01
5G Cellular Network design using NS3 Simulation
In this project, the students will simulate the key functionalities associated with the RAN and/or core network functions across the layers of the protocol stack for an LTE or 5G o…
Electronics · Course
Telecom Transformation using AI and ML
A comprehensive course on AI and ML for telecom use cases using modern network architecture design and development. This course is highly suited for professionals in Telecom industry for upskiling
- Duration
- 48 weeks
- Content
- 114 hours
- Projects
- 5
- Mode
- Online
OverviewWhat you'll
What you'll
be able to do.
- The students will have a thorough knowledge of Telecommunication Network transformation.
- They can specialize in the domain and gain complete in-depth knowledge of it.
- The students are exposed to the modern trends & the standard practices followed in the industry right now.
- After completing this course, the students will gain a better understanding of the concepts of Machine Learning and Artificial Intelligence for the Telecommunication Network Transformation application.
- The software used in this course such as NS3, MobileInsight, PyTorch, sklearn, Keras, TensorFlow, and Kubernetes will help the students to gain a better understanding of what is being taught to them and make them industry-ready.
Tools you'll use
NS3MobileInsightQXDMPyTorchscikit-learnKerasTensorFlowKubernetes
Curriculum24 modules,
24 modules,
built with industry.
01Telecom Transformation using AI and ML
- Introduction to 4G LTE and 5G NRW01
- Evolution from 4G LTE/LTE-A to 5G NR: IMT-2020 requirements for 5G
- 5G use-cases
- (a) Enhanced mobile broadband (eMBB)
- (b) Ultra-reliable low latency communication (uRLLC)
- (c) Machine-type Communication or massive IoT (mIoT, MTC)
- Gaps in LTE/LTE-A – to be bridged en route to 5G
- Key design principles of 4G and 5G NRW02
- Carrier aggregation
- Transmit diversity and spatial multiplexing
- Small cell deployment and network densification
- Cellular network deployment in unlicensed bands
- Resource allocation and quality of service
- Differences between LTE protocol stack and NR protocol stack
- PHY layer enhancements in 5G NRW03
- 5G NR frequency bands: sub-6 GHz and mmwave frequency bands
- 5G NR frame structure – Time Domain, the notion of TTIs, flexible TTIs, slots, flexible slots, aggregated slots, mini-slots, UL/DL multiplexing, and time-division duplexing (TDD/FDD), mmwave.
- 5G NR frame structure – Frequency Domain, the notion of sub-carrier spacing numerology and bandwidth parts.
- Massive MIMO, Analog, digital and hybrid beamforming, frequency localization in 5G NR waveform, and channel coding schemes.
- Cell-based association in LTE vs beam-based association in 5G NR.
- MAC layer and higher RAN layer enhancements in 5G NRW04
- Resource allocation (Physical Resource Blocks and Bandwidth part allocation), frequency/time multiplexing
- HARQ process, retransmissions, and Asynchronous HARQ
- Grant-free resource allocation for uplink.
- Carrier Aggregation: Primary cell and secondary cells.
- Dual connectivity architecture, Non-stand-alone, and stand-alone 5G architectures
- The notion of bearers and flows, QoS Class Index (QCI), QoS Flow Index (QFI)
- RLC windowing, RLC ARQ retransmission impact on latency.
- PDCP duplications and Service Data Adaptation Protocol.
- 5G Core Network and application designW05
- Enhanced EPC core and CUPS architecture
- 5G Next Generation (NG) Core network architecture and components.
- Network Slicing.
- 5G Core Interfaces and protocols
- 5G Core basic procedures.
- Cross-layer inter-dependencies: Impact of RAN on TCP/IP and application performances
- 360 video and Virtual Reality, Field-of-View based VR, drones/UAVs.
- Augmented Reality (AR) and AR gaming over a cellular network
- Wearables, IoT apps.
- Mobile edge cloud architecture
- Advanced topics and course summaryW06
- Disaggregated RAN architecture
- Open-RAN and virtual RAN design principles, Software-Defined Networking and Network Function Virtualization in RAN.
- RAN Intelligent Controller: Real-time and non-real-time.
- Machine Learning micro-services
- Introduction to ML and analyticsW07
- Fundamentals of machine learning.
- Challenges and potentials of machine learning in networks
- Application examples in networks
- Introduction to 5G and 4G performance counters.
- Measurement object classes
- Introduction to 5G and 4G Performance Measurements
- Introduction to 5G and 4G Key Performance Indicators
- 5G Performance Measurements – L1/L2 (DU)W08
- RLC PDU delay measurements
- Radio Resource utilization
- UE throughput measurements
- MCS and Transport Block related transmission and retransmission measurements
- F1-U PDU drop measurements
- L1 RSRP and SINR measurements
- CQI and MCS distributions
- RACH preambles
- Slices and sessions
- 5G Performance Measurements – L2/L3 (CU)W09
- RRC connected users
- DRB Session time and QoS flow time
- Handovers and types of handovers: Inter-system and intra-system handovers
- Radio Link failures and other causes
- PDCP Packet delay measurements
- PDCP bit-rate measurements
- PDCP data volume measurements
- Slices and session-specific measurements
- 5G and 4G Performance Measurements – KPIsW10
- Accessibility KPI
- Integrity KPI
- Utilization KPI
- Retainability KPI
- Mobility KPI
- Energy Efficiency KPI
- RRC connection measurements
- E-RAB related measurements
- Handover related measurements
- Cell-level radio bearer QoS related measurements
- Radio utilization measurements
- Carrier Aggregation measurements
- Power, Energy, Environment measurements
- Accessibility, Integrity, Utilization, Retainability KPIs
- Channel models, path loss and propagation loss modelsW11
- Channel models for 0.5 – 100 GHz
- Antenna Modeling
- Pathloss and penetration modeling
- Slow fading and Fast fading model
- Channel models for link-level evaluations and calibration
- NWDAF and MDAF analyticsW12
- UE Mobility prediction
- Network Performance
- QoS, Experience, and sustainability
- User Data Congestion
- NF Load
- M-plane analytics
- Introduction to ML for RANW13
- Supervised learning
- Unsupervised learning
- Semi-supervised learning
- Reinforcement learning
- Neural Networks
- Deep learning
- Applicability for cellular RAN predictions
- KPI Prediction techniques using regression/classificationW14
- Throughput prediction
- Latency prediction
- Packet loss prediction
- Basics of supervised and unsupervised learning
- Random Forest regression/classification for KPI prediction
- Support Vector Regression for KPI prediction
- RNN and Deep Learning and introduction to RLW15
- ARIMA time series for KPI prediction
- RNN LSTM and auto-encoders for KPI prediction
- Deep Learning for KPI prediction
- Markov Decision process
- Model-free vs Model-based RL
- Reinforcement LearningW16
- Off-policy vs On-policy RL
- Offline vs Online RL
- Q learning: Vanilla Q-learning and DQN
- Variants of Q learning
- Differences between Q learning, SARSA, PPO
- RL decision trees
- RL actions
- Bayesian Optimization and other ML techniquesW17
- Introduction to Bayesian optimization and acquisition functions
- Black box configuration in RAN
- Optimization of configuration functions
- Power and beam tilt optimization
- Review and RecapW18
- Supervised ML
- Unsupervised ML
- Reinforcement Learning
- Recurrent Neural Networks
- Regression, Classification and Decision trees
- Bayesian Optimization
- Building embedded ML intelligence in a cellular networkW19
- O-RAN architecture
- Service Management and Orchestration framework for M-plane cellular network configuration management
- Non-RT RIC and offline ML training
- Near-RT RIC and reinforcement learning/online ML
- Full ML life cycle
- Virtualization and SDN
- AI/ML use-case in cellular networksW20
- Traffic Steering
- Quality of Service/Quality of Experience
- Network slicing
- MU-MIMO Beamforming
- SoN optimization
- V2X, UAV, and connected vehicles
- Load optimization and UPF selection
- Slice-specific Schedulers
- Root-cause analysis
- Case studies (Part A): Traffic Steering, QoS and slicingW21
- Non-RT RIC for offline ML models
- KPI target and declarative guidance
- Near-RT RIC for RL/online models and inference host
- SMO components as inference hosts
- SMF as inference host for UPF selection
- UPF as inference host for traffic delivery
- Case studies (Part B): Root-cause diagnostics and anomaly detectionW22
- PM and KPI Data-driven analysis
- Root-cause diagnostics
- What-if analysis
- SoN optimization and Radio link failure minimization
- Channel State Information Predictions from imperfect channel feedback
- Case studies (Part C): RAN-assisted IP analyticsW23
- RAN PMs and KPIs for throughput
- Using network/edge cloud APIs to communicate RAN KPIs to core/application network – using NWDAF and Application Function
- Video bit-rate adaptation at application server using ML and deep learning techniques.
- QoE metrics: Freezing, Stalling, Session Rebuffering, etc.
- RAN PMs and KPIs for latency
- Using network/edge cloud APIs to communicate RAN KPIs to core/application network – using NWDAF and Application Function
- IP packet size adaptation using ML/RL techniques.
- QoE metrics: Screen freezes/blanks, etc.
- Cellular Network deploymentsW24
- RAN PMs and KPIs for latency
- Using network/edge cloud APIs to communicate with the RAN KPIs to core/application network – using NWDAF and Application Function
- IP packet size adaptation using ML/RL techniques.
- QoE metrics: Screen freezes/blanks, etc.
- Introduction to 4G LTE and 5G NRW01
Industry projects5 projects
5 projects
you'll ship.
Our projects are designed by experts in the industry to reflect industry standards. By working through our projects, Learners will gain a practical understanding of what they will take on at a larger-scale in the industry. In total, there are 5 Projects that are available in this program.
02
RAN Data analytics using MobileInsight or QXDM
RAN Data analytics using MobileInsight or QXDM
03
Transport layer performance of Facebook 360 using Wireshark PCAP logs
Transport layer performance of Facebook 360 using Wireshark PCAP logs
04
Cloud-native microservices for real-time prediction and optimization
Cloud-native microservices for real-time prediction and optimization
05
ML Models for Network KPI prediction
ML Models for Network KPI prediction
FeeOne fee.
One fee.
Everything included.
Become job ready with our comprehensive industry focused curriculum for freshers & early career professionals
Premium
₹2,00,000
Inclusive of all charges- 1 Year Access
- Personalized Page
- Live Individual & Group Sessions
- Personalized & Hands-On
- Job-Oriented Industry Relevant Curriculum
Related
You might also like.
Next stepNot sure it's
Not sure it's
the right fit?
A counsellor will walk you through the syllabus, projects and fee options — and tell you honestly if another course suits you better.