CN-122026956-A - Multi-band self-adaptive communication method and system based on power line carrier

CN122026956ACN 122026956 ACN122026956 ACN 122026956ACN-122026956-A

Abstract

The invention provides a multi-band self-adaptive communication method and a system based on a power line carrier, which are characterized in that a multi-band communication model is built, a multi-band observation sequence is acquired and input into the multi-band communication model, a state probability distribution and an expected transmission rate matrix of each band are output through a perception layer, a decision layer is based on the state probability distribution and the expected transmission rate matrix and is combined with an auction algorithm to acquire a matching result, a scheduling layer is based on the matching result and is combined with a Lyapunov drift optimization mechanism to acquire a service packet list, the service packet list is input into the decision layer, parameters in the perception layer and the scheduling layer are updated, the updated multi-band communication model is acquired, self-adaptive communication is carried out according to the multi-band communication model, and the multi-band self-adaptive communication method with accurate prediction of the channel and efficient matching of the service and the band is adopted, so that the quality of the self-adaptive communication is improved.

Inventors

CAI WENTING
YANG CHEN
YANG XIANZHI
LIN ZHENFU
CHEN JIAN
CHEN WENQIN

Assignees

南方电网数字电网研究院股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260211

Claims (8)

1. The multi-band self-adaptive communication method based on the power line carrier is characterized by comprising the following steps of: constructing a multi-band communication model, wherein the multi-band communication model comprises a perception layer, a decision layer and a scheduling layer; Acquiring a multi-dimensional observation sequence, inputting the multi-dimensional observation sequence into a multi-band communication model, and outputting state probability distribution and expected transmission rate matrix of each band based on a perception layer; the decision layer is based on state probability distribution and expected transmission rate matrix, and acquires a matching result by combining an auction algorithm; The scheduling layer acquires a service packet list based on the matching result and by combining a Lyapunov drift optimization mechanism; And inputting the service packet list into a decision layer, updating parameters in a perception layer and a scheduling layer, acquiring an updated multi-band communication model, and carrying out self-adaptive communication based on the multi-band communication model.
2. The power line carrier-based multiband adaptive communication method according to claim 1, wherein acquiring a multidimensional observation sequence and inputting the multidimensional observation sequence into a multiband communication model, outputting a state probability distribution and an expected transmission rate matrix of each band based on a perception layer, comprises: The multi-dimensional observation sequence comprises a short-term average signal-to-noise ratio, a noise pulse count and a channel idle ratio; Constructing a perception layer based on a hidden Markov model, and acquiring a historical multi-dimensional observation sequence to train parameters of the hidden Markov model; and acquiring a real-time multidimensional observation sequence, inputting the real-time multidimensional observation sequence into a trained hidden Markov model, and outputting the state probability distribution and the expected transmission rate matrix of each frequency band by combining a Viterbi algorithm.
3. The power line carrier-based multi-band adaptive communication method according to claim 2, wherein the training of hidden markov model parameters by collecting historical multi-dimensional observation sequences comprises: the hidden Markov model parameters comprise a state transition matrix and an observation probability matrix; wherein, the channel states in the state transition matrix comprise excellent, good, bad and blocking four types of channel states; the observation probability matrix is expressed as probability distribution of corresponding multi-dimensional observation sequences under different channel states.
4. The power line carrier based multi-band adaptive communication method according to claim 1, wherein the decision layer obtains a matching result based on a state probability distribution and a desired transmission rate matrix in combination with an auction algorithm, comprising: Constructing an auction mechanism, which comprises defining a service package to be sent as an auction party, defining an available frequency band as an auction item, and defining a decision unit in a decision layer as an auction party; acquiring bidding functions of the service package on each frequency band based on the state probability distribution and the expected transmission rate matrix; And solving the bidding function based on an auction algorithm, acquiring a corresponding relation between the service package and the frequency band, and a modulation coding scheme corresponding to successful matching, and combining matching utility information of the service package to form a matching result, wherein the matching utility information is expressed as a bid corresponding to successful matching.
5. The method for power line carrier based multi-band adaptive communication according to claim 4, wherein obtaining a bid function of a service packet for each band based on a state probability distribution and a desired transmission rate matrix, comprises: acquiring the time delay sensitivity of each service packet, and constructing a utility function based on the expected transmission rate and the time delay sensitivity; acquiring a bidding function based on the state probability distribution and the utility function; the bidding function is specifically expressed as ; Wherein the said Represented as bidding of each service package on each frequency band, said Expressed as a state probability distribution, of Expressed as a utility function, said Represented as a desired transmission matrix, said Expressed as delay sensitivity.
6. The multi-band adaptive communication method based on the power line carrier according to claim 1, wherein the scheduling layer obtains a service packet list based on a matching result in combination with a lyapunov drift optimization mechanism, comprising: extracting history matching utility information of the service package in the history matching result, and acquiring average matching utility information based on the history matching utility information; Constructing a virtual queue based on the service packets, queuing successfully matched service packets from the virtual queue, and reserving unsuccessfully matched service packets in the virtual queue to obtain the length of the virtual queue; Defining scheduling parameters, setting inherent value of service packets and admission marks, and constructing an optimization target formula by combining average matching utility information and virtual queue length; Wherein, the admission mark of the service packet comprises 0 and 1, wherein 0 is expressed as not allowing the service packet to participate in matching, and 1 is expressed as allowing the service packet to participate in matching; and solving the service packet list with the maximum optimization target formula, obtaining a service packet list allowed to enter a decision layer, and dynamically adjusting the delay sensitivity of the service packet.
7. The method for adaptive communication in multiple frequency bands based on power line carrier according to claim 1, wherein the step of inputting the service packet list into the decision layer, updating parameters in the sensing layer and the scheduling layer, obtaining an updated multiple frequency band communication model, and performing adaptive communication based on the multiple frequency band communication model comprises the steps of Inputting the service packet list and the dynamically adjusted delay sensitivity into a decision layer, and acquiring an updating matching result by combining the real-time state probability distribution and the expected transmission rate matrix by the decision layer; extracting real-time matching utility information in the updated matching result, and updating the virtual queue length and average matching utility information in the scheduling layer based on the real-time matching utility information; And acquiring a communication result, and adjusting the confidence level of the state transition matrix and the observation probability matrix in the perception layer based on the communication result.
8. A multi-band adaptive communication system based on a power line carrier for implementing the method of any of claims 1-7, comprising: the construction module is used for constructing a multi-band communication model based on the perception layer, the decision layer and the scheduling layer; The sensing module inputs the multi-dimensional observation sequence into the multi-band communication model, and outputs state probability distribution and expected transmission rate matrix of each band based on the sensing layer; The decision module is based on the state probability distribution and the expected transmission rate matrix, and is used for acquiring a matching result by combining an auction algorithm; the scheduling module is used for acquiring a service packet list based on the matching result and combining with a Lyapunov drift optimization mechanism; the feedback module inputs the service package list into the decision layer, acquires an updating matching result, updates parameters in the sensing layer and the scheduling layer based on the updating matching result, and acquires an updated multi-band communication model; And the communication module is used for carrying out self-adaptive communication based on the multi-band communication model.

Description

Multi-band self-adaptive communication method and system based on power line carrier Technical Field The invention relates to the technical field of communication, in particular to a multi-band self-adaptive communication method and system based on a power line carrier. Background The power line communication technology is used as a communication mode for realizing data transmission by utilizing the existing power line, has the advantages of no need of rewiring, wide coverage range, low cost and the like, and is widely applied to the scenes of household intelligent power grid, intelligent home control and the like. However, the power line channel environment is complex and changeable, and has the problems of noise interference, signal attenuation and the like, so that the channel state has stronger time-varying property and is difficult to predict; meanwhile, the requirements of different service types on transmission delay, speed and reliability are different, and the traditional power line communication method often has certain defects, including single channel state sensing means and multiple dependence on signal parameters with single dimension, such as signal-to-noise ratio monitoring, the real state of a channel cannot be comprehensively indicated, so that the channel prediction precision is low, and the time-varying channel characteristics are difficult to adapt; meanwhile, the matching mechanism of the service and the frequency band is deficient, and the traditional communication method mostly adopts a fixed allocation mode, and the adaptability of the service demand difference and the channel resource state is not fully considered, so that the resource utilization rate is low. Therefore, a multi-band adaptive communication method with accurate channel prediction and efficient matching between service and frequency band is needed, so as to improve the quality of adaptive communication. Disclosure of Invention In view of the above-mentioned problems, in combination with the first aspect of the present invention, an embodiment of the present invention provides a multi-band adaptive communication method based on a power line carrier, the method including: constructing a multi-band communication model, wherein the multi-band communication model comprises a perception layer, a decision layer and a scheduling layer; Acquiring a multi-dimensional observation sequence, inputting the multi-dimensional observation sequence into a multi-band communication model, and outputting state probability distribution and expected transmission rate matrix of each band based on a perception layer; the decision layer is based on state probability distribution and expected transmission rate matrix, and acquires a matching result by combining an auction algorithm; The scheduling layer acquires a service packet list based on the matching result and by combining a Lyapunov drift optimization mechanism; And inputting the service packet list into a decision layer, updating parameters in a perception layer and a scheduling layer, acquiring an updated multi-band communication model, and carrying out self-adaptive communication based on the multi-band communication model. As a further aspect of the present invention, a multi-dimensional observation sequence is acquired and input into a multi-band communication model, and a state probability distribution and an expected transmission rate matrix of each band are output based on a perception layer, including: The multi-dimensional observation sequence comprises a short-term average signal-to-noise ratio, a noise pulse count and a channel idle ratio; Constructing a perception layer based on a hidden Markov model, and acquiring a historical multi-dimensional observation sequence to train parameters of the hidden Markov model; and acquiring a real-time multidimensional observation sequence, inputting the real-time multidimensional observation sequence into a trained hidden Markov model, and outputting the state probability distribution and the expected transmission rate matrix of each frequency band by combining a Viterbi algorithm. As a further aspect of the present invention, collecting a historical multi-dimensional observation sequence for training hidden markov model parameters includes: the hidden Markov model parameters comprise a state transition matrix and an observation probability matrix; wherein, the channel states in the state transition matrix comprise excellent, good, bad and blocking four types of channel states; the observation probability matrix is expressed as probability distribution of corresponding multi-dimensional observation sequences under different channel states. As a further aspect of the present invention, a decision layer obtains a matching result based on a state probability distribution and an expected transmission rate matrix in combination with an auction algorithm, including: Constructing an auction mechanism, which comprises defining a service package