Search

CN-122001428-A - Signal processing system based on 6G communication

CN122001428ACN 122001428 ACN122001428 ACN 122001428ACN-122001428-A

Abstract

The invention relates to the technical field of communication signal processing, in particular to a signal processing system based on 6G communication, which comprises the following steps of obtaining a composite transmitting signal with a self-adaptive signal structure, utilizing the composite transmitting signal to irradiate a target communication area and receive a feedback signal from user equipment, generating enhanced channel state information, and dynamically forming a wave beam for carrying out directional enhanced transmission on service data symbols in a subsequent transmission time slot, wherein wave beam parameters of the wave beam are adapted to the user equipment state indicated by endogenous state sensing information. The invention ensures that the base station not only obtains the environment channel condition when receiving the feedback, but also synchronously knows the service strength which can be born by the equipment, and avoids the problems of excessive scheduling, performance overload and the like in the traditional communication system. The transition from perceiving only the external channel to simultaneously perceiving the terminal capabilities is achieved.

Inventors

  • XIE WENMING
  • WANG MENGTING

Assignees

  • 上海拿帖软件有限公司

Dates

Publication Date
20260508
Application Date
20260318

Claims (10)

  1. 1. A signal processing system based on 6G communication, the system comprising: the method comprises the steps of inputting a service data stream to be transmitted to a signal structure dynamic generation model at a base station side to obtain a composite transmitting signal with a self-adaptive signal structure, wherein the self-adaptive signal structure at least comprises pilot patterns which are unevenly distributed on a time-frequency two-dimensional resource grid and service data symbols surrounded by the pilot patterns in a time-frequency domain; Illuminating a target communication area by utilizing the composite transmitting signal and receiving a feedback signal from user equipment; the feedback signal includes enhanced channel state information, the generation of the enhanced channel state information by the user equipment including: performing initial channel estimation based on a pilot pattern in the received composite transmission signal, and simultaneously acquiring endogenous state sensing information of the user equipment, and performing feature level fusion on the endogenous state sensing information and the initial channel estimation result to generate enhanced channel state information, wherein the endogenous state sensing information at least comprises a real-time calculation load state, equipment residual capacity and power consumption constraint mode of the user equipment; And generating a dynamic shaping wave beam aiming at the user equipment by utilizing a wave beam shaping weight dynamic calculation model according to the enhanced channel state information, wherein the dynamic shaping wave beam is used for carrying out directional enhanced transmission on the service data symbol in a subsequent transmission time slot, and the wave beam parameter of the dynamic shaping wave beam is adapted to the user equipment state indicated by the endogenous state sensing information.
  2. 2. The signal processing system based on 6G communication according to claim 1, wherein the base station side further comprises analyzing the service data stream and extracting at least one data stream attribute characteristic thereof, wherein the data stream attribute characteristic at least comprises a service type priority of data, a modulation and coding scheme level and a time delay sensitivity mark; The method comprises the steps of inputting the data stream attribute characteristics into the signal structure dynamic generation model, wherein the signal structure dynamic generation model is an encoder-decoder structure constructed based on a depth neural network, encoding the data stream attribute characteristics by an encoder of the signal structure dynamic generation model to generate hidden space vectors representing signal structure constraints, and decoding and generating a mapping strategy of a time-frequency two-dimensional resource grid by a decoder of the signal structure dynamic generation model according to the hidden space vectors.
  3. 3. The 6G communication based signal processing system of claim 2, wherein the mapping policy indicates: (a) The position and shape of the data resource blocks used for carrying the business data symbols; (b) And the pilot frequency pattern is configured to unevenly surround the resource block where the service data symbol is located on the time domain and/or the frequency domain to form an enclosure situation.
  4. 4. A signal processing system based on 6G communication according to claim 3, wherein the composite transmission signal is synthesized by modulating and mapping the traffic data stream onto the traffic data symbols and mapping a predetermined pilot sequence onto the resource elements indicated by the pilot pattern according to the mapping strategy.
  5. 5. The signal processing system of claim 1, wherein the base station directs the composite transmit signal to the target communication area via a massive antenna array in a beam scanning manner, wherein the transmit power of the composite transmit signal is set to an adjustable enhanced power offset on a pilot resource element carrying the pilot pattern to improve the received signal-to-noise ratio of the pilot pattern on the user equipment side.
  6. 6. The signal processing system based on 6G communication according to claim 5 wherein said user equipment receives said composite transmission signal, and performs time-frequency domain joint channel estimation according to said pilot pattern carried in the received signal to obtain an initial channel estimation result, said initial channel estimation result including at least a channel frequency response matrix and a spatial covariance matrix, and simultaneously said user equipment collects said endogenous state sensing information in real time through its built-in system state monitoring module, said endogenous state sensing information including at least a real-time occupancy rate and a temperature of a central processor, a real-time occupancy rate and a temperature of a graphic processor, an available memory bandwidth, a battery remaining capacity, and a current power consumption mode flag.
  7. 7. The signal processing system based on 6G communication of claim 6 wherein said user device inputs said initial channel estimation result and said endogenous state awareness information into a locally deployed feature fusion neural network model, said feature fusion neural network model mapping said initial channel estimation result to a first high-dimensional feature vector and said endogenous state awareness information to a second high-dimensional feature vector; The feature fusion neural network model performs weighted splicing and nonlinear transformation on the first high-dimensional feature vector and the second high-dimensional feature vector, outputs a fused joint feature vector as the enhanced channel state information, wherein the dimension of the enhanced channel state information is lower than the sum of the dimensions of the first high-dimensional feature vector and the second high-dimensional feature vector, and the user equipment encodes and transmits the enhanced channel state information to the base station through a preset uplink feedback channel.
  8. 8. The signal processing system according to claim 1, wherein the base station inputs the received enhanced channel state information to the beamforming weight dynamic calculation model, wherein the beamforming weight dynamic calculation model is a dual-branch neural network based on an attention mechanism, and comprises: a first characteristic analysis branch for analyzing a physical channel characteristic component from the enhanced channel state information; A second state analysis branch, configured to analyze a user equipment state feature component from the enhanced channel state information; And the beam forming weight dynamic calculation model dynamically weights and modulates the physical channel characteristic components according to the user equipment state characteristic components through an attention fusion layer in the beam forming weight dynamic calculation model to generate a joint optimization characteristic vector.
  9. 9. The signal processing system of claim 8 wherein the output layer of the beamforming weight dynamic computation model maps and generates a complex weight matrix of the dynamic beamforming beam according to the joint optimization feature vector and synchronously outputs a recommended transmission parameter set, wherein the recommended transmission parameter set at least comprises a recommended modulation order and code rate, and the recommended transmission parameter set is directly derived from the user equipment state feature component.
  10. 10. The signal processing system of claim 9, wherein the base station uses the complex weight matrix to beam-shape traffic data symbols addressed to the user equipment in one or more subsequent transmission time slots, and adjusts modulation and coding schemes for the traffic data symbols in accordance with the recommended transmission parameter set to complete the directional enhanced transmission, wherein the beamwidth and transmission power of the dynamically shaped beam are configured in an inverse relationship with the computational load and power consumption constraints indicated by the user equipment state characteristic components.

Description

Signal processing system based on 6G communication Technical Field The invention relates to the technical field of communication signal processing, in particular to a signal processing system based on 6G communication. Background With the development of the sixth generation mobile communication (6G) system, the communication demand is expanding from the traditional acceleration of human-human communication to the diversification fusion of human-machine, object-object and the like, and the service form presents the trend of coexistence of multiple extreme characteristics of large bandwidth, high reliability, low time delay, high energy efficiency and the like. Under the background, how to construct an intelligent signal processing architecture which simultaneously satisfies multi-service cooperation, resource efficient scheduling and terminal difference adaptation becomes a key challenge of current 6G system design. The pilot frequency design and resource mapping of the current wireless signals mostly adopt a static configuration mode, and time-frequency patterns with structural suitability are difficult to flexibly generate according to parameters such as service types, modulation modes or time delay sensitivity, so that pilot frequency overhead is overlarge or channel estimation performance is insufficient, and overall link efficiency is further influenced. The traditional feedback path usually only pays attention to the channel quality, ignores the factors such as calculation load, battery power, power consumption limitation and the like of the user equipment in different states, and is easy to cause the situation that a scheduling instruction is not matched with the actual processing capacity of the terminal, so that the problems of data backlog, overload failure and the like are caused. The current beam forming is mainly optimized according to channel directivity estimation, and dynamic properties such as terminal processing capacity, heat dissipation capacity, power supply state and the like cannot be comprehensively considered, so that partial high-precision beams are increased in system burden when resources are tense or equipment state is poor, and overall stability is reduced. Disclosure of Invention The invention provides a signal processing system based on 6G communication, which integrates a three-dimensional cooperative mechanism of service perception, terminal state perception and channel state perception, realizes full-link intelligent optimization from signal generation, feedback control to space transmission, and constructs a next generation signal processing system oriented to a 6G complex communication scene. A signal processing system based on 6G communication, the system comprising performing the following: S1, inputting a service data stream to be transmitted to a signal structure dynamic generation model at a base station side to obtain a composite transmitting signal with a self-adaptive signal structure, wherein the self-adaptive signal structure at least comprises pilot patterns unevenly distributed on a time-frequency two-dimensional resource grid and service data symbols surrounded by the pilot patterns in a time-frequency domain; S2, irradiating a target communication area by utilizing the composite emission signal, and receiving a feedback signal from user equipment; the feedback signal includes enhanced channel state information, the generation of the enhanced channel state information by the user equipment including: performing initial channel estimation based on a pilot pattern in the received composite transmission signal, and simultaneously acquiring endogenous state sensing information of the user equipment, and performing feature level fusion on the endogenous state sensing information and the initial channel estimation result to generate enhanced channel state information, wherein the endogenous state sensing information at least comprises a real-time calculation load state, equipment residual capacity and power consumption constraint mode of the user equipment; And S3, generating a dynamic shaping wave beam aiming at the user equipment by utilizing a wave beam shaping weight dynamic calculation model according to the enhanced channel state information, wherein the dynamic shaping wave beam is used for carrying out directional enhanced transmission on the service data symbol in a subsequent transmission time slot, and the wave beam parameter of the dynamic shaping wave beam is adapted to the user equipment state indicated by the endogenous state sensing information. The beam parameters include complex weight matrix, beam width, transmit power, etc. Optionally, the method further comprises analyzing the service data stream and extracting at least one data stream attribute characteristic thereof, wherein the data stream attribute characteristic at least comprises the service type priority, the modulation and coding scheme level and the time delay sensitivity mark of the d