Multiple access method and related device
AssigneeBeijing University of Posts and Telecommunications
InventorsKai Niu, Zijian Liang, Ping Zhang, Jincheng Dai, Chao Dong, Xiaodong Xu, Chen Dong
Life of this patent
Estimating the term…
Abstract
A multiple access method includes: obtaining, by a user equipment (UE), first source information; extracting a semantic feature of the first source information to obtain a first semantic feature sequence; performing a joint source-channel coding on the first semantic feature sequence to obtain a first semantic information sequence; mapping the first semantic information sequence into preset time-frequency resources of an uplink multiple access channel; and transmitting the first semantic information sequence through the uplink multiple access channel.
Claims8
A multiple access method based on a semantic domain, implemented by a multi-user transmission system, the multi-user transmission system comprising a base station and a plurality of user equipment (UEs), the method comprising:
- obtaining, by a UE, first source information;
- extracting, by the UE, a semantic feature of the first source information to obtain a first semantic feature sequence;
- performing, by the UE, a joint source-channel coding on the first semantic feature sequence to obtain a first semantic information sequence;
- mapping, by the UE, the first semantic information sequence into preset time-frequency resources;
- and
- transmitting, by the UE, the first semantic information sequence through an uplink multiple access channel;
- obtaining, by the base station, a first superposition semantic information sequence of a plurality of the first semantic information sequences of the plurality of UEs of the multi-user transmission system from the uplink multiple access channel;
- wherein the first superposition semantic information sequence represents a superposition of the plurality of the first semantic information sequences of the plurality of UEs of the multi-user transmission system;
- and
- for each of the plurality of UEs, obtaining, by the base station, first reconstructed source information corresponding to the UE according to the first superposition semantic information sequence;
- wherein performing, by the UE, a joint source-channel coding on the first semantic feature sequence to obtain a first semantic information sequence comprises:
- transforming, by the UE, the first semantic feature sequence to obtain a first side information sequence;
- estimating probability information of the first semantic feature sequence based on the first side information sequence to obtain a first probability information sequence;
- obtaining a first channel input symbol dimension sequence according to the first probability information sequence;
- and
- performing the joint source-channel coding on the first semantic feature sequence on the basis of the first channel input symbol dimension sequence to obtain the first semantic information sequence;
- and
- wherein obtaining, by the base station, first reconstructed source information according to the first superposition semantic information sequence for each of the plurality of UEs comprises:
- for each of the plurality of UEs,
- transforming, by the base station, the first side information sequence of the UE to obtain a first reconstructed probability information sequence;
- obtaining a first reconstructed channel input symbol dimension sequence according to the first reconstructed probability information sequence;
- obtaining a first reconstructed semantic feature sequence through a joint source-channel decoding according to the first reconstructed channel input symbol dimension sequence and the first superposition semantic information sequence;
- and
- obtaining the first reconstructed source information according to the first reconstructed semantic feature sequence.
A multiple access system, comprising a base station and a plurality of user equipment (UEs), wherein,
- each of the plurality of UEs comprises:
- a first obtaining module, configured to obtain first source information;
- a semantic feature extracting module, configured to exact a semantic feature of the first source information to obtain a first semantic feature sequence;
- a joint source-channel coding module, configured to perform a joint source-channel coding on the first semantic feature sequence to obtain a first semantic information sequence;
- a time-frequency resource mapping module, configured to map the first semantic information sequence into preset time-frequency resources;
- and
- a transmission module, configured to transmit the first semantic information sequence through an uplink multiple access channel;
- the base station comprises:
- a second obtaining module, configured to obtain a first superposition semantic information sequence of a plurality of the first semantic information sequences of the plurality of UEs of the multi-user transmission system from the uplink multiple access channel;
- wherein the first superposition semantic information sequence represents a superposition of the plurality of the first semantic information sequences of the plurality of UEs of the multi-user transmission system;
- and
- a reconstructing module, configured to for each of the plurality of UEs, obtain first reconstructed source information corresponding to the UE according to the first superposition semantic information sequence;
- wherein the joint source-channel coding module is further configured to:
- transform the first semantic feature sequence to obtain a first side information sequence;
- estimate probability information of the first semantic feature sequence based on the first side information sequence to obtain a first probability information sequence;
- obtain a first channel input symbol dimension sequence according to the first probability information sequence;
- and
- perform the joint source-channel coding on the first semantic feature sequence on the basis of the first channel input symbol dimension sequence to obtain the first semantic information sequence;
- and
- wherein the reconstructing module is further configured to:
- transform, for each of the plurality of UEs, the first side information sequence of the UE to obtain a first reconstructed probability information sequence;
- obtain a first reconstructed channel input symbol dimension sequence according to the first reconstructed probability information sequence;
- obtain a first reconstructed semantic feature sequence through a joint source-channel decoding according to the first reconstructed channel input symbol dimension sequence and the first superposition semantic information sequence;
- and
- obtain the first reconstructed source information according to the first reconstructed semantic feature sequence.
Dependencies are inferred from the claim text.