earticle

논문검색

Research on Ant Colony Algorithm Optimization Neural Network Weights Blind Equalization Algorithm

초록

영어

The project of ant colony algorithm optimization neural network combining blind equalization algorithm is proposed. The better initial weights of neural networks are provided because of the randomness, ergodicity and positive feedback of the ant colony algorithm. And then, a combination of optimal weights are found through BP algorithm, which is fast local search speed. Thus blind equalization performance is improved. Computer simulation show that, the novel blind equalization algorithm speeds up the convergence rate, reduces the remaining steady-state error and bit error rate, which is compared with the Neural Network Blind Equalization Algorithm(NNBE) and Genetic Algorithm optimization Neural Network Blind Equalization Algorithm(GA-NNBE) .

목차

Abstract
 1. Introduction
 2. The Basic Ideas on Ant Colony Algorithm Optimizing Neural Network Weight in Blind Equalization Algorithm
 3. Ant Colony Algorithm Optimizing the Blind Equalization of Neural Network Initial Weight and its Realization
  3.1. The Selection of Neural Network
  3.2. Ant Colony Algorithm Optimizing the Initial Weight of Neural Network
  3.3. Ant Colony Algorithm Optimizing the Neural Network in Blind Equalization Algorithm
 4. Experimental Simulation
 5. Conclusions
 References

저자정보

  • Yanxiang Geng College of Information and Engineering Tianjin University of Commerce, Tianjin, 300134, China
  • Liyi Zhang College of Information and Engineering Tianjin University of Commerce, Tianjin, 300134, China; College of Information and Engineering Tianjin University, Tianjin, 300072, China
  • Yunshan Sun College of Information and Engineering Tianjin University of Commerce, Tianjin, 300134, China; College of Information and Engineering Tianjin University, Tianjin, 300072, China
  • Yao Zhang College of Information and Engineering Tianjin University of Commerce, Tianjin, 300134, China
  • Nan Yang College of Information and Engineering Tianjin University of Commerce, Tianjin, 300134, China
  • Jiawei Wu College of Information and Engineering Tianjin University of Commerce, Tianjin, 300134, China

참고문헌

자료제공 : 네이버학술정보

    함께 이용한 논문

      ※ 원문제공기관과의 협약기간이 종료되어 열람이 제한될 수 있습니다.

      0개의 논문이 장바구니에 담겼습니다.