ICACT20260346 Question.1
Questioner: yangyn@dsp.ac.cn 2026-02-09 오후 12:03:12 |
ICACT20260346 Answer.1
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 12:03:12
Chrome  Click!! |
| How does the two-stage relay selection strategy in UW-LFRR prioritize between path stability and computational efficiency, and were any trade-offs observed in practice? |
UW-LFRRs two-stage relay selection strategy prioritizes path stability by first attempting to graft traffic back to the original backbone (Phase I) with minimal detour (at most 2 extra hops) and low computational complexity O(|Ku|), only resorting to a more complex search within the branch-off region (Phase II) with O(|Ku| log|Ku|) complexity when the backbone is completely disrupted. This approach achieves high path stability with minimal computational cost in most failure scenarios, while Phase II ensures connectivity when needed, resulting in a practical balance between stability and efficiency. |
ICACT20260346 Question.10
Questioner: zhaoyb@dsp.ac.cn 2026-02-09 오후 12:04:39 |
ICACT20260346 Answer.10
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 12:04:39
Chrome Click!! |
| In the recursive multi-failure recovery framework, multiple nodes may simultaneously detect failures at different locations and promote themselves as RNs. If these repair paths overlap spatially, is there a potential risk of routing loops? |
No, there is no potential risk of routing loops in this scenario. UW-LFRRs recursive multi-failure recovery framework ensures that each repair segment either follows a tree path in an SPT or a shortest path on the purified topology Gu, both of which are loop-free by construction. When multiple nodes promote themselves as RNs, the new repair node updates the failure set and rebuilds the local purified view, pruning all paths through newly failed nodes. Combined with finite-lifetime temporary entries and the latest-first selection rule, this ensures that the same packet cannot traverse a persistent loop under an unchanged rule set, even when multiple repair paths overlap spatially. |
ICACT20260346 Question.11
Questioner: shil@dsp.ac.cn 2026-02-09 오후 12:04:48 |
ICACT20260346 Answer.11
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 12:04:48
Chrome Click!! |
| In the two-stage relay selection strategy of UW-LFRR, how does Phase I differ from Phase II in terms of search objective and computational complexity, especially under different failure scenarios? |
Phase I aims to quickly graft traffic back to the original backbone with minimal detour (at most 2 extra hops) and has computational complexity O(|Ku|), making it suitable for most failure scenarios where the backbone is partially available. Its search objective is to find a grafting point on the original shortest path (Porig) that minimizes the distance from the repair node to the boundary relay (BR). Phase II, in contrast, is triggered only when the backbone is completely disrupted within the branch-off region (Cmain = ∅), and it searches for a relatively optimal detour over all of Ku with complexity approximately O(|Ku| log|Ku|). Its search objective is to find a pair (BR, GR) that minimizes the total cost (dGu(RN, BR) + 1 + dCfallback(GR, DST)), ensuring connectivity when the backbone is fully cut. |
ICACT20260346 Question.12
Questioner: shil@dsp.ac.cn 2026-02-09 오후 12:04:57 |
ICACT20260346 Answer.12
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 12:04:57
Chrome Click!! |
| How does UW-LFRR’s first-packet paving mechanism limit the spread of temporary forwarding state and ensure that control overhead remains low in underwater acoustic networks? |
UW-LFRRs first-packet paving mechanism limits the spread of temporary forwarding state by installing entries only at nodes that would otherwise misroute traffic (i.e., where the original next hop toward DST lies in the failure region and differs from the next hop toward BR). It uses a finite lifetime (TTL) for entries, which decreases to zero and causes the entry to be removed, ensuring state doesnt persist unnecessarily. The latest-first policy ensures that the most recent failure view overrides older paths, preventing stale state from persisting. This localized, bounded installation of state, combined with the finite TTL and latest-first selection, ensures that temporary state remains confined to a small area around the failure, keeping control overhead very low (close to baseline HL-OLSR). |
ICACT20260346 Question.13
Questioner: Lee19711936516@163.com 2026-02-09 오후 2:27:45 |
ICACT20260346 Answer.13
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 2:27:45
Chrome Click!! |
| What are the core steps through which the local purification topology G ᵤ constructed by UW-LFRR is obtained? What kind of constraint does it have on the search space for subsequent fault repair? |
|
ICACT20260346 Question.14
Questioner: Lee19711936516@163.com 2026-02-09 오후 2:28:45 |
ICACT20260346 Answer.14
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 2:28:45
Chrome Click!! |
| In the recursive recovery scenario of multi node failures, the new repair node of UW-LFRR will perform Sanity Check before triggering secondary repair. What is the core logic of this check? If the inspection finds that the target node DST is not in the new branch set K , what forwarding strategy will the system adopt? |
|
ICACT20260346 Question.15
Questioner: hejf@dsp.ac.cn 2026-02-09 오후 10:46:52 |
ICACT20260346 Answer.15
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 10:46:52
Chrome Click!! |
| How does the proposed fast reroute scheme guarantee loop-free recovery while relying only on localized (neighbor) information in an underwater ad hoc network? |
|
ICACT20260346 Question.16
Questioner: md02anas02@gmail.com 2026-02-10 오전 3:47:33 |
ICACT20260346 Answer.16
Answer by Auhor tanghr@dsp.ac.cn 2026-02-10 오전 3:47:33
Chrome Click!! |
| Which specific version of NS-3 was used to simulate the underwater acoustic channel models? |
|
ICACT20260346 Question.17
Questioner: hominpark@etri.re.kr 2026-02-10 오전 11:15:13 |
ICACT20260346 Answer.17
Answer by Auhor tanghr@dsp.ac.cn 2026-02-10 오전 11:15:13
Chrome Click!! |
| When handling multi-failures as recursive local repairs, given the high underwater propagation latency, isnt there a risk of TTL expiration or excessive accumulated delay before the packet reaches its destination? |
|
ICACT20260346 Question.2
Questioner: yangyn@dsp.ac.cn 2026-02-09 오후 12:03:29 |
ICACT20260346 Answer.2
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 12:03:29
Chrome Click!! |
| What criteria guided the selection of simulation parameters in NS-3 (e.g., 49-node grid topology and 4 kbps bandwidth), and how do they reflect real-world underwater conditions? |
The NS-3 simulation parameters (49-node grid topology, 1000m spacing, 1100m transmission range, 4 kbps bandwidth) were selected to reflect realistic underwater acoustic communication constraints. These parameters represent typical array-based UANET deployments where nodes move slowly, bandwidth is limited to a few kbps, and propagation speed is approximately 1500 m/s, creating a simulation environment that closely mirrors real-world underwater communication challenges while maintaining manageable computational complexity. |
ICACT20260346 Question.3
Questioner: yangyn@dsp.ac.cn 2026-02-09 오후 12:03:40 |
ICACT20260346 Answer.3
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 12:03:40
Chrome Click!! |
| What are the most promising extensions for UW-LFRR, such as integrating machine learning for failure prediction or adapting to partially connected networks? |
Promising extensions for UW-LFRR include integrating machine learning for failure prediction to anticipate node failures before they occur, potentially reducing reactive repairs, and adapting the framework for partially connected networks by incorporating more sophisticated failure impact modeling. Additionally, incorporating energy-aware routing metrics could further optimize performance for energy-constrained underwater nodes, and extending the approach to handle dynamic topologies with higher mobility could broaden its applicability to more challenging underwater scenarios. |
ICACT20260346 Question.4
Questioner: pangb@dsp.ac.cn 2026-02-09 오후 12:03:49 |
ICACT20260346 Answer.4
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 12:03:49
Chrome Click!! |
| I noticed that the paper talks about the data plane and control plane. Does this mean the proposed system is based on SDN? If so, how does an SDN network ensure timely state feedback and flow table updates in an underwater acoustic environment? |
No, UW-LFRR is not based on SDN. Its a data-plane-driven mechanism that operates within the existing routing framework without requiring a centralized SDN controller. The system avoids network-wide flooding by using local information and only installing temporary forwarding entries for affected flows, which is crucial for underwater environments with slow propagation and high latency, thus bypassing the SDN-related challenges of timely state feedback in underwater acoustic networks. |
ICACT20260346 Question.5
Questioner: pangb@dsp.ac.cn 2026-02-09 오후 12:03:58 |
ICACT20260346 Answer.5
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 12:03:58
Chrome Click!! |
| In the overview, you mention constructing a purified local view. Who is responsible for building this? Also, during the presentation, it was stated that before a failure occurs, each node maintains a network-wide topology snapshot. How is this achieved in practice? |
Each node is responsible for building its own purified local view. Before failures, each node maintains a network-wide topology snapshot Gsnap by collecting link-state information from neighbors through periodic control messages (as in typical link-state routing protocols), which is standard practice in routing protocols like OLSR. This snapshot is kept updated within the routing convergence period, and when a failure occurs, the node uses its local view to identify the affected region (Ku) and construct the purified topology Gu = Gsnap \ Forbiddenu, where Forbiddenu includes failed nodes and their branch-off sets. |
ICACT20260346 Question.6
Questioner: xueyy@dsp.ac.cn 2026-02-09 오후 12:04:06 |
ICACT20260346 Answer.6
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 12:04:06
Chrome Click!! |
| In UW-LFRR, how can routing loops be avoided during the failure recovery process? |
UW-LFRR avoids routing loops by constructing a purified local topology view that excludes failed nodes and their branch-off sets (Forbiddenu), ensuring all logical paths consist only of healthy, reachable nodes. The two-stage relay selection strategy prioritizes grafting traffic back to the original backbone when possible (Phase I), which inherently avoids loops, and Phase II searches within the branch-off region using a shortest-path search on the purified topology, which guarantees loop-free paths. Additionally, the first-packet paving mechanism ensures temporary forwarding entries are only installed where necessary. |
ICACT20260346 Question.7
Questioner: xueyy@dsp.ac.cn 2026-02-09 오후 12:04:16 |
ICACT20260346 Answer.7
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 12:04:16
Chrome Click!! |
| How does the “first-packet path paving” mechanism work, and what advantages does it have over traditional signaling-based approaches? |
The first-packet path paving mechanism works by having the first failure-affected packet carry a compact control triplet ⟨DST, BR, FN⟩ that installs temporary forwarding entries along the detour path. Subsequent packets of the same flow then follow the repaired route without additional control headers. This approach eliminates network-wide flooding (reducing control overhead), installs temporary state only where necessary (limiting state spread), and avoids per-packet computation overhead by installing the path once rather than recomputing it for every packet. |
ICACT20260346 Question.8
Questioner: xueyy@dsp.ac.cn 2026-02-09 오후 12:04:24 |
ICACT20260346 Answer.8
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 12:04:24
Chrome Click!! |
| In UW-LFRR’s two-stage relay selection strategy, what scenarios are the first stage and the second stage respectively suitable for? |
Phase I (backbone-oriented fast search) is suitable for most failure scenarios where the backbone is still partially available, i.e., when there are usable backbone nodes within the branch-off region (Ku) that can serve as grafting points. This phase achieves a bounded path stretch (at most 2 extra hops) with low computational complexity. Phase II (branch-off fallback search) is triggered only when Phase I fails (i.e., when the backbone is completely disrupted within Ku, making Cmain = ∅), and it searches for a short detour over all of Ku to ensure connectivity when the backbone is fully cut. |
ICACT20260346 Question.9
Questioner: wangsy@dsp.ac.cn 2026-02-09 오후 12:04:31 |
ICACT20260346 Answer.9
Answer by Auhor tanghr@dsp.ac.cn 2026-02-09 오후 12:04:31
Chrome Click!! |
| UW-LFRR relies on "purified local topology view", but underwater acoustic links have large fluctuations and neighbor information is prone to expire: What mechanism do you use to ensure that the local topology view is sufficiently up-to-date? If the neighbor table/link quality estimation lags behind, how much impact will it have on the roundabout selection and loop avoidance? |
UW-LFRR relies on the quasi-static nature of array-/formation-based UANETs, where nodes move slowly with missions or currents, so the relative topology remains almost unchanged within one convergence period. Each node maintains its topology snapshot Gsnap through periodic link-state updates within the routing convergence period. The impact of outdated neighbor information is mitigated by the failure detection based on missing ACKs or failure information in packet headers, which immediately triggers the local repair process, and the branch-off region (Ku) is defined based on the current failure set rather than the entire network. |
