CN113839744B - Blind detection method of generalized wireless optical MIMO system based on deep learning - Google Patents
Blind detection method of generalized wireless optical MIMO system based on deep learning Download PDFInfo
- Publication number
- CN113839744B CN113839744B CN202111108640.0A CN202111108640A CN113839744B CN 113839744 B CN113839744 B CN 113839744B CN 202111108640 A CN202111108640 A CN 202111108640A CN 113839744 B CN113839744 B CN 113839744B
- Authority
- CN
- China
- Prior art keywords
- layer
- blind
- deep learning
- input
- wireless optical
- 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.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 58
- 230000003287 optical effect Effects 0.000 title claims abstract description 26
- 238000013135 deep learning Methods 0.000 title claims abstract description 23
- 239000013598 vector Substances 0.000 claims abstract description 25
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000013528 artificial neural network Methods 0.000 claims abstract description 13
- 230000006870 function Effects 0.000 claims description 20
- 210000002569 neuron Anatomy 0.000 claims description 17
- 238000013507 mapping Methods 0.000 claims description 14
- 238000000034 method Methods 0.000 claims description 14
- 239000011159 matrix material Substances 0.000 claims description 11
- 230000008054 signal transmission Effects 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 8
- 230000004913 activation Effects 0.000 claims description 7
- 230000005540 biological transmission Effects 0.000 claims description 6
- 239000000654 additive Substances 0.000 claims description 3
- 230000000996 additive effect Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 2
- 238000004891 communication Methods 0.000 abstract description 7
- 230000000694 effects Effects 0.000 abstract description 3
- 238000001228 spectrum Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
- H04L1/0036—Systems modifying transmission characteristics according to link quality, e.g. power backoff arrangements specific to the receiver
- H04L1/0038—Blind format detection
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0045—Arrangements at the receiver end
- H04L1/0047—Decoding adapted to other signal detection operation
- H04L1/0048—Decoding adapted to other signal detection operation in conjunction with detection of multiuser or interfering signals, e.g. iteration between CDMA or MIMO detector and FEC decoder
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0045—Arrangements at the receiver end
- H04L1/0054—Maximum-likelihood or sequential decoding, e.g. Viterbi, Fano, ZJ algorithms
-
- 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)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Radio Transmission System (AREA)
Abstract
The invention relates to the technical field of wireless optical communication, and particularly discloses a blind detection method of a generalized wireless optical MIMO system based on deep learning r Amplifying and extracting the characteristics of a received signal y output by the PD receiver in parallel to obtain an input signalThen input signal is input through blind detector which is trained in advanceAnd detecting and directly outputting corresponding bit information. The blind detector is built based on a deep learning neural network, and learns correct bit vectors and input signals after training is completedThe relation between the two signals can be real-time compared with the input signalAnd (6) detecting. The invention can obtain the error code performance equivalent to the Joint-ML detection scheme with low computation complexity equivalent to the ZF-ML detection scheme under the condition of no real-time channel information, and obtain the optimal detection effect same as the Joint-ML detection scheme。
Description
Technical Field
The invention relates to the technical field of wireless optical communication, in particular to a blind detection method of a generalized wireless optical MIMO system based on deep learning.
Background
In recent years, wireless optical communication (OWC) technology has attracted much attention due to its characteristics of abundant spectrum resources, green and low power consumption. Among them, wireless optical communication using infrared, visible light, or ultraviolet Light Emitting Diodes (LEDs) is considered to be a very potential technology capable of satisfying the demand for high-speed communication. However, due to the limited modulation bandwidth of LEDs, the implementation and development of OWC systems is limited in practical applications. To date, there are many techniques for increasing the capacity of a band-limited OWC system, and Multiple Input Multiple Output (MIMO) transmission is a technique capable of effectively increasing the system capacity. In order to further exploit the advantages of MIMO systems, researchers have further proposed generalized wireless optical MIMO (gomimo) transmission techniques.
In order to successfully implement the GOMIMO system, an efficient MIMO detection scheme method should be adopted. Similar to the conventional MIMO system, Joint-maximum likelihood (Joint-ML) detection is generally the optimal detection scheme for the GOMIMO system. However, Joint-ML detection has extremely high computational complexity, which makes it limited in practical applications. In addition, joint detection based on zero-forcing equalization and ML (ZF-ML) is also a practical low-complexity detection scheme suitable for the GOMIMO system. However, both Joint-ML and ZF-ML detection schemes require that accurate real-time channel information be known in advance for successful implementation, and therefore the system must perform channel estimation in advance. Real-time channel estimation in the GOMIMO system usually needs to be completed by transmitting additional training symbols, and the use of the additional training symbols will inevitably cause the reduction of the system spectrum efficiency and the increase of the system delay.
Disclosure of Invention
The invention provides a blind detection method of a generalized wireless optical MIMO system based on deep learning, which solves the technical problems that: how to achieve the error performance equivalent to the Joint-ML detection scheme without the need of real-time channel information.
In order to solve the technical problems, the invention provides a blind detection method of a generalized wireless optical MIMO system based on deep learning, which comprises the following steps:
s1, for N r Preprocessing the received signal y output by the PD receiver in parallel to obtain an input signal
S2, adopting the trained blind detector to input signalsAnd detecting and outputting corresponding bit information.
Preferably, the blind detector comprises 1 input layer, 4 hidden layers, 1 output layer and 1 decision layer which are sequentially arranged;
the input layer is provided with N r A number of neurons, equal to the number of PD receivers, for converting an input signalInputting the data to the hidden layers one by one; the hidden layer is sequentially provided with four layers, and L is correspondingly arranged 1 、L 2 、L 3 、L 4 A neuron for learning an input signal by trainingAnd statistical characteristics of additive noise; the output layer is provided with S neurons, the S neurons correspond to the total bit number carried by each input vector and are used for generating fuzzy bit information of which the numerical value interval is (0,1) of each input vector; the decision layer is provided with S neurons and is used for judging the S fuzzy bit information generated by the output layer to generate S0 or 1 bit information.
Further, the preprocessing process in step S1 specifically includes:
amplifying the received signal y by alpha times and according to N t The mapping relation of the LED transmitter during information transmission generates a characteristic matrix F, and finally generates an input signalN r ≥N t 。
Preferably, the functional relationship among the input layer, the hidden layer and the output layer is expressed as follows:
wherein k ═ 1 denotes a first layer, i.e., the input layer; k 2 ≦ k ≦ 5 denotes the second to fifth layers, i.e., the 4-layer hidden layers, and k ≦ 6 denotes the 6-th layer, i.e., the output layer; z is a radical of k Represents the output vector of the k-th layer, W k-1 Is a weight matrix of the k-1 th layer and the k layer, b k-1 Is the bias vector of the k layer; the activation function of the hidden layer is a ReLU function f ReLU (x) Max (0, x); the activation function of the output layer is a Sigmoid function f Sigmoid (x)=1/(1+exp -x )。
Preferably, let z be q In order to be a vague bit, the bit is,the decision rule of the decision layer is as follows:
preferably, it is assumed that N is r A PD receiver and N t In the mapping process of each LED transmitter, only N is used a The LED transmitters are activated for signal transmission, and N is more than or equal to 1 a ≤N t Then the feature matrix F is represented as N t ×N r Wherein each row and each column has N a The assignment of one element to 1 indicates a signal transmission, and the assignment of the other elements to 0 indicates no signal transmission.
Preferably, the loss function selected by the blind detector is a mean square error:
wherein,a bit vector estimated for the decision layer and b is the label of the correct bit vector transmitted, i.e. the received signal y.
Preferably, in the off-line training stage of the blind detector, the blind detection related parameters are tested and determined at different positions in a specific space, then the neural network in the blind detector is trained, and the trained blind detector is stored; when N is present r When the PD receiver is located at a certain position, the feedforward deep neural network module directly calls a trained blind detector corresponding to the position to detect signals.
The invention provides a blind detection method of a generalized wireless optical MIMO system based on deep learning, which is characterized in that a preprocessing module is arranged to carry out N pairs of N pairs according to the mapping relation between an LED transmitter and a PD receiver r Amplifying and extracting the characteristics of a received signal y output by the PD receiver in parallel to obtain an input signalThen the input signal is input through a blind detector in a feedforward deep neural network module which is trained in advanceAnd detecting and directly outputting corresponding bit information. The blind detector is built based on a deep learning neural network, and learns correct bit vectors and input signals after training through an input layer, a hidden layer, an output layer, a decision layer, parameters and function settingsThe relation between the two signals, can be real-time compared with the input signalAnd (6) detecting. The invention solves the problem of dependence of traditional Joint-ML detection and ZF-ML detection on channel information, and can obtain the same optimal detection effect as the Joint-ML detection scheme under the condition of no real-time channel information; meanwhile, the blind detection method of the GOMIMO system based on deep learning solves the problem of high technical complexity of the Joint-ML detection scheme, and can successfully realize the optimal detection of the GOMIMO system with low computation complexity equivalent to that of the ZF-ML detection scheme. The blind detection method of the GOMIMO system based on deep learning provided by the invention can obtain the error code performance equivalent to the Joint-ML detection scheme under the condition of not needing real-time channel information, and has the following main advantages:
1. real-time channel estimation is not needed, so that extra training symbols are not needed for real-time channel estimation, the frequency spectrum efficiency of the GOMIMO system can be improved, and the communication delay is reduced;
2. with a low computational complexity comparable to the ZF-ML detection scheme.
Drawings
Fig. 1 is a schematic block diagram illustrating an application of a generalized wireless optical MIMO system and a blind detection method thereof according to an embodiment of the present invention;
fig. 2(a) and (b) are mapping tables of a PD transmitter and an LED receiver under GOSM (generalized optical spatial modulation) and GOSMP (generalized optical spatial multiplexing), respectively, according to an embodiment of the present invention;
fig. 3 is a flow chart of preprocessing and blind detector detection provided by an embodiment of the present invention.
Detailed Description
The embodiments of the present invention will be described in detail below with reference to the accompanying drawings, which are given solely for the purpose of illustration and are not to be construed as limitations of the present invention, including reference to and illustration of the accompanying drawings, which are not to be construed as limitations of the scope of the invention, since many variations thereof are possible without departing from the spirit and scope of the invention.
FIG. 1 shows the application of the blind detection method of the present invention to N r ×N t Schematic block diagram on GOMIMO system, and the GOMIMO system comprises N r A PD receivesMachine and N t An LED transmitter. Wherein, in the GOMIMO mapping process, only N is included a Each (1 is less than or equal to N) a ≤N t ) The LED transmitter is activated for signal transmission. As shown in fig. 1, the input bitstream is first divided into two parts: one part is used for generating a constellation symbol vector c by traditional signal modulation, and the other part is used for generating a space index vector v by LED index selection. Based on the obtained constellation symbol vector c and spatial index vector v, GOMIMO (GOSM or gos mp) mapping is performed to generate a transmission signal vector x. FIGS. 2(a) and (b) show N, respectively t 4 and N a Mapping tables of GOSM (generalized optical spatial modulation) and GOSMP (generalized optical spatial multiplexing) corresponding to 2. The GOMIMO detection in fig. 1 refers to the blind detection method of the present invention.
Specifically, as shown in fig. 3, the blind detection method for the generalized wireless optical MIMO system provided in this embodiment is based on deep learning, and includes the steps of:
s1, for N r Preprocessing the received signal y output by the PD receiver in parallel to obtain an input signal
S2, input signal pair by adopting trained blind detectorAnd detecting and outputting corresponding bit information.
As shown in fig. 3, the preprocessing process in step S1 specifically includes:
amplifying the received signal y by alpha times and according to N t Generating characteristic matrix F by mapping relation of LED transmitter in information transmission, and finally generating input signalN r ≥N t . The signal amplification factor α can be set by pre-testing, and the feature matrix F for feature extraction is determined by GOMIMO mapping. In N r A PD receiver and N t In the mapping process of each LED transmitterOf only N a The LED transmitters are activated for signal transmission, and N is more than or equal to 1 a ≤N t Then the feature matrix F is represented as N t ×N r Wherein each row and each column has N a The assignment of one element to 1 indicates a signal transmission, and the assignment of the other elements to 0 indicates no signal transmission. Taking fig. 2 as an example, the feature matrix corresponding to the two mapping manners of fig. 2 is:
more specifically, the blind detector includes 1 input layer, 4 hidden layers, 1 output layer, and 1 decision layer, which are sequentially arranged.
As shown in FIG. 3, the input layer is provided with N r A number of neurons, equal to the number of PD receivers, for converting an input signalInputting the data into the hidden layers one by one; four layers are sequentially arranged on the hidden layer, and L is correspondingly arranged 1 、L 2 、L 3 、L 4 A neuron for learning an input signal by trainingAnd statistical characteristics of additive noise; the output layer is provided with S neurons, the S neurons correspond to the total bit number carried by each input vector and are used for generating fuzzy bit information of which the numerical value interval is (0,1) of each input vector; the decision layer is provided with S neurons and is used for judging the S fuzzy bit information generated by the output layer to generate S0 or 1 bit information. Each dot in the neural network module stores a scalar value for each neuron, and each neuron has an activation function.
The functional relationship of the input layer, the hidden layer and the output layer is expressed as follows:
wherein k ═ 1 denotes a first layer, i.e., an input layer; k is 2 ≦ k ≦ 5 for the second to fifth layers, i.e., the 4-layer hidden layer, and k is 6 for the 6 th layer, i.e., the output layer; z is a radical of k Represents the output vector of the k-th layer, W k-1 Is a weight matrix of the k-1 th layer and the k layer, b k-1 Is the bias vector of the k layer; the activation function of the hidden layer is a ReLU function f ReLU (x) Max (0, x); the activation function of the output layer is Sigmoid function f Sigmoid (x)=1/(1+exp -x )。
Suppose z q In order to be a vague bit, the bit is,the decision rule of the decision layer is as follows:
the input layer, the hidden layer and the output layer of the decision layer in the blind detector are typical deep neural networks, the back propagation training is carried out by taking a loss function as a standard through pre-generated symbols and corresponding labels, the weight value in the network is iterated, and the process is repeated for a plurality of times until the loss function is converged, namely the training of the neural networks is completed. In a GOMIMO system, the used training symbols are the direct received signals of the GOMIMO system, and the labels are the corresponding correct bit information. The selected loss function is the Mean Square Error (MSE) which can be expressed as:
wherein,is the bit vector estimated by the decision layer and b is the correct bit vector transmitted, i.e. the label of the received signal y.
Taking a room as an example, in the off-line training stage of the blind detector, firstly, the room isTesting different positions in the room and determining blind detection related parameters, then training a neural network in the blind detector, and storing the trained blind detector; when N is present r When a PD receiver is positioned at a certain position, a trained blind detector corresponding to the position is directly called to detect signals.
In summary, in the blind detection method for the generalized wireless optical MIMO system based on deep learning according to the embodiments of the present invention, the preprocessing module is configured to perform N pairs according to the mapping relationship between the LED transmitter and the PD receiver r Amplifying and extracting the characteristics of a received signal y output by the PD receiver in parallel to obtain an input signalThen the input signal is input through a blind detector in a feedforward deep neural network module which is trained in advanceAnd detecting and directly outputting corresponding bit information. The blind detector is built based on a deep learning neural network, and learns correct bit vectors and input signals after training through an input layer, a hidden layer, an output layer, a decision layer, parameters and function settingsThe relation between the two signals, can be real-time compared with the input signalAnd (6) detecting. The invention solves the problem of dependence of traditional Joint-ML detection and ZF-ML detection on channel information, and can obtain the same optimal detection effect as the Joint-ML detection scheme under the condition of no real-time channel information; meanwhile, the blind detection scheme of the GOMIMO system based on deep learning provided by the invention solves the problem of higher technical complexity of the Joint-ML detection scheme, and can successfully realize the optimal detection of the GOMIMO system with low computation complexity equivalent to that of the ZF-ML detection scheme. The deep learning-based blind detection method for the GOMIMO system can be used without the need ofThe method obtains the error code performance equivalent to the Joint-ML detection scheme under the condition of real-time channel information, and has the main advantages that:
1. real-time channel estimation is not needed, so that extra training symbols are not needed for real-time channel estimation, the frequency spectrum efficiency of the GOMIMO system can be improved, and the communication delay is reduced;
2. with a low computational complexity comparable to the ZF-ML detection scheme.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.
Claims (8)
1. A blind detection method of a generalized wireless optical MIMO system based on deep learning is characterized by comprising the following steps:
s1, for N r Preprocessing the received signal y output by the PD receiver in parallel to obtain an input signal
2. The blind detection method for the generalized wireless optical MIMO system based on deep learning of claim 1, wherein: the blind detector comprises 1 input layer, 4 hidden layers, 1 output layer and 1 decision layer which are sequentially arranged;
the input layer is provided with N r A number of neurons, equal to the number of PD receivers, for converting an input signalInputting the data to the hidden layers one by one; the hidden layer is sequentially provided with four layers, and L is correspondingly arranged 1 、L 2 、L 3 、L 4 A neuron for learning an input signal by trainingAnd statistical characteristics of additive noise; the output layer is provided with S neurons, the S neurons correspond to the total bit number carried by each input vector and are used for generating fuzzy bit information of which the numerical value interval is (0,1) of each input vector; the decision layer is provided with S neurons and is used for judging the S fuzzy bit information generated by the output layer to generate S0 or 1 bit information.
3. The blind detection method of the deep learning-based generalized wireless optical MIMO system according to claim 2, wherein the preprocessing process in the step S1 specifically includes:
4. The blind detection method for the generalized wireless optical MIMO system based on deep learning of claim 3, wherein the functional relationship among the input layer, the hidden layer and the output layer is expressed as follows:
wherein k ═ 1 denotes a first layer, i.e., the input layer; k 2 ≦ k ≦ 5 denotes the second to fifth layers, i.e., the 4-layer hidden layers, and k ≦ 6 denotes the 6-th layer, i.e., the output layer; z is a radical of k Representing the output of the k-th layerVector, W k-1 Is a weight matrix of the k-1 th layer and the k layer, b k-1 Is the bias vector of the k layer; the activation function of the hidden layer is a ReLU function f ReLU (x) Max (0, x); the activation function of the output layer is a Sigmoid function f Sigmoid (x)=1/(1+exp -x )。
6. the blind detection method for the generalized wireless optical MIMO system based on deep learning according to claim 4, wherein: suppose in N r A PD receiver and N t In the mapping process of each LED transmitter, only N is used a The LED transmitters are activated for signal transmission, and N is more than or equal to 1 a ≤N t Then the feature matrix F is represented as N t ×N r Wherein each row and each column has N a The assignment of one element to 1 indicates a signal transmission, and the assignment of the other elements to 0 indicates no signal transmission.
8. The blind detection method for the deep learning-based generalized wireless optical MIMO system according to claim 7, wherein: in the off-line training stage of the blind detector, testing and determining blind detection related parameters at different positions in a specific space, training a neural network in the blind detector, and storing the trained blind detector; when N is present r When the PD receiver is located at a certain position, the feedforward deep neural network module directly calls a trained blind detector corresponding to the position to detect signals.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111108640.0A CN113839744B (en) | 2021-09-22 | 2021-09-22 | Blind detection method of generalized wireless optical MIMO system based on deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111108640.0A CN113839744B (en) | 2021-09-22 | 2021-09-22 | Blind detection method of generalized wireless optical MIMO system based on deep learning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113839744A CN113839744A (en) | 2021-12-24 |
CN113839744B true CN113839744B (en) | 2022-09-02 |
Family
ID=78960369
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111108640.0A Active CN113839744B (en) | 2021-09-22 | 2021-09-22 | Blind detection method of generalized wireless optical MIMO system based on deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113839744B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114363218B (en) * | 2022-01-07 | 2023-07-28 | 合肥工业大学 | Communication reachable rate detection method based on end-to-end learning |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102104957A (en) * | 2009-12-21 | 2011-06-22 | 华为技术有限公司 | Method and device for indicating multiple user-multiple input multiple output (MU-MIMO) mode |
CN103259631A (en) * | 2013-05-13 | 2013-08-21 | 杭州电子科技大学 | Frequency spectrum blind detecting method based on characteristic functions in MIMO system |
CN103763222A (en) * | 2014-01-16 | 2014-04-30 | 西安电子科技大学 | Channel ambiguity removing method in MIMO signal blind detection process |
CN103916344A (en) * | 2014-03-07 | 2014-07-09 | 南京邮电大学 | Virtual multiple input and multiple output system signal blind detection method of wireless sensor network |
CN106100789A (en) * | 2016-06-13 | 2016-11-09 | 南京邮电大学 | Multilamellar MIMO wireless sensor network blind checking method based on sub-clustering |
CN108390705A (en) * | 2018-03-29 | 2018-08-10 | 东南大学 | The extensive mimo system detection method of deep neural network based on BP algorithm structure |
CN109921882A (en) * | 2019-02-20 | 2019-06-21 | 深圳市宝链人工智能科技有限公司 | A kind of MIMO coding/decoding method, device and storage medium based on deep learning |
CN110518945A (en) * | 2019-08-20 | 2019-11-29 | 东南大学 | A kind of MIMO detection method and device based on deep learning Yu SDR algorithm |
CN111769862A (en) * | 2020-06-28 | 2020-10-13 | 南京理工大学 | Joint detection algorithm in spatial modulation network based on deep learning |
CN112821938A (en) * | 2021-01-08 | 2021-05-18 | 重庆大学 | Total throughput and energy consumption optimization method of air-space-ground satellite communication system |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101668336B (en) * | 2008-09-04 | 2012-05-02 | 电信科学技术研究院 | Method and device for MIMO transmission of high-speed packet access evolution system |
CN103812602B (en) * | 2012-11-09 | 2019-05-28 | 北京三星通信技术研究有限公司 | The method and apparatus of blind examination public search space and the specific search space UE |
-
2021
- 2021-09-22 CN CN202111108640.0A patent/CN113839744B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102104957A (en) * | 2009-12-21 | 2011-06-22 | 华为技术有限公司 | Method and device for indicating multiple user-multiple input multiple output (MU-MIMO) mode |
CN103259631A (en) * | 2013-05-13 | 2013-08-21 | 杭州电子科技大学 | Frequency spectrum blind detecting method based on characteristic functions in MIMO system |
CN103763222A (en) * | 2014-01-16 | 2014-04-30 | 西安电子科技大学 | Channel ambiguity removing method in MIMO signal blind detection process |
CN103916344A (en) * | 2014-03-07 | 2014-07-09 | 南京邮电大学 | Virtual multiple input and multiple output system signal blind detection method of wireless sensor network |
CN106100789A (en) * | 2016-06-13 | 2016-11-09 | 南京邮电大学 | Multilamellar MIMO wireless sensor network blind checking method based on sub-clustering |
CN108390705A (en) * | 2018-03-29 | 2018-08-10 | 东南大学 | The extensive mimo system detection method of deep neural network based on BP algorithm structure |
CN109921882A (en) * | 2019-02-20 | 2019-06-21 | 深圳市宝链人工智能科技有限公司 | A kind of MIMO coding/decoding method, device and storage medium based on deep learning |
CN110518945A (en) * | 2019-08-20 | 2019-11-29 | 东南大学 | A kind of MIMO detection method and device based on deep learning Yu SDR algorithm |
CN111769862A (en) * | 2020-06-28 | 2020-10-13 | 南京理工大学 | Joint detection algorithm in spatial modulation network based on deep learning |
CN112821938A (en) * | 2021-01-08 | 2021-05-18 | 重庆大学 | Total throughput and energy consumption optimization method of air-space-ground satellite communication system |
Non-Patent Citations (4)
Title |
---|
"Joint Detection for Generalized Optical MIMO: A Deep Learning Approach";Xin Zhong, Chen Chen, Lin Zeng, Ruochen Zhang;《2021 IEEE 16th Conference on Industrial Electronics and Applications (ICIEA)》;20210830;全文 * |
"Virtual MIMO Blind Detection ClusteredWSN System";Chaodi Jin,Yun Zhang,Shujuan Yu;《2015 Asia-Pacific Microwave Conference (APMC)》;20160225;全文 * |
"基于自适应PI控制方法的海上浮式风机独立变桨控制研究";陈晨;《中国优秀硕士学位论文全文数据库 工程科技辑》;20170331;全文 * |
"空间调制MIMO系统盲检测研究与设计";续澄澄;《中国优秀硕士学位论文全文数据库 信息科技辑》;20210228;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN113839744A (en) | 2021-12-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114051701B (en) | Apparatus and method for machine learning assisted precoding | |
CN112600772B (en) | OFDM channel estimation and signal detection method based on data-driven neural network | |
CN110336594B (en) | Deep learning signal detection method based on conjugate gradient descent method | |
CN109039534A (en) | A kind of sparse CDMA signals detection method based on deep neural network | |
CN107124378B (en) | Visible light channel joint equalization method based on orthogonal mapping and probabilistic neural network | |
CN110098870B (en) | Optical generalized spatial modulation method based on OB-MMSE detection algorithm | |
CN109905170A (en) | A kind of nonlinear distortion compensation algorithm and visible light communication device based on K-DNN | |
CN113626960B (en) | Anti-mode coupling few-mode signal complex format analysis method and device | |
Askri et al. | DNN assisted sphere decoder | |
Lee et al. | A deep learning approach to universal binary visible light communication transceiver | |
CN113839744B (en) | Blind detection method of generalized wireless optical MIMO system based on deep learning | |
CN110311876A (en) | The implementation method of underwater sound OFDM receiver based on deep neural network | |
CN113381828A (en) | Sparse code multiple access random channel modeling method based on condition generation countermeasure network | |
CN112242969A (en) | Novel single-bit OFDM receiver based on model-driven deep learning | |
CN116318311B (en) | Transmission method based on reconfigurable intelligent surface anti-phase index modulation | |
KR102064301B1 (en) | Signal detection apparatus using ensemble machine learning based on MIMO system and method thereof | |
CN114598582A (en) | Ultra-high order signal modulation format rapid identification method based on transfer learning | |
CN114513394B (en) | Signal modulation format identification method, system and device based on attention mechanism diagram neural network and storage medium | |
Hua et al. | Signal detection in uplink pilot-assisted multi-user MIMO systems with deep learning | |
Ye et al. | Bilinear convolutional auto-encoder based pilot-free end-to-end communication systems | |
CN113938198B (en) | Optical fiber transmission system, LDA-based method and module for simplifying nonlinear equalizer | |
Le-Tran et al. | Deep learning-based collaborative constellation design for visible light communication | |
CN117176517A (en) | Method and system for implementing channel estimation for MIMO-OFDM visible light communication by adopting genetic algorithm and deep learning | |
CN116249128A (en) | Joint beam forming method based on end-to-end learning in multi-RIS auxiliary communication system | |
CN112104400B (en) | Combined relay selection method and system based on supervised machine learning |
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 |