RA (Repeat Accumulate) code has been widely used in wireless communication systems. But the bp (Belief Propagation) decoding algorithm which is commonly used in RA Code has high complexity. In order to reduce the comp...
详细信息
RA (Repeat Accumulate) code has been widely used in wireless communication systems. But the bp (Belief Propagation) decoding algorithm which is commonly used in RA Code has high complexity. In order to reduce the complexity of bp algorithm while maintaining coding performance, we proposed Chebyshev-Polynomial-Fitting based bp (CPF-bp) decoding algorithm. In CPF-bp decoding algorithm, we used polynomial which is generated by chebyshev fitting algorithm instead of hyperbolic tangent operation in bp decoding algorithm to reduce the implementation complexity. Experiment and simulation results show that this low-complexity algorithm removes hyperbolic tangent operation in conventional bp algorithm with same amount of multiplier under the parameters setting in the paper. The performance degradation of CPF-bp algorithm is negligible which is less than 0.6dB at BER of 4 10 compared with conventional bp algorithm.
In order to evaluate the performance of the logistics enterprise scientifically, and by using of advantage of artificial intelligence (bp neural network) with self-organizing, self-learning, anti-interference, etc., a...
详细信息
In order to evaluate the performance of the logistics enterprise scientifically, and by using of advantage of artificial intelligence (bp neural network) with self-organizing, self-learning, anti-interference, etc., a performance evaluation model of logistics enterprise was established. Then the step of the model was introduced in detail, bp neural network was constructed and bp algorithm was designed. Finally, an example was given for testing the effectiveness and feasibility of the model, the example shows that the evaluation results are basically consistent with the actual situation.
According to the floatation processing characteristic with time-variation, uncertainty and complicated nonlinear relations, a prediction method of concentrate grade and prediction model of ore dressing date is propose...
详细信息
ISBN:
(纸本)9781612848396
According to the floatation processing characteristic with time-variation, uncertainty and complicated nonlinear relations, a prediction method of concentrate grade and prediction model of ore dressing date is proposed. This article establish a prediction model of ore dressing date based on Jordan neural network including input of influence factors and dynamic time sequence feedback of concentrate grade, by combining bp algorithm with the temporal difference methods. The results applied in industry indicate that predictive precision is high, error is small, and stability is high. It has practical value, the application is successful.
Personal credit scoring plays an important role in keeping away from credit risks in consumer credit market of commercial *** to the insufficiency of bp neural network,this paper constructs a PSO neural network model ...
详细信息
Personal credit scoring plays an important role in keeping away from credit risks in consumer credit market of commercial *** to the insufficiency of bp neural network,this paper constructs a PSO neural network model for personal credit scoring by using PSO to train the network's *** with bp network,PSO network gets higher classification accuracy in testing samples and shows stronger ***'s more,PSO network is more applicable for personal credit scoring in a dynamic credit environment.
Low-density parity-check (LDPC) codes deliver very good performance when decoded with the bp algorithm. Unfortunately, this algorithm has the disadvantage to be complex for implementation. In this paper, we investigat...
详细信息
Low-density parity-check (LDPC) codes deliver very good performance when decoded with the bp algorithm. Unfortunately, this algorithm has the disadvantage to be complex for implementation. In this paper, we investigate the bp algorithm and its simplifications (bp-Based, normalized bp-Based, offset bp-Based). Previous works on normalized and offset bp-Based proposed theoretical values for the offset and normalization factors, we perform simulations which allow to validate these theoretical results.
According to the current situation of finish machining by free grinding tools, GA-bp algorithm, which integrates neural network and genetic algorithm together, is applied to finish machining. A model based on genetic ...
详细信息
According to the current situation of finish machining by free grinding tools, GA-bp algorithm, which integrates neural network and genetic algorithm together, is applied to finish machining. A model based on genetic neural network has been built to forecast surface quality of machined parts and to choose optimized processing parameters. Finish machining by fluid magnetic grinding tools has been illustrated by comparing theoretical calculations with measured values. The result that the error is only about 5% shows that the algorithm is better. After simple modification, such algorithm could be successfully applied in other finish machining forms by free grinding tools, consequently, reasonable machining methods could be chosen.
In order to improve the efficiency and decoding performance of LDPC-coded BICM system, the joint de-modulation Belief Propagation(JMbp) algorithm is proposed in this paperDue to the good sparse structure of LDPC code,...
详细信息
In order to improve the efficiency and decoding performance of LDPC-coded BICM system, the joint de-modulation Belief Propagation(JMbp) algorithm is proposed in this paperDue to the good sparse structure of LDPC code, the JMbp algorithm can be derived from the generalized optimal bp algorithm based on the generalized distributive law(GDL)The JMbp algorithm has comparably low complexity while provides better decoding performance when compared with the traditional BICM-bp algorithm in highary modulated BICM systemsAnd the simulation results confirm that the JMbp algorithm outperforms the traditional BICM-bp algorithm over both the AWGN and Rayleigh fading channel.
The costs of pre-building graphic in the construction project management is a very important job,based on neural network theory and the characteristics of the costs of building graphic,raised GA neural network to solv...
详细信息
The costs of pre-building graphic in the construction project management is a very important job,based on neural network theory and the characteristics of the costs of building graphic,raised GA neural network to solve the cost of building graphic problem,applied bp network model and used GA to optimize the weight based on bp *** calculation examples showed that the accuracy of the cost estimation met the requirements basically.
The intrusion detection technology took one of computer network information security measures important methods, the invasion detection took one kind of dynamic safe protection technology,has provided to the internal ...
详细信息
The intrusion detection technology took one of computer network information security measures important methods, the invasion detection took one kind of dynamic safe protection technology,has provided to the internal attack, exterior attack and the disoperation real-time protection,receives before the endangerment in the network system interception and response *** on the characteristic to the computer network data to withdraw, proposed used the Genetic algorithms and the Neural Network unifies the invasion detection *** Genetic algorithms have the computation to be simple,the optimized effect good *** the bp algorithm using the Genetic algorithms the local minimum point,thus achieves a minimum point of RMS error,also solved the bp algorithm to restrain the slow question;At the same time also solved has alone used GA algorithm often not to be able to seek in the short time to approaches the optimal solution this *** confirmed the invasion detection effect through the computer experiment,enhanced the recognition rate,causes reporting mistakenly rate and failing to report rate reduces.
This paper is devoted to the study and the implementation of a joint source/channel decoding algorithm of VLC (Variable Length Code) and LDPC (Low Density Parity Check codes) codes. The principle of joint decoding...
详细信息
This paper is devoted to the study and the implementation of a joint source/channel decoding algorithm of VLC (Variable Length Code) and LDPC (Low Density Parity Check codes) codes. The principle of joint decoding is the information transfer between the source and channel decoder. Then, we can’t apply the classic VLC decoding algorithm. So, we implemented MAP decoding algorithm basing on trellis description. Simulations results showed that the performances of this algorithm are better than those of a classic decoding. Then, we introduced a LDPC code into the considered communication chain (source coded using a VLC code) and we proceed to an iterative joint decoding of VLC and LDPC codes. Simulation results show that iterative joint source/channel decoding provide an important gain compared to classic communication schema.
暂无评论