0% found this document useful (0 votes)
5 views48 pages

TS WN Lesson 8

The document provides an overview of telecommunication services, specifically focusing on Voice over IP (VoIP) technology, its components, and protocols. It discusses the differences between circuit-switched and packet-switched voice communication, VoIP endpoints, and the importance of Quality of Service (QoS) metrics. Additionally, it highlights security measures for VoIP and the challenges posed by NAT traversal, concluding with key takeaways on VoIP functionality and deployment considerations.
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
5 views48 pages

TS WN Lesson 8

The document provides an overview of telecommunication services, specifically focusing on Voice over IP (VoIP) technology, its components, and protocols. It discusses the differences between circuit-switched and packet-switched voice communication, VoIP endpoints, and the importance of Quality of Service (QoS) metrics. Additionally, it highlights security measures for VoIP and the challenges posed by NAT traversal, concluding with key takeaways on VoIP functionality and deployment considerations.
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Telecommunication Systems and

Wireless Networks

Telecommunication Services and Voice over IP


Telecommunication services: where VoIP fits

Services

Voice calling • Messaging • Video meetings • Contact centers • Emergency services

Control & media protocols

Signaling (SIP / IMS) • Media transport (RTP/RTCP) • Security (TLS/SRTP)

Networks

Access (Wi-Fi, DSL, cable, fiber, 4G/5G) • IP core (routing, QoS, peering) • Legacy interconnect (PSTN, gateways)

VoIP = voice service delivered over IP (not dedicated circuit-switched PSTN)


Circuit switching vs packet switching (voice)

Circuit-switched (PSTN) Packet-switched (IP/VoIP)

• Dedicated end-to-end path (reserved resources) • Voice encoded → packetized → routed like other
• Predictable delay and jitter IP traffic
• Signaling + media tied to the circuit • Statistical multiplexing (shared links)
• Efficient for constant-rate voice; less flexible for • Variable delay/jitter; loss possible (best effort)
data • Needs buffering, QoS, and robust codecs
VoIP endpoints: ATA vs IP phone vs softphone

ATA (Analog Telephone Adapter) IP Phone Softphone

• App on PC/mobile
• Reuse legacy handset • Ethernet/Wi-Fi endpoint
• Uses headset/mic
• Good for home migration • Often supports PoE
• Great for remote work +
• Gateway: analog digital • Enterprise features (PBX/UC)
WebRTC
VoIP split: signaling vs media
Setup → Talk → Teardown

Control plane (Signaling)

• SIP messages: REGISTER, INVITE, 200 OK, ACK, BYE


• Negotiates media parameters via SDP

User plane (Media)

• RTP carries audio/video packets


• RTCP carries feedback: loss, jitter, timing
SIP
Session Initiation Protocol (signaling)
RTP / RTCP
Real-Time Transport Protocol (media) + control feedback
QoS for VoIP: what can go wrong (and what we do)

Key metrics Typical mitigations

• One-way delay (latency): humans notice


conversational lag • Packet prioritization (DiffServ/queuing) on managed
• Jitter: variation in delay → needs jitter buffer networks
• Packet loss: causes gaps/robotic audio; PLC can mask • Adaptive jitter buffer and packet-loss concealment
small loss • Codec choice: bandwidth vs robustness vs CPU
• Echo: worse with higher delays → echo cancellation • RTCP feedback + congestion control (esp. for video)
matters

Rule of thumb: keep one-way delay well below 150 ms for “good” voice; 150–400 ms can be acceptable with care and echo control.
Codecs & packetization: why overhead matters

Common audio codecs (examples) Header overhead (IPv4/UDP/RTP)

• IP header: 20 bytes
• G.711 (PCM): high quality, ~64 kbps payload
• UDP header: 8 bytes
• Opus: wideband, adapts bitrate/packet loss
• RTP header: 12 bytes
• G.729 / AMR: lower bitrate, more compression
• Total: 40 bytes (before link-layer headers)

Packetization interval (typical)


Example (G.711, 20 ms):
~10–30 ms of audio per RTP packet (trade-off: overhead vs Payload: 64 kbps × 0.02 s = 1280 bits = 160
delay) bytes
Headers: ~40 bytes
Overhead ≈ 40 / (160 + 40) = 20%
Security & NAT traversal (practical VoIP)

Security goals Why NAT is tricky


• Protect signaling (who is calling whom?) • SIP/SDP carry IP:port info that NAT may rewrite
• Protect media (what is being said?) • RTP uses dynamic UDP ports
• Prevent spoofing/replay
Typical approach
Common building blocks • STUN: discover public mapping
• SIP over TLS (sips:) • TURN: relay when direct path fails
• SRTP/SRTCP for media confidentiality + integrity • ICE: tries candidate paths and picks the best

Operational note: emergency calling and lawful intercept requirements may affect VoIP deployments.
Recap & quick check

Key takeaways

• VoIP runs voice over IP networks; quality depends on delay/jitter/loss.


• SIP is signaling (setup/modify/terminate); RTP is the media transport.
• RTP adds sequencing/timestamps; RTCP adds feedback and synchronization.
• Real deployments need security (TLS/SRTP) and NAT traversal (ICE).

Quick check (1–2 minutes)

• 1) In SIP, which message confirms the final response to INVITE?


• 2) In RTP, which fields help detect loss and reorder packets?
• 3) Why is TCP usually a bad fit for real-time voice media?

You might also like