of one or a plurality of channel state information reference signal (CSI-RS) resources which are used for channel measurement. The parameters of each group of channel state information (CSI) reports in the plurality of groups of channel state information (CSI) reports further comprise one of the following parameters: distribution information of one or a plurality of interference measurement resources corresponding to each channel state information reference signal (CSI-RS) resource in one or the plurality of channel state information reference signal (CSI-RS) resources; a parameter which is used for indicating the corresponding relationship between one or the plurality of channel state information reference signal (CSI-RS) resources and the interference measurement resources; a parameter which is used for indicating the corresponding relationship between the channel state information reference signal (CSI-RS) resources and channel state information reference signal (CSI-RS) resources used for interfering calculation; and/or a parameter used for indicating the corresponding relationship between one or a plurality of channel state information reference signal (CSI-RS) resources and component carriers in carrier aggregation.

The invention discloses a data reading and writing method and a data reading and writing device based on the implementation of a distributed consistence protocol, computer equipment and a computer storage medium. The method comprises the steps of when any service node receives a data writing request sent by a first client, executing a data writing operation and updating a data version number recorded by the service node, and synchronizing data to be written to the other service nodes, and returning the data version number back to the first client; when any service node receives the data reading request carrying the first data version number sent by the second client, comparing the first data version number with the second data version number recorded by the service node, and responding tothe data reading request according to a comparison result. Therefore, the problem that in the prior art, when the data is read based on the distributed consistency protocol, the expense for reading the data is large due to the fact that more than half of the service nodes need to reach an agreement is solved.

A federated learning global model training method based on differential privacy and quantification comprises the following steps that a central server issues a pre-trained federated learning global model, a local model gradient of each local user is generated, and noise adding, threshold quantification and compression quantification is sequentially performed on the local model gradient of each local user, and the compressed and quantified local model gradient is uploaded to the central server, the central server performs weighted aggregation on the uploaded local model gradient by the central server, updates a global model, issues the global model to each local user, and training is ended when the privacy budget value of each local user is exhausted or the federal learning global model is converged. According to the method, on the premise that the precision of the federal learning global model is not lost, the privacy of local users is protected, the communication overhead in the transmission process is reduced, and the training efficiency of the federal learning global model is improved.

The invention discloses a ruthenium oxide-copper sulfide composite material, which solves defects of high price and less application in industrial production of ruthenium oxide in the prior art. A preparation method of the ruthenium oxide-copper sulfide composite material comprises the steps of (1) ultrasonic dissolving hydrous copper chloride into ethylene glycol, adding thiourea into the above reaction system, stirring for 0.5h-1h, carrying out a hydro-thermal reaction for 8h-12h under a condition of 150 DEG C-200 DEG C after stirring, performing suction filtration, washing and alcohol washing, and drying for 24h at a temperature of 150 DEG C-200 DEG c so as to acquire a copper sulfide nano material; (2) dissolving hydrous ruthenium trichloride into 40ml-60ml of water so as to acquire a ruthenium trichloride solution with the concentration being 10mg/ml-20mg/ml; and (3) adding the copper sulfide nano material prepared in the step (1) into the ruthenium trichloride solution prepared in the step (2), stirring for 0.5h-1h, then carrying out a hydro-thermal reaction for 8h-12h under a condition of 150 DEG C-200 DEG C, performing suction filtration, washing and alcohol washing, and drying for 24h under the condition of 150 DEG C-200 DEG C so as to acquire a ruthenium oxide-copper sulfide composite material.

The invention discloses a local node repair method based on a complete graph minimum bandwidth regeneration code. The implementation steps are as follows: constructing a complete graph minimum bandwidth regeneration code; dividing nodes in the distributed storage system into a plurality of repair groups; constructing a complete graph minimum bandwidth regeneration code in each repair group, packaging the to-be-stored data in blocks and storing the to-be-stored data in each repair group in a coding manner, and when the distributed storage system has single-node failure, connecting a new node with a survival node in the repair group where a failure node is located to complete repair. According to the method, the single failure node in the distributed storage system can be repaired, the repair bandwidth overhead and the disk I/O overhead in the node repair process are reduced, coding and decoding operations are not needed in the repair process, and the repair efficiency is improved.

The invention discloses a configurable and extensible pipelined multiply accumulator and belongs to the field of integrated circuit design. According to the structural design, a control selector is added, the configurable and extensible pipelined multiply accumulator is capable of implementing the calculating functions such as multiply, multiply add and multiply accumulation through different configurations according to the application requirements of users accordingly, and the calculation function of different bit widths can be implemented; meanwhile, a plurality of registers are added according to the structural design, the internal cascade and pipeline structure is realized, the electrical connection of feedback signals are adopted in the entire structural design, cascade of a plurality of configurable and extensible pipelined multiply accumulators can be achieved, efficient parallel accumulation calculation can be implemented, and the application requirements of large-capacity data and high-speed signal processing can be met; meanwhile, the structural design is compact, the calculating speed is increased greatly and efficiency is improved greatly, the extension is flexible, and user configuration is realized.

The invention discloses a smart card carrier tape manufacturing method including forming a catalytic ink graphic layer on a substrate by printing the catalytic ink onto the substrate, forming a metal seeding layer by means of replacement, and performing metal planting and coating planting processes to complete the manufacturing of the smart card carrier tapes. In the smart card carrier tape manufacturing method, manufacturing processes of the smart card carrier tapes in the prior art are simplified and consumption of metal materials and processing cost are reduced; since the etching process is removed, the problem of high material consumption caused by the etching process in the prior art is solved, severe environment pollution caused by a great quantity of waste water generated during the etching process in the prior art is overcome, production cost is reduced while competitive rate of products is increased, the purposes of saving work and materials are achieved while waste water and pollutant are reduced, and accordingly environments are protected while green production is realized.

The invention discloses a press for high-speed pressing of powder metallurgy on the basis of mechanical spring energy storage. The press comprises a press body frame, a heavy hammer lifting mechanism, a heavy hammer impact mechanism, a hammer handle locking and releasing mechanism, a mechanical spring, a suspension part thereof, a spring sleeve, a fastener thereof, an upper die punching and supporting table and a power mechanism, wherein the heavy hammer lifting mechanism sliding along a straight line, the hammer handle locking and releasing mechanism, the heavy hammer impact mechanism and the upper die punching and supporting table are successively connected on a guide rod in series; in the direction vertical to the guide rod, two ends of a shaft pin type sensor pass through and are clamped into the through holes of a suspension part frame; and the middle part passes through to be clamped in a hanger through hole. The high-speed press for powder metallurgy of the invention is suitable for producing powder metallurgy high-speed pressing parts with smaller volume, has small volume, small weight, simple structure, convenient assembly and reliable performance, overcomes the problems of complex structure and high manufacturing cost of the existing hydraulic high-speed pressing equipment, and is easy to popularize and apply in the powder metallurgy industry.

The invention discloses a frequency spectrum detection system positioning method based on an ATDOA algorithm, and the method comprises the steps: at least six base stations are deployed in a three-dimensional space to carry out the positioning monitoring of a radiation source, the coordinates of the radiation source are obtained, the positions of the base stations is fixed and known, and clocks are allowed to be asynchronous among the base stations; the radiation source sends sound pulses at a fixed frequency, can be measured by a monitoring station and moves within a monitoring range at a relatively high speed. The problem of clock synchronization is solved for a monitoring system applying a TDOA algorithm for positioning. An ATDOA algorithm is applied to frequency spectrum monitoring, and the ATDOA algorithm comprises the steps of identifying the frequency of signals sent by a radiation source, obtaining the time difference of the signals sent by the radiation source received by a plurality of observation stations twice, distributing tasks of the observation stations and a central station in the process, and solving obtained data information through a genetic algorithm. Therefore, strict clock synchronization required by the TDOA method is avoided, the positioning precision of the system is effectively improved, and the positioning power consumption is reduced.

The invention provides a device and method for treating ship sewage by combining a modified non-woven fabric with a circulating activated sludge method. The device comprises a water inlet pump, a backflow pump, an aeration pump, a drainage pump, a biological selection region, a pre-reaction region, a main reaction region and aeration heads, and further comprises a modified non-woven fabric filtering component, the modified non-woven fabric filtering assembly is a plate frame type assembly and is obliquely arranged at the rear section of the main reaction region; the modified non-woven fabric is prepared by taking polypropylene non-woven fabric with the pore diameter of 30 microns as a base film, adopting a surface coating method and using glutaraldehyde cross-linked polyvinyl alcohol for coating the base film; the circulating activated sludge method is combined with a membrane separation technology, and the modified non-woven fabric film is used as a substitute film component, good andstable treatment effects on the domestic sewage of the ship are achieved, the structure is simple, the equipment is durable, maintenance is convenient, and operation is automatic. The non-woven fabric film can effectively reduce the cost of the film assembly, and the film pollution resistance is improved after modification and is very suitable for ship domestic sewage treatment.

The invention discloses a cooling chip module and belongs to the field of structural design of cooling chip modules. According to the cooling chip module, an upper cooling chip and a lower cooling chip which adopt completely the same structure are designed, so that the problem that in the prior art, the cooling chip module production cost is relatively high because an upper cooling chip and a lower cooling chip are of different structures is solved, the upper cooling chip and the lower cooling chip can be manufactured only by one die and one production line, accordingly, the cost in producing the cooling chip module is greatly lowered, and further, the manufacturing cost is lowered for manufacturers.

The invention provides a wavelength allocation method and system. The method comprises the following steps of S1.dividing a user traffic request behavior period into a plurality of switching periods;S2.grouping all ONUs within each switching period to form a plurality of ONU communities by an optical line terminal (OLT) according to the traffic request bandwidth of each optical network unit (ONU); and S3.allocating the corresponding available wavelength to each ONU community. According to the method and the system, each user traffic request behavior period is divided into the switching periods according to the allocation condition of the request traffic bandwidth, and all ONUs are grouped within each switching period to form the ONU communities, wherein the ONU communities in each switching period and the wavelength allocated to each group of ONUs are changeless, so that the switching times of a user can be minimized and the tuning expenditure can be reduced.