📝 Kademlia identifier space (9 MCQs)

📖 Explanation: When the receiver assumes a monotonically increasing sequence, the sudden jump from 65535 to 0 appears as a regression. Without wrap‑around logic, it treats the new packet as already received, discarding it. This creates a perceived loss and can disrupt ordering, leading to gaps in the media stream.

📖 Explanation: RTCP provides periodic reports that inform the sender about packet loss, jitter, and round‑trip time. Losing 5 % of these reports means the sender receives incomplete feedback, leading it to assume the network is healthier than it actually is. Consequently, it may keep or raise its bitrate, aggravating congestion and degrading call quality.

📖 Explanation: The expected number of lost packets equals $N \times p = 10{,}000 \times 0.02 = 200$. If each loss adds 5 ms of buffering, the total extra delay is $200 \times 5\text{ ms} = 1{,}000\text{ ms}$ or 1 second. Therefore the jitter buffer must be enlarged by roughly one second to absorb the additional variability.

📖 Explanation: RTP (Real‑time Transport Protocol) is responsible for the timely delivery of audio, video, or other media streams. RTCP (RTP Control Protocol) operates alongside RTP, exchanging statistics, quality reports, and participant information. This separation allows media to flow continuously while control messages are sent less frequently to monitor and adapt the session.

📖 Explanation: SIP registrars can be deployed in a distributed fashion, allowing multiple servers to share the registration load. This design prevents any single node from becoming a bottleneck, reduces latency for registration queries, and scales more gracefully as the number of endpoints grows, directly easing network resource consumption.

📖 Explanation: The compressed frame of 500 KB equals about 512 000 bytes. Dividing by the payload limit of 1 500 bytes yields roughly 341 packets per frame. Raw frames of 2 MB (≈2 097 152 bytes) at 30 fps require about 505 Mbps, whereas compressed frames need about 123 Mbps, giving a reduction factor close to 4×.

📖 Explanation: A larger identifier space expands the pool of possible Call‑IDs, making the chance that two independently generated IDs coincide drop dramatically. According to the birthday‑paradox principle, the collision probability falls roughly with the square of the space size, so expanding it renders accidental collisions practically impossible.

📖 Explanation: Lossless compression preserves the original data, so any bit error can be detected and often corrected, enhancing error resilience. However, preserving all information typically yields larger file sizes than lossy alternatives, increasing the required bandwidth. The higher bandwidth can introduce additional queuing delay, raising overall latency, which must be managed in real‑time applications.

📖 Explanation: Appending a timestamp (e.g., epoch time) to the Call‑ID creates a value that changes with each new session, guaranteeing temporal uniqueness even if the random portion repeats. This prevents accidental reuse of an identifier within the window where a previous session might still be active, thereby avoiding misrouting or cross‑talk.