Abstract: Embodiments of the present invention provide an end-to-end lightweight method and apparatus for license plate recognition. The method comprises: obtaining an image to be recognized; obtaining a number of a license plate in the image to be recognized and position coordinates of the license plate in the image to be recognized on the basis of the image to be recognized and a pre-trained target license plate recognition model, wherein the target license plate recognition model comprises a target feature extraction network, a target region candidate localization network, a target super-resolution generation network and a target recurrent neural network. Because in this solution, once an image to be recognized is input into the target license plate recognition model, the target license plate recognition model can output the license plate number and position coordinates of the license plate in the image to be recognized, one realizes an end-to-end model.

Abstract: A method for feeding back a HARQ result includes that: a mapping relationship between each downlink carrier group and each uplink carrier group is determined; a target manner for acquiring a target uplink carrier from the each uplink carrier group is determined; and the mapping relationship and the target manner are sent to the terminal to enable the terminal to feed back a target HARQ result through the target uplink carrier after determining the target uplink carrier in a present uplink carrier group according to the target manner.

Abstract: The embodiment of the present application provides a method and apparatus for spectrum prediction in a cognitive radio network. The method comprises: receiving a target access request from a secondary user; obtaining real-time occupancy information of bands through a subspace filtering-based spectrum sensing approach; updating a spectrum resource database of radio environment maps (REM) according to the real-time occupancy information of hands; establishing a prediction model for bands occupied by primary users according to the updated spectrum resource database of radio environment maps (REM); and allowing an access of the secondary user to a target band according to the prediction model for bands occupied by primary users. The method and apparatus provided by embodiments of the present application allow a secondary user to have a targeted spectrum access. The spectrum utilization efficiency of the radio network can thus be improved.

Abstract: A sensing recognition method and device based on wireless communication signals are disclosed. The method comprises the steps of obtaining channel state information from a received wireless communication signal; extracting a channel state feature value from the channel state information; and outputting a sensing result mapped with the channel state feature value according to the channel state feature value. The disclosed method and device can improve the accuracy of sensing recognition and achieve better recognition effect.

Abstract: A method for controlling interference includes: dividing a control region in a time domain transmission unit to obtain first and second power regions, the transmission power of the first power region being higher than that of the second power region; dividing all current channels to obtain first and second channel groups; transmitting data carried by all the channels included in the second channel group by means of the first power region, and transmitting data carried by all the channels included in the first channel group by means of the second power region. The method can greatly reduce interference to a PDSCH of a small cell from a PDCCH of a macro base station when the macro base station and the small cell use different parameter sets on the same carrier in a heterogeneous network, without reducing the spectrum utilization rates of the macro base station and the small cell.

Abstract: A method and an apparatus for deploying a network slice in a mobile communication system are disclosed. The method comprises the steps of mapping an underlying infrastructure network into an underlying network layer in a complex multilayer network model, and mapping a network slice into a network layer on the underlying network layer; and deploying the network slice based on the complex multilayer network model. In the process of deploying a network slice in a mobile communication system, the invention can accurately and flexibly manage the network resources and enhance the utilization of network resources, thereby avoiding the resource competition and ensuring the network communication quality.

Abstract: Embodiments provides a low-latency data switching device and method. The device includes at least two boundary hardware modules and at least one optical forwarding module. The first boundary hardware module is configured for matching, with a first data forwarding table, signaling message information for a signaling message sent by a host router, acquiring a service request sent by the host router, and searching a second data forwarding table for an optical channel for the service request. The at least one optical forwarding module is configured for mapping the service request to a second service request, and forwarding the second service request in sequence via the optical channel, the second boundary hardware module is configured for acquiring next-hop routing information of the data switching device for the service request, and forwarding the service request to the next-hop router of the data switching device.

Abstract: Disclosed is a method for processing downlink signalling of an SDN virtualization platform based on OpenFlow. The method comprises: after the preprocessing of downlink signalling is executed, further executing same, so as to judge whether more Match items also exist in a Match item list; if it is judged that more Match items also exist, acquiring an item from the Match item list; if it is judged that there are no more Match items, ending the processing; after completing the step of acquiring an item from the Match item list if it is judged that more Match items also exist, further judging whether more Action items also exist; if there are no more Action items, combining a new Action item list with the Match items to generate downlink signalling, and issuing same to a virtual network switch; and returning to the step to continue judging whether more Match items also exist; and if more Action items also exist, acquiring the next Action item.

Abstract: A sensing recognition method and device based on wireless communication signals are disclosed. The method comprises the steps of obtaining channel state information from a received wireless communication signal; extracting a channel state feature value from the channel state information; and outputting a sensing result mapped with the channel state feature value according to the channel state feature value. The disclosed method and device can improve the accuracy of sensing recognition and achieve better recognition effect.

Abstract: A method and an apparatus for deploying a network slice in a mobile communication system are disclosed. The method comprises the steps of mapping an underlying infrastructure network into an underlying network layer in a complex multilayer network model, and mapping a network slice into a network layer on the underlying network layer; and deploying the network slice based on the complex multilayer network model. In the process of deploying a network slice in a mobile communication system, the invention can accurately and flexibly manage the network resources and enhance the utilization of network resources, thereby avoiding the resource competition and ensuring the network communication quality.

Abstract: A hand motion identification method includes: obtaining a to-be-identified video; performing area-locating and tracking of a hand for the to-be-identified video, and extracting a red-green-blue (RGB) video and a depth information video of the hand; detecting the RGB video and the depth information video of the hand, to obtain a feature point; representing the feature point by using a 3D Mesh motion scale-invariant feature transform (MoSIFT) feature descriptor; and comparing the 3D Mesh MoSIFT feature descriptor of the feature point with a 3D Mesh MoSIFT feature descriptor in a positive sample obtained through beforehand training, to obtain a hand motion category in the to-be-identified video.

Abstract: The present disclosure provides a three-dimensional facial recognition method and system. The method includes: performing pose estimation on an input binocular vision image pair by using a three-dimensional facial reference model, to obtain a pose parameter and a virtual image pair of the three-dimensional facial reference model with respect to the binocular vision image pair; reconstructing a facial depth image of the binocular vision image pair by using the virtual image pair as prior information; detecting, according to the pose parameter, a local grid scale-invariant feature descriptor corresponding to an interest point in the facial depth image; and generating a recognition result of the binocular vision image pair according to the detected local grid scale-invariant feature descriptor and training data having attached category annotations. The present disclosure can reduce computational costs and required storage space.

Abstract: The present application discloses a vehicle searching method and device, which can perform the steps of: calculating an appearance similarity distance between a first image of a target vehicle and several second images containing the searched vehicle; selecting several images from the several second images as several third images; obtaining corresponding license plate features of license plate areas in the first image and each of the third images with a preset Siamese neural network model; calculating a license plate feature similarity distance between the first image and each of the third images according to license plate feature; calculating a visual similarity distance between the first image and each of the third images according to the appearance similarity distance and the license plate feature similarity distance; obtaining a the first search result of the target vehicle by arranging the several third images in an ascending order of the visual similarity distances.

Abstract: The embodiments of the present application disclose a method and device for sending and receiving an optical signal. The method for sending an optical signal comprises: performing serial-to-parallel conversion on a data signal to be transmitted to obtain an I path data sequence, a Q path data sequence and a PPM path data sequence which are in parallel; performing mPQ-encoding on the I path data sequence, the Q path data sequence and the PPM path data sequence to obtain an I path and a Q path of an mPQ-encoded digital signal; shaping the I path and the Q path of the mPQ-encoded digital signal by Nyquist-filtering to obtain an I path and Q path of a filtered digital signal; performing digital-to-analog conversion on the I path and the Q path of the filtered digital signal and mapping the converted I path and Q path onto an optical carrier to obtain a target optical signal and send the same.

Abstract: The invention relates to a weather recognition method and device based on image information detection, including: obtaining an image extracting multiple first image features of the image with respect to each preset type of weather using a number of first preset algorithms preset correspondingly for different preset types of weather; inputting the multiple first image features to a preset multi-kernel classifier, the multi-kernel classifier performing classification according to the image features to identify the weather in which the image was taken. The multi-kernel classifier is realized by: selecting a first preset number of image samples for each of the preset types of weather; for the image samples of this type of weather, extracting the first image features of each image sample according to the first preset algorithm corresponding to this preset type of weather; and performing machine learning for the first image features according to a preset multi-kernel learning algorithm.

Abstract: Embodiments of the present disclosure may provide a service transmission method and device. The service transmission method may include: determining, by a user terminal, whether a service of the user terminal is a service needing global cooperation; transmitting, by the user terminal, the service with a global Base Band Unit (BBU) pool mode if the service of the user terminal is the service needing the global cooperation; and transmitting, by the user terminal, the service with a non-global BBU pool mode if the service of the user terminal is the service, which does not need the global cooperation.

Abstract: Methods and devices for determining scoring models of a three-dimensional animation scene frame are provided.

Abstract: Embodiments of the present invention disclose a node scheduling method, where the method includes: configuring, for a node, a working period matched with a performance indicator of the node, where duration of the working period is an integer multiple of duration of a super frame, and the super frame is a super frame corresponding to a beacon frame in a network in which the node is located; and sending, to the node, an updated beacon frame that includes information about the working period, so that the node acquires the information about the working period from the updated beacon frame, receives a working beacon frame, establishes a super frame corresponding to the working beacon frame until the working period ends, and then repeats the receiving a working beacon frame and establishing a super frame corresponding to the working beacon frame until the updated beacon frame is received again.

Abstract: The present disclosure provides a big-data-mining-based wireless channel modeling method comprising: obtaining image information of a measurement environment and a channel impulse response data sample under a preset condition; obtaining at least one multipath wave and a channel parameter of each of the multipath wave according to the channel impulse response data sample using a channel parameter estimation algorithm; and clustering the at least one multipath wave according to the channel parameter of each of the multipath wave using a clustering algorithm to obtain at least one cluster; obtaining at least one scattering object in the measurement environment according to the image information of the measurement environment; matching each of the cluster with each of the scattering object to obtain a cluster kernel which is a cluster matching with the scattering object; establishing, a base wireless channel model under the preset condition according to all of the cluster kernel.

Abstract: A device for controlling the fusion of a heterogeneous system, which receives a first data stream from a MAC layer of a WLAN through a LLC sub-layer and a second data stream from a MAC layer of a cellular network through a PDCP and a RLC sub-layer, and for fusing the first data stream and the second data stream into one data stream and transmitting the fused one data stream to a network layer. The device demultiplexes the data stream from the network layer into a third data stream corresponding to the MAC layer of the WLAN and a fourth data stream corresponding to the MAC layer of the cellular network, and transmits the third data stream to the MAC layer of the WLAN through the LLC and the fourth data stream to the MAC layer of the cellular network through the PDCP and the RLC sub-layer.