Multi-point communication method and apparatus for describing and predicting event

WO2024193738A3PCT designated stage expired Publication Date: 2024-11-14CENT FOR EXCELLENCE IN BRAIN SCI & INTELLIGENCE TECH CHINESE ACAD OF SCI
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
PCT/CN2024/104292
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-08-21
Filing Date
2024-07-08
Publication Date
2024-11-14

AI Technical Summary

Technical Problem

When existing computers describe complex events, due to the asynchronous nature of data operations, computing requirements increase exponentially, limiting the intelligence and thinking speed of artificial intelligence, and consuming a large amount of logical operations and communication resources.

Method used

Multi-point communication devices are used to form connections through random paths and potential energy oscillations of communication terminals to achieve synchronous description of event parameters and avoid the high consumption of pre-established channels and global broadcast channels between multiple communication terminals.

Benefits of technology

It realizes multi-level description of complex events, sequential event description, prediction, recall, short-term memory and long-term memory functions, reduces the consumption of computing resources, and improves the description efficiency and accuracy of artificial intelligence.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2024104292_14112024_PF_FP_ABST
    Figure CN2024104292_14112024_PF_FP_ABST
Patent Text Reader

Abstract

The present application relates to multi-point synchronous communication technology. Provided are a multi-point communication apparatus and an application thereof in the field of artificial intelligence. Specifically, provided in the present application are a multi-point communication method and apparatus for describing and predicting an event. By means of a multi-point synchronous communication principle, the apparatus realizes the functions of multi-level description of a complex event, the description of a sequential event, prediction, recalling, and short-term memory and long-term memory. By means of the apparatus and the multi-point communication method used thereby, all parameters of a complex event are synchronously described in the same operation set; in addition, the pre-establishment of a channel between multiple communication ends is avoided, and the defect of the high consumption of communication resources for acquiring a channel by means of global broadcasting is also avoided.
Need to check novelty before this filing date? Find Prior Art

Description

A multipoint communication method and device for describing and predicting events Technical Field

[0001] The present invention relates to an information communication method, a method and device for describing complex events, and adopts a multi-point communication method to describe the objective material world, and in particular to a multi-point communication method and device for describing and predicting complex events. Background Art

[0002] The description of an event requires the participation of multiple parameters, and the asynchronous nature of data operations makes it impossible for existing computers to directly merge all the parameters describing an event into the same operation set. The accurate description of a single event consumes a large amount of logical operations and communication resources between various operation modules, resulting in the description of complex events requiring exponentially growing computing requirements, which limits the development of artificial intelligence's intelligence, thinking speed, and low power consumption.

[0003] Therefore, solving the synchronous calculation of various parameters describing events is the key to transforming the development of artificial intelligence.

[0004] Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a multi-point communication method and device for describing and predicting events, so as to solve the problem of non-cooperative operation of various parameters in event description and realize the description of complex events.

[0006] Technical Solution

[0007] A multipoint communication device for describing and predicting events, the multipoint communication device comprising a plurality of random paths and a plurality of communication terminals;

[0008] Any communication end is connected to a plurality of random pathways, and the random pathways connected to different communication ends have mutually contacting / close locations;

[0009] The potential energy of the communication terminal oscillates when it is activated, and the communication terminal is synchronized with the potential energy of the random path to which it is connected;

[0010] When the potential energy difference between the two contacting / approaching random paths exceeds the minimum connection threshold, a connection occurs;

[0011] The potential energy amplitude of the activated communication terminal is less than the minimum potential energy threshold required for connecting different random paths;

[0012] Different characterizing parameters of an event activate different communication terminals. The activated communication terminals oscillate asynchronously with each other through potential energy, forming a superimposed potential energy difference between random pathways that exceeds the minimum connection threshold, thereby prompting the random pathways connected to each activated communication terminal to connect with each other, forming a network that describes the event.

[0013] In another preferred embodiment, the potential energy of the communication terminal in the activated state oscillates, and the potential energy of the communication terminal and the random path connected thereto are synchronized;

[0014] When the potential energy difference between the mutually contacting or close parts of the random pathways exceeds a minimum connection threshold, the random pathways are connected at the parts;

[0015] The potential energy amplitude of the communication terminal in the activated state is less than the minimum connection threshold required for connecting different random paths.

[0016] Furthermore, the multipoint communication device includes a plurality of random paths and a plurality of communication terminal sets;

[0017] When two random paths are in contact / close to each other, and when the potential energy difference between the two random paths exceeds the threshold ΔE TS When , the two random paths that are in contact / close to each other are connected;

[0018] Each communication terminal set includes several communication terminals. Each communication terminal set receives the same type of external input signals. Different signal parameter values ​​under the same type of signal activate different communication terminals in the communication terminal set.

[0019] Each communication end is connected to several random paths;

[0020] After the communication terminal is activated, its internal potential energy oscillates. The potential energy in the random path connected to the communication terminal oscillates along with the potential energy oscillation of the communication terminal. The potential energy amplitude of the communication terminal and the connected random path is defined as E bit , E bit <ΔE TS <2E bit ;

[0021] The event is decomposed into several characterization signals. Each characterization signal is input into the communication terminal set corresponding to the signal type. The communication terminal corresponding to the signal parameter value is activated. The activated communication terminal and the random path potential energy connected to it generate an amplitude E. bit Oscillation within the range;

[0022] The event corresponds to several different communication terminals activated by the characterization signal and generates asynchronous potential energy oscillations at the same time. The absolute value of the potential energy difference between any two activated communication terminals exceeds ΔE TS When the two activated communication terminals are connected to each other, the random paths are in contact with each other or are close to each other and are connected.

[0023] Furthermore, the spatial positions of the starting point and the end point of the random path in the multipoint communication device are randomly distributed, and the number of input terminals and the number of output terminals of the random path are one or more.

[0024] Furthermore, the absolute value of the potential energy difference between any two activated communication terminals exceeds ΔE TS When the random paths connected to the two activated communication terminals touch or approach each other, a temporary connection occurs; the absolute value of the potential energy difference between any two activated communication terminals exceeds 2E bit When the two activated communication terminals are connected to each other, the random paths are in contact with each other or close to each other to form a long-term connection.

[0025] Furthermore, the potential energy between the connected random paths and communication terminals tends to be synchronized.

[0026] Furthermore, the multipoint communication device also includes a potential energy amplifier, which is connected to a plurality of random paths; the potential energy amplifier can accept the input of low-amplitude potential energy fluctuations from the connected random paths and output high-amplitude potential energy fluctuations to the random paths.

[0027] Furthermore, the potential energy amplitude of the random path connected to the potential energy amplifier is defined as E Bit , E Bit >E bit , preferably, 2E bit >E Bit >E bit ; More preferably, 2E bit >ΔE TS >E Bit >E bit .

[0028] Furthermore, the several characterization signals corresponding to the event include light signals, force signals, sound signals, temperature signals, and acceleration signals. The light signal includes spectrum and light intensity, and the parameter values ​​of the light signal, temperature signal, and force signal contain spatial coordinate information.

[0029] Furthermore, a weak connection path gathering area is provided in the multipoint communication device. In this area, weak connection paths are only briefly connected to each other, and no long-term connection occurs between the weak connection paths.

[0030] Furthermore, the communication end is connected to a plurality of random paths through weak connection paths, wherein the weak connection paths are only briefly connected to each other and are not permanently connected to each other.

[0031] Furthermore, the events are obtained from the real world or directly from the virtual world through a detector; the information obtained from events in the virtual world and the real world is classified, and different parameter value information in a certain type of parameter information of the same or similar nature respectively activates different communication terminals in the multi-point communication device and the connected random paths or random path clusters.

[0032] Furthermore, the functions of potential energy oscillations at different communication terminals are different, which leads to differences in the probability of forming a network between different communication terminals.

[0033] Furthermore, the oscillation frequencies of the potential energy oscillation functions of different communication terminals are different, which leads to differences in the probability of forming a network between different communication terminals.

[0034] Furthermore, the detector is selected from one or more of light signal, force signal, sound signal, temperature signal, acceleration signal, electric field signal, and magnetic field signal detectors; the light signal includes spectrum and light intensity, the sound signal includes sound frequency and loudness, the force signal and acceleration signal include intensity and direction, and the electric field signal and magnetic field signal include field intensity and direction.

[0035] Furthermore, the detector includes a micro-detector array, and the connection between the signal output end of each micro-detector in the micro-detector array and the specific communication end is constant.

[0036] Furthermore, the event signal input by the detector to the multipoint communication device comes from detection of the real world, or from a signal of a virtual event output by an electronic computer program, or from a signal output by other multipoint communication devices.

[0037] Furthermore, the random pathway is selected from nerve cells; or a communication wave band of a specific frequency; or a semiconductor pathway having the properties of the random pathway described in the present technical solution; or a physical electrical circuit having a similar function; or one or more of the above.

[0038] A method for generating an internal network in a multipoint communication device involves multiple detectors detecting the same event and obtaining parameters of different dimensions. Each dimension parameter is input into a corresponding set of communication terminals, and corresponding communication terminals are activated according to the parameter values ​​within that dimension. Parameters of different dimensions synchronously activate communication terminals within different sets of communication terminals, causing the potential energy of multiple activated communication terminals within the multipoint communication device to oscillate asynchronously. The activated communication terminals are interconnected via random pathways, forming a network describing the event.

[0039] The low-level description network serves as the communication end of the high-level description network and is used to describe complex events.

[0040] Furthermore, the random passages in the homogeneous body are wrapped with a supporting medium, the potential energy of the supporting medium is defined as zero-point potential energy, and the potential energy inside the random passages oscillates around the zero-point potential energy value.

[0041] Furthermore, standard events are used for description training to generate a multi-level description network in the multipoint communication device.

[0042] Furthermore, the standard event refers to a set of events with a common dimensional parameter x=a and several sets of non-common dimensional parameters y=b, c, d, e,..., and the event set description function is F(x=a, y=b, c, d, e,...); after standard event training, a set of sub-networks is formed, and any sub-network in the sub-network set is connected to a common dimensional parameter unit with a common dimensional parameter x=a, and each sub-network is also randomly connected to one or more non-common dimensional parameter units with non-common dimensional parameters y=b, c, d, e,...

[0043] Furthermore, a plurality of sets of F (x = a / b / c / d / e / …, y = b, c, d, e, …, z = b, c, d, e, …, …) standard events are used to train the description device to obtain a plurality of sub-network sets, wherein each sub-network set is defined as a level II parameter unit according to the different common dimensional parameter units connected to each sub-network set (for example, divided into (x = a, and / or y = b, c, d, e, …, and / or z = b, c, d, e, …, …), (x = b, and / or y = b, c, d, e, …, and / or z = b, c, d, e, …, …), (x = c, and / or y = b, c, d, e, …, and / or z = b, c, d, e, …, …), … network nodes);

[0044] The II-level parameter unit is trained with standard events to obtain the III-level parameter unit, and so on, and N (N=2, 3, 4, 5, 6, ...)-level parameter units are obtained through training.

[0045] Furthermore, the capacity of describing events at each level within the multipoint communication device is N i , i is the level of the highest event; n0 is the number of internal communication terminals of the multi-point communication device connected to the input signal terminal of the external sensor; R i is the density function of available channels within the multipoint communication device,

[0046] Furthermore, the total capacity of the multipoint communication device is N,

[0047] Furthermore, through sensors, the multipoint communication device is raised in accordance with the method of raising a newborn baby in human civilization.

[0048] The present application also provides a method for describing sequential events in a multipoint communication device, wherein a frame event refers to an event that does not evolve; a sequential event includes a plurality of frame events, wherein the frame events have the same parameter value, and the frame events also have one or more different parameter values, and the changes in the one or more different parameters between the frame events are sequential and continuous; the steps for describing the sequential event include:

[0049] Step 1. The preamble event is described in the multipoint communication device;

[0050] Step 2. The communication end corresponding to the portion of the parameters in the pre-frame event that do not overlap with the post-frame event is closed, and the closed communication end is disconnected from the original communication end network described by the pre-frame event; the communication end corresponding to the portion of the parameters in the post-frame event that overlap with the pre-frame event remains activated; the communication end corresponding to the portion of the parameters in the post-frame event that do not overlap with the pre-frame event is synchronously activated and connected to the network described by the pre-frame event, completing the transition from the pre-frame event to the post-frame event state.

[0051] Step 3. After the description of the subsequent frame event is completed, the subsequent frame event is used as the new preceding frame event to start the description of the subsequent frame event, and each frame event within the complete sequential event is described frame by frame to form a complete sequential event description network.

[0052] Furthermore, a network connection channel has been formed between the preceding frame event and the following frame event in the temporal event network for the transition from the preceding frame event activation state to the following frame event activation state. The transition from the preceding frame event activation state to the following frame event activation state in the formed temporal event network is the optimal solution for the internal communication of the multi-point communication device driven by the potential energy difference. The internal network of the multi-point communication device encodes the description of each frame in the temporal event between frames through the optimal solution of the potential energy difference. The transition from the preceding frame event activation state to the following frame event activation state is the inevitable result of the potential energy difference driving the communication of each communication end. The multi-point communication device naturally has temporal sequence in the temporal event description network.

[0053] The present application also provides a method for predicting unknown events in a multi-point communication device. The multi-point communication device describes a sequential event. After the activation network migrates from the start frame event network to the end frame event network, the potential energy of each communication end of the activated end frame event network and the internal communication network form a certain three-dimensional potential energy distribution structure in the three-dimensional spatial structure of the multi-point communication device. When the potential energy difference with other communication ends exceeds the communication threshold ΔETS, the activation state of the existing description network is migrated to the activation state of a new network formed by autonomous induction, thereby predicting the next unknown progress direction of the preceding sequential event, i.e., autonomous thinking.

[0054] Preferably, the other communication terminals are selected from randomly activated communication terminals, or are selected from communication terminals activated in real time by detectors due to external environmental events.

[0055] The present application also provides a multi-point communication device, comprising a neural cell cluster, a neural development and maintenance system for maintaining the growth and development of the neural cells; and an array voltage controller, wherein the array voltage controller is connected to some cells in the neural cell cluster via neural electrodes and is capable of regulating the electric potential in each connected neural cell.

[0056] It also includes a data source capable of outputting event information. The different information parameters of the event information output by the data source are converted into asynchronously changing electric potentials through the array voltage controller, and then corresponding asynchronously changing electric potentials are simultaneously generated in different nerve cells through the connected neural electrodes, thereby generating a potential difference between the connected nerve cells.

[0057] Furthermore, the nerve cell cluster includes a plurality of nerve cell aggregation areas connected to each other, and different nerve cell aggregation areas are respectively used to receive parameter information of different properties input by the nerve electrodes.

[0058] Furthermore, the array voltage controller generates a potential difference between several neural electrodes with signal input, including at least two gears. The smaller potential difference is used to promote a short-term connection between the nerve cells connected to different electrodes with signal input, and the larger potential difference is used to promote a continuous connection between the nerve cells connected to different electrodes with signal input.

[0059] Furthermore, a second array voltage controller is included. In this technical solution, the second array voltage controller is named "hippocampus". The hippocampus is connected to a number of nerve cells and can receive input of low-amplitude potential energy fluctuations of the connected nerve cells and output high-amplitude potential energy fluctuations (E) to the nerve cells or other nerve cells connected to the hippocampus. Bit ), which promotes long-term connectivity between nerve cells that receive input from high-amplitude potential energy fluctuations.

[0060] Furthermore, the hippocampus continuously outputs high-amplitude potential energy fluctuations over a large range, which is used to actively activate multi-point communication devices to describe and predict high-level complex events.

[0061] Furthermore, the data source is selected from one or both of a sensor and a computer system that generates event information.

[0062] Furthermore, the sensor is used to detect the physical environment in real time. The sensor includes a plurality of array sensors, and the sub-sensors in each array sensor act in correspondence with a plurality of specific nerve cells.

[0063] Furthermore, the multipoint communication device also includes an execution system, which is connected to a plurality of nerve cells in the nerve cell cluster through nerve electrodes and is controlled by the connected nerve cells.

[0064] Furthermore, the sensor is provided on the execution system.

[0065] Furthermore, the sensor is selected from one or more of a mechanical sensor, a temperature sensor, a sound sensor, an optical sensor, a radio sensor, an acceleration sensor, and an air pressure sensor.

[0066] The present application also provides a method for generating an internal network of a multipoint communication device, comprising:

[0067] Multiple detectors detect the same event, thereby obtaining parameters of different dimensions of the event.

[0068] The parameters of each dimension are input into the corresponding communication terminal set, and the corresponding communication terminal is activated according to the value of different parameters under this dimension.

[0069] Parameters of different dimensions synchronously activate communication terminals in different communication terminal sets, causing the potential energy of multiple activated communication terminals in the multipoint communication device to oscillate asynchronously. The activated communication terminals are then connected to each other through random pathways, forming a network describing the event.

[0070] The low-level description network serves as the communication end of the high-level description network and is used to describe complex events.

[0071] The present application also provides a method for describing sequential events in a multipoint communication device, comprising:

[0072] A frame event refers to an event that does not evolve; a sequential event includes several frame events, each of which has the same parameter value, and each of which has one or more different parameter values. The changes in one or more different parameters between each frame event are sequential and continuous. The steps for describing the sequential event include:

[0073] Step 1. Complete description of the preamble event in the multipoint communication device;

[0074] Step 2. Closing the communication end corresponding to the parameters in the pre-frame event that do not overlap with the post-frame event, and the closed communication end is disconnected from the original communication end network describing the pre-frame event;

[0075] maintaining the communication terminals corresponding to the parameters of the post-frame event that overlap with the pre-frame event in an activated state; synchronously activating the communication terminals corresponding to the parameters of the post-frame event that do not overlap with the pre-frame event, and connecting the activated communication terminals to the network describing the pre-frame event to form a network describing the post-frame event, thereby transitioning from describing the pre-frame event to describing the post-frame event;

[0076] Step 3. After the description of the subsequent frame event is completed, the subsequent frame event is used as the new preceding frame event to start the description of the subsequent frame event.

[0077] Each frame event within a complete temporal event is described frame by frame, thereby forming a complete description network of the temporal event.

[0078] In another preferred embodiment, the temporal events are dynamic events, developing events, and events that change over time.

[0079] In another preferred example, a network connection channel has been formed between the pre-frame event and the post-frame event in the description network of the sequential events, and the transition from the activation state of the pre-frame event to the activation state of the post-frame event has been formed. The transition from the activation state of the pre-frame event to the activation state of the post-frame event in the description network of the sequential events that has been formed is the optimal solution for communication between the communication terminals driven by the potential energy difference. The internal network of the multi-point communication device encodes the description of each frame in the sequential event between frames through the optimal solution of the potential energy difference. The transition from the activation state of the pre-frame event to the activation state of the post-frame event is the inevitable result of the potential energy difference drive. The description network of the sequential events formed in the multi-point communication device naturally has sequentiality.

[0080] In another preferred embodiment, a multipoint communication device describes a sequential event, and after the activation network migrates from the description network of the start frame event to the description network of the end frame event, the potential energy of each communication end of the description network of the activated end frame event and the internal potential energy of the communication network form a certain three-dimensional distribution potential energy in the three-dimensional spatial structure of the multipoint communication device, and the difference between the potential energy of the three-dimensional distribution and the potential energy of other communication ends exceeds the communication threshold ΔE TS , thereby achieving the migration from the activation state of the existing description network to the activation state of the new description network formed by autonomous induction, and then achieving the prediction of the next unknown direction of the preceding temporal events, that is, autonomous thinking;

[0081] The other communication terminals are selected from randomly activated communication terminals, or are selected from communication terminals activated in real time by detectors due to external environmental events. Beneficial effects

[0082] The present invention provides a multipoint communication device and method. This device utilizes a multipoint synchronous communication method to achieve multi-level description of complex events, description of sequential events, prediction, recall, and short-term and long-term memory. This device and the multipoint synchronous communication method employed achieve synchronous description of all parameters of complex events within a single operation set, while avoiding the drawbacks of pre-establishing channels between multiple communication terminals and the high communication resource consumption associated with global broadcast acquisition channels. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] FIG1 is a schematic diagram of a complex event description principle according to the present invention;

[0084] FIG2 is a schematic diagram of a communication network formed by describing complex events using multi-level communication terminals according to the present invention. DETAILED DESCRIPTION

[0085] The present invention will be further described below with reference to specific embodiments and Figures 1 and 2 .

[0086] Logical operations are the underlying foundation of existing electronic computers. However, due to the progressive computational steps of 0 / 1 logic gates within integrated circuits, even if multiple parameters can be processed in parallel on a chip, the sub-computational units that run each parameter are actually independent of each other. Ultimately, the data from all these parallel operations is merged to obtain the result, either through instructions or chip architecture design. This computational approach cannot accommodate excessive parameters within a single sub-computational unit, and cannot perform true multi-parameter synchronous computation at the microscopic level. Consequently, when describing complex multi-parameter events, the computational effort of existing electronic computers increases exponentially with the number of parameters describing the complex event, requiring the computer to consume significant computing and storage resources to process these complex events.

[0087] Potential energy represents the tendency of matter to move. By utilizing this tendency to spontaneously decrease, high potential energy can be input into the system without releasing it, enabling detection of potential energy differences exceeding a threshold in any region of the system with virtually no energy loss. This allows for associations between high-potential and low-potential energy ends with virtually no energy loss, no logical calculations, and no time delay. This avoids the drawbacks of pre-establishing channels between multiple communication terminals and the high communication resource consumption of global broadcast acquisition channels. Applying this tendency for potential energy to spontaneously decrease to describe an unlimited number of parameters for complex events is one of the core principles of the technical solution of this patent application.

[0088] Example 1 Multipoint Communication Method

[0089] Existing communication modes between multiple transceivers include various modes, including broadcast and point-to-point channels. The broadcast mode requires broadcasting to the entire system, which consumes a lot of energy. Furthermore, within the same spectrum, only one channel can be established at any given moment. While point-to-point channels avoid the energy consumption of broadcasting to the entire system, they require pre-established point-to-point channels. Therefore, a non-broadcast, non-pre-established channel communication mode between multiple points still exists.

[0090] In this embodiment, the multipoint communication device includes a plurality of random paths and a plurality of communication terminal sets. When the input and output terminals of two random paths are in contact or close to each other, and when the potential energy difference between the two random paths exceeds a threshold value ΔE TS When , the input and output ends of the two random paths that are in contact / close to each other are connected.

[0091] The potential energy of multiple activated communication terminals within a multipoint communication device oscillates asynchronously, and the potential energy within the random paths connected to the corresponding communication terminals also oscillates accordingly. The amplitude of the potential energy oscillation of the communication terminal itself is defined as E bit , the potential energy difference ΔE between the two communication ends bit The minimum potential energy difference between two contacting / approaching random paths to achieve connection is ΔE TS , ΔE TS >E bit ; preferably, 2E bit >ΔE TS >E bit .

[0092] The potential energy difference ΔE between any two communication terminals bit >ΔE TS When the potential energy difference ΔE between the random paths connected to the two communication terminals is bit >ΔE TS At this time, the two random paths are connected at the contact / close position, realizing random path connection between multiple communication terminals, and then realizing the establishment of communication channels between multiple communication terminals within the multi-point communication device, forming a communication network connecting multiple communication terminals.

[0093] Example 2 Application of Multipoint Communication Technology in Event Description

[0094] An event requires a description of multiple parameters, or a collection of multiple parameters is defined as an event. Describing an event requires recording all parameters of the event and categorizing and storing them according to the logical relationships within the event. Therefore, synchronously recording all event parameters and establishing relationships between them is the most efficient method for describing an event. This embodiment utilizes the multipoint communication device described in Example 1. Various event parameters are converted by sensors into input signals that activate multiple communication terminals within the multipoint communication device. The activated multiple communication terminals generate potential energy oscillations, and channels are established using the multipoint communication method described in Example 1. The multiple communication terminals and the resulting communication network serve as a parameter description set to collectively describe the event.

[0095] Example 3: Layering of parameter description sets

[0096] In Example 2, the multiple communication terminals and the formed communication network are used as a parameter description set to jointly realize the description of the event. After a path is formed between the multiple communication terminals described in the event and the formed communication network, the potential energy within the parameter description set is balanced.

[0097] In this embodiment, the event described above is defined as a Level I event. The corresponding communication endpoint within the parameter description set is defined as a Level I communication endpoint, and the corresponding parameter description set is defined as a Level I parameter description set. The Level I communication endpoint within this Level I parameter description set is in potential equilibrium with all points in the communication network. In this embodiment, the Level I parameter description set is defined as a Level II communication endpoint. The multipoint communication device described in Example 2 describes the combined event (Level II event) of multiple Level I events to obtain a Level II parameter description set, which is then defined as a Level III communication endpoint. Similarly, N Level communication endpoints and N Level parameter description sets are defined.

[0098] Specifically, a multipoint communication device with several Level II communication terminals obtained through training is used to describe Level II events. The Level II events contain the contents of multiple Level I events and have more descriptive parameters than Level I events. Level II events are converted into more input signals to activate Level I communication terminals. When the percentage of activated Level I communication terminals in a Level II communication terminal gradually increases, the overall potential energy E of the Level II communication terminal increases. bit2 The amplitude gradually increases, and the potential energy E of multiple level II communication terminals bit2 Asynchronous oscillation occurs when the potential energy difference ΔE between each level II communication terminal bit2 >ΔE TSWhen the random paths randomly set between the Level II communication terminals are connected to each other, a communication network is established between multiple communication points within the multi-point communication device, thereby completing the description of the Level II event and obtaining a parameter description set formed by the Level II event description (including the channel network formed by the Level II communication terminals and each other). The parameter description set formed by the Level II event description is defined as the Level III communication terminal. In addition, some Level I communication terminals that are not included in the Level II communication terminals are activated in the Level II event, and their own potential energy E bit It will also be combined with the potential energy E of the Level II communication terminal. bit2 It exhibits asynchronous oscillation, forming a channel network in which level I communication terminals and level II communication terminals are mixed with each other, that is, the level I communication terminal bypasses the level II communication terminal where it is located, and independently establishes a network connection with other level II communication terminals, forming part of the level III communication terminal.

[0099] Similarly, different levels of communication terminals and corresponding channel networks can be trained according to the complexity of the described events, and the internal compatibility of high-level communication terminals includes cross-level low-level communication terminals. The number of communication terminals and corresponding channel networks trained at different levels increases exponentially with the increase in level. Specifically:

[0100] Assume that the number of level I communication terminals is N1, the number of level II communication terminals is N2, and the network redundancy parameter is R n , the R n It depends on the network density of each level within the multipoint communication device and is positively correlated with the random path length, random path density, number of nodes, and number of random path branches.

[0101] Assume that the multipoint communication device has been trained for events of different levels with the maximum carrying capacity; then,

[0102] The number of level I communication terminals is N1.

[0103] The number of level II communication terminals is N2.

[0104] The number of level III communication terminals is N3.

[0105] The number of communication terminals at level i is N i ,

[0106] The capacity of each level of the multipoint communication device to describe events is N.

[0107] Where i is the highest event level; n0 is the number of internal communication terminals of the multi-point communication device connected to the external sensor input signal terminal; R iIt is a parameter of the network redundancy within a multipoint communication device. When the density of available channels within a multipoint communication device is high, there are multiple coexisting channel networks describing the same event.

[0108] Example 4: Method for describing temporal events

[0109] A sequential event refers to a series of multi-parameter overlapping events. A sequential event includes several frame events. The same parameter values ​​exist between each frame event. At the same time, there are one or more different parameters between each frame event. The changes of one or more different parameters between each frame event are sequential.

[0110] For example, the event of "an apple falling from a tree" is a time-series event, and the scene at any moment in the process of the apple falling is a frame event; in this example, the still apple tree, the still earth, the red color of the apple, the blue sky, and the white clouds in the sky are all the same parameter values ​​that exist between each frame event, and the three-dimensional spatial coordinate position that constantly changes when the apple falls is the continuously changing and mutually related different parameter values ​​that exist between each frame event.

[0111] The description of sequential events is based on the description of frame events. The preceding frame event is described in the multipoint communication device, which then describes the following frame event. Because the preceding and following frame events have multiple overlapping parameters and one or more non-overlapping parameters, during the description of the following frame event, the communication end and network described by most of the parameters of the following frame event are directly inherited from the communication end and network of the preceding frame event (i.e., the communication end and network for the overlapping parameters of the following frame event and the preceding frame event are the same). The communication end corresponding to the non-overlapping parameters of the following frame event is activated and incorporated into the communication end network described by the preceding frame event. The communication end corresponding to the non-overlapping parameters of the following frame event is closed and separated from the communication end network described by the preceding frame event, thus closing the description of the preceding frame event and completing the description of the following frame event. Subsequently, after the description of the post-frame event is completed, the post-frame event is used as the new pre-frame event of the next stage to start describing the post-frame event, and each frame event within the complete sequential event is described frame by frame to complete the description of the sequential event.

[0112] Time-series events are described frame by frame. Essentially, the communication endpoints corresponding to the common parameters of each frame event are consistently incorporated into the entire time-series event description network, while the communication endpoints corresponding to the differential parameters of each frame event are sequentially incorporated into the corresponding frame event description network according to the temporal progression of the events. Therefore, the entire process of describing time-series events is a process in which non-overlapping communication endpoints gradually access a proliferative cognitive network dominated by an overlapping communication endpoint network, following the development order of the time-series events. After a multipoint communication device completes the description of a time-series event, the resulting parameter description set includes a common subset and several differential subsets. Based on the description results of the time-series event, the multipoint communication device merges the description network, automatically merging the common parameters of the multiple events being described. This is a direct reflection of the description network's physical extraction of the common parameters of the multiple entities being described, and achieves compression of duplicate data. This means that communication endpoints at all levels within the multipoint communication device are shared across the descriptions of different events.

[0113] Example 5: Migration of frame events

[0114] As in the above-described embodiment, after training with sequential events, a multipoint communication device has already formed a network describing these sequential events. Within this network, the communication endpoint corresponding to the newly introduced differential parameter for the post-frame event is the optimal communication channel within the multipoint communication device, corresponding to the network potential energy intensity and the three-dimensional spatial distribution of the potential energy region in the pre-frame event activation state. This channel represents the channel where the absolute value of the superposition function of the pre-frame network potential energy and the communication endpoint potential energy corresponding to the newly introduced differential parameter is the highest. This indicates that the transition from the pre-frame event activation state to the post-frame event activation state is the optimal solution for driving communication in the multipoint communication device by the potential energy difference. After training with sequential events, a network channel for this transition has already been formed within the multipoint communication device. Therefore, this transition is the inevitable result of the multipoint communication device's operation driven by the potential energy difference. Therefore, the multipoint communication device's internal network describes each frame in a sequential event using an optimal solution for potential energy differences, encoding the frames between them. In this technical solution, the multipoint communication device's description of sequential events inherently involves sequential time. The multipoint communication device's reactivation of each frame event according to this sequential time is a recollection of these sequential events.

[0115] Example 6 Prediction

[0116] As described in the above embodiment, the principle of frame event migration is the same as that of the previously described embodiment. However, all time-series event description networks trained in multipoint communication devices always have a starting frame and an ending frame. When the activated network migrates from the starting frame to the ending frame, the potential energy of each communication end and the internal communication network in the activated last frame forms a certain three-dimensional potential energy distribution structure in the three-dimensional spatial structure within the multipoint communication device. Although there is no trained internal communication channel within the multipoint communication device at this time, the three-dimensional distribution potential energy structure, in addition to the network channel itself, may still have other potential energy differences with the internal network that exceed the communication threshold ΔE. TS , forming the communication end of a pathway, thereby enabling the migration from the activation state of an existing network to the activation state of a new network formed through autonomous induction, i.e., the prediction of a preceding temporal event. The new network formed through autonomous induction is the predicted event of the preceding event, the result of the computation of complex events by the trained multi-point communication device, i.e., autonomous thinking.

[0117] In this embodiment, a random activation mode of the communication end can also be introduced. The potential energy difference between the randomly activated communication end and the last frame event exceeds the communication threshold ΔE TS , realizing the formation of a new multipoint communication device internal network between the multipoint communication device internal network corresponding to the last frame event and the randomly activated communication end, thereby forming a predicted event;

[0118] In this embodiment, it can also be considered that the sensors connected to the multipoint communication device continuously convert external environmental events into signals and input them into different communication terminals, randomly activating different communication terminals and corresponding network combinations of activated frame events to form a new activated internal network of the multipoint communication device, thereby achieving the formation of predicted events through induction of the last frame event.

[0119] In this embodiment, a potential energy accumulation mode can also be introduced. The potential energy amplitude within the internal network of the multipoint communication device activated by the end frame event gradually increases over time, and the threshold between the communication terminals must exceed the communication threshold ΔE. TS , forming a new internal network of the multi-point communication device in an activated state, realizing the formation of predicted events through the induction of the last frame event.

[0120] Example 7 Short-term memory and long-term memory

[0121] The potential energy of multiple activated communication terminals within a multipoint communication device oscillates asynchronously, and the potential energy within the random paths connected to the corresponding communication terminals also oscillates accordingly. The amplitude of the potential energy oscillation of the communication terminal itself is defined as E bit , the potential energy difference ΔE between the two communication ends bit The minimum potential energy difference between two contacting / approaching random paths to achieve connection is ΔE TS , 2E bit >ΔE TS >E bit .

[0122] The potential energy difference ΔE between any two communication terminals bit >ΔE TS When the potential energy difference ΔE between the random paths connected to the two communication terminals is bit >ΔE TS At this time, the two random paths are connected at the contact / close position, realizing random path connection between multiple communication terminals, thereby establishing communication channels between multiple communication terminals within the multi-point communication device, forming a communication network connecting multiple communication terminals.

[0123] As described in the above embodiment, there is a communication threshold ΔE between random paths TS ; The potential energy difference ΔE between any two random paths that are in contact / close to each other bit >ΔE TS When the random paths connected to the two activated communication terminals touch or approach each other, a temporary connection occurs; the absolute value of the potential energy difference between any two activated communication terminals exceeds 2E bit When the two activated communication terminals are connected to each other, the random paths are in contact with each other or close to each other to form a long-term connection.

[0124] In the preferred embodiment of this invention, the multipoint communication device further includes a potential energy amplifier, which is connected to a plurality of random paths. The potential energy amplifier can receive input of low-amplitude potential energy fluctuations from the connected random paths and output high-amplitude potential energy fluctuations to the random paths. The potential energy amplifier outputs high-amplitude potential energy fluctuations to the random paths. The potential energy amplitude of the random paths is defined as E Bit , 2E bit >E Bit >E bit Therefore, the potential energy fluctuation output by the potential amplifier causes the potential energy difference between the communication ends to exceed 2E bit , enabling long-term connections between communication terminals.

[0125] Example 8 A multipoint communication device structure

[0126] The device structure of this embodiment includes a neural cell cluster, a neural electrode array, an electrode array voltage controller, a sensor, a neural development and maintenance system, and may also include an execution system:

[0127] The neural electrodes include receiving-type neural electrodes with one end connected to a single nerve cell or a nerve cell cluster and the other end connected to a sensor signal input end; and output-type neural electrodes with one end connected to a single nerve cell or a nerve cell cluster and the other end connected to an execution system.

[0128] The sensor inputs the detected physical signals of the same type but different values ​​into different receiving neural electrodes;

[0129] The electrode array voltage controller can, when detecting that one or more sensors input electrical signals into the neural electrode array, regulate the electric potential of all or part of the neural electrodes with signal input to produce periodic or non-periodic oscillations simultaneously or within a controllable time, thereby generating an electric potential difference between different neural electrodes with signal input, thereby promoting communication between the nerve cells connected to different electrodes with signal input.

[0130] Furthermore, the electrode array voltage controller generates at least two potential differences between different neural electrodes receiving signal inputs: a smaller potential difference is used to promote a brief connection (instantaneous memory) between nerve cells connected to different electrodes receiving signal inputs, and a larger potential difference is used to promote a sustained connection (long-term memory) between nerve cells connected to different electrodes receiving signal inputs. In this technical solution, the duration of the brief connection and sustained connection are relative values. The actual connection duration of the two determines the degree to which the device chooses to accept different events and does not affect the operation of the device.

[0131] Furthermore, a second array voltage controller is included. In this technical solution, the second array voltage controller is named "hippocampus". The hippocampus is connected to a number of nerve cells and can receive input of low-amplitude potential energy fluctuations of the connected nerve cells and output high-amplitude potential energy fluctuations (E) to the nerve cells or other nerve cells connected to the hippocampus. Bit ), which promotes continuous connection between nerve cells that receive high-amplitude potential energy fluctuation input and forms long-term memory.

[0132] Furthermore, the hippocampus selectively outputs high-amplitude potential energy fluctuations over a large range and continuously, synchronously activating multiple networks for parallel operations, thereby activating multi-point communication devices to describe and predict higher-level complex events.

[0133] Furthermore, the sensor is selected from one or more of a mechanical sensor, a temperature sensor, a sound sensor, an optical sensor, a radio sensor, an acceleration sensor, and an air pressure sensor; one or more of each sensor is set; and some or all of the sensors are set on the execution system.

[0134] In this technical solution, various random pathways can be selected, such as neural cells that can be electrically induced to fuse, electronic devices with similar functions, and modified chip etched circuits. Multipoint communication devices created using different types of random pathways may differ only in their efficiency, but their underlying operating principles and results are essentially the same. In implementing this technical solution, specific devices should not be used to limit its scope; as long as the functions described in this technical solution are achieved, a multipoint communication device capable of describing and predicting events can be created.

[0135] Example 9: Potential Energy Communication Working Mechanism of the Central Nervous System

[0136] Based on the multi-point communication method for describing and predicting events proposed in this patent application, we speculate that the neural system in the brain of an organism evolved in a parallel space-time universe uses a potential energy communication mechanism to operate. The specific working method is as follows:

[0137] The events of the physical world are input into different brain regions of the brain in a short period of time through visual neural signals, auditory neural signals, temperature neural signals, acceleration neural signals, and force neural signals, respectively activating neurons in different brain regions. For neurons in the brain, there is a minimum potential energy difference (i.e., potential energy difference threshold). Only when the potential energy difference between two neurons exceeds the potential energy difference threshold will neurons be connected and can information be transmitted to each other. The relationship between the specific potential energy difference threshold and the potential energy oscillation of a single neuron is shown in Examples 1 and 7. The neurons activated in each brain region experience asynchronous potential energy oscillations with each other. Due to the asynchronous potential energy oscillations, the potential energy difference between different activated neurons is caused to exceed the potential energy difference threshold at which neurons are connected. Different activated neurons are connected to each other to form a neural network that describes events in the physical world. The corresponding multi-level network architecture is the same as the architecture and formation mechanism described in Example 3.

[0138] Accordingly, the reason why the hippocampus in the brain can consolidate short-term memory and then convert it into long-term memory is essentially because the hippocampus is actually a "transformer" of neuronal potential energy. Neurons are connected to the hippocampus, and neurons input low potential energy oscillations into the hippocampus. After the hippocampus receives such low potential energy oscillations, the hippocampus will output high potential energy oscillations to the neuron or other neurons. It is with the participation of the hippocampus that the originally asynchronous low potential energy amplitude oscillations between multiple neurons are converted into asynchronous high potential energy amplitude oscillations between multiple neurons. The potential energy difference formed between neurons with high potential energy oscillations is much higher than the potential energy difference threshold mentioned above. The higher potential energy difference between neurons promotes the formation of permanent connections between neurons, i.e., long-term memory. The specific principle is the same as that of Example 7.

[0139] Based on the above information, combined with the contents of Examples 1 to 7, the principle of potential energy oscillation can be used to input memories into the brain, erase memories, integrate more neuronal cells into a mature brain to expand brain capacity, interconnect multiple brains, synchronize the brain with chip operations, and connect various brain regions to a virtual computer system.

[0140] In the potential energy communication mode, the transmission of neuronal information, the role of Ranvier nodes, and the role of neurotransmitters:

[0141] Information transmission: One view holds that when the brain thinks, information is transmitted from one neuron to the next: the nerve impulse from an activated neuron propagates along the axon until it reaches the end of the axon, releasing neurotransmitters, which then activate the next neuron, completing the information transfer. However, in potential energy communication theory, information is not transmitted from one activated neuron to another as described above. Instead, the information is broken down between two or more neurons. The oscillation of the potential energy of each neuron promotes connections between them, and the information is then reassembled into a complete message in the form of a neural network.

[0142] The role of nodes of Ranvier: Some believe that the nodes of Ranvier on the surface of nerve axons accelerate the transmission of nerve impulses. However, in potential energy communication theory, the more critical role of nodes of Ranvier is to modulate the oscillation function of the overall electrical potential of the neuron during nerve impulse transmission. The nodes of Ranvier on the axons of different neuronal types have different lengths, densities, and arrangements. These differences in nodes of Ranvier among different neurons result in different modulations of the neuronal potential oscillation function when the axons generate nerve impulses, achieving matching of the potential oscillation functions of specific neuronal types, thereby enabling communication and connection between specific neuronal types.

[0143] The role of neurotransmitters: Regarding the role of neurotransmitters, some believe that the neurotransmitters released by the axon of the previous neuron are transmitted to the dendrites of the next neuron, thereby activating the next neuron. In the theory of potential energy communication, the exact opposite view can be drawn: the more important role of neurotransmitters is to terminate the connection between the neuron that releases the neuron and the next neuron at the potential energy level. In other words, the conduction process of the nerve impulse in the axon is the process of maintaining the potential energy oscillation of the neuron. When the nerve impulse reaches the end of the axon, the neuron can no longer maintain the potential energy oscillation of the neuron. The neuron loses its meaning in the entire potential energy communication process, thereby releasing neurotransmitters to cut off the connection between the neuron and the next level neuron at the potential energy level, and at the same time initiate the potential energy oscillation of the next level neuron, that is, the nerve impulse. Specifically, after the information is decomposed, different neurons are activated. The activated neurons do not directly contact each other. Instead, each activated neuron is connected to the lower-level neurons. The activated neurons connect to other activated neurons through the connected lower-level neurons. Therefore, the potential energy oscillation of the activated neuron should be synchronized with the lower-level neurons to which it is connected, so as to achieve a potential energy difference between the lower-level neurons connected to different activated neurons, and form a neural network between multiple lower-level neurons through this potential energy difference. Therefore, the activated neuron and the lower-level neurons to which it is connected can be regarded as a whole with synchronized potential energy. When the axon of the activated neuron transmits a pulse, the pulse plays a role in maintaining the overall potential energy oscillation of the entire activated neuron and its connected lower-level neurons. When the nerve pulse of the activated neuron is transmitted to the end of the axon, the activated neuron has lost the ability to continue to maintain the potential energy oscillation. At this time, the axon releases neurotransmitters, which cut off the connection between the lower-level neurons and the activated neuron at the potential energy level. At the same time, the neurotransmitter activates the next neuron. The activated lower-level neuron takes over the role of continuing to maintain the potential energy oscillation of this neuron and the lower-level neurons, until multiple activated neurons are interconnected to form a neural cognitive network for information.

[0144] As used herein, the terms "minimum potential energy threshold" and "minimum connection threshold" are used interchangeably and have the same meaning.

[0145] As used herein, the term "contact / near" means "in contact or near."

Claims

1. A multipoint communication device for describing and predicting events, characterized in that: The multipoint communication device includes a plurality of random paths and a plurality of communication terminals; The communication end is connected to a plurality of random paths, and the random paths connected to different communication ends have mutually contacting / close parts; The potential energy of the communication terminal oscillates when it is activated, and the potential energy between the communication terminal and the random path to which it is connected is synchronized; When the potential energy difference between the random paths that are in contact / close to each other exceeds a minimum connection threshold, connection occurs; The potential energy amplitude of the activated communication terminal is less than the minimum potential energy threshold required for connecting different random paths; Different characterization parameters of an event activate different communication terminals. The activated communication terminals oscillate asynchronously with each other in potential energy, forming a superimposed potential energy difference between random pathways that exceeds the minimum connection threshold, thereby prompting the random pathways connected to each activated communication terminal to connect with each other, forming a network that describes the event.

2. The multipoint communication device for describing and predicting events as claimed in claim 1, characterized in that: The multipoint communication device includes a plurality of random paths and a plurality of communication end sets; when two random paths are in contact / close to each other, and when the potential energy difference between the two random paths exceeds a threshold value ΔE TS When , the two random paths that are in contact / close to each other are connected; Each communication terminal set includes several communication terminals. Each communication terminal set receives the same type of signals input from the outside. Different signal parameter values ​​under the same type of signals activate different communication terminals in the communication terminal set. Each communication end is connected to several random paths; After the communication end is activated, its internal potential energy oscillates, and the potential energy in the random path connected to the communication end oscillates with the potential energy oscillation of the communication end; the potential energy amplitude of the communication end and the connected random path is defined as E bit , E bit <ΔE TS <2E bit ; The event is decomposed into several representative signals. Each representative signal is input into the communication terminal set corresponding to the signal of this type. The communication terminal corresponding to the signal parameter value is activated. The activated communication terminal and the random path potential energy connected to it generate an amplitude E bit Oscillation within the range; The different communication terminals activated by the several characterizing signals corresponding to the event simultaneously generate asynchronous potential energy oscillations, and the absolute value of the potential energy difference between any two activated communication terminals exceeds ΔE TS When the two activated communication terminals are connected to each other, the random paths are in contact with each other / are close to each other and connected.

3. The multipoint communication device for describing and predicting events as claimed in claim 2, characterized in that: The spatial positions of the start point and the end point of the random path in the multipoint communication device are randomly distributed, and the number of input terminals and the number of output terminals of the random path are one or more.

4. The multipoint communication device for describing and predicting events as claimed in claim 2, characterized in that: The absolute value of the potential energy difference between any two activated communication terminals exceeds ΔE TS When the random paths connected to the two activated communication terminals touch / approach each other for a short time, the absolute value of the potential energy difference between any two activated communication terminals exceeds 2E bit When the two activated communication terminals are connected to each other, the random paths are in contact with each other / are close to each other to form a long-term connection.

5. The multipoint communication device for describing and predicting events as claimed in claim 4, characterized in that: The multipoint communication device also includes a potential energy amplifier, which is connected to several random channels; the potential energy amplifier can receive input of low-amplitude potential energy fluctuations from the connected random channels, and output high-amplitude potential energy fluctuations to the random channels or other random channels.

6. The multipoint communication device for describing and predicting events as claimed in claim 5, characterized in that: The potential energy amplitude output by the potential amplifier is defined as E Bit , E Bit >E bit .

7. The multipoint communication device for describing and predicting events as claimed in claim 6, characterized in that: 2E bit >E Bit >E bit 。 8. The multipoint communication device for describing and predicting events as claimed in claim 7, characterized in that: 2E bit >ΔE TS >E Bit >E bit 。 9. The multipoint communication device for describing and predicting events as claimed in claim 1, characterized in that: The event signal comes from the detection of the real world, or from the signal of a virtual event output by an electronic computer program, or from the signal output by other multi-point communication devices; The information in the event is classified, and different parameter value information in a certain type of parameter information of the same or similar nature respectively activates different communication terminals in the multipoint communication device and the connected random paths or random path clusters.

10. The multipoint communication device for describing and predicting events as claimed in claim 9, characterized in that: Different communication terminals have different functions of potential energy oscillation, which leads to differences in the probability of forming a network between different communication terminals.

11. The multipoint communication device for describing and predicting events as claimed in claim 9, characterized in that: The oscillation frequencies of the potential energy oscillation functions of different communication terminals are different, which leads to differences in the probability of forming a network between different communication terminals.

12. The multipoint communication device for describing and predicting events as claimed in claim 1, wherein: The random pathway is selected from nerve cells; or selected from a communication wave band of a specific frequency; or selected from a semiconductor pathway having the properties of the random pathway described in the technical solution; or a physical electrical circuit with a similar function, or one or more of the above.

13. The multipoint communication device for describing and predicting events as claimed in claim 1, characterized in that: The random path is a neuron, and the communication end is a neuron connected to an external sensor or a signal output end. The device includes a nerve cell cluster, a nerve development and maintenance system for maintaining the growth and development of the nerve cells; and an array voltage controller, which is connected to the nerve cells in the nerve cell cluster through nerve electrodes to transmit electrical signals and can regulate the electric potential in each connected nerve cell; It also includes a data source capable of outputting event information. The different information parameters of the event information output by the data source are converted into asynchronously changing electric potentials through an array voltage controller, and then corresponding asynchronously changing electric potentials are simultaneously generated in different nerve cells through the connected neural electrodes, and the superposition difference of the electric potentials generated between the connected nerve cells promotes neural connection, thereby converting the event into a neural connection network. In another preferred embodiment, the asynchronously changing electric potential further generates corresponding asynchronously changing electric potentials in different nerve cells simultaneously through the connected neural electrodes; a superimposed difference in electric potential is generated between the nerve cells, thereby prompting the nerve cells to connect, thereby converting the event into a neural connection network.

14. The multipoint communication device for describing and predicting events as claimed in claim 13, characterized in that: The nerve cell cluster includes a plurality of nerve cell clustering regions connected to each other, and different nerve cell clustering regions are respectively used to receive parameter information of different properties input by nerve electrodes.

15. The multipoint communication device for describing and predicting events as claimed in claim 13, characterized in that: The array voltage controller generates a potential difference between a number of neural electrodes with signal inputs, including at least two gears. A smaller potential difference is used to promote a short-term connection between the neural cells connected to different electrodes with signal inputs, and a larger potential difference is used to promote a continuous connection between the neural cells connected to different electrodes with signal inputs.

16. The multipoint communication device for describing and predicting events as claimed in claim 13, characterized in that: It also includes a second array voltage controller, which is named "hippocampus". The hippocampus is connected to several of the nerve cells, can receive input of low-amplitude potential energy fluctuations from the connected nerve cells, and output high-amplitude potential energy fluctuations to the nerve cell or other nerve cells connected to the hippocampus, thereby promoting long-term connection between the nerve cells that receive the high-amplitude potential energy fluctuation input.

17. The multipoint communication device for describing and predicting events as claimed in claim 13, characterized in that: The data source is selected from one or both of sensors and computer systems that generate event information.

18. A method for generating an internal network of a multipoint communication device, characterized in that: Multiple detectors detect the same event and obtain parameters of different dimensions. Each dimension parameter is input into a corresponding communication terminal set. The corresponding communication terminal is activated according to the different parameter values ​​under the dimension. Different dimension parameters synchronously activate communication terminals in different communication terminal sets. The potential energy of multiple activated communication terminals in the multi-point communication device oscillates asynchronously. The activated communication terminals are connected to each other through random paths to form a description network of the event. The low-level description network serves as the communication end of the high-level description network and is used to describe complex events.

19. The method for generating an internal network of a multipoint communication device as claimed in claim 18, wherein: Standard events are used for description training to generate a multi-level description network in a multi-point communication device.

20. The method for generating an internal network of a multipoint communication device as claimed in claim 18, wherein: Through sensors, a device describing this complex event is nurtured in accordance with the method of raising newborn babies in human civilization.

21. A method for describing sequential events in a multipoint communication device, characterized in that: Frame events refer to events that are not evolutionary; sequential events include several frame events, and there are the same parameter values ​​between the frame events, and there are one or more different parameter values ​​between the frame events, and the changes of one or more different parameters between the frame events are sequential and continuous; the description steps of sequential events include: Step 1. The preamble frame event is described in the multipoint communication device; Step 2. The communication end corresponding to the parameter part of the pre-frame event that does not overlap with the post-frame event is closed, and the closed communication end is separated from the original communication end network described by the pre-frame event; the communication end corresponding to the parameter part of the post-frame event that overlaps with the pre-frame event remains activated; the communication end corresponding to the parameter part of the post-frame event that does not overlap with the pre-frame event is synchronously activated and connected to the network described by the pre-frame event, completing the transition from the pre-frame event to the post-frame event description state; Step 3. After the description of the post-frame event is completed, the post-frame event is used as the new pre-frame event to start the description of the post-frame event, and each frame event within the complete sequential event is described frame by frame to form a complete sequential event description network.

22. The method for describing sequential events by a multipoint communication device as claimed in claim 21, wherein: A network connection channel has been formed between the preceding frame event and the following frame event in the temporal event network for the transition from the activation state of the preceding frame event to the activation state of the following frame event. The transition from the activation state of the preceding frame event to the activation state of the following frame event in the formed temporal event network is the optimal solution for communication between the communication terminals driven by the potential energy difference. The internal network of the multi-point communication device encodes the description of each frame in the temporal event through the optimal solution of the potential energy difference between frames. The transition from the activation state of the preceding frame event to the activation state of the following frame event is the inevitable result of the potential energy difference drive. The multi-point communication device naturally has temporal sequence in the temporal event description network.

23. The method for describing sequential events by a multipoint communication device as claimed in claim 21, wherein: The multi-point communication device describes the sequential events. After the activation network migrates from the start frame event network to the end frame event network, the potential energy of each communication end of the activated end frame event network and the internal communication network forms a certain three-dimensional potential energy distribution structure in the three-dimensional space structure of the multi-point communication device. The potential energy difference with other communication ends exceeds the communication threshold ΔEts, realizing the migration from the activation state of the existing description network to the activation state of the new description network formed by autonomous induction, and realizing the prediction of the next unknown development direction of the preceding sequential events, that is, autonomous thinking; The other communication terminals are selected from randomly activated communication terminals, or selected from communication terminals activated in real time by external environmental events through detectors.

Citation Information

Patent Citations

  • Visual and auditory perception integrated multitask collaborative identification method and system

    CN108804715A

  • Robot degenerated environment obstacle avoidance method based on impulsive neural network internal plasticity

    CN114037050A

  • Multi-point communication method and device for describing and predicting event

    CN116996844A

  • Framework for the evolution of electronic neural assemblies toward directed goals

    US20110161268A1

  • Adaptation of snns through transient synchrony

    US20230368004A1