第一篇   基础与网络篇

第1章  “5G+AI”概述·································2

1.1  新基建下的“5G+AI”技术发展·························3

1.1.1  新基建的内涵和外延···············································3

1.1.2  新基建对5G和AI发展的影响······························6

1.2 5G时代的AI技术趋势······································10

1.2.1 AI部署云边协同····················································10

1.2.2 AI注智实时持续····················································12

1.2.3 AI应用民主灵活····················································13

1.2.4 AI决策高度仿真····················································14

1.3  我国5G产业与技术发展···································16

1.3.1  我国5G技术发展历程··········································16

1.3.2 5G改变社会···························································17

1.4  我国AI产业与技术发展····································22

1.4.1  人工智能发展概述·················································22

1.4.2  我国人工智能技术的发展·····································24

第2章  AI与5G网络智能切片····················29

2.1 5G业务多样化与网络需求弹性化····················29

2.2 5G网络智能切片概述········································31

2.2.1 5G网络智能切片的概念与特征···························32

2.2.2 5G网络智能切片端到端结构·······························33

2.2.3 5G网络智能切片的RAN侧技术挑战················34

2.2.4 5G网络智能切片的AI平台和分析系统·············35

2.2.5 5G网络智能切片的智能部署·······························36

2.2.6 5G网络智能切片的标准化增强···························37

2.3  应用于5G网络切片中的AI技术·····················38

2.3.1 5G网络智能切片的设计流程·······························38

2.3.2  基于GA-PSO优化的网络切片编排算法············43

2.3.3 5G网络切片使能智能电网···································53

2.3.4  应用于NWDAF中的联邦学习技术····················59

第3章  AI与智能物联网······························63

3.1 5G时代IoT海量数据实时处理·························63

3.2  边缘计算与云边协同··········································65

3.2.1  边缘计算·················65

3.2.2  云边协同·················67

3.3  应用于智能IoT中的AI技术····························72

3.3.1  联邦迁移学习·························································72

3.3.2 RPnet网络与车牌识别··········································74

3.3.3  对抗生成网络与移动目标检测·····························76

3.3.4 Android手机去中心化的分布式机器学习···········78

3.3.5  “AI＋移动警务”················································79

第4章  AI与5G网络多量纲计费················80

4.1 5G时代变得日益复杂的网络计费····················80

4.2 5G多量纲计费概述············································82

4.2.1  与4G计费量纲对标··············································83

4.2.2 5G计费因子确定···················································85

4.2.3 5G计费欺诈预防···················································86

4.2.4 5G流量异常监测···················································87

4.3  应用于智能计费中的AI技术····························89

4.3.1 ST-DenNetFus算法与网络需求弹性分析············89

4.3.2  强化学习（RL）与客户意图分析························92

第二篇? 客户与管理篇

第5章  AI与客户体验管理··························98

5.1  客户感知网络质量与客观KPI指标差异··········98

5.2 CEM概述···························································102

5.2.1 CEM基本概念·····················································102

5.2.2  客户网络体验感知量化·······································104

5.2.3 CEMC与端到端客户服务体验改善··················106

5.3  应用于CEM中的AI技术·······························108

5.3.1 ADS算法与用户网络感知原因定位··················109

5.3.2 Chatbot技术与客服体验优化·····························111

5.3.3  基于KDtree、LSTM以及多算法融合的网络容量预测··································113

5.3.4 NPS度量与用户业务感知提升··························114

第6章  AI与客户关系管理（CRM）·········118

6.1 5G需求差异化与服务精准化··························118

6.2 CRM概述··························································120

6.2.1 CRM基本概念·····················································120

6.2.2 AI注智客户差异化服务营销······························121

6.3  应用于CRM中的AI技术·······························122

6.3.1 BERT技术在客服NLP中的应用······················122

6.3.2  基于用户单侧通话记录检测的诈骗电话识别···················································127

6.3.3  应用于用户差异化营销中的人脸识别应用技术···············································131

6.3.4  应用于户外广告屏的人体属性识别技术···········134

6.3.5 MPMD加权回归方法在客户画像中的应用实现··············································139

6.3.6  “CRNN+OpenCV”与用户身份证信息自动录入···········································146

6.3.7  基于OCR识别的用户签名信息核对·················148

6.3.8  基于中心性和图相似性算法的智能推荐应用···················································148

6.3.9  基于LDA和MLLT的语音识别特征变换矩阵估计方法································150

6.3.10  基于MFCC和Kaldi-chain声学模型的语音情绪分析···································153

第7章  AI与流程管理································156

7.1  智能流程管理与企业降本增效························156

7.2 AIRPA助力数字化转型····································157

7.2.1 RPA概述··············157

7.2.2 RPA开发运行流程··············································161

7.2.3 RPA开发工具······················································163

7.2.4 RPA管控调度······················································164

7.2.5 RPA任务执行引擎··············································166

7.3  应用于智能流程管理中的AI技术··················167

7.3.1 YOLO模型检测和分类票据·······························167

7.3.2  用OpenCV去除印章···········································169

7.3.3 CRNN识别票据关键信息···································170

7.3.4  基于模板的OCR识别·········································171

第8章  AI与商业智能································173

8.1 5G与运营商业务决策和业务流程优化··········173

8.2  构建基于通信AI的全面战略管理决策体系··················································176

8.3  应用于智能决策中的AI技术··························177

8.3.1  纳什均衡算法与携号转网最优市场决策···········177

8.3.2  Transfer Learning（迁移学习）技术与客户携转风险识别······························183

8.3.3  基于多源指标关联分析的业务沙盘推演···········186

8.3.4  基于社群发现的用户转网预警分析···················192

第三篇? 运维与安全篇

第9章  AI与网络智能运维························198

9.1 5G网络复杂化与运维模式创新······················198

9.2 AIOps概述·························································200

9.2.1 AIOps概念与关键业务流程·······························200

9.2.2 AIOps与智能运维学件·······································202

9.3  应用于智能运维中的AI技术··························204

9.3.1  基于动态阈值的网络运维异常检测···················204

9.3.2  基于DBSCAN和Apriori算法的传输网告警根因定位···································209

9.3.3  集成学习算法与网络故障预测···························214

9.3.4  时序算法与网络黄金指标预测···························216

9.3.5  基于异构知识关联的运维知识图谱构建···········218

第10章AI与机房智慧管控·······················221

10.1 5G时代的中心机房智慧管控························221

10.2  机房资源调度与监控管理概述······················223

10.2.1  机房环境物理指标·············································223

10.2.2  “IoT+AI”辅助机房管理自动化·····················224

10.2.3  机房安防布控与违规预警·································225

10.3  应用于机房智能化中的AI技术····················225

10.3.1  机器学习方法辅助数据中心降低能源消耗·····················································225

10.3.2 Faster-RCNN目标检测算法监控机柜资源占用··············································229

10.3.3  基于计算机视觉方法的机房火情监测·············233

第11章AI与智能安防······························235

11.1  “5G+AI”安防发展趋势·······························236

11.2  应用于智能安防中的5G技术·······················239

11.2.1  无线视频监控部署·············································239

11.2.2  三域一体立体化防控·········································241

11.2.3  海量数据实时响应·············································242

11.3  应用于智能安防中的AI技术························244

11.3.1 AI安防模型························································244

11.3.2 AI服务实现························································250

11.3.3  资源混编调度·····················································252

第12章5G时代的AI能力平台化············255

12.1 AI平台建设与能力沉积·································255

12.2 AI平台建设理念与思路·································256

12.3  AI平台建设功能设计····································261

12.3.1  云化引擎设计·····················································261

12.3.2 API算法体系······················································262

12.3.3 AI能力生产方式················································262

12.3.4 AI能力输出方式················································265

12.3.5  与生产环境对接·················································266

12.4  AI平台建设的技术设计·································267

参考文献······················································269