Regarding a single-layered PLN network with feedback connections as an associative memory network,the complexity of recognition is discussed.We have the main result:if the size of the network N is m,then the complexit...Regarding a single-layered PLN network with feedback connections as an associative memory network,the complexity of recognition is discussed.We have the main result:if the size of the network N is m,then the complexity of recognition is an exponential function of m.The necessary condition under which the complexity of recognition is polynomial is given.展开更多
Due to the increase in the types of business and equipment in telecommunications companies,the performance index data collected in the operation and maintenance process varies greatly.The diversity of index data makes...Due to the increase in the types of business and equipment in telecommunications companies,the performance index data collected in the operation and maintenance process varies greatly.The diversity of index data makes it very difficult to perform high-precision capacity prediction.In order to improve the forecasting efficiency of related indexes,this paper designs a classification method of capacity index data,which divides the capacity index data into trend type,periodic type and irregular type.Then for the prediction of trend data,it proposes a capacity index prediction model based on Recurrent Neural Network(RNN),denoted as RNN-LSTM-LSTM.This model includes a basic RNN,two Long Short-Term Memory(LSTM)networks and two Fully Connected layers.The experimental results show that,compared with the traditional Holt-Winters,Autoregressive Integrated Moving Average(ARIMA)and Back Propagation(BP)neural network prediction model,the mean square error(MSE)of the proposed RNN-LSTM-LSTM model are reduced by 11.82%and 20.34%on the order storage and data migration,which has greatly improved the efficiency of trend-type capacity index prediction.展开更多
In this paper we propose a new discrete bidirectional associative memory (DBAM) which is derived from our previous continuous linear bidirectional associative memory (LBAM). The DBAM performs bidirectionally the opti...In this paper we propose a new discrete bidirectional associative memory (DBAM) which is derived from our previous continuous linear bidirectional associative memory (LBAM). The DBAM performs bidirectionally the optimal associative mapping proposed by Kohonen. Like LBAM and NBAM proposed by one of the present authors,the present BAM ensures the guaranteed recall of all stored patterns,and possesses far higher capacity compared with other existing BAMs,and like NBAM, has the strong ability to suppress the noise occurring in the output patterns and therefore reduce largely the spurious patterns. The derivation of DBAM is given and the stability of DBAM is proved. We also derive a learning algorithm for DBAM,which has iterative form and make the network learn new patterns easily. Compared with NBAM the present BAM can be easily implemented by software.展开更多
Previous research on deep-space networks based on delay-tolerant networking(DTN)has mainly focused on the performance of DTN protocols in simple networks;hence,research on complex networks is lacking.In this paper,we ...Previous research on deep-space networks based on delay-tolerant networking(DTN)has mainly focused on the performance of DTN protocols in simple networks;hence,research on complex networks is lacking.In this paper,we focus on network evaluation and protocol deployment for complex DTNbased deep-space networks and apply the results to a novel complex deep-space network based on the Universal Interplanetary Communication Network(UNICON-CDSN)proposed by the National Space Science Center(NSSC)for simulation and verification.A network evaluation method based on network capacity and memory analysis is proposed.Based on a performance comparison between the Licklider Transmission Protocol(LTP)and the Transmission Control Protocol(TCP)with the Bundle Protocol(BP)in various communication scenarios,a transport protocol configuration proposal is developed and used to construct an LTP deployment scheme for UNICON-CDSN.For the LTP deployment scheme,a theoretical model of file delivery time over complex deep-space networks is built.A network evaluation with the method proposed in this paper proves that UNICONCDSN satisfies the requirements for the 2020 Mars exploration mission Curiosity.Moreover,simulation results from a universal space communication network testbed(USCNT)designed by us show that the LTP deployment scheme is suitable for UNICON-CDSN.展开更多
In order to improve the bidirectional associative memory(BAM) performance, a modified BAM model(MBAM) is used to enhance neural network(NN)’s memory capacity and error correction capability, theoretical analysis and ...In order to improve the bidirectional associative memory(BAM) performance, a modified BAM model(MBAM) is used to enhance neural network(NN)’s memory capacity and error correction capability, theoretical analysis and experiment results illuminate that MBAM performs much better than the original BAM. The MBAM is used in computer numeric control(CNC) machine fault diagnosis, it not only can complete fault diagnosis correctly but also have fairly high error correction capability for disturbed Input Information sequence.Moreover MBAM model is a more convenient and effective method of solving the problem of CNC electric system fault diagnosis.展开更多
This paper analyzes the relationship between capacity and dynamics in recurrent correlation neural network, and points out that in some conditions the recurrent correlation neural network has high memory capacity. The...This paper analyzes the relationship between capacity and dynamics in recurrent correlation neural network, and points out that in some conditions the recurrent correlation neural network has high memory capacity. Then this paper presents several methods for improving the performance.展开更多
In this paper,the memory capacity of Probabilistic Logic Neuron(PLN)network is discussed. We obtain two main results:(1)the method for constructing a PLN network with a given memory capacity;(2)the relationship betwee...In this paper,the memory capacity of Probabilistic Logic Neuron(PLN)network is discussed. We obtain two main results:(1)the method for constructing a PLN network with a given memory capacity;(2)the relationship between the memory capacity and the size of a PLN network.We show that the memory capacity of a PLN network depends on not only the number of input ports of its element but also the number of elements themselves.The results provide a new method for designing a PLN network.展开更多
文摘Regarding a single-layered PLN network with feedback connections as an associative memory network,the complexity of recognition is discussed.We have the main result:if the size of the network N is m,then the complexity of recognition is an exponential function of m.The necessary condition under which the complexity of recognition is polynomial is given.
基金supported by Research on Big Data Technology for New Generation Internet Operators(H04W180609)the second batch of Sichuan Science and Technology Service Industry Development Fund Projects in 2018(18KJFWSF0388).
文摘Due to the increase in the types of business and equipment in telecommunications companies,the performance index data collected in the operation and maintenance process varies greatly.The diversity of index data makes it very difficult to perform high-precision capacity prediction.In order to improve the forecasting efficiency of related indexes,this paper designs a classification method of capacity index data,which divides the capacity index data into trend type,periodic type and irregular type.Then for the prediction of trend data,it proposes a capacity index prediction model based on Recurrent Neural Network(RNN),denoted as RNN-LSTM-LSTM.This model includes a basic RNN,two Long Short-Term Memory(LSTM)networks and two Fully Connected layers.The experimental results show that,compared with the traditional Holt-Winters,Autoregressive Integrated Moving Average(ARIMA)and Back Propagation(BP)neural network prediction model,the mean square error(MSE)of the proposed RNN-LSTM-LSTM model are reduced by 11.82%and 20.34%on the order storage and data migration,which has greatly improved the efficiency of trend-type capacity index prediction.
文摘In this paper we propose a new discrete bidirectional associative memory (DBAM) which is derived from our previous continuous linear bidirectional associative memory (LBAM). The DBAM performs bidirectionally the optimal associative mapping proposed by Kohonen. Like LBAM and NBAM proposed by one of the present authors,the present BAM ensures the guaranteed recall of all stored patterns,and possesses far higher capacity compared with other existing BAMs,and like NBAM, has the strong ability to suppress the noise occurring in the output patterns and therefore reduce largely the spurious patterns. The derivation of DBAM is given and the stability of DBAM is proved. We also derive a learning algorithm for DBAM,which has iterative form and make the network learn new patterns easily. Compared with NBAM the present BAM can be easily implemented by software.
基金supported by the Strategic leading project of the Chinese Academy of Sciences (Grant No. XDA15014603)。
文摘Previous research on deep-space networks based on delay-tolerant networking(DTN)has mainly focused on the performance of DTN protocols in simple networks;hence,research on complex networks is lacking.In this paper,we focus on network evaluation and protocol deployment for complex DTNbased deep-space networks and apply the results to a novel complex deep-space network based on the Universal Interplanetary Communication Network(UNICON-CDSN)proposed by the National Space Science Center(NSSC)for simulation and verification.A network evaluation method based on network capacity and memory analysis is proposed.Based on a performance comparison between the Licklider Transmission Protocol(LTP)and the Transmission Control Protocol(TCP)with the Bundle Protocol(BP)in various communication scenarios,a transport protocol configuration proposal is developed and used to construct an LTP deployment scheme for UNICON-CDSN.For the LTP deployment scheme,a theoretical model of file delivery time over complex deep-space networks is built.A network evaluation with the method proposed in this paper proves that UNICONCDSN satisfies the requirements for the 2020 Mars exploration mission Curiosity.Moreover,simulation results from a universal space communication network testbed(USCNT)designed by us show that the LTP deployment scheme is suitable for UNICON-CDSN.
文摘In order to improve the bidirectional associative memory(BAM) performance, a modified BAM model(MBAM) is used to enhance neural network(NN)’s memory capacity and error correction capability, theoretical analysis and experiment results illuminate that MBAM performs much better than the original BAM. The MBAM is used in computer numeric control(CNC) machine fault diagnosis, it not only can complete fault diagnosis correctly but also have fairly high error correction capability for disturbed Input Information sequence.Moreover MBAM model is a more convenient and effective method of solving the problem of CNC electric system fault diagnosis.
文摘This paper analyzes the relationship between capacity and dynamics in recurrent correlation neural network, and points out that in some conditions the recurrent correlation neural network has high memory capacity. Then this paper presents several methods for improving the performance.
文摘In this paper,the memory capacity of Probabilistic Logic Neuron(PLN)network is discussed. We obtain two main results:(1)the method for constructing a PLN network with a given memory capacity;(2)the relationship between the memory capacity and the size of a PLN network.We show that the memory capacity of a PLN network depends on not only the number of input ports of its element but also the number of elements themselves.The results provide a new method for designing a PLN network.
