Pith. sign in

REVIEW 2 major objections 5 minor 47 references

Energy Efficient Routing and Network Coding in Core Networks

T0 review · 2 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read In non-bypass IP-over-WDM core networks, XOR-coding opposite-direction flows at intermediate nodes cuts total network power by up to 33 percent, and the savings follow a closed-form expression in average hop count.

desk verdict Useful closed-form result, but the 27–33% savings ride on an unmeasured port-power premium and need hardware data before being taken at face value. read the letter →

arxiv 1908.07903 v1 pith:RUSMLAYR submitted 2019-08-21 cs.NI

Mohamed Musa , Taisir Elgorashi , Jaafar Elmirghani This is my paper · ORCID
classification cs.NI
keywords networkcodingenergyefficiencyIPoverWDMcorenetworksmixedintegerlinearprogrammingminimum-hoproutingzeropaddingtrafficpartitioning
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to show that a small hardware change in core-network routers can deliver large energy savings: at an intermediate node, two conventional ports handling opposite-direction traffic can be replaced by one network-coding port that XOR-combines the two flows and multicasts the encoded flow back to both endpoints. Using a mixed-integer linear program and a minimum-hop heuristic, the paper reports daily-average power savings of 27 percent on the NSFNET and 33 percent on the USNET under non-bypass IP-over-WDM routing, with similar benefits for ring and line topologies and none for full mesh. The paper also derives closed-form bounds, centered on the formula $\varphi = (0.45h - 1)/h$ for zero-padding coding at port power ratio $r = 1.1$, showing savings grow with average hop count and saturate near 45 percent for ring/line networks and 22.5 percent for stars. If true, the result means ordinary unicast backbone traffic can be made noticeably cheaper to carry without new fibers or wavelengths, using only router-port redesign and a routing policy (minimum hops) that operators already understand.

What carries the argument

The load-bearing object is the network-coding-enabled port: a conventional port augmented with a second receiver, a small storage/buffer for synchronization, an XOR unit, a splitter/coupler to multicast the encoded flow, and an amplifier. It replaces two conventional ports at an intermediate node on a bidirectional flow. The argument is carried by a MILP whose objective sums router-port, transponder, switch, mux/demux, and EDFA power, with linearized constraints that identify coding opportunities and count NC ports by the maximum of the two opposite flows. The closed-form analysis reduces the whole system to the ratio $r$ and the average hop count $h$, producing $\varphi = (0.45h - 1)/h$ for the paper's $r=1.1$ and asymptotes of 45% (ring, line) and 22.5% (star) as network size grows.

What would settle it

Measure the end-to-end power draw of the proposed network-coding port (two receivers, storage, XOR unit, coupler, amplifier) against a conventional port at the same line rate; if the coding-enabled port draws 1.6 kW or more while the conventional port draws 1 kW, the zero-padding savings disappear, and if the bypass coded transponder draws more than about 660 W at 400 Gbps, the bypass savings disappear.

Watch

Extended reading notes

Core claim

The central discovery is that XOR network coding, normally associated with multicast or wireless broadcast, pays off in wired unicast core networks because opposite-direction flows that pass through the same intermediate node can be merged at the IP layer. In the non-bypass architecture the paper proposes, a single NC port at the intermediate node replaces two conventional router ports: it receives both flows, synchronizes and stores them, XORs them, and transmits the encoded packet back to both neighbors, who decode using their own stored copy. The MILP model optimizes routes, coding locations, and port counts, and the paper reports that this substitution reduces daily-average network power by 27% on the NSFNET and 33% on the USNET against conventional non-bypass routing with the same traffic matrices. The savings are governed by the ratio $r$ of NC-port-plus-transponder power to conventional-port-plus-transponder power; with $r = 1.1$ the closed form becomes $\varphi = (0.45h - 1)/h$, and savings persist as long as $r < 2$ for zero padding and beyond for packet partitioning. The paper claims the minimum-hop heuristic reproduces the MILP savings closely, so the benefit does not require solving the NP-hard optimization.

Load-bearing premise

The entire non-bypass saving figure rests on the unmeasured assumption that a network-coding-enabled port draws only 10 percent more power than a conventional port (1.1 kW vs 1 kW); the paper itself shows that if that premium reaches about 60 to 100 percent, the zero-padding savings vanish.

Editorial extensions

If this is right

  • Network operators could retrofit XOR coding at the IP layer of existing non-bypass routers and reap 27-33% daily-average power savings without changing wavelengths, fibers, or traffic demands.
  • Savings scale with average hop count, so ring, line, and other high-hop topologies benefit most, while full-mesh and star gains are small or zero.
  • The minimum-hop heuristic captures most of the MILP-optimal savings, meaning the benefit is attainable without solving an NP-hard routing optimization at run time.
  • The closed form $\varphi = (0.45h - 1)/h$ gives a design-time rule of thumb: estimate average hop count and port power ratio to predict whether coding pays.
  • Packet partitioning extends the benefit to asymmetric bidirectional traffic and survives a higher port power premium than zero padding.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper evaluates only core topologies; the same coding mechanism would plausibly transfer to metro or data-center fabrics with high hop counts, though that extension is not tested here.
  • The closed form gives planners a cheap metric: because savings rise with average hop count, topology changes that lengthen paths increase coding's payoff, at the cost of added latency.
  • The decisive open question is the real power draw of an NC port; a component-level measurement of the second receiver, buffer, XOR, coupler, and amplifier would confirm or overturn the 27-33% figures.
  • The model assumes a proportional power profile for partially used ports; relaxing that to discrete port activation could shift the break-even port ratio and is a natural next model.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 5 minor

Summary. This paper proposes a network-coding-enabled IP port architecture for core IP-over-WDM networks, in which opposite-direction unicast flows are XOR-coded at intermediate nodes, replacing two conventional ports with one NC port. The authors formulate a MILP for non-bypass and bypass routing, propose a minimum-hop heuristic, and derive closed-form savings expressions as a function of the port power ratio r and the average hop count h. They report daily-average power savings of 27% and 33% for NSFNET and USNET in the non-bypass case, 28% for the bypass case, and verify the MILP results with analytic bounds and closed forms for regular topologies.

Significance. If the hardware power assumptions were adequately supported, the paper would make a useful contribution: it extends network coding from wireless multicast settings to wired unicast core networks, provides a detailed MILP formulation, and reduces the savings to an explicit closed-form expression in r and h that is checked against the MILP and heuristic. The derivation in Section VII is not fitted to the MILP output, and the paper includes explicit sensitivity analyses (Figs. 16 and 18), which are strengths because they show where the claimed benefits disappear. The optimization model and the analytic derivation are internally consistent modulo typographical errors. The main limitation is that the headline quantitative savings depend on an unmeasured 10% premium for NC ports in the non-bypass case and on a 360 W coded-transponder assumption in the bypass case; without supporting evidence or a fully parameterized presentation, the results are conditional rather than definitive.

major comments (2)
  1. [Section IV / Table IV / Eq. (49) / Fig. 16] The central non-bypass savings figures (27% for NSFNET and 33% for USNET) and the closed-form expression φ = (h−1)(1−r/2)/h with r ≈ 1.1 rest entirely on the 'moderate estimate' that an NC port consumes 1.1 kW instead of 1 kW. No component-level power model or measurement is given for the extra receiver, coupler/splitter, amplifier, XOR logic, and synchronization buffer shown in Fig. 3, and Table V makes the same kind of assumption for the 360 W coded transponder in the bypass case. Since Fig. 16 shows that zero-padding savings vanish at r ≈ 1.6, the quantitative headline is directly determined by the least-supported parameter in the paper. The authors should either add a bill-of-materials or datasheet-based model for Px and for the coded transponder, or present the headline results as explicit functions of r with a justified uncertainty range, and adjust the abstract and conclusions accordingly.
  2. [Section VI / Fig. 17 / Abstract] The abstract states '28% power savings' for the bypass case without specifying the baseline. The text clarifies that this 28% is measured against the non-bypass architecture; against the conventional bypass architecture the saving is 18% at the same 10% coded-transponder premium. Using the higher-power non-bypass baseline in the headline overstates the benefit of coding in the bypass case. Please report the conventional-bypass comparison as the primary baseline or explicitly qualify the abstract's bypass-savings claim.
minor comments (5)
  1. [Eq. (49)] Equation (49) as printed reads φ = (0.45h − 1)/h, but the derivation and Eqs. (50)–(57) use φ = 0.45(h − 1)/h. This is not a notational nit: for the star topology with h = 2 the printed form gives a negative saving, while the intended formula gives 22.5%.
  2. [Table V] The coded transponder power in Table V is listed as 360 kW; it should be 360 W, otherwise it exceeds the total network power reported in Fig. 17 by orders of magnitude.
  3. [Eq. (26)] In Section V, Equation (26) defines Y^m_nk = (w^m_nk − w^m_nk)/B as written, which is identically zero. The intended expression is (w^m_nk − w^m_kn)/B, the difference of the two opposite-direction flows through node m.
  4. [Section V, after Eq. (29)] The paragraph after Eq. (29) says 'the first term of equation (29) is used to calculate the number of conventional ports in the zero padding case.' This is confusing because Eq. (29) is the partitioning-model port count; please rephrase to distinguish the zero-padding total-port formula in Eq. (13) from the partitioning formula in Eq. (29).
  5. [Abstract and Section VII.B] The abstract says the savings are 'proportional to the average hop count,' but Eq. (49) gives a concave, saturating function 0.45(h − 1)/h; 'increasing with' or 'dependent on' would be more accurate.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the closed-form savings expression is derived analytically from declared power parameters; the unmeasured r=1.1 NC-port premium is an assumption that affects the magnitude of the result, not a circular input.

full rationale

The paper's central quantitative claim—daily average savings of 27% on NSFNET and 33% on USNET—follows from a MILP model and from the closed-form expression φ = (0.45h−1)/h (Eq. 49, properly read as 0.45(h−1)/h). This expression is obtained by algebraic manipulation of the cost accounting in Eqs. (39)–(45), with the ratio r = (p_t+p_x)/(p_t+p_p) introduced as a declared parameter and then set to approximately 1.1 using the component power values in Table IV. Nothing in the derivation takes the target savings percentages as inputs, and the MILP is an optimization over routing and port counts, not a curve fit to a predetermined savings value. The agreement between the analytic formula, the MILP, and the minimum-hop heuristic is therefore an internal consistency check among three models that share the same stated power assumptions; it does not independently validate the assumed 1.1 kW NC-port power, but it is not circular. The paper's self-citations, mainly to prior works [28], [29], [36], and [37], describe earlier related frameworks and methods, but the model, constraints, and analytic derivations in this paper are presented in full and do not rely on an unverified self-cited result as the load-bearing step. The one substantive concern, as the paper itself acknowledges through its sensitivity analysis (Figs. 16 and 18), is that the headline savings are sensitive to the unmeasured assumption that an NC port consumes only 10% more power than a conventional port (r ≈ 1.1) and that a coded bypass transponder consumes 360 W. Those are empirical or modeling uncertainties, not circular reasoning: the predictions are conditional on the assumptions, and the paper makes the dependence explicit. Overall, no step in the claimed derivation chain reduces by construction to its own inputs, so the appropriate circularity score is 0.

Assumptions & free parameters 3 free parameters · 4 assumptions · 2 invented entities

The central claims depend on three classes of inputs: (1) assumed power premiums for the network-coding port and transponder (10% for the non-bypass case, 360 W for the bypass case), which the sensitivity analysis shows directly controls whether savings exist; (2) modeling choices such as relaxed port-count ceilings and proportional power consumption for partial wavelengths, which tend to make the conventional baseline more expensive and the network-coding case look better; and (3) a traffic model with unpublished random matrices. None of these are hidden, they are stated in Sections IV and VII, but they are all load-bearing. The delivery of this arXiv version contains no shipped code, data, or measurement, only re-implementable descriptions.

free parameters (3)
  • NC port power premium (r = 1.1) = 1.1 kW NC port vs 1 kW conventional port
    The claimed savings hinge on this 10% premium. The paper calls it a moderate estimate (Section IV) but provides no measurement or detailed bill of materials. Section V later shows savings vanish for zero padding when r approaches 2.
  • Coded transponder power (bypass case) = 360 W (also 500 W in one scenario)
    For the bypass scenario, the savings are computed for a coded transponder power of 360 W, an assumed value matching the GreenTouch 2020 projection style. The sensitivity analysis in Fig. 18 spans 330 to 660 W but no component-level justification is given.
  • Traffic model averages per time zone = 20 to 120 Gbps average per node pair, with demands 10 to 230 Gbps
    The daily traffic profile in Fig. 7 with a 22:00 peak is taken from prior work, but the random traffic matrices are generated in an unspecified way (no seed published). The savings averages depend on this traffic distribution.
assumptions (4)
  • domain assumption The MILP relaxation of port counts (Equations 11-12 use non-integer values instead of ceilings) does not materially distort the savings.
    The paper states the ceilings are relaxed due to relaxation of the MILP model (Section III). Underestimating port counts can overstate savings, and no rounding sensitivity analysis is provided.
  • domain assumption Proportional power consumption for partially used wavelengths and ports, achievable by grooming.
    Section IV states partially used wavelengths consume part of a router port and transponder proportional to traffic. This assumption is favorable to the savings calculation and is acknowledged by the authors.
  • domain assumption Control and management overhead, and the power of the additional storage/XOR processing at end nodes, are negligible.
    Section II states the control overhead is negligible and that storage can be part of existing processing power. The closed-form formulas (Equations 39-49) include only router ports and transponders, omitting EDFAs, switches, and mux/demux, while the MILP includes them, which is why analytic savings are higher than MILP savings (the paper notes this in Section VII).
  • standard math The non-bypass network is fully configured with one IP router port per node pair adjacency, as implied by Equation (14) with Ym summing over neighbours.
    The conventional port count model is the standard way to count router line cards in non-bypass IP-over-WDM, and is a reasonable modeling choice.
invented entities (2)
  • NC enabled port (IP layer)
    purpose: Combines two incoming receiver paths with an XOR unit and one transmitter with a coupler, replacing two conventional ports at intermediate nodes.
    The paper proposes this architecture (Fig. 3) and assumes a 1.1 kW power consumption, but provides no measured prototype, no detailed component power breakdown, and no external benchmark. It is a design concept rather than an experimentally validated entity.
  • Coded transponder (bypass/optical layer)
    purpose: Performs XOR coding in the optical layer at intermediate nodes for the bypass scenario.
    The power consumption of 360 W at 400 Gbps is assumed (Table V) without measurement or detailed justification, and the authors state the saving depends on this value.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Energy Efficient Routing and Network Coding in Core Networks." pith.science (2026). https://pith.science/paper/RUSMLAYR

@misc{pith2026190807903,
  author       = {Pith},
  title        = {Pith review of: Energy Efficient Routing and Network Coding in Core Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RUSMLAYR}},
  note         = {Machine review of arXiv:1908.07903}
}
read the original abstract

We propose network coding as an energy efficient data transmission technique in core networks with non-bypass and bypass routing approaches. The improvement in energy efficiency is achieved through reduction in the traffic flows passing through intermediate nodes. A mixed integer linear program (MILP) is developed to optimize the use of network resources, and the results show that our proposed network coding approach introduces up to 33% power savings for the non-bypass case compared with the conventional architectures. For the bypass case, 28% power savings are obtained considering futuristic network components power consumption. A heuristic based on the minimum hop count routing shows power savings comparable to the MILP results. Furthermore, we study how the change in network topology affects the savings produced by network coding. The results show that the savings are proportional to the average hop count of the network topology. We also derive power consumption analytic bounds and closed form expressions for networks that implement network coding and thus also verify the results obtained by the MILP model.

Figures

Figures reproduced from arXiv: 1908.07903 by the authors.

Figure 2
Figure 2. Conventional architecture (a), and Network coding [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Network coding node architecture t=1 t=2 t=3 t=4 t=5 t=6 n1 n2 n3 n4 n1 n2 n3 n4 n1 n2 n3 n4 n1 n2 n3 n4 n1 n2 n3 n4 n1 n2 n3 n4 1 3 2 3 3 2 A B C 1 A B C 2 A 1 B C 3 B+1 A+2 C A A+2 B+1 1 A+2+C 3+B+1 B 3+B+1 A+2+C 2 C A+2+C 3+B+1 3 [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. Network coding at two intermediate nodes [PITH_FULL_IMAGE:figures/full_fig_p003_4.png] view at source ↗
Figures from the paper (17 more)
Figure 6
Figure 6. Figure 6: The USNET topology 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 0 20 40 60 80 100 120 Time of the day Average Traffic Demand Per Node Pair (Gbps) EST area CST area MST area PST area [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 5
Figure 5. Figure 5: The NSFNET topology We model traffic demands by using the average traffic demands of the network as shown in [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 7
Figure 7. Figure 7: Average traffic demands at different times of the day[4] [PITH_FULL_IMAGE:figures/full_fig_p006_7.png]
Figure 8
Figure 8. Figure 8: The NSFNET network power consumption with and [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]
Figure 9
Figure 9. Figure 9: The USNET network power consumption with and [PITH_FULL_IMAGE:figures/full_fig_p007_9.png]
Figure 10
Figure 10. Figure 10: Comparison between the number of the two port types [PITH_FULL_IMAGE:figures/full_fig_p008_10.png]
Figure 12
Figure 12. Figure 12: Ring and Star topologies power consumption with and [PITH_FULL_IMAGE:figures/full_fig_p008_12.png]
Figure 13
Figure 13. Figure 13: Maximum, minimum and average daily power savings [PITH_FULL_IMAGE:figures/full_fig_p009_13.png]
Figure 14
Figure 14. Figure 14: Approaches to ports calculation the variable Y m nk to represent the difference between the two packet sizes calculated as Y m nk = w m nk − w m nk B , (26) and changing the variable Xm nk to represent the minimum of the opposite flows, this will mean constraint (18) …
Figure 15
Figure 15. Figure 15: Power consumption of the NSFNET under NC with [PITH_FULL_IMAGE:figures/full_fig_p010_15.png]
Figure 16
Figure 16. Figure 16: Power consumption versus ports ratio using the MILP [PITH_FULL_IMAGE:figures/full_fig_p010_16.png]
Figure 17
Figure 17. Figure 17: Power consumption of the NSFNET with bypass at [PITH_FULL_IMAGE:figures/full_fig_p011_17.png]
Figure 18
Figure 18. Figure 18: Power consumption of the network for different coded [PITH_FULL_IMAGE:figures/full_fig_p011_18.png]
Figure 20
Figure 20. Figure 20: Effect of network coding on large network sizes [PITH_FULL_IMAGE:figures/full_fig_p013_20.png]
Figure 19
Figure 19. Figure 19: Power consumption of the analytical and MILP models [PITH_FULL_IMAGE:figures/full_fig_p013_19.png]
Figure 21
Figure 21. Figure 21: MILP results, the upper and lower bounds of the zero [PITH_FULL_IMAGE:figures/full_fig_p016_21.png]
Figure 22
Figure 22. Figure 22: MILP results, the upper and lower bounds of the [PITH_FULL_IMAGE:figures/full_fig_p017_22.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 47 canonical work pages

  1. [1]

    GeSI SMARTer 2030:

    GeSI, “GeSI SMARTer 2030: ”ICT Solutions for 21st Century Chal- lenges”,” 2015

    work page 2015

  2. [2]

    Greening of the internet,

    M. Gupta and S. Singh, “Greening of the internet,” Karlsruhe, Germany, pp. 19–26, 2003

    work page 2003

  3. [3]

    Energy Efficiency in Telecom Optical Networks,

    Y . Zhang, P. Chowdhury, M. Tornatore, and B. Mukherjee, “Energy Efficiency in Telecom Optical Networks,” vol. 12, no. 4, pp. 441–458, 2010

    work page 2010

  4. [4]

    Ip over wdm networks employing renewable energy sources,

    X. Dong, T. El-Gorashi, and J. M. Elmirghani, “Ip over wdm networks employing renewable energy sources,” Lightwave Technology, Journal of, vol. 29, no. 1, pp. 3–14, 2011. 18

    work page 2011

  5. [5]

    Green IP Over WDM Networks With Data Centers,

    X. Dong, T. El-Gorashi, and J. M. H. Elmirghani, “Green IP Over WDM Networks With Data Centers,” Journal of Lightwave Technology, vol. 29, no. 12, pp. 1861–1880, Jun. 2011. [Online]. Available: http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=5756625

    work page 2011

  6. [6]

    Future energy efficient data centers with disaggregated servers,

    H. M. M. Ali, T. E. El-Gorashi, A. Q. Lawey, and J. M. Elmirghani, “Future energy efficient data centers with disaggregated servers,” Jour- nal of Lightwave Technology , vol. 35, no. 24, pp. 5361–5380, 2017

    work page 2017

  7. [7]

    On the Energy Ef fi ciency of Physical Topology Design for IP Over WDM Networks,

    X. Dong, T. E. H. El-gorashi, and J. M. H. Elmirghani, “On the Energy Ef fi ciency of Physical Topology Design for IP Over WDM Networks,” vol. 30, no. 12, pp. 1931–1942, 2012

    work page 1931

  8. [8]

    Distributed energy effi- cient clouds over core networks,

    A. Lawey, T. El-Gorashi, and J. Elmirghani, “Distributed energy effi- cient clouds over core networks,” Lightwave Technology, Journal of , vol. 32, no. 7, pp. 1261–1281, April 2014

    work page 2014

Show all 47 references
  1. [9]

    Energy-efficient future high-definition tv,

    N. Osman, T. El-Gorashi, L. Krug, and J. Elmirghani, “Energy-efficient future high-definition tv,” Lightwave Technology, Journal of , vol. 32, no. 13, pp. 2364–2381, July 2014

    2014

  2. [10]

    Bittorrent content distri- bution in optical networks,

    A. Lawey, T. El-Gorashi, and J. Elmirghani, “Bittorrent content distri- bution in optical networks,” Lightwave Technology, Journal of, vol. 32, no. 21, pp. 4209–4225, Nov 2014

    2014

  3. [11]

    Energy efficient iot virtualization framework with peer to peer net- working and processing,

    Z. T. Al-Azez, A. Q. Lawey, T. E. El-Gorashi, and J. M. Elmirghani, “Energy efficient iot virtualization framework with peer to peer net- working and processing,” IEEE Access, vol. 7, pp. 50 697–50 709, 2019

    2019

  4. [12]

    Patient-centric cellular networks optimization using big data analytics,

    M. S. Hadi, A. Q. Lawey, T. E. El-Gorashi, and J. M. Elmirghani, “Patient-centric cellular networks optimization using big data analytics,” IEEE Access, vol. 7, pp. 49 279–49 296, 2019

    2019

  5. [13]

    Big data analytics for wireless and wired network design: A survey,

    M. S. Hadi, A. Q. Lawey, T. E. El-Gorashi, and J. M. Elmirghani, “Big data analytics for wireless and wired network design: A survey,” Computer Networks, vol. 132, pp. 180–199, 2018

    2018

  6. [14]

    Energy efficient big data networks: Impact of volume and variety,

    A. M. Al-Salim, A. Q. Lawey, T. E. El-Gorashi, and J. M. Elmirghani, “Energy efficient big data networks: Impact of volume and variety,” IEEE Transactions on Network and Service Management, vol. 15, no. 1, pp. 458–474, 2017

    2017

  7. [15]

    Energy efficient virtual network embedding for cloud networks,

    L. Nonde, T. El-Gorashi, and J. Elmirghani, “Energy efficient virtual network embedding for cloud networks,” Lightwave Technology, Jour- nal of, vol. 33, no. 9, pp. 1828–1849, May 2015

    2015

  8. [16]

    Optimized energy aware 5g network function virtualization,

    A. N. Al-Quzweeni, A. Q. Lawey, T. E. Elgorashi, and J. M. Elmirghani, “Optimized energy aware 5g network function virtualization,” IEEE Access, vol. 7, pp. 44 939–44 958, 2019

    2019

  9. [17]

    Greentouch greenmeter core network energy- efficiency improvement measures and optimization,

    J. Elmirghani, T. Klein, K. Hinton, L. Nonde, A. Lawey, T. El-Gorashi, M. Musa, and X. Dong, “Greentouch greenmeter core network energy- efficiency improvement measures and optimization,” IEEE/OSA Journal of Optical Communications and Networking , vol. 10, no. 2, pp. A250– A269, 2018

    2018

  10. [18]

    Bounds on greentouch greenmeter network energy efficiency,

    M. O. Musa, T. E. El-Gorashi, and J. M. Elmirghani, “Bounds on greentouch greenmeter network energy efficiency,” Journal of Light- wave Technology, vol. 36, no. 23, pp. 5395–5405, 2018

    2018

  11. [19]

    Network information flow,

    R. Ahlswede, C. Ning, S. Y . R. Li, and R. W. Yeung, “Network information flow,” Information Theory, IEEE Transactions on , vol. 46, no. 4, pp. 1204–1216, 2000

    2000

  12. [20]

    Fragouli and E

    C. Fragouli and E. Soljanin, Network coding applications . Now Publishers Inc, 2007

    2007

  13. [21]

    Network coding in passive optical networks,

    M. Belzner and H. Haunstein, “Network coding in passive optical networks,” in Optical Communication, 2009. ECOC ’09. 35th European Conference on, 2009, pp. 1–2

    2009

  14. [22]

    Network coding in next- generation passive optical networks,

    K. Fouli, M. Maier, and M. Medard, “Network coding in next- generation passive optical networks,”Communications Magazine, IEEE, vol. 49, no. 9, pp. 38–46, 2011

    2011

  15. [23]

    Network- Coding-Based Energy Management for Next-Generation Passive Optical Networks,

    X. Liu, K. Fouli, R. Kang, M. Maier, and S. Member, “Network- Coding-Based Energy Management for Next-Generation Passive Optical Networks,” vol. 30, no. 6, pp. 864–875, 2012

    2012

  16. [24]

    Network Protection Codes Against Link Failures Using Network Coding,

    S. A. Aly and A. E. Kamal, “Network Protection Codes Against Link Failures Using Network Coding,” in Global Telecommunications Conference, 2008. IEEE GLOBECOM 2008. IEEE , 2008, pp. 1–6

    2008

  17. [25]

    Overlay Protection Against Link Failures Using Network Coding,

    A. E. Kamal, A. Ramamoorthy, L. Long, and L. Shizheng, “Overlay Protection Against Link Failures Using Network Coding,” Networking, IEEE/ACM Transactions on, vol. 19, no. 4, pp. 1071–1084, 2011

    2011

  18. [26]

    Optimal dedicated protection approach to shared risk link group failures using network coding,

    P. Babarczi, J. Tapolcai, P.-H. Ho, and M. M ´edard, “Optimal dedicated protection approach to shared risk link group failures using network coding,” in Communications (ICC), 2012 IEEE International Confer- ence on. IEEE, 2012, pp. 3051–3055

    2012

  19. [27]

    Linear network coding and parallel transmission increase fault tolerance and optical reach,

    X. Chen, A. Engelmann, A. Jukan, and M. M ´edard, “Linear network coding and parallel transmission increase fault tolerance and optical reach,” IEEE/OSA Journal of Optical Communications and Networking, vol. 9, no. 4, pp. 244–256, 2017

    2017

  20. [28]

    Bounds for energy-efficient survivable ip over wdm networks with network coding,

    M. Musa, T. Elgorashi, and J. Elmirghani, “Bounds for energy-efficient survivable ip over wdm networks with network coding,” IEEE/OSA Journal of Optical Communications and Networking , vol. 10, no. 5, pp. 471–481, 2018

    2018

  21. [29]

    Energy efficient survivable ip over wdm networks with network coding,

    M. Musa, T. Elgorashi, and J. Elmirghani, “Energy efficient survivable ip over wdm networks with network coding,” IEEE/OSA Journal of Optical Communications and Networking , vol. 9, no. 3, pp. 207–217, 2017

    2017

  22. [30]

    Network coding for optical-layer multicast,

    E. D. Manley, J. S. Deogun, and X. Lisong, “Network coding for optical-layer multicast,” in Broadband Communications, Networks and Systems, 2008. BROADNETS 2008. 5th International Conference on , 2008, pp. 452–459

    2008

  23. [31]

    All-Optical Network Coding,

    E. D. Manley, J. S. Deogun, L. Xu, and D. R. Alexander, “All-Optical Network Coding,”Optical Communications and Networking, IEEE/OSA Journal of, vol. 2, no. 4, pp. 175–191, 2010

    2010

  24. [32]

    Optical Physical-Layer Network Coding,

    Z. Liu, M. Li, L. Lu, C.-K. Chan, S.-C. Liew, and L.-K. Chen, “Optical Physical-Layer Network Coding,” IEEE Photonics Technology Letters, vol. 24, no. 16, pp. 1424–1427, Aug. 2012. [Online]. Available: http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=6226828

    2012

  25. [33]

    Ge- netic representations for evolutionary minimization of network coding resources,

    M. Kim, V . Aggarwal, U.-M. OReilly, M. M ´edard, and W. Kim, “Ge- netic representations for evolutionary minimization of network coding resources,” Applications of Evolutionary Computing , pp. 21–31, 2007

    2007

  26. [34]

    Evolutionary computation with spatial receding horizon control to minimize network coding resources,

    X.-B. Hu and M. S. Leeson, “Evolutionary computation with spatial receding horizon control to minimize network coding resources,” The Scientific World Journal, vol. 2014, 2014

    2014

  27. [35]

    Optimization-based linear network coding for general connections of continuous flows,

    Y . Cui, M. M ´edard, E. Yeh, D. Leith, and K. Duffy, “Optimization-based linear network coding for general connections of continuous flows,” in Communications (ICC), 2015 IEEE International Conference on . IEEE, 2015, pp. 4492–4498

    2015

  28. [36]

    Energy Efficient Core Networks Using Network Coding,

    M. Musa, T. El-Gorashi, and J. Elmirghani, “Energy Efficient Core Networks Using Network Coding,” in Transparent Optical Networks (ICTON), 2015 17th International Conference on , July 2015

    2015

  29. [37]

    Network coding for energy efficiency in bypass ip/wdm networks,

    M. Musa, T. E. El-Gorashi, and J. M. Elmirghani, “Network coding for energy efficiency in bypass ip/wdm networks,” in 2016 18th Interna- tional Conference on Transparent Optical Networks (ICTON) . IEEE, 2016

    2016

  30. [38]

    XORs in the air: practical wireless network coding,

    S. Katti, H. Rahul, W. Hu, D. Katabi, M. Muriel, #233, Dard, and J. Crowcroft, “XORs in the air: practical wireless network coding,” SIGCOMM Comput. Commun. Rev., vol. 36, no. 4, pp. 243–254, 2006

    2006

  31. [39]

    Cisco crs-1 16-slot single-shelf system

    Cisco, “Cisco crs-1 16-slot single-shelf system.” [Online]. Avail- able: http://www.cisco.com/c/en/us/products/collateral/routers/crs-1-16- slot-single-shelf-system/product˙data˙sheet09186a008022d5f3.html

  32. [40]

    Cisco ONS 15454 10-Gbps Mul- tirate Transponder Card

    Cisco, “Cisco ONS 15454 10-Gbps Mul- tirate Transponder Card.” [Online]. Available: http://www.cisco.com/c/en/us/products/collateral/optical-networking/ ons-15454-m12-multiservice-transport-platform- mstp/product˙data˙sheet0900aecd80121bf7.html

  33. [41]

    Data sheet of Glimmerglass Intelligent Optical System 500

    Cisco, “Data sheet of Glimmerglass Intelligent Optical System 500.” [Online]. Available: http://www.glimmerglass.com/products/intelligent- optical-system-500

  34. [42]

    Data sheet of Cisco ONS 15454 100- GHz 4-CH Multi/Demultiplexer

    Cisco, “Data sheet of Cisco ONS 15454 100- GHz 4-CH Multi/Demultiplexer.” [Online]. Available: http://www.cisco.com/c/en/us/products/collateral/optical-networking/ ons-15454-series-multiservice-provisioning- platforms/product˙data˙sheet09186a00801a5572.html 19

  35. [43]

    Cisco ONS 15501 Erbium Doped Fiber Amplifier Data Sheet

    Cisco, “Cisco ONS 15501 Erbium Doped Fiber Amplifier Data Sheet.” [Online]. Available: http://www.cisco.com/en/US/products/hw/optical/ps2011/ products˙data˙sheet09186a008008870d.html

  36. [44]

    Energy-Minimized Design for IP Over WDM Networks,

    G. Shen and R. S. Tucker, “Energy-Minimized Design for IP Over WDM Networks,” Journal of Optical Communications and Networking, vol. 1, no. 1, p. 176, Jun. 2009

    2009

  37. [45]

    Greentouch final results from green meter research study,

    “Greentouch final results from green meter research study,” http://www.ourenergypolicy.org/wp-content/uploads/2015/06/ GreenTouch Green Meter Final Results 18 June 2015.pdf, accessed: 30-09-2016

    2015

  38. [46]

    Power savings versus network performance in dynamically provisioned wdm networks,

    P. Wiatr, P. Monti, and L. Wosinska, “Power savings versus network performance in dynamically provisioned wdm networks,” IEEE Com- munications Magazine, vol. 50, no. 5, 2012

    2012

  39. [47]

    Multi- hour network planning based on domination between sets of traffic matrices,

    P. Pavon-Marino, B. Garcia-Manrubia, and R. Aparicio-Pardo, “Multi- hour network planning based on domination between sets of traffic matrices,” Computer Networks, vol. 55, no. 3, pp. 665–675, 2011. Mohamed Musa received the BSc degree (first-class Honours) in Electrical and Ele...

    2011

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.