CAJ | 학술논문

针对Viterbi译码算法的计算复杂度随着卷积码约束长度的增加呈指数增加,译码延迟过大,只适用于约束长度较小的卷积码译码的缺陷,提出了适用于大约束度的卷积码译码方法.采用了改进粒子群优化算法,弥补传统粒子群优化算法在解决离散问题方面的缺陷--对卷积码快速译码.该方法通过设定种群规模M来确定译码路径数,极大地缩小了译码网格中的路径搜索范围,使译码延迟减小,更适用于约束长度较大的卷积码.还提出了译码宽度自适应的卷积码译码方法,对Viterbi译码算法进行了改进,把固定的译码路径宽度改进为随信道噪声的变化而变化,大大降低译码计算复杂度.仿真实验表明提出的2种译码方法的有效性.
침대Viterbi역마산법적계산복잡도수착권적마약속장도적증가정지수증가,역마연지과대,지괄용우약속장도교소적권적마역마적결함,제출료괄용우대약속도적권적마역마방법.채용료개진입자군우화산법,미보전통입자군우화산법재해결리산문제방면적결함--대권적마쾌속역마.해방법통과설정충군규모M래학정역마로경수,겁대지축소료역마망격중적로경수색범위,사역마연지감소,경괄용우약속장도교대적권적마.환제출료역마관도자괄응적권적마역마방법,대Viterbi역마산법진행료개진,파고정적역마로경관도개진위수신도조성적변화이변화,대대강저역마계산복잡도.방진실험표명제출적2충역마방법적유효성.
To overcome the disadvantage of Viterhi decoding algorithm, in which its complexity exponentially increases with the increasing constraint length of convolutional codes, and the decoding delay was too large to fit the decoding of longer constraint length convolutional codes, a fast decoding of convolutional codes for longer constraint length, based on improved particle swarm optimization algorithm, was proposed. The proposed method reduces the searching area in the grid of decoding and shortens the decoding delay by setting the population size M to determine the number of decoding path, therefore was more suitable for longer constraint length convolufional codes. Another method of decoding convolutional codes based on serf-adapting of decoding width was also proposed. Simulation results show that the proposed both methods have advantages in reducing the computational complexity and the decoding time.