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