摘要:
This application provides a neural network circuit, including: a first sample-and-hold circuit, a reference voltage generation circuit, a first comparator circuit, and a first output circuit. The first sample-and-hold circuit is configured to generate a first analog voltage based on a first output current output by a first neural network computation array. The reference voltage generation circuit is configured to generate a reference voltage based on a first control signal. The first control signal is determined based on first computation precision. The first comparator circuit is connected to the first sample-and-hold circuit and the reference voltage generation circuit, and is configured to output a first level signal based on the first analog voltage and the reference voltage. The first output circuit is configured to sample the first level signal based on a second control signal, and output a first computation result that meets the first computation precision, and the second control signal is used to control a frequency at which the first level signal is sampled. The sampling frequency of the first output circuit and duration of the first level signal can be adjusted to adapt to computation precision of the neural network. This avoids that a computation precision requirement cannot be met in a case of low precision, and that a waste of power consumption is caused in a case of high precision
摘要:
Embodiments of the present invention provide a communication method, a network device, and a terminal device. The method includes: obtaining, by a network device, state information of current communication with a terminal device; determining, by the network device, a transmission time sequence of the terminal device based on the state information, where the transmission time sequence of the terminal device indicates duration from a time point at which the terminal device receives data to a time point at which the terminal device sends a response to the data; and sending, by the network device to the terminal device, indication information used to indicate the transmission time sequence. In the embodiments of the present invention, the network device can flexibly determine the transmission time sequence of the terminal device based on the state information of the communication with the terminal device, to avoid a prior-art problem caused by a single transmission time sequence, and improve system performance.
摘要:
A radio resource scheduling method, apparatus and system are provided. The method includes: in a case that two or more Absolute Grant (AG) tables exist, a base station obtains an indication sent by a base station control node for indicating which AG table among the two or more AG tables is used by the base station, in which the AG table used by the base station as indicated in the indication is the same as an AG table used by a User Equipment (UE); and the base station performs High Speed Uplink Packet Access (HSUPA) radio resource scheduling for the UE according to the received indication sent by the base station control node. The technical solutions can ensure that the network side and the UE side have the same understanding about the granting information, thereby ensuring correct scheduling of radio resources.
摘要:
Embodiments of the present invention disclose a method and an apparatus for processing channel quality. The method for processing channel quality includes: receiving a channel quality indicator (CQI) value sent by a user equipment; and compensating the received CQI value by using a channel parameter related to channel quality of the user equipment, to enable a signal noise ratio of a channel of the user equipment to converge at a target value. The technical solution provided by the embodiments of the present invention helps increase the transmission rate of user data, improve the Qos for a service, and increase the air interface capacity of a system while lessening load on an uplink channel.
摘要:
A switching method and device are disclosed by the present invention. The switching method according to the embodiments of the present invention includes: receiving from a base station a switching indication information that requires a UE to switch a TTI; sending to the base station a switching confirmation information with which the switching indication information is confirmed; and switching the TTI. The switching method according to the embodiments of the present invention includes: sending to a UE a switching indication information that requires the UE to switch a TTI; and receiving from the UE a switching confirmation information with which the switching indication information is confirmed. The UE according to the embodiments of the present invention includes: a first receiving module, a sending module, and a processing module. The base station according to the embodiments of the present invention includes: a first sending module and a first receiving module. By using the switching method and device according to the embodiments of the present invention, the delay of switching the TTI can be shortened and the TTI can be switched quickly.
摘要:
The present invention provides a transaction method and a node design in a blockchain system. A client initiates a transaction by using an SQL write command including transaction data or queries transaction data by using an SQL query command. A node parses an SQL command, executes a simulated transaction, and checks block data based on a transaction result of the simulated transaction after obtaining the block data. After the check succeeds, the node converts the block data into an SQL write command used to write the block data, thereby writing the block data to an RDB database. In embodiments of the present invention, an RDB data block is connected to the blockchain system, so that data query and processing capabilities of the blockchain system are improved.
摘要:
A method for allocating the number of code words and power according to physical channels carrying data includes: determining initial power and an initial number of code words; adjusting the initial power and the initial number of code words in balanced mode according to resource utilization of data transmission physical channels to select a transport block size (TBS) suitable for data transmission, where the power and the number of code words corresponding to the TBS suitable for data transmission are allocated power and number of code words. An apparatus for allocating the number of code words and power according to physical channels carrying data is also provided. With the present invention, it is guaranteed that the number of code words and the power may be allocated in balanced mode.
摘要:
Embodiments of this application disclose an optical splitting device and an optical splitting system, so that the optical splitting device can be connected to more ONTs, and investment costs and construction difficulty are reduced in a scenario with relatively high user density. The optical splitting device includes a housing, at least one first optical splitter, a multi-core input optical interface, a multi-core output optical interface, and at least one single-core output optical interface. The multi-core input optical interface, the multi-core output optical interface, and the at least one single-core output optical interface are disposed on an outer wall of the housing, the at least one first optical splitter is disposed in the housing, and each first optical splitter includes an input end, a first output end, and at least one second output end. The multi-core input optical interface is connected to an input end of the at least one first optical splitter, the first output end of each first optical splitter is connected to the multi-core output optical interface, and the second output end of each first optical splitter is connected to the single-core output optical interface in a one-to-one correspondence.