CN112105089A

CN112105089A - Communication signal correlation method based on response time probability distribution

Info

Publication number: CN112105089A
Application number: CN202010995522.5A
Authority: CN
Inventors: 马盛元; 魏平; 李万春; 张花国
Original assignee: University of Electronic Science and Technology of China
Current assignee: NORTH AUTOMATIC CONTROL TECHNOLOGY INSTITUTE
Priority date: 2020-09-21
Filing date: 2020-09-21
Publication date: 2020-12-18
Anticipated expiration: 2040-09-21
Also published as: CN112105089B

Abstract

The invention belongs to the technical field of communication, and particularly relates to a communication signal correlation method based on response time probability distribution. The invention provides a communication signal association algorithm based on response time probability distribution aiming at a specific multi-node Ad-Hoc communication system. The invention uses the response mechanism widely existed in the communication system as prior information, the transmitting party transmits a request signal to the receiving party, and the receiving party replies a response message within a time interval from the characteristic probability distribution, therefore, the concerned association problem can be summarized as finding out the pairing mode of the transmitting signal and the response signal which best meets the constraint condition. Under the condition of given constraint, the method can dig out a certain association relation from the intercepted TOAs, and the probability of complete correct association is more than 90%.

Description

Communication signal correlation method based on response time probability distribution

Technical Field

The invention belongs to the technical field of communication, and particularly relates to a communication signal correlation method based on response time probability distribution.

Background

In communication countermeasure, it is often desirable to obtain an association relationship between nodes and nodes in a communication system of interest. In order to accomplish the task, a non-cooperative communication reconnaissance device is generally required to be used for reconnaissance of communication signals, measurement of time, frequency and space main characteristic parameters of the signals is completed, sorting of the signals is further completed according to pulse description words, and then correlation of the signals or nodes is performed. However, usually, the detecting and receiving party can only obtain the transmitting signal of the transmitting node, the information of the receiving node cannot be obtained, and the association needs to bind the transmitting node and the receiving node, so that the signal association itself is an underdetermined problem, but it does not mean that the association problem is not solved, some characteristics of the communication system itself can provide some additional constraints for solving the problem, and the solution space can be reduced by using the constraints, which is the solution method of the signal association problem.

Disclosure of Invention

The invention provides a communication signal association algorithm based on response time probability distribution aiming at a specific multi-node Ad-Hoc communication system. The invention uses the response mechanism widely existed in the communication system as prior information, the transmitting party transmits a request signal to the receiving party, and the receiving party replies a response message within a time interval from the characteristic probability distribution, therefore, the concerned association problem can be summarized as finding out the pairing mode of the transmitting signal and the response signal which best meets the constraint condition.

The technical scheme adopted by the invention is as follows:

firstly, establishing a communication system model: consider N nodes p₁，p₂...，p_NThe Ad-hoc network is formed, and M request signals Q ═ Q from the nodes are detected in a period of time t₁，Q₂...，Q_MAnd M response signals S ═ S₁，S₂...，S_MTOA of, i.e. T (Q)_i) Or T (S)_i) Indicating a request signal Q_iOr a response signal S_iThe 2M nodes, i.e., P (Q), can be known through signal sorting_i) Or P (S)_i) Indicating a request signal Q_iOr a response signal S_iThe node to which it belongs. The communication association problem of the nodes is converted into the problem of pairing the 2M signals two by two.

Possible result set of pairings λ ═ { λ ═ λ₁，λ₂，...，λ_nWhere n denotes that there are n correlation results, λ_iDenotes the i-th correlation result, P (λ)_i) Representing the prior probability of the ith correlation result, which can be assumed

x represents the current TOA and the result of the analysis, i.e. an observation, P (x) represents the prior probability of the current observation, P (λ)_i| x) represents the posterior probability of the ith correlation result under the current observation, P (x | λ)_i) Indicating the probability of occurrence of the current observation under the ith correlation result.

The following derivation was performed:

taking lambda as the parameter to be estimated, lambda has n options, and the parameter is selected based on the observation. M represents M sets of associated pairs, t_j|λ_iIs shown at λ_iThe time interval between the jth group of association pairs in the association manner of (1), p (t)_j|λ_i) Which represents the probability density for the corresponding time interval, at is a unit of time density. Using MAP criterion, i.e. maximum a posteriori probability

Corresponding to

Is the correlation result.

According to the above derivation, the method of the present invention comprises the following steps:

the method comprises the following steps: selecting corresponding time probability model and threshold value

According to a specific communication protocol and in combination with a queuing theory, a probability density function F (t) obeyed by a time slot interval between signals of a response message can be obtained, and a threshold value F is set₀In the following solution, only the terms belonging to { t | F (t) > F are considered₀An associated pair of inter-interval times.

Step two: converting an input TOA into a directed graph

According to F, as shown in FIG. 2₀The selected time range is used for drawing the time radius of the transmitting signal and the response signal, the transmitting signal and the receiving signal which can be overlapped at the time point have one directed edge in the directed graph, namely, the left graph time domain graph can be converted into the directed graph of the right graph, and the set G (Q) of the directed graph edges is_iS_j|Q_i∈Q，S_j∈S}。

Calculating the arrival time difference of two signals connected with each directed edge, substituting the time difference into f (t), and taking the obtained probability density as the weight W (Q) of the edge_i，S_j)＝f(T(S_i)-T(Q_i))。

Step three: traversing various association scenarios

Definition I (Q)_i，S_j) E {0, 1} represents Q_i，S_jThe two signals are in or out of pairing, the value 0 represents that the pairing is not in, and the value 1 represents that the pairing is in. The exclusivity of each signal is satisfied at the same time: for any signal S_j，

For any signal Q_i，

And calculating the Score of each association condition, and selecting a group of association relations with the maximum Score as final output.

The invention has the beneficial effects that: the invention provides a method for associating from an input TOA stream according to interval time distribution probability, under the condition of given constraint, a certain association relation can be mined from the intercepted TOA, and the probability of complete correct association is more than 90%.

Drawings

FIG. 1 is a TOA stream input for a model;

FIG. 2 shows a detailed solution process;

FIG. 3 is a statistical graph of associated error probabilities.

Detailed Description

The technical scheme of the invention is described in detail below with reference to the accompanying drawings and embodiments:

examples

In this example, there are 5 nodes in the Ad-hoc network, the density of the signal pair is set to be 10/s, the observation time length is 1s, and the probability distribution of the response time is

Fig. 1 shows a set of input TOA streams, and it can be seen that there is a problem of fuzzy pairing between three sets of transmit signals and reply signals, numbered 2 and 3, 5 and 6, and 8 and 9, and the association relationship cannot be directly determined.

Fig. 2 shows a processing flow of the model, (a) the graph extracts the weight of the directed edge, and it can be seen that the 3 groups of signals really have the problem of fuzzy association, (b) the graph removes the edge without fuzzy association and only analyzes the edge with fuzzy association, (c) the graph finds out the three fuzzy regions, numbers the three fuzzy regions, calculates Score for each region and determines the association relationship, and (d) the graph obtains the final association relationship.

The statistics of the correlation error rate after 10000 times of simulation are shown in fig. 3, and it can be seen that the probability of complete correlation being correct is above 90%. The correlation can be better derived from the input TOA stream.

And (4) conclusion: the invention provides a method for extracting association relation from input TOA flow according to interval time distribution probability, under the condition of given constraint, a certain association relation can be mined from detected and collected TOA, and the probability of complete correct association is more than 90%. However, since the correlation itself is an underdetermined problem, there is still a case where about 10% of the analysis has a correlation error of 1 pair.

Claims

1. Method for correlating a communication signal based on a probability distribution of response times for a communication system having N nodes p₁，p₂...，p_NForming an Ad-hoc network defining M request signals Q ═ Q detected from these nodes during time t₁，Q₂...，Q_MAnd M response signals S ═ S₁，S₂...，S_MTOA of, i.e. T (Q)_i) Or T (S)_i) Indicating a request signal Q_iOr a response signal S_iThe TOA of (2) converts the communication association problem of the nodes into the problem of pairing the 2M signals pairwise; the association method is characterized by comprising the following steps:

s1, setting the time slot interval between the response message and the signal to obey the probability density function F (t), and setting the threshold value F₀Consider only those belonging to { t | F (t) > F₀An associated pair of inter-interval times;

s2, according to F₀The time range of the time domain of the_iS_j|Q_i∈Q，S_j∈S}；

Calculating the arrival time difference of two signals connected with each directed edge, substituting the time difference into f (t), and taking the obtained probability density as the weight W (Q) of the edge_i，S_j)＝f(T(S_i)-T(Q_i))；

S3, traversing various association conditions:

selecting a group of association relations with the largest Score as final output; i (Q)_i，S_j) E {0, 1} represents Q_i，S_jThe two signals are in or out of pairing, the value 0 represents that the pairing is not in, and the value 1 represents that the pairing is in.