CN115290130A - Distributed information estimation method based on multivariate probability quantification - Google Patents

Distributed information estimation method based on multivariate probability quantification Download PDF

Info

Publication number
CN115290130A
CN115290130A CN202211229494.1A CN202211229494A CN115290130A CN 115290130 A CN115290130 A CN 115290130A CN 202211229494 A CN202211229494 A CN 202211229494A CN 115290130 A CN115290130 A CN 115290130A
Authority
CN
China
Prior art keywords
probability
quantization
multivariate
data
function
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202211229494.1A
Other languages
Chinese (zh)
Other versions
CN115290130B (en
Inventor
黄川�
崔曙光
何萌
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chinese University of Hong Kong Shenzhen
Original Assignee
Chinese University of Hong Kong Shenzhen
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chinese University of Hong Kong Shenzhen filed Critical Chinese University of Hong Kong Shenzhen
Priority to CN202211229494.1A priority Critical patent/CN115290130B/en
Publication of CN115290130A publication Critical patent/CN115290130A/en
Application granted granted Critical
Publication of CN115290130B publication Critical patent/CN115290130B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D21/00—Measuring or testing not otherwise provided for
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y20/00—Information sensed or collected by the things
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y20/00—Information sensed or collected by the things
    • G16Y20/10—Information sensed or collected by the things relating to the environment, e.g. temperature; relating to location
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y40/00—IoT characterised by the purpose of the information processing
    • G16Y40/10—Detection; Monitoring
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y40/00—IoT characterised by the purpose of the information processing
    • G16Y40/20—Analytics; Diagnosis
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y40/00—IoT characterised by the purpose of the information processing
    • G16Y40/60—Positioning; Navigation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/38—Services specially adapted for particular environments, situations or purposes for collecting sensor information
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Biomedical Technology (AREA)
  • Environmental & Geological Engineering (AREA)
  • Toxicology (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Arrangements For Transmission Of Measured Signals (AREA)

Abstract

The invention discloses a distributed information estimation method based on multivariate probability quantization, which comprises the following steps: s1, constructing a distributed information estimation scene: comprises a fusion center FC located at the center of wireless communication network and a plurality of wireless communication devices distributed in the wireless communication networkA sensor at the edge of the network; s2, constructing a multivariate probability quantizer for quantizing local observation data by a sensor; s3, optimizing design parameters of multi-element quantization probability function
Figure 493582DEST_PATH_IMAGE001
(ii) a S4, designing a quantitative fusion estimator by a fusion center FC and optimizing to obtain an optimal estimation function
Figure 495036DEST_PATH_IMAGE002
(ii) a S5. Based on
Figure 910580DEST_PATH_IMAGE001
And
Figure 48300DEST_PATH_IMAGE002
the distributed information estimation is quantized with multivariate probability. The method can adapt to the condition that the quantization result has a plurality of elements and keep higher estimation performance.

Description

Distributed information estimation method based on multivariate probability quantization
Technical Field
The invention relates to distributed information estimation, in particular to a distributed information estimation method based on multivariate probability quantization.
Background
Distributed information estimation based on quantized data has been an active area of research. In a typical distributed estimation framework, local sensors send local observation data of raw information to a fusion center. The fusion center receives data sent from different local sensors and estimates unknown original information by using an estimation algorithm. However, due to bandwidth/energy limitations, the local observed data on the sensors typically needs to be quantified before transmission to the fusion center. The use of the same quantizer for all sensors is a widely adopted solution because it simplifies the design problem.
However, many conventional solutions mainly consider the problem of quantizer optimization in an environment where ideally no observation noise exists. And further research into quantizer designs that take into account observed noise conditions is lacking. Furthermore, many performance analyses and theories regarding the optimal quantizer only consider the case of binary quantization, i.e., the length of the quantized data on the sensor is limited to 1 bit. For the case of quantizing the observed data into multi-bit data on the sensor, that is, when there is a plurality of possibilities for the quantized result, the corresponding quantizer design scheme is also lack of research.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a distributed information estimation method based on multivariate probability quantization, which can adapt to the situation that a quantization result has multivariate and keep higher estimation performance.
The purpose of the invention is realized by the following technical scheme: a distributed information estimation method based on multivariate probability quantization comprises the following steps:
s1, constructing a distributed information estimation scene: the system comprises a fusion center FC positioned in the center of a wireless communication network and a plurality of sensors distributed at the edge of the wireless communication network;
each sensor is responsible for the raw information required by the fusion center
Figure 100002_DEST_PATH_IMAGE001
Observing to obtain own local observation data, performing multivariate probability quantization operation on the local observation data, converting continuous observation data into binary discrete data which can be used for digital communication, and sending the binary discrete data to a fusion center FC; the FC fusion center estimates original information according to the quantitative data sent by all the sensors;
s2, constructing a multivariate probability quantizer for quantizing local observation data by a sensor;
s3, optimizing design parameters of multi-element quantization probability function
Figure 100002_DEST_PATH_IMAGE002
；
S4, designing a quantitative fusion estimator by a fusion center FC and optimizing to obtain an optimal estimation function
Figure 100002_DEST_PATH_IMAGE003
；
S5. Based on
Figure 503964DEST_PATH_IMAGE002
And
Figure 280159DEST_PATH_IMAGE003
the distributed information estimation is quantized with multivariate probability.
The beneficial effects of the invention are: the multivariate quantization probability method still keeps the capability of approximately linearly decreasing with the total bit number under the condition that the total bit number of the network is changed, and the performance of the multivariate quantization probability method is efficiently estimated in the distributed wireless sensor network.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a diagram of a distributed information estimation scenario;
FIG. 3 is a block diagram of a multivariate probability quantizer;
FIG. 4 is a diagram of a quantization function structure;
FIG. 5 is a block diagram of a quantized fusion estimator;
fig. 6 is a diagram illustrating MSE estimated by the network for the original information under the condition that the total quantization bit number of the entire network changes.
Detailed Description
The technical solutions of the present invention are further described in detail below with reference to the accompanying drawings, but the scope of the present invention is not limited to the following.
Aiming at the problem of information estimation based on a bandwidth/energy limited distributed wireless sensor in a future wireless communication network, the invention designs a distributed information estimation scheme based on multivariate probability quantization: the method comprises the steps of designing a multivariate probability quantizer positioned on a sensor and a corresponding multivariate quantization probability function optimization algorithm; the quantitative fusion estimator located on the fusion center is designed and a corresponding estimation function optimization algorithm is designed. Consider a generalized scenario in which a distributed wireless sensor network estimates an unknown raw information, the network containing a Fusion Center (FC) located at a central node of the network and multiple sensors distributed at different locations at the edge of the network. The original information may be any kind of data required by the network, and is determined by the specific requirements of the network, such as common positioning information or weather information. Each sensor observes the original information and obtains its own local observation data, and usually, in an actual environment, due to the influence of environmental noise on the observation, an error exists between the local observation data and the original information. For a sensor with limited bandwidth/energy, the local observation data needs to be quantized first, and the continuous observation data is converted into binary discrete data which can be used for modern digital communication, so that the own observation data can be smoothly sent to the FC. FC can only estimate raw information using quantized data sent from all sensors. The measurement index of the estimation performance of the original information generally uses Mean Squared Error (MSE) of the original information and the estimation value thereof, and smaller MSE means more accurate estimation and better estimation performance;
as shown in fig. 1, a distributed information estimation method based on multivariate probability quantization includes the following steps:
s1, constructing a distributed information estimation scene: as shown in fig. 2, the system comprises a fusion center FC located at the center of the wireless communication network and a plurality of sensors distributed at the edge of the wireless communication network;
each sensor is responsible for the raw information required by the fusion center
Figure 100002_DEST_PATH_IMAGE004
Observing to obtain local observation data of the user, performing multivariate probability quantization operation on the local observation data, converting continuous observation data into binary discrete data which can be used for digital communication, and sending the binary discrete data to a fusion center FC; the FC fusion center estimates original information according to the quantized data sent by all the sensors;
s2, constructing a multivariate probability quantizer for quantizing local observation data by a sensor:
multiple sensors distributed at the edge of the network, together with the original information required by the fusion center
Figure 512426DEST_PATH_IMAGE004
And (5) observing to respectively obtain own local observation. When all sensor observations are considered here, the observation noise affected by the environment is independently and equally distributed. Therefore, to reduce the difficulty in sensor design and deployment, we also consider the use of the exact same multivariate probability quantizer structure on all sensors, including any adjustable design parameters on the quantizer.
Because independent and equally distributed observation noise on all sensors is considered and the same multivariate quantizer is used. We take any sensor (ignoring sensor numbers) as an example here to describe the process of observation and data quantization on the sensor, and the structure, function and design scheme of the multivariate probability quantizer. As shown in FIG. 3, the sensor observes the raw information
Figure 100002_DEST_PATH_IMAGE005
Obtaining local observations
Figure 100002_DEST_PATH_IMAGE006
By using
Figure 100002_DEST_PATH_IMAGE007
To express the observed value
Figure 606503DEST_PATH_IMAGE006
Relative to what is observed
Figure 79597DEST_PATH_IMAGE005
To describe the randomness between them. The sensor obtains the observed value
Figure 495535DEST_PATH_IMAGE006
Then inputting the quantized data into a multi-element probability quantizer and outputting the final quantization result
Figure 100002_DEST_PATH_IMAGE008
Quantifying the results
Figure 100002_DEST_PATH_IMAGE009
Is a one contains
Figure 100002_DEST_PATH_IMAGE010
Binary data of bits. Inside the quantizer, the observed value of the input
Figure 625819DEST_PATH_IMAGE006
Is first fed into a multivariate quantization probability controlFunction(s)
Figure 100002_DEST_PATH_IMAGE011
Is mapped to one
Figure 100002_DEST_PATH_IMAGE012
Probability vector of dimension
Figure 100002_DEST_PATH_IMAGE013
Probability vector
Figure 100002_DEST_PATH_IMAGE014
All of the elements in (1) are
Figure 100002_DEST_PATH_IMAGE015
Interval values are taken while satisfying the condition that the sum of the additions is 1, i.e.
Figure 100002_DEST_PATH_IMAGE016
（1）
Wherein, the first and the second end of the pipe are connected with each other,
Figure 100002_DEST_PATH_IMAGE017
is a function of design parameters
Figure 100002_DEST_PATH_IMAGE018
A variable function of control having
Figure 100002_DEST_PATH_IMAGE019
（2）
Wherein
Figure 100002_DEST_PATH_IMAGE020
Comprises a
Figure 100002_DEST_PATH_IMAGE021
All of the parameters that can be adjusted in (c),
Figure 100002_DEST_PATH_IMAGE022
is the number of design parameters. From the formula(2) It can be seen that by changing the design parameters
Figure 100002_DEST_PATH_IMAGE023
By changing the parameter function accordingly
Figure 602215DEST_PATH_IMAGE021
The functional expressions and structures of (c). From
Figure 333411DEST_PATH_IMAGE021
Output probability vector
Figure 100002_DEST_PATH_IMAGE024
Then fed into the quantization function
Figure 100002_DEST_PATH_IMAGE025
In (A), (B)
Figure 985497DEST_PATH_IMAGE025
Detailed in the following and in fig. 4), a decimal one-dimensional discrete value is output
Figure 100002_DEST_PATH_IMAGE026
Then we go through decimal
Figure 100002_DEST_PATH_IMAGE027
Quantization result converted into binary
Figure 100002_DEST_PATH_IMAGE028
. For quantization function
Figure 100002_DEST_PATH_IMAGE029
Its output
Figure 100002_DEST_PATH_IMAGE030
A is common to
Figure 100002_DEST_PATH_IMAGE031
Different results, it is desirable to make
Figure 582569DEST_PATH_IMAGE027
Taking the probability of each result entirely from
Figure 381898DEST_PATH_IMAGE031
Probability vector of dimension
Figure 100002_DEST_PATH_IMAGE032
Control, i.e. effecting
Figure 100002_DEST_PATH_IMAGE033
（3）
Wherein
Figure 100002_DEST_PATH_IMAGE034
Representing observed values at a given input multivariate probability quantizer
Figure 100002_DEST_PATH_IMAGE035
Under the premise of (1), output is quantized
Figure 100002_DEST_PATH_IMAGE036
Take a value of
Figure 100002_DEST_PATH_IMAGE037
The probability of (c).
To implement the above-described function of probability quantization of local observation data by the multivariate probability quantizer, we apply quantization function in the multivariate probability quantizer
Figure 72992DEST_PATH_IMAGE029
The structure shown in fig. 4 is designed.
Wherein the function is quantized
Figure 100002_DEST_PATH_IMAGE038
Is inputted by
Figure 100002_DEST_PATH_IMAGE039
Probability vector of dimension
Figure 100002_DEST_PATH_IMAGE040
The output being a one-dimensional discrete value
Figure 100002_DEST_PATH_IMAGE041
It is composed of
Figure 100002_DEST_PATH_IMAGE042
The serial sublayers with the same structure are composed, and the specific structural functions are as follows:
inputting: input device
Figure 817351DEST_PATH_IMAGE039
Probability vector of dimension
Figure 100002_DEST_PATH_IMAGE043
And an initial quantization value
Figure 100002_DEST_PATH_IMAGE044
。
M sub-layer, M =1,2, \8230;, M: input of M sub-layer (if M =1, its input is
Figure 449189DEST_PATH_IMAGE045
And
Figure 100002_DEST_PATH_IMAGE046
) Is output from the previous sublayer (m-1 th sublayer)
Figure 380105DEST_PATH_IMAGE047
Vector of dimensions
Figure 100002_DEST_PATH_IMAGE048
And quantized value
Figure 100002_DEST_PATH_IMAGE049
(ii) a First, in the m-th sublayer, to be inputted
Figure 100002_DEST_PATH_IMAGE050
Divided into two sub-vectors of equal length, each containing
Figure 207597DEST_PATH_IMAGE050
All elements of the first half and all elements of the second half, i.e. two
Figure 100002_DEST_PATH_IMAGE051
Subvectors of dimensions
Figure 100002_DEST_PATH_IMAGE052
And
Figure 100002_DEST_PATH_IMAGE053
(ii) a Then, the m sub-layer utilizes
Figure 100002_DEST_PATH_IMAGE054
And
Figure 100002_DEST_PATH_IMAGE055
outputting the quantized value
Figure 100002_DEST_PATH_IMAGE056
In which
Figure 100002_DEST_PATH_IMAGE057
（4）
Figure 100002_DEST_PATH_IMAGE058
Is [0,1 ]]Random noise, function, evenly distributed over intervals
Figure 100002_DEST_PATH_IMAGE059
An input non-negative number will output a1, whereas an input negative number will output a 0. Definition of
Figure 100002_DEST_PATH_IMAGE060
The mth sublayer output
Figure 100002_DEST_PATH_IMAGE061
Vector of dimensions
Figure 100002_DEST_PATH_IMAGE062
In which
Figure 100002_DEST_PATH_IMAGE063
（5）
And (3) outputting: quantization function
Figure 100002_DEST_PATH_IMAGE064
Is output quantized value
Figure 100002_DEST_PATH_IMAGE065
Is that it is
Figure 100002_DEST_PATH_IMAGE066
Quantized values output by the sub-layers
Figure 100002_DEST_PATH_IMAGE067
I.e. by
Figure 100002_DEST_PATH_IMAGE068
。
With the structure as in FIG. 4, the function is quantized
Figure 484643DEST_PATH_IMAGE064
Implements the quantized value of making its output
Figure 100002_DEST_PATH_IMAGE069
Taking the probability of each possible outcome entirely from the probability vector
Figure 100002_DEST_PATH_IMAGE070
To control, the function in equation (3) is implemented.
S3, optimizing design parameters of multi-element quantization probability function
Figure 100002_DEST_PATH_IMAGE071
;
Probability control function is quantized by adjusting the multivariate in the multivariate probability quantizer on the sensor as shown in equations (2) and (3)
Figure 100002_DEST_PATH_IMAGE072
Design parameters of
Figure 100002_DEST_PATH_IMAGE073
Can be varied accordingly
Figure 731341DEST_PATH_IMAGE072
And further changing the probability distribution of the quantized data on the sensor relative to the local observed data and the original information. This means that we can optimize the multivariate quantization probability function for the original information subject to different random distributions and the observation noise with different random characteristics under different observation environments
Figure 872472DEST_PATH_IMAGE072
Design parameters of
Figure 698346DEST_PATH_IMAGE073
To obtain an optimal quantized data probability distribution adapted to the current environment. By using Bayesian estimation theory, we consider minimizing the channel by an algorithm
Figure 63468DEST_PATH_IMAGE073
And determining an estimated MSE lower bound which can be achieved by using quantized data in the fusion center after the sensor quantizes the local observation so as to find out the optimal design parameter suitable for the current observation environment
Figure 100002_DEST_PATH_IMAGE074
And corresponding use-optimized design parameters on the sensor
Figure 138740DEST_PATH_IMAGE074
Is optimized to a multivariate quantization probability function
Figure 83562DEST_PATH_IMAGE071
。
We assume a common
Figure 100002_DEST_PATH_IMAGE075
The individual sensors are distributed throughout the network, a multivariate probability quantizer structure as shown in fig. 3 is used on all sensors, and we use the exact same multivariate quantization probability function in all multivariate probability quantizers
Figure 761013DEST_PATH_IMAGE072
And corresponding design parameters
Figure 562616DEST_PATH_IMAGE073
So as to reduce the design cost and difficulty of the whole network.
Figure 594025DEST_PATH_IMAGE075
Each sensor is used for respectively comparing the original information
Figure 100002_DEST_PATH_IMAGE076
Observing to obtain own local observation data with observation noise
Figure 100002_DEST_PATH_IMAGE077
. We consider here the presence of independent identically distributed observation noise on each sensor and define a probability density function
Figure 100002_DEST_PATH_IMAGE078
To describe the distribution of observed noise: for the first
Figure 100002_DEST_PATH_IMAGE079
A sensor, its local observation data
Figure 100002_DEST_PATH_IMAGE080
Is quantized into a plurality of probability quantizers through a local multivariate probability quantizer
Figure 100002_DEST_PATH_IMAGE081
Binary data of bits
Figure 100002_DEST_PATH_IMAGE082
And sent to the FC of the hub. Therefore, it is possible to
Figure 762445DEST_PATH_IMAGE075
A sensor jointly generates
Figure 500594DEST_PATH_IMAGE075
An
Figure 676360DEST_PATH_IMAGE081
Quantized data of bits
Figure 100002_DEST_PATH_IMAGE083
And sent to the FC, which needs to estimate the original information using all the received quantized data.
Through Bayesian probability theory, quantitative data on all sensors is given
Figure 100002_DEST_PATH_IMAGE084
Are independently and simultaneously distributed. Therefore, based on the probability distribution of the quantized data, first, the quantized data is calculated when the FC receives the quantized data
Figure 726225DEST_PATH_IMAGE083
For the original information
Figure 12850DEST_PATH_IMAGE084
The lower bound of the estimated MSE that can be achieved by the estimation, i.e.
Figure 100002_DEST_PATH_IMAGE085
Figure 100002_DEST_PATH_IMAGE086
（6）
Wherein
Figure 100002_DEST_PATH_IMAGE087
，
Figure 100002_DEST_PATH_IMAGE088
Indicating that FC receives all the quantized data sent from the sensor,
Figure 100002_DEST_PATH_IMAGE089
representing FC utilization quantized data
Figure 100002_DEST_PATH_IMAGE090
Pair that can be realized
Figure 100002_DEST_PATH_IMAGE091
Respectively, to the left of the inequality in the formula
Figure 100002_DEST_PATH_IMAGE092
Representing original information
Figure 841391DEST_PATH_IMAGE091
And its estimated value
Figure 100002_DEST_PATH_IMAGE093
The MSE between the first and second MSE,
Figure 100002_DEST_PATH_IMAGE094
represents the operation of calculating mathematical expectation, the right side of the inequality in the formula (6) represents the lower bound of the calculated MSE,
Figure 100002_DEST_PATH_IMAGE095
is the number of combinations in the mathematical definition,
Figure 100002_DEST_PATH_IMAGE096
（7）
is based on a multivariate quantization probability function
Figure 100002_DEST_PATH_IMAGE097
Design parameters of
Figure 100002_DEST_PATH_IMAGE098
And originalInformation
Figure 100002_DEST_PATH_IMAGE099
A series of intermediate calculation terms for the decision. As can be seen from equation (6), in the original information
Figure 496713DEST_PATH_IMAGE099
Probability distribution and observed noise distribution of
Figure 100002_DEST_PATH_IMAGE100
FC with both determinants utilizes quantized data pairs
Figure 502715DEST_PATH_IMAGE099
MSE estimated, i.e.
Figure 100002_DEST_PATH_IMAGE101
The lower bound of which is entirely defined by the multivariate quantization probability function
Figure 124190DEST_PATH_IMAGE097
Design parameters of
Figure 571351DEST_PATH_IMAGE098
And (6) determining. Thus, the right side of the inequality in equation (6) is minimized algorithmically by
Figure 100002_DEST_PATH_IMAGE102
The determined FC uses the lower bound of the MSE estimated by the quantized data on the original information to find the optimal design parameter adapted to the current observation environment
Figure 100002_DEST_PATH_IMAGE103
And corresponding optimal multivariate quantization probability function
Figure 100002_DEST_PATH_IMAGE104
。
Based on the above analysis, we consider an iterative algorithm that is computationally based on a series of samples of raw information collected from the actual observation environment and sensor local observation dataThe method approximately solves the optimization problem about the design parameters in each iteration and gradually approaches the optimal design parameters in the iteration process
Figure 216484DEST_PATH_IMAGE103
。
Initialization: total number of sensors is
Figure 100002_DEST_PATH_IMAGE105
Sample set
Figure 100002_DEST_PATH_IMAGE106
,
Figure 100002_DEST_PATH_IMAGE107
Is the total number of samples contained in the sample set,
Figure 100002_DEST_PATH_IMAGE108
is a sample of the original information that was,
Figure 100002_DEST_PATH_IMAGE109
indicates the serial number of the sample, and
Figure 100002_DEST_PATH_IMAGE110
representing sensor versus raw information samples
Figure 490208DEST_PATH_IMAGE108
Total observations
Figure 100002_DEST_PATH_IMAGE111
Obtained secondarily
Figure 670302DEST_PATH_IMAGE111
An observation sample; setting initial design parameters
Figure 100002_DEST_PATH_IMAGE112
Setting the tolerance threshold of iteration to
Figure 100002_DEST_PATH_IMAGE113
Setting upThe initial iteration count is
Figure 100002_DEST_PATH_IMAGE114
。
The method comprises the following steps: in the first place
Figure 100002_DEST_PATH_IMAGE115
In the second iteration, the pairs are obtained according to the formula (7)
Figure 100002_DEST_PATH_IMAGE116
Definition of (2) to
Figure 100002_DEST_PATH_IMAGE117
Is calculated by
Figure 100002_DEST_PATH_IMAGE118
Multivariate quantization probability function design parameters obtained in sub-iteration
Figure 100002_DEST_PATH_IMAGE119
A determined series of intermediate calculation terms
Figure 100002_DEST_PATH_IMAGE120
Wherein
Figure 100002_DEST_PATH_IMAGE121
（8）
Step two: using as in formula (8)
Figure 100002_DEST_PATH_IMAGE122
And sample set
Figure 100002_DEST_PATH_IMAGE123
The lower MSE bound on the right side of the inequality in equation (6) is approximately calculated as
Figure 100002_DEST_PATH_IMAGE124
Figure 100002_DEST_PATH_IMAGE125
（9）
Then, by using the interior point method and the gradient descent method, the following minimization problem is solved
Figure 100002_DEST_PATH_IMAGE126
（10）
Obtaining new design parameters after solving
Figure 100002_DEST_PATH_IMAGE127
。
Step three: calculating and viewing convergence criteria
Figure 100002_DEST_PATH_IMAGE128
If it is true.
If not, the explanation needs to continue iteration, and needs to be
Figure 100002_DEST_PATH_IMAGE129
Step A2 is carried out to carry out the next iteration, and the iteration count is updated
Figure 100002_DEST_PATH_IMAGE130
(ii) a If the convergence condition is established, outputting
Figure 100002_DEST_PATH_IMAGE131
As an optimal design parameter.
And (3) outputting: optimum design parameters
Figure 828095DEST_PATH_IMAGE103
The optimal design parameters can be obtained through the algorithm
Figure 250986DEST_PATH_IMAGE103
And corresponding to the optimal multivariate quantization probability function
Figure 100002_DEST_PATH_IMAGE132
。
S4. Fusion center FCSetting a metric fusion estimator and optimizing to obtain an optimal estimation function
Figure 100002_DEST_PATH_IMAGE133
；
As mentioned in the foregoing description of the preferred embodiment,
Figure 293897DEST_PATH_IMAGE075
a sensor jointly generates
Figure 991595DEST_PATH_IMAGE075
An
Figure 147770DEST_PATH_IMAGE081
Of bits
Figure 275650DEST_PATH_IMAGE083
And sent to the FC, which needs to estimate the original information using all the received quantized data. Therefore, we first present a quantitative fusion estimator design on FC.
As shown in fig. 5, FC receives a slave
Figure 100002_DEST_PATH_IMAGE134
Transmitted from a sensor
Figure 681224DEST_PATH_IMAGE134
Quantized data
Figure 100002_DEST_PATH_IMAGE135
And inputting it into the quantization fusion estimator, outputting the original information
Figure 100002_DEST_PATH_IMAGE136
Is estimated value of
Figure 100002_DEST_PATH_IMAGE137
. In the quantitative fusion estimator, of the input
Figure 100002_DEST_PATH_IMAGE138
An
Figure 100002_DEST_PATH_IMAGE139
Binary quantized data of bits
Figure 100002_DEST_PATH_IMAGE140
Is first converted into
Figure 819163DEST_PATH_IMAGE138
Is arranged at
Figure 100002_DEST_PATH_IMAGE141
Median decimal discrete data
Figure 100002_DEST_PATH_IMAGE142
；
Figure 16795DEST_PATH_IMAGE138
Decimal data
Figure 100002_DEST_PATH_IMAGE143
Then the data is sent to an Onehot function for carrying out one-hot coding operation to obtain the corresponding data
Figure 578227DEST_PATH_IMAGE138
One-hot coded vector
Figure 100002_DEST_PATH_IMAGE144
. By the first
Figure 100002_DEST_PATH_IMAGE145
Decimal data
Figure 100002_DEST_PATH_IMAGE146
By way of example only, it is possible to use,
Figure 100002_DEST_PATH_IMAGE147
is its corresponding one-hot coded vector,
Figure 100002_DEST_PATH_IMAGE148
is composed of
Figure 100002_DEST_PATH_IMAGE149
binary data of bit length, to
Figure 100002_DEST_PATH_IMAGE150
In total of
Figure 379086DEST_PATH_IMAGE149
The bits bit are numbered in sequence as
Figure 100002_DEST_PATH_IMAGE151
Bit, decimal data
Figure 100002_DEST_PATH_IMAGE152
The value range of (A) is just the serial number of all bits, and the one-hot coding means that only one bit is coded
Figure 100002_DEST_PATH_IMAGE153
To
Figure 100002_DEST_PATH_IMAGE154
The bit will be set to 1 and all the rest of the bit bits will be set to 0, i.e. the bit is set to 1
Figure 100002_DEST_PATH_IMAGE155
，
Figure 100002_DEST_PATH_IMAGE156
（11）
Then, the process of the present invention is carried out,
Figure 100002_DEST_PATH_IMAGE157
one-hot coded vector
Figure 100002_DEST_PATH_IMAGE158
Is fed into an averager to obtain their mean vectors
Figure 100002_DEST_PATH_IMAGE159
；
After that
Figure 100002_DEST_PATH_IMAGE160
Is then fed into the estimation function
Figure 100002_DEST_PATH_IMAGE161
In the method, an estimated value of the original information is output
Figure 100002_DEST_PATH_IMAGE162
，
Figure 100002_DEST_PATH_IMAGE163
Is also a design parameter
Figure 100002_DEST_PATH_IMAGE164
A function of control, i.e.
Figure 100002_DEST_PATH_IMAGE165
Figure 100002_DEST_PATH_IMAGE166
（12）
Wherein
Figure 100002_DEST_PATH_IMAGE167
Comprises a
Figure 100002_DEST_PATH_IMAGE168
All adjustable parameters.
When the same multivariate probability quantizer as shown in FIG. 3 is used on all sensors, and with the same optimal design parameters
Figure 100002_DEST_PATH_IMAGE169
Of the multivariate quantization probability function
Figure 100002_DEST_PATH_IMAGE170
Then the quantized data sent by all sensors to the FC is relative to the original information
Figure 100002_DEST_PATH_IMAGE171
Are conditionally independent and identically distributed;
at the moment, based on Bayesian estimation theory and probability model, the estimation value of FC to the original information is calculated
Figure 100002_DEST_PATH_IMAGE172
With the original information
Figure 100002_DEST_PATH_IMAGE173
MSE between is
Figure 100002_DEST_PATH_IMAGE174
Figure 100002_DEST_PATH_IMAGE175
（13）
Figure 100002_DEST_PATH_IMAGE176
In order to estimate the MSE,
Figure 100002_DEST_PATH_IMAGE177
（14）
following the pair in equation (7)
Figure 100002_DEST_PATH_IMAGE178
By quantifying the optimal design parameters of the probability function
Figure 100002_DEST_PATH_IMAGE179
And original information
Figure 100002_DEST_PATH_IMAGE180
A determined series of intermediate calculation terms;
the optimal design parameters of the probability function in the multivariate quantization are obtained from the formula (13)
Figure 452476DEST_PATH_IMAGE179
In the determined caseThe MSE estimated for the original information at FC is entirely determined by the estimation function
Figure 100002_DEST_PATH_IMAGE181
Variable design parameters of
Figure 100002_DEST_PATH_IMAGE182
Controlling;
s404, a series of samples based on original information collected from actual observation environment and local observation data of the sensor and an optimal multivariate quantization probability function on the sensor
Figure 100002_DEST_PATH_IMAGE183
Solving the optimum estimation function design parameters that minimize the estimated MSE on FC
Figure 100002_DEST_PATH_IMAGE184
。
Initialization: total number of input sensors
Figure 100002_DEST_PATH_IMAGE185
Optimal multivariate quantization probability function on sensor
Figure 16531DEST_PATH_IMAGE183
And their corresponding optimum design parameters
Figure 100002_DEST_PATH_IMAGE186
Sample set
Figure 100002_DEST_PATH_IMAGE187
Following the pair in equation (14)
Figure 100002_DEST_PATH_IMAGE188
Definition of (2) to arbitrary
Figure 100002_DEST_PATH_IMAGE189
Will be composed of
Figure 100002_DEST_PATH_IMAGE190
And original information samples
Figure 100002_DEST_PATH_IMAGE191
Intermediate item of decision
Figure 100002_DEST_PATH_IMAGE192
Is approximately calculated as
Figure 100002_DEST_PATH_IMAGE193
（15）
Setting initial design parameters
Figure 100002_DEST_PATH_IMAGE194
Setting the tolerance threshold of iteration to
Figure 100002_DEST_PATH_IMAGE195
Setting an initial iteration count to
Figure 100002_DEST_PATH_IMAGE196
。
The method comprises the following steps: in the first place
Figure 100002_DEST_PATH_IMAGE197
In the second iteration, define
Figure 100002_DEST_PATH_IMAGE198
For the estimated MSE at FC, using equation (13),
Figure 100002_DEST_PATH_IMAGE199
and the design parameters obtained in the first iteration
Figure 100002_DEST_PATH_IMAGE200
Will be
Figure 100002_DEST_PATH_IMAGE201
Is approximately calculated as
Figure 100002_DEST_PATH_IMAGE202
Figure 100002_DEST_PATH_IMAGE203
（16）
Using the interior point method and the gradient descent method, the following minimization problem is solved
Figure 100002_DEST_PATH_IMAGE204
（17）
Obtaining new design parameters
Figure 100002_DEST_PATH_IMAGE205
。
Step two: calculating and viewing convergence criteria
Figure 100002_DEST_PATH_IMAGE206
Whether the result is true; if not, continue iteration, will
Figure 100002_DEST_PATH_IMAGE207
Step B2 is carried out to carry out the next iteration and the iteration count is updated
Figure 100002_DEST_PATH_IMAGE208
(ii) a If the convergence condition is established, outputting
Figure 100002_DEST_PATH_IMAGE209
As an optimal design parameter.
And (3) outputting: optimum design parameters
Figure 100002_DEST_PATH_IMAGE210
。
The optimal design parameters can be obtained through the algorithm
Figure 791940DEST_PATH_IMAGE210
And corresponding optimal estimation function
Figure 100002_DEST_PATH_IMAGE211
。
S5. Based on
Figure 100002_DEST_PATH_IMAGE212
And
Figure 100002_DEST_PATH_IMAGE213
the distributed information estimation is quantized with multivariate probability.
Based on the above two algorithms, we obtain the optimal multivariate quantization probability functions on all sensors respectively
Figure 627565DEST_PATH_IMAGE212
And optimal estimation function on FC
Figure 470756DEST_PATH_IMAGE213
. Next, we briefly describe the whole process of the whole network to estimate the original information. The functional structures of the multivariate probability quantizer on the sensor and the quantized fusion estimator on the FC are described in detail above, and are not described herein again.
First of all, the first step is to,
Figure 100002_DEST_PATH_IMAGE214
individual sensor pair raw information
Figure 100002_DEST_PATH_IMAGE215
The observation was performed separately. By the first
Figure 100002_DEST_PATH_IMAGE216
An example of a sensor, it
Figure 905630DEST_PATH_IMAGE215
Obtaining own local observation data after observation
Figure 100002_DEST_PATH_IMAGE217
And will be
Figure 605601DEST_PATH_IMAGE217
Is sent into as shown in the figure3 (note that at this time, the multivariate quantization probability function in fig. 3
Figure DEST_PATH_IMAGE218
Has been optimized
Figure 269801DEST_PATH_IMAGE212
Replacement), and finally output
Figure DEST_PATH_IMAGE219
binary quantization data of bit
Figure DEST_PATH_IMAGE220
Is sent to the FC. All of
Figure 182262DEST_PATH_IMAGE214
A sensor jointly generates
Figure 349938DEST_PATH_IMAGE214
Quantized data
Figure DEST_PATH_IMAGE221
. FC receiving signals from all sensors
Figure 427003DEST_PATH_IMAGE214
Quantizes the data and feeds them into a quantized fusion estimator as shown in fig. 5 (note also that in this case, the estimation function in fig. 5
Figure DEST_PATH_IMAGE222
Has been optimized
Figure 47340DEST_PATH_IMAGE213
Substitution), and finally outputs an estimated value of the original information
Figure 497913DEST_PATH_IMAGE137
。
In the embodiment of the present application, considering the estimation performance of the proposed distributed information estimation method based on multivariate probability quantization on original information in a practical environment, as described above, we use the Mean Square Error (MSE) of the original information and its estimation value as an evaluation criterion, and a smaller MSE indicates better estimation performance. Specifically, we have tested the MSE performance of the network on the original information estimate when the total quantized bit number of the entire network changes and compared it with the current optimal binary quantization SQMLF method and the lower bound of the theoretical minimum MSE that can be reached on the original information estimate under binary quantization (one-bit quantization). As can be seen from fig. 6, although the SQMLF method has almost at all times completely approached the lower MSE bound for the original information estimation using the constraint of binary quantization, the estimated MSE for the original information using the distributed wireless sensor network of the multivariate probability quantization method is much smaller than both, which means that the multivariate probability quantization method we propose is better than any binary quantization method without considering the constraint on the number of bits to quantize data on the sensor. Furthermore, it can be observed that the multivariate quantization probability method still maintains the capability of approximately linearly decreasing with the total number of quantization bits under the condition that the total number of bits of the network is changed. The method verifies the efficient estimation performance of the multivariate probability quantification method in the distributed wireless sensor network and the adaptability and the expansibility of the total quantification bit number dynamic change of the network in the actual environment.
The foregoing description shows and describes a preferred embodiment of the invention, but as aforementioned, it is to be understood that the invention is not limited to the form disclosed herein, but is not to be construed as excluding other embodiments and from various other combinations, modifications, and environments and is capable of changes within the scope of the inventive concept as expressed herein, commensurate with the above teachings or the skill or knowledge of the relevant art. And that modifications and variations may be effected by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (10)

1. A distributed information estimation method based on multivariate probability quantization is characterized in that: the method comprises the following steps:
s1, constructing a distributed information estimation scene: the system comprises a fusion center FC positioned in the center of a wireless communication network and a plurality of sensors distributed at the edge of the wireless communication network;
each sensor is responsible for the raw information required by the fusion center
Figure DEST_PATH_IMAGE001
Observing to obtain own local observation data, performing multivariate probability quantization operation on the local observation data, converting continuous observation data into binary discrete data which can be used for digital communication, and sending the binary discrete data to a fusion center FC; the FC fusion center estimates original information according to the quantized data sent by all the sensors;
s2, constructing a multivariate probability quantizer for quantizing local observation data by a sensor;
s3, optimizing design parameters of multi-element quantization probability function
Figure DEST_PATH_IMAGE002
；
S4, designing a quantitative fusion estimator by a fusion center FC and optimizing to obtain an optimal estimation function
Figure DEST_PATH_IMAGE003
；
S5. Based on
Figure 315014DEST_PATH_IMAGE002
And
Figure 31822DEST_PATH_IMAGE003
the distributed information estimation is quantized with multivariate probability.
2. The distributed information estimation method based on multivariate probability quantization as claimed in claim 1, characterized in that: when each sensor observes the original information, the observation noise influenced by the environment is independently and identically distributed, and all the sensors use the same multivariate probability quantizer structure.
3. The distributed information estimation method based on multivariate probability quantization as claimed in claim 1, characterized in that: the step S2 includes:
s201, observing original information by a sensor
Figure DEST_PATH_IMAGE004
Obtaining local observations
Figure DEST_PATH_IMAGE005
By using
Figure DEST_PATH_IMAGE006
To express the observed value
Figure 857564DEST_PATH_IMAGE005
Relative to what is observed
Figure 580670DEST_PATH_IMAGE004
To describe randomness therebetween;
s202, obtaining an observed value by a sensor
Figure 988517DEST_PATH_IMAGE005
Then inputting the data into a multi-element probability quantizer and outputting a final quantization result
Figure DEST_PATH_IMAGE007
Quantification of the results
Figure DEST_PATH_IMAGE008
Is a one contains
Figure DEST_PATH_IMAGE009
Binary data of bits:
inside the quantizer, the observed value of the input
Figure 160299DEST_PATH_IMAGE005
Is first fed into a multivariate quantization probability control function
Figure DEST_PATH_IMAGE010
Is mapped to one
Figure DEST_PATH_IMAGE011
Probability vector of dimension
Figure DEST_PATH_IMAGE012
Probability vector
Figure DEST_PATH_IMAGE013
All of the elements in (A) are in
Figure DEST_PATH_IMAGE014
Interval values are taken while satisfying the condition that the sum of the additions is 1, i.e.
Figure DEST_PATH_IMAGE015
（1）
Wherein, the first and the second end of the pipe are connected with each other,
Figure DEST_PATH_IMAGE016
is a function of design parameters
Figure DEST_PATH_IMAGE017
A variable function of control having
Figure DEST_PATH_IMAGE018
（2）
Wherein
Figure DEST_PATH_IMAGE019
Comprises a
Figure DEST_PATH_IMAGE020
All of which are adjustableThe parameters of the number of the first and second antennas,
Figure DEST_PATH_IMAGE021
is the number of design parameters; by varying design parameters
Figure DEST_PATH_IMAGE022
By changing the parameter function accordingly
Figure 627665DEST_PATH_IMAGE020
The functional expressions and structures of (a);
from
Figure 888882DEST_PATH_IMAGE020
Output probability vector
Figure DEST_PATH_IMAGE023
Then fed into the quantization function
Figure DEST_PATH_IMAGE024
In, one-dimensional discrete value of one decimal is output
Figure DEST_PATH_IMAGE025
Then we go through decimal
Figure DEST_PATH_IMAGE026
Quantization result converted into binary
Figure DEST_PATH_IMAGE027
(ii) a For quantization function
Figure DEST_PATH_IMAGE028
Its output
Figure DEST_PATH_IMAGE029
A share of
Figure DEST_PATH_IMAGE030
Different results, it is desirable to make
Figure 164618DEST_PATH_IMAGE026
Taking the probability of each result entirely from
Figure 220299DEST_PATH_IMAGE030
Probability vector of dimension
Figure DEST_PATH_IMAGE031
Control, i.e. effecting
Figure DEST_PATH_IMAGE032
（3）
Wherein
Figure DEST_PATH_IMAGE033
Representing observations at a given input multivariate probability quantizer
Figure DEST_PATH_IMAGE034
Under the premise of (1), output is quantized
Figure DEST_PATH_IMAGE035
Take a value of
Figure DEST_PATH_IMAGE036
The probability of (c).
4. The distributed information estimation method based on multivariate probability quantization as claimed in claim 3, characterized in that: the quantization function
Figure DEST_PATH_IMAGE037
Is inputted by
Figure DEST_PATH_IMAGE038
Probability vector of dimension
Figure DEST_PATH_IMAGE039
The output being a one-dimensional discrete value
Figure DEST_PATH_IMAGE040
It is composed of
Figure DEST_PATH_IMAGE041
The serial sublayers with the same structure are composed, and the specific structural functions are as follows:
inputting: input device
Figure 516239DEST_PATH_IMAGE038
Probability vector of dimension
Figure DEST_PATH_IMAGE042
And an initial quantization value
Figure DEST_PATH_IMAGE043
；
M sub-layer, M =1,2, \ 8230, M: the input of M sub-layer is output of the last sub-layer
Figure DEST_PATH_IMAGE044
Vector of dimensions
Figure DEST_PATH_IMAGE045
And quantized value
Figure DEST_PATH_IMAGE046
(ii) a First, in the m-th sublayer, to be inputted
Figure DEST_PATH_IMAGE047
Divided into two sub-vectors of equal length, each containing
Figure 161240DEST_PATH_IMAGE047
All elements of the first half and all elements of the second half, i.e. two
Figure DEST_PATH_IMAGE048
Subvectors of dimensions
Figure DEST_PATH_IMAGE049
And
Figure DEST_PATH_IMAGE050
(ii) a Then, the m sub-layer utilizes
Figure DEST_PATH_IMAGE051
And
Figure DEST_PATH_IMAGE052
outputting the quantized value
Figure DEST_PATH_IMAGE053
Wherein
Figure DEST_PATH_IMAGE054
（4）
Figure DEST_PATH_IMAGE055
Is [0,1 ]]Random noise, function, evenly distributed over intervals
Figure DEST_PATH_IMAGE056
Inputting a non-negative number and outputting 1, otherwise, inputting a negative number and outputting 0; definition of
Figure DEST_PATH_IMAGE057
The mth sublayer output
Figure DEST_PATH_IMAGE058
Vector of dimensions
Figure DEST_PATH_IMAGE059
Wherein
Figure DEST_PATH_IMAGE060
(5) And (3) outputting: quantization function
Figure DEST_PATH_IMAGE061
Is output quantized value
Figure DEST_PATH_IMAGE062
Is that it is
Figure DEST_PATH_IMAGE063
Quantized values output by the sub-layers
Figure DEST_PATH_IMAGE064
I.e. by
Figure DEST_PATH_IMAGE065
。
5. The distributed information estimation method based on multivariate probability quantization as claimed in claim 1, characterized in that: in the step S3, bayesian estimation theory is used, and the method is considered
Figure DEST_PATH_IMAGE066
After the determined sensor quantifies the local observation, the fusion center finds the optimal design parameter suitable for the current observation environment by using the estimated MSE lower bound which can be reached by the quantified data
Figure DEST_PATH_IMAGE067
And corresponding use-optimized design parameters on the sensor
Figure 872843DEST_PATH_IMAGE067
Is optimized to a multivariate quantization probability function
Figure DEST_PATH_IMAGE068
。
6. The distributed information estimation method based on multivariate probability quantization as claimed in claim 1, characterized in that: the step S3 includes:
are all provided with
Figure DEST_PATH_IMAGE069
The independent sensors are distributed in the whole network, each sensor adopts the step S2 to construct a multi-element probability quantizer structure, and all the multi-element probability quantizers use the same multi-element quantization probability function
Figure DEST_PATH_IMAGE070
And corresponding design parameters
Figure DEST_PATH_IMAGE071
；
Figure 413939DEST_PATH_IMAGE069
Each sensor is used for respectively comparing the original information
Figure DEST_PATH_IMAGE072
Observing to obtain own local observation data with observation noise
Figure DEST_PATH_IMAGE073
；
Considering the presence of independent and identically distributed observation noise on each sensor, and defining a probability density function
Figure DEST_PATH_IMAGE074
To describe the distribution of observed noise: for the first
Figure DEST_PATH_IMAGE075
A sensor, its local observation data
Figure DEST_PATH_IMAGE076
Is quantized into a plurality of probability quantizers through a local multivariate probability quantizer
Figure DEST_PATH_IMAGE077
Binary data of bits
Figure DEST_PATH_IMAGE078
And is sent to the FC of the network center, so
Figure 885591DEST_PATH_IMAGE069
A sensor jointly generates
Figure 957452DEST_PATH_IMAGE069
An
Figure 928819DEST_PATH_IMAGE077
Quantized data of bits
Figure DEST_PATH_IMAGE079
And sending the information to the FC, wherein the FC needs to estimate the original information by using all received quantized data;
quantitative data on all sensors at a given time by Bayesian probability theory
Figure DEST_PATH_IMAGE080
Also in the case of (2), since they are independently and identically distributed, based on the probability distribution of the quantized data, it is first calculated when the FC receives the quantized data
Figure 919778DEST_PATH_IMAGE079
For the original information
Figure 713946DEST_PATH_IMAGE080
The lower bound of the estimated MSE that the estimation can achieve, i.e. the lower bound
Figure DEST_PATH_IMAGE081
Figure DEST_PATH_IMAGE082
（6）
Wherein
Figure DEST_PATH_IMAGE083
，
Figure DEST_PATH_IMAGE084
Indicating that FC receives all the quantized data sent from the sensor,
Figure DEST_PATH_IMAGE085
representing FC utilization quantized data
Figure DEST_PATH_IMAGE086
Pair that can be realized
Figure DEST_PATH_IMAGE087
Respectively, to the left of the inequality in the formula
Figure DEST_PATH_IMAGE088
Representing original information
Figure 678579DEST_PATH_IMAGE087
And its estimated value
Figure DEST_PATH_IMAGE089
The MSE between the two is the MSE,
Figure DEST_PATH_IMAGE090
shows the operation of calculating mathematical expectation, the right side of the inequality in the formula (6) shows the lower bound of the MSE,
Figure DEST_PATH_IMAGE091
is the number of combinations in the mathematical definition,
Figure DEST_PATH_IMAGE092
（7）
is based on a multivariate quantization probability function
Figure DEST_PATH_IMAGE093
Design parameters of
Figure DEST_PATH_IMAGE094
And original information
Figure DEST_PATH_IMAGE095
The determined series of intermediate calculation terms, as can be seen from equation (6), are in the original information
Figure 130551DEST_PATH_IMAGE095
Probability distribution and observed noise distribution of
Figure DEST_PATH_IMAGE096
FC utilizes quantized data pairs
Figure 844077DEST_PATH_IMAGE095
MSE estimated, i.e.
Figure DEST_PATH_IMAGE097
Whose lower bound is fully defined by the multivariate quantization probability function
Figure 388191DEST_PATH_IMAGE093
Design parameters of
Figure 333013DEST_PATH_IMAGE094
Determining;
minimizing the right side of the inequality in equation (6) by an algorithm
Figure DEST_PATH_IMAGE098
The determined FC uses the lower bound of the MSE estimated for the original information by the quantized data to find the adaptation to the currentOptimal design parameters under observation environment
Figure DEST_PATH_IMAGE099
And corresponding optimal multivariate quantization probability function
Figure DEST_PATH_IMAGE100
。
7. The distributed information estimation method based on multivariate probability quantization as claimed in claim 1, characterized in that: in the step S3, optimal design parameters suitable for the current observation environment are obtained
Figure 403606DEST_PATH_IMAGE099
And corresponding optimal multivariate quantization probability function
Figure 208139DEST_PATH_IMAGE100
The process comprises the following steps:
a1, setting the total number of sensors as
Figure DEST_PATH_IMAGE101
Sample set
Figure DEST_PATH_IMAGE102
,
Figure DEST_PATH_IMAGE103
Is the total number of samples contained in the sample set,
Figure DEST_PATH_IMAGE104
is a sample of the original information that was,
Figure DEST_PATH_IMAGE105
indicates the serial number of the sample, and
Figure DEST_PATH_IMAGE106
representing sensor versus raw information samples
Figure 203996DEST_PATH_IMAGE104
Total observations
Figure DEST_PATH_IMAGE107
Obtained by
Figure 483667DEST_PATH_IMAGE107
(ii) an observation sample; setting initial design parameters
Figure DEST_PATH_IMAGE108
Setting the tolerance threshold of iteration to
Figure DEST_PATH_IMAGE109
Setting an initial iteration count to
Figure DEST_PATH_IMAGE110
；
A2 in the first place
Figure DEST_PATH_IMAGE111
In the second iteration, the pairs are obtained according to the formula (7)
Figure DEST_PATH_IMAGE112
Definition of (2) to
Figure DEST_PATH_IMAGE113
Is calculated by
Figure DEST_PATH_IMAGE114
Multivariate quantization probability function design parameters obtained in sub-iteration
Figure DEST_PATH_IMAGE115
A series of intermediate calculation items of the decision
Figure DEST_PATH_IMAGE116
Wherein
Figure DEST_PATH_IMAGE117
（8）
A3, using the formula (8)
Figure DEST_PATH_IMAGE118
And sample set
Figure DEST_PATH_IMAGE119
The lower MSE bound on the right side of the inequality in equation (6) is approximately calculated as
Figure DEST_PATH_IMAGE120
Figure DEST_PATH_IMAGE121
（9）
Then using the interior point method and the gradient descent method, the following minimization problem is solved
Figure DEST_PATH_IMAGE122
（10）
Obtaining new design parameters after solving
Figure DEST_PATH_IMAGE123
；
A4, calculating and checking convergence conditions
Figure DEST_PATH_IMAGE124
Whether or not:
if not, the explanation needs to continue iteration, and needs to be
Figure DEST_PATH_IMAGE125
Step A2 is carried out to carry out the next iteration, and the iteration count is updated
Figure DEST_PATH_IMAGE126
(ii) a If the convergence condition is established, outputting
Figure DEST_PATH_IMAGE127
As an optimal design parameter, and determining a corresponding optimal multivariate quantization probability function
Figure DEST_PATH_IMAGE128
。
8. The distributed information estimation method based on multivariate probability quantization as claimed in claim 1, characterized in that: the step S4 includes:
s401.FC receives a slave
Figure DEST_PATH_IMAGE129
Transmitted from a sensor
Figure 874679DEST_PATH_IMAGE129
Quantized data
Figure DEST_PATH_IMAGE130
And inputting it into the quantization fusion estimator, outputting the original information
Figure DEST_PATH_IMAGE131
Is estimated value of
Figure DEST_PATH_IMAGE132
；
In the quantitative fusion estimator, of the input
Figure DEST_PATH_IMAGE133
An
Figure DEST_PATH_IMAGE134
Binary quantized data of bits
Figure DEST_PATH_IMAGE135
Is first converted into
Figure 224014DEST_PATH_IMAGE133
Is arranged at
Figure DEST_PATH_IMAGE136
Mean decimal discrete data
Figure DEST_PATH_IMAGE137
；
Figure 67686DEST_PATH_IMAGE133
Decimal data
Figure DEST_PATH_IMAGE138
Then the data is sent to an Onehot function for carrying out one-hot coding operation to obtain the corresponding data
Figure 151049DEST_PATH_IMAGE133
One-hot coded vector
Figure DEST_PATH_IMAGE139
；
S402, with the first
Figure DEST_PATH_IMAGE140
Decimal data
Figure DEST_PATH_IMAGE141
By way of example only, it is possible to use,
Figure DEST_PATH_IMAGE142
is its corresponding one-hot coded vector,
Figure DEST_PATH_IMAGE143
is composed of
Figure DEST_PATH_IMAGE144
binary data of bit length to
Figure DEST_PATH_IMAGE145
In total
Figure 651694DEST_PATH_IMAGE144
The bits bit are numbered in sequence as
Figure DEST_PATH_IMAGE146
Bit, decimal data
Figure DEST_PATH_IMAGE147
The value range of (1) is just the serial number of all bits, and the one-hot coding means that only one bit is coded
Figure DEST_PATH_IMAGE148
To (1)
Figure DEST_PATH_IMAGE149
The bit will be set to 1 and all the rest of the bit bits will be set to 0, i.e. the bit is set to 1
Figure DEST_PATH_IMAGE150
，
Figure DEST_PATH_IMAGE151
（11）
Then, the process of the present invention is carried out,
Figure DEST_PATH_IMAGE152
one-hot coded vector
Figure DEST_PATH_IMAGE153
Is fed into an averager to obtain their mean vectors
Figure DEST_PATH_IMAGE154
；
After that
Figure DEST_PATH_IMAGE155
Is then fed into the estimation function
Figure DEST_PATH_IMAGE156
In the method, an estimated value of the original information is output
Figure DEST_PATH_IMAGE157
，
Figure DEST_PATH_IMAGE158
Is also a design parameter
Figure DEST_PATH_IMAGE159
A function of control, i.e.
Figure DEST_PATH_IMAGE160
Figure DEST_PATH_IMAGE161
（12）
Wherein
Figure DEST_PATH_IMAGE162
Comprises a
Figure DEST_PATH_IMAGE163
All of the adjustable parameters;
s403. When the same multivariate probability quantizer is used on all sensors, and the method is used with the same optimal design parameters
Figure DEST_PATH_IMAGE164
Of the multivariate quantization probability function
Figure DEST_PATH_IMAGE165
Then the quantized data sent by all sensors to the FC is relative to the original information
Figure DEST_PATH_IMAGE166
Are conditionally independent and identically distributed;
at the moment, based on Bayesian estimation theory and probability model, the estimation value of FC to the original information is calculated
Figure DEST_PATH_IMAGE167
With the original information
Figure DEST_PATH_IMAGE168
MSE between
Figure DEST_PATH_IMAGE169
Figure DEST_PATH_IMAGE170
（13）
Figure DEST_PATH_IMAGE171
In order to estimate the MSE,
Figure DEST_PATH_IMAGE172
（14）
following the pair in equation (7)
Figure DEST_PATH_IMAGE173
By quantifying the optimal design parameters of the probability function
Figure DEST_PATH_IMAGE174
And original information
Figure DEST_PATH_IMAGE175
A determined series of intermediate calculation terms;
the optimal design parameters of the probability function in the multivariate quantization are obtained from the formula (13)
Figure 422464DEST_PATH_IMAGE174
The MSE estimated on FC for the original information is determined entirely by the estimation function
Figure DEST_PATH_IMAGE176
Variable design parameters of
Figure DEST_PATH_IMAGE177
Controlling;
s404, a series of samples based on original information collected from actual observation environment and local observation data of the sensor and an optimal multivariate quantization probability function on the sensor
Figure DEST_PATH_IMAGE178
Solving the optimum estimation function design parameters that minimize the estimated MSE on FC
Figure DEST_PATH_IMAGE179
。
9. The distributed information estimation method based on multivariate probability quantization as claimed in claim 8, wherein: the step S404 includes:
b1, total number of input sensors
Figure DEST_PATH_IMAGE180
Optimal multivariate quantization probability function on sensor
Figure 943313DEST_PATH_IMAGE178
And their corresponding optimum design parameters
Figure DEST_PATH_IMAGE181
Sample set
Figure DEST_PATH_IMAGE182
Following the pair in equation (14)
Figure DEST_PATH_IMAGE183
Definition of (2) to arbitrary
Figure DEST_PATH_IMAGE184
Will be composed of
Figure DEST_PATH_IMAGE185
And original information samples
Figure DEST_PATH_IMAGE186
Intermediate item of decision
Figure DEST_PATH_IMAGE187
Is approximately calculated as
Figure DEST_PATH_IMAGE188
（15）
Setting initial design parameters
Figure DEST_PATH_IMAGE189
Setting the tolerance threshold of iteration to
Figure DEST_PATH_IMAGE190
Setting an initial iteration count to
Figure DEST_PATH_IMAGE191
；
B2 in the first
Figure DEST_PATH_IMAGE192
In the second iteration, define
Figure DEST_PATH_IMAGE193
For the estimated MSE at FC, using equation (13),
Figure DEST_PATH_IMAGE194
and the design parameters obtained in the first iteration
Figure DEST_PATH_IMAGE195
Will be
Figure DEST_PATH_IMAGE196
Is approximately calculated as
Figure DEST_PATH_IMAGE197
Figure DEST_PATH_IMAGE198
（16）
Using the interior point method and the gradient descent method, the following minimization problem is solved
Figure DEST_PATH_IMAGE199
（17）
Obtaining new design parameters
Figure DEST_PATH_IMAGE200
；
B3, calculating and checking convergence conditions
Figure DEST_PATH_IMAGE201
Whether the result is true or not; if not, continue iteration, will
Figure DEST_PATH_IMAGE202
Step B2 is carried out to carry out the next iteration and the iteration count is updated
Figure DEST_PATH_IMAGE203
(ii) a If the convergence condition is established, outputting
Figure DEST_PATH_IMAGE204
As optimal design parameters and determining corresponding optimal estimation functions
Figure DEST_PATH_IMAGE205
。
10. The distributed information estimation method based on multivariate probability quantization as claimed in claim 1, characterized in that: the step S5 includes:
Figure DEST_PATH_IMAGE206
individual sensor pair raw information
Figure DEST_PATH_IMAGE207
And (3) respectively observing:
for the first
Figure DEST_PATH_IMAGE208
A sensor, it is to
Figure 653868DEST_PATH_IMAGE207
Obtaining own local observation data after observation
Figure DEST_PATH_IMAGE209
And will be
Figure 694505DEST_PATH_IMAGE209
Fed into a multivariate probability quantizer, multivariate quantization probability functions
Figure DEST_PATH_IMAGE210
Get the best
Figure DEST_PATH_IMAGE211
And finally output
Figure DEST_PATH_IMAGE212
binary quantization data of bit
Figure DEST_PATH_IMAGE213
Is sent to FC(ii) a All of
Figure 13008DEST_PATH_IMAGE206
A sensor jointly generates
Figure 240727DEST_PATH_IMAGE206
Quantized data
Figure DEST_PATH_IMAGE214
；
FC receiving signals from all sensors
Figure 462630DEST_PATH_IMAGE206
Quantizing the data and feeding them to a quantization fusion estimator, estimating a function
Figure DEST_PATH_IMAGE215
Get the best
Figure DEST_PATH_IMAGE216
Finally, an estimated value of the original information is output
Figure DEST_PATH_IMAGE217
。
CN202211229494.1A 2022-10-08 2022-10-08 Distributed information estimation method based on multivariate probability quantization Active CN115290130B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211229494.1A CN115290130B (en) 2022-10-08 2022-10-08 Distributed information estimation method based on multivariate probability quantization

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211229494.1A CN115290130B (en) 2022-10-08 2022-10-08 Distributed information estimation method based on multivariate probability quantization

Publications (2)

Publication Number Publication Date
CN115290130A true CN115290130A (en) 2022-11-04
CN115290130B CN115290130B (en) 2022-12-16

Family

ID=83819224

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211229494.1A Active CN115290130B (en) 2022-10-08 2022-10-08 Distributed information estimation method based on multivariate probability quantization

Country Status (1)

Country Link
CN (1) CN115290130B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130297298A1 (en) * 2012-05-04 2013-11-07 Sony Computer Entertainment Inc. Source separation using independent component analysis with mixed multi-variate probability density function
CN107451101A (en) * 2017-07-21 2017-12-08 江南大学 It is a kind of to be layered integrated Gaussian process recurrence soft-measuring modeling method
CN110035405A (en) * 2019-04-18 2019-07-19 电子科技大学 A kind of efficient fusion method of Multisensor Distributed based on random set theory
CN114235257A (en) * 2021-12-24 2022-03-25 长春工业大学 Bolt axial force online accurate monitoring method and device based on data fusion
CN114900817A (en) * 2022-07-12 2022-08-12 香港中文大学（深圳） Distributed wireless sensor network information estimation method based on binary quantization

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130297298A1 (en) * 2012-05-04 2013-11-07 Sony Computer Entertainment Inc. Source separation using independent component analysis with mixed multi-variate probability density function
CN107451101A (en) * 2017-07-21 2017-12-08 江南大学 It is a kind of to be layered integrated Gaussian process recurrence soft-measuring modeling method
CN110035405A (en) * 2019-04-18 2019-07-19 电子科技大学 A kind of efficient fusion method of Multisensor Distributed based on random set theory
CN114235257A (en) * 2021-12-24 2022-03-25 长春工业大学 Bolt axial force online accurate monitoring method and device based on data fusion
CN114900817A (en) * 2022-07-12 2022-08-12 香港中文大学（深圳） Distributed wireless sensor network information estimation method based on binary quantization

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
郭黎利等: "无线传感器网络中基于多比特量化的极大似然分布式估计方法", 《电子学报》 *

Also Published As

Publication number Publication date
CN115290130B (en) 2022-12-16

Similar Documents

Publication Publication Date Title
Luo An isotropic universal decentralized estimation scheme for a bandwidth constrained ad hoc sensor network
CN113537514A (en) High-energy-efficiency federal learning framework based on digital twins
CN113381952A (en) Multi-antenna system channel estimation method based on deep learning
CN113259968A (en) Intelligent calculation method for power distribution network equipment based on information freshness
Kar et al. Distributed detection in noisy sensor networks
Gupta et al. SBL-based adaptive sensing framework for WSN-assisted IoT applications
US11888703B1 (en) Machine learning algorithms for quality of service assurance in network traffic
Zhang et al. Deep learning based user association in heterogeneous wireless networks
CN115290130B (en) Distributed information estimation method based on multivariate probability quantization
CN114900817B (en) Distributed wireless sensor network information estimation method based on binary quantization
WO2022089217A1 (en) Key generation method and apparatus, electronic device, and storage medium
Liao et al. A linearly convergent robust compressed push-pull method for decentralized optimization
Gray Quantization in task-driven sensing and distributed processing
Shen et al. An efficient sensor quantization algorithm for decentralized estimation fusion
Vucic et al. Fixed point iteration for max-min SIR balancing with general interference functions
Malka et al. Decentralized low-latency collaborative inference via ensembles on the edge
CN108848519B (en) Heterogeneous network user access method based on cross entropy learning
Wang et al. Distributed online quantile regression over networks with quantized communication
WO2020190340A1 (en) Successive approximation analog to digital converter
CN116471612A (en) Method for jointly optimizing quantization level and client selection in federal learning network
Wang et al. Distortion outage minimization in Nakagami fading using limited feedback
CN114337881A (en) Wireless spectrum intelligent sensing method based on multi-unmanned aerial vehicle distribution and LMS
CN108768470B (en) Information transmission method of relay cooperative communication system under feedback limitation
WO2020087293A1 (en) Communication receiver and method for processing signal
Madkar et al. Machine learning based efficient routing protocol in wireless sensor network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant