ScienceON 검색
ScienceON '특허' 검색
MORE| No | 제목 / 내용 |
|---|---|
| 1 |
IMAGE PROCESSOR / FUJI HEAVY IND LTD PROBLEM TO BE SOLVED: To provide an image processor which increases the speed of image processing by learning the procedure of image processing that enables construction of a desired image processing circuit without being limited by an image filter.SOLUTION: The image processor 1 includes a processor 2, a shift register 3, learning units 4, a multiplexer 5, and a learning means 6. Each learning unit 4x comprises a computing unit having function processing circuits arranged in parallel, and a selector for selecting data to be input to each of the function processing circuits. Each of the selectors in the plurality of units 4 connected in series, selects, in accordance with a selection signal SS showing the procedure of image processing, the data to be input to each of the function processing circuits, and the learning means 6 optimizes, through learning, the selection signal SS allocated to the selector of each learning unit 4. |
| 2 |
METHOD FOR THEFT PROTECTION OF MACHINE LEARNING MODULES, AND PROTECTION SYSTEM / SIEMENS AKTIENGESELLSCHAFT [According to the invention, for theft protection of a machine learning module (NN) which is trained, on the basis of training data (TD), to derive, from operating signals (BS) of a machine (M), control signals for controlling the machine (M), a distribution of the training data (TD) in a representation space of the operating signals (BS) is [determined] in spatial resolution.] The machine learning module (NN) is also expanded by an additional input layer (IL'), and the expanded machine learning module (NN) is transmitted to a user. When input signals (BS, SS) are fed into the expanded machine learning module (NN), a location of a relevant input signal (BS, SS) in the representation space is [determined] by the additional input layer (IL'). Moreover, depending on the distribution of the training data (TD), a coverage value specifying coverage of the location of each input signal (BS, SS) by the training data (TD) is determined in each case. Finally, depending on the determined coverage values, in particular in the event of low coverage values, an alarm signal (A) is output. |
| 3 |
Anomaly detection in SS7 control network using reconstructive neural networks / Oracle International Corporation Herein are machine learning (ML) techniques for unsupervised training with a corpus of signaling system 7 (SS7) messages having a diversity of called and calling parties, operation codes (opcodes) and transaction types, numbering plans and nature of address indicators, and mobile country codes and network codes. In an embodiment, a computer stores SS7 messages that are not labeled as anomalous or non-anomalous. Each SS7 message contains an opcode and other fields. For each SS7 message, the opcode of the SS7 message is stored into a respective feature vector (FV) of many FVs that are based on respective unlabeled SS7 messages. The FVs contain many distinct opcodes. Based on the FVs that contain many distinct opcodes and that are based on respective unlabeled SS7 messages, an ML model such as a reconstructive model such as an autoencoder is unsupervised trained to detect an anomalous SS7 message. |
| 4 |
ANOMALY DETECTION IN SS7 CONTROL NETWORK USING RECONSTRUCTIVE NEURAL NETWORKS / Ahmadi, Hamed Herein are machine learning (ML) techniques for unsupervised training with a corpus of signaling system 7 (SS7) messages having a diversity of called and calling parties, operation codes (opcodes) and transaction types, numbering plans and nature of address indicators, and mobile country codes and network codes. In an embodiment, a computer stores SS7 messages that are not labeled as anomalous or non-anomalous. Each SS7 message contains an opcode and other fields. For each SS7 message, the opcode of the SS7 message is stored into a respective feature vector (FV) of many FVs that are based on respective unlabeled SS7 messages. The FVs contain many distinct opcodes. Based on the FVs that contain many distinct opcodes and that are based on respective unlabeled SS7 messages, an ML model such as a reconstructive model such as an autoencoder is unsupervised trained to detect an anomalous SS7 message. |
| 5 |
SYSTEM AND METHOD FOR MAPPING SS7 BEARER CHANNELS / TriaSys Technologies Corporation A system and method for associating Signaling System 7 logical circuits and bearer channels are presented. The system may include an event detector configured to receive an SS7 signaling message on an SS7 signaling link, parse a logical circuit from the SS7 signaling message, receive an SS7 bearer channel, and detect a bearer channel event on the SS7 bearer channel. A statistical learning model block is configured to calculate a correlation confidence value between said bearer channel and said logical circuit. The method may include parsing a logical circuit ID from a signaling message on an SS7 signal link, identifying a bearer channel associated with a bearer event on a bearer circuit, and calculating a current correlation confidence value between the logical circuit ID and the bearer channel. |
| 6 |
OPERATOR SUPPORTING SYSTEM / HITACHI LTD (PURPOSE) To make highly accurate positioning simply by finding a linear weighted average value of the result of control from measured value of positioning, and outputting the difference between the target command and the weighted average value as the next correction. (CONSTITUTION) The learning correction of the result of positioning fn is given by Fn =(1-α).Fn-1 +αfn by a microcomputer 4, where Fn is linear weighted average value of the result of positioning and α is smoothing coefficient. Here, target command of control basing on the result of learning SS1 and correction DS1 for target of positioning SS are expressed by expressions I, II. Here, k is correction coefficient. Fn can be found by finding simple average stop position x, obtaining provisional correction DS1 from DS1=SS-x and supposing and giving α. Therefrom, k is obtained by the expression II. The difference between the target command SS1 and weighted average value Fn is outputted from a positioning device 2 as next correction. |
| 7 |
SYSTEM AND METHOD FOR CONVERGING CIRCUIT SWITCHED AND PACKED SWITCHED COMMUNICATIONS / KOUCHRI, Farrokh Mohammadzadeh, US A method and system are provided for converging time division multiplexing (TDM) communication networks and packet based networks. Signaling may be monitored between TDM network ele ments by a Convergence Resource Manager (CRM) to identify calls that may be re-directed over a packet network to an end node with another associated CRM; the re-directed call bypassing much of the TDM network. Installation of a CRPd typically does not require any configuration changes to the TDM network such as point codes. A self learning CRM may also be provided so that a routing database may be automatically built. An originating SS7 message may be modified by a self learning CRM to add a Tag which identifies the originating node. As the modified message traverses the SS7 network, any other self learning CRMS in the network report to the originating self learning CRM that the message has been seen. The last reporting self learning CRM is then associated with the dialed number for future routing of originating calls over the packet network, directly to the last reporting self learning CRM and end network node, thus flattening the TDM network. |
| 8 |
SYSTEM AND METHOD FOR CONVERGING CIRCUIT SWITCHED AND PACKED SWITCHED COMMUNICATIONS / KOUCHRI, Farrokh Mohammadzadeh, US A method and system are provided for converging time division multiplexing (TDM) communication networks and packet based networks. Signaling may be monitored between TDM network ele ments by a Convergence Resource Manager (CRM) to identify calls that may be re-directed over a packet network to an end node with another associated CRM; the re-directed call bypassing much of the TDM network. Installation of a CRPd typically does not require any configuration changes to the TDM network such as point codes. A self learning CRM may also be provided so that a routing database may be automatically built. An originating SS7 message may be modified by a self learning CRM to add a Tag which identifies the originating node. As the modified message traverses the SS7 network, any other self learning CRMS in the network report to the originating self learning CRM that the message has been seen. The last reporting self learning CRM is then associated with the dialed number for future routing of originating calls over the packet network, directly to the last reporting self learning CRM and end network node, thus flattening the TDM network. |
| 9 |
System and method for mapping SS7 bearer channels / Triasys Technologies Corporation A system and method for associating Signaling System 7 logical circuits and bearer channels are presented. The system may include an event detector configured to receive an SS7 signaling message on an SS7 signaling link, parse a logical circuit from the SS7 signaling message, receive an SS7 bearer channel, and detect a bearer channel event on the SS7 bearer channel. A statistical learning model block is configured to calculate a correlation confidence value between said bearer channel and said logical circuit. The method may include parsing a logical circuit ID from a signaling message on an SS7 signal link, identifying a bearer channel associated with a bearer event on a bearer circuit, and calculating a current correlation confidence value between the logical circuit ID and the bearer channel. |
| 10 |
SYSTEM AND METHOD FOR CONVERGING CIRCUIT SWITCHED AND PACKED SWITCHED COMMUNICATIONS / Kouchri, Farrokh Mohammadzadeh A method and system are provided for converging time division multiplexing (TDM) communication networks and packet based networks. Signaling may be monitored between TDM network ele ments by a Convergence Resource Manager (CRM) to identify calls that may be re-directed over a packet network to an end node with another associated CRM; the re-directed call bypassing much of the TDM network. Installation of a CRPd typically does not require any configuration changes to the TDM network such as point codes. A self learning CRM may also be provided so that a routing database may be automatically built. An originating SS7 message may be modified by a self learning CRM to add a Tag which identifies the originating node. As the modified message traverses the SS7 network, any other self learning CRMS in the network report to the originating self learning CRM that the message has been seen. The last reporting self learning CRM is then associated with the dialed number for future routing of originating calls over the packet network, directly to the last reporting self learning CRM and end network node, thus flattening the TDM network. |
ScienceON '논문' 검색
MOREScienceON '보고서' 검색
MORE| No | 제목 / 내용 |
|---|---|
| 1 |
나노전자소자 기술을 응용한 신경세포 모방 뉴런소자 및 시스템 원천기술 개발 유승주 , 서울대학교 , 2021 |
| 2 |
글로벌 디지털 헬스케어 기술 동향 이상후 , 서울대학교 , 2019 |
| 3 |
PF 기반 뉴런 소자 및 회로 개발 이종호 , 서울대학교 , 2021 |
| 4 |
Flow cytometry 데이터 처리를 위한 머신러닝 기반 데이터 분석 프레임워크 이계민 , 서울과학기술대학교 , 2017 |