摘要
为提高航天测控软件的质量与可靠性,提出一种基于改进的PSO-SVM(Particle Swarm Optimization-Support Vector Machine,粒子群优化支持向量机)方法的航天测控软件缺陷预测模型。针对航天测控软件领域特征,构造了基于软件生命周期的软件度量集,并收集了实际航天测控软件的度量和缺陷数据,通过对软件历史版本数据的学习,在软件当前版本的生命周期早期数据的基础上进行缺陷预测。实例应用结果表明,采用历史版本软件数据对当前软件版本进行缺陷预测,从全局来看可达90%的预测准确度。因此,该方法可用于对航天测控软件的缺陷预测。
To ensure the quality and reliability of space TT＆C （Tracking, Telemetry and Command） software, a software defect prediction model using improved PSO-SVM （Particle Swarm Optimization-Support Vector Machine） algorithm is presented. Software metrics based on life-cycle are constructed according to the area characteristics of TT＆C software. Actual TT＆C software metrics and defect data are measured and learned by the prediction model. Application results over the early version data in the life-cycle of the current version show that the proposed TT＆C software defect prediction model reaches a global prediction accuracy of 90%. Therefore, it is applicable for defect prediction for space TT＆C software.
出处
《飞行器测控学报》
CSCD
2015年第1期102-108,共7页
Journal of Spacecraft TT&C Technology
关键词
软件缺陷
缺陷预测
粒子群优化支持向量机(PSO-SVM)
航天测控
软件度量
software defect
defect prediction
Particle Swarm Optimization-Support Vector Machine （PSO-SVM）
space Tracking,Telemetry and Command （TT＆C） software
software metrics
