Review of Network Coding based Routing Algorithm for Wireless Mesh Networks

brrrclergymanNetworking and Communications

Jul 18, 2012 (5 years and 1 month ago)

761 views

Review of Network Coding based Routing Algorithm for Wireless Mesh
Networks

1
Xing Shao,
*2
Ru-chuan Wang
1
College of Computer, Nanjing University of Posts and Telecommunications,
Nanjing 210003, China, shawxing@gmail.com
*2
College of Computer, Nanjing University of Posts and Telecommunications,
Nanjing 210003, China, wangrc@njupt.edu.cn


Abstract
Network coding has been widely acknowledged as an effective approach to improve the throughput
of wireless mesh networks (WMNs) and a few network coding based routing algorithms for wireless
mesh networks have been proposed. This paper investigates the state of the art of network coding
based routing algorithms for WMNs that exploits network coding in routing algorithms. In addition,
the taxonomy of network coding based routing for WMNs is presented and the typical network coding
based routing algorithms for WMNs are reviewed by analyzing their specific characteristics,
advantages and drawbacks. Some open problems in designing network coding based routing for
wireless mesh networks are discussed finally.

Keywords: Network Coding, Routing Algorithm, Wireless Mesh Networks



1. Introduction

Recently, wireless mesh networks (WMNs) [1], as a kind of wireless multi-hop network, have
received increasing attention due to their attractive advantages, e.g. low cost, ease of deployment and
wide range of application scenarios, and widely accepted as a promising solution to provide wireless
broadband access.
Routing algorithm of WMNs, as one of the key technologies that influence their performance
[2][3][4], has attracted much attention of researchers. Reference [2] presents a hierarchical hybrid
routing for WMNs. The load balanced routing is investigated in [3]. The capacity problem is discussed
in oriented routing in [4]. However, the wireless links in WMNs have inherent broadcast and lossy
characteristics, and their poor quality is a great challenge. For example, 50% of the operational links in
Roofnet have loss rates higher than 30% [5]. To address this issue, some novel routing algorithms (e.g.
opportunistic routing [6], network coding based routing [7][8]) and routing metrics (e.g. ETX[9],
ETT[10], iAware[11]) have been proposed.
Network coding (NC), first proposed in [12], changes the classic store/forward mechanism in
computer network and allows nodes to mix multiple packets into one packet. Therefore, network
coding reducing the number of transmissions, could improve the throughput of network significantly
and save the bandwidth resource. Hereafter, some novel coding mechanism were proposed [13][14]
[15][16][17]. Besides, the inherent broadcast nature of wireless channel in WMNs makes network
coding particularly advantageous in terms of bandwidth efficiency and enables opportunistic encoding
and decoding. The excellent quality of network coding and the broadcast property of wireless medium
motivate researchers to combine network coding and routing of WMNs [18].
So far, a few network coding based routing algorithms for WMNs have been proposed. In this paper,
the taxonomy of network coding based routing algorithm is presented. Then the representative
principles, benefits and design challenges of these approaches are introduced. Additionally, a table is
exploited to list the characteristics of these solutions.
The rest of this paper is organized as follows. Section 2 elaborates the basic principles of network
coding. The inter-flow NC based routing algorithms[19][20][21][22][23][24] and NC based hybrid
routing [25][26][28][29][30][31] are reviewed in section 3 and 4 respectively. Section 5 presents the
comparison of typical NC based routings. Section 6 concludes this paper.
Review of Network Coding based Routing Algorithm for Wireless Mesh Networks
Xing Shao, Ru-chuan Wang
Journal of Convergence Information Technology(JCIT)
Volume6, Number12, December 2011
doi:10.4156/jcit.vol6.issue12.19
146

2. Network Coding Principles


Figure 1. Example of Network Coding in Butterfly Diagram

To explain the idea underlying network coding, the famous butterfly examples shown in Fig.1 is
used to illustrate the basic principle of network coding.
Consider the scenario in Fig.1, where node S1 wants to send packet a to both D1 and D2, and node
S2 wants to send packet b to the same two receivers. Assume each link has the packet loss ratio of zero,
and capacity of one packet per unit of time. Then if the intermediate nodes uses store/forward approach,
R1 will be the bottleneck, since it only sends a to R1 or b to R2 in sequence and the total number of
transmissions is 6. If R1 codes a and b (e.g. a⊕b) and send a⊕b to R2, D1 and D2 could obtain a and b
in every time unit. Therefore, network coding reduces the number of needed transmissions.

3. The Taxonomy of NC based Routing

Network Coding based Routing Algorithm for
Wireless Mesh Networks
Inter-Flow Network Coding
based Routing Algorithm
Network Coding based
Hybrid Routing Algorithm
COPE [19]
ROCX[20]
CAMP[21]
RCR[22]
DCAR[23]
I
2
MIX [25]
MMSR [24]
Inter-Flow
& Intra-Flow
Inter-Flow
& OR
Intra-Flow
& OR
MORE [29]
CodeOR[30]
CCACK [31]
XCOR[26]
CORE[28]
Coding-Aware

Figure 2. Taxonomy of Network Coding based Routings for WMNs

For the sake of clarity, Fig.2 illustrates the taxonomy of NC based routing. In Fig.2, network
coding based routing are divided into two sub-categories: inter-flow network coding based
approaches and network coding based hybrid approaches. The later one is divided into three
sub-categories: hybrid of inter-flow and intra-flow NC approach, hybrid of inter-flow NC and
opportunistic routing (OR) [6], hybrid of intra-flow NC and OR. The listed solutions will be
analyzed in details in the following sections.

4. Inter-Flow NC based Routing

4.1 COPE

COPE [19] is the first practical and representative network coding based routing for WMNs,
before which research on network coding is mainly theoretical and focuses on multicast issue.
In COPE, two typical topologies (chain topology and “X” topology) existing coding
opportunities are investigated. These two coding topologies are components of many other
coding topologies and the basic of coding opportunity detection.
Review of Network Coding based Routing Algorithm for Wireless Mesh Networks
Xing Shao, Ru-chuan Wang
147
In COPE, every node uses ETX [9] as route metric to discover routes and then finds the coding
opportunities among routes according to the basic coding topologies when forwarding packets. If node
finds there exists coding opportunity, it will code the packets. However, there are two drawbacks in
coding opportunity detection of COPE. The coding topologies in COPE are limited within two-
hop range. In addition, COPE separates the process of route discovery and coding opportunity
detection which neglects many potential coding opportunities.

4.2 Coding-Aware Series

Though COPE leverages network coding to improve network throughput, the coding opportunities
in COPE are critically depend on the traffic pattern and exploited passively among routes. In a word,
COPE could not proactively change routing of flows to create more coding opportunities, which
motivates some improvements [20][21][22][23] of COPE, i.e. coding aware approaches.

4.2.1 ROCX

ROCX (Routing with Opportunistically Coded Exchanges) [20] first proposed the concept of
coding-aware, which means that node could detect potential coding opportunities and change routes to
increase coding opportunities. Besides, ROCX used route metric called ECX that captures the expected
number of coded transmissions needed for a successful exchange of packets between two nodes via an
intermediate node. Based on ECX, ROCX uses linear programming to optimize the routing.

4.2.2 CAMP

CAMP (Coding-Aware Multi-Path Routing)[21] combines the idea of coding-aware with multi-path,
which uses ETX as route metric to discover multiple paths to destination and split traffic among these
paths dynamically based on path reliability and coding opportunity. The most important point is that
CAMP could actively create instead of passively waiting for coding opportunity by switching its path
among the multiple discovered paths to maximize gain without the usage of opportunistic overhearing.
To measure the tradeoff between the achieved coding gain and the loss from changing the best path
to the suboptimal one, CAMP proposed concept of path switching gain, which is used to quantitatively
decide which path should be switched to utilize the coding opportunity judiciously.

4.2.3 RCR

Most network coding based routings assume that flows participating in network coding have the
same data rate, which is unpractical. RCR (Rate-adaptive Coding-aware Routing) [22] attempted to
solve the problem of rate assignment in multi-path coding-aware routing algorithm.
For this purpose, RCR proposed a node-centric routing metric, called RTN (Required Transmission
Number). The traffic splitting among paths in RCR, on one hand increases coding opportunities routes,
on the other hand meets the requirements of coding rate adaptation and congestion avoiding.

4.2.4 DCAR

DCAR (Distributed Coding-Aware Routing) [23] pointed out that COPE has two fundamental
limitations due to the separation of coding discovery and routing discovery. Coding opportunity in
COPE, on one hand crucially depends on the routes, on the other hand is limited within a two-hop
region only. To address these two limitations, DCAR proposed a coding-aware path discover
mechanism, called “Coding+Routing”, and first stated the necessary and sufficient conditions of
network coding to facilitate the detection of coding opportunity. DCAR also proposed routing metric,
CRM (Coding-aware Routing Metric), which facilitates the performance comparison between “coding-
possible” and “coding-impossible” paths. DCAR could discover high throughput paths with coding
opportunities and detect coding opportunities on the entire path, thus eliminating the “two-hop” coding
limitations in COPE.

4.3 MMSR
Review of Network Coding based Routing Algorithm for Wireless Mesh Networks
Xing Shao, Ru-chuan Wang
148

MMSR (Markovian Metric Source Routing)[24] proposed a solution from a novel prospective.
To model the reduction of channel resource consumption due to network coding, MMSR proposed
Markovian metric, which takes into account the effect of network coding in reducing the transmissions
and leads to routing decisions that can better take advantage of network coding. To facilitate the routes
computation using Markovian route metric, called ERC (Expect Resource Consumption), MMSR also
introduced the construction of Dot Graph, which reflects the conditioned metric is lower than
unconditioned metric because of reduction in resource consumption for the usage of network coding
and makes the route computation uniform.

5. NC based Hybrid Routing

5.1 Hybrid of Inter-Flow and Intra-Flow NC

Researches have shown that inter-flow network coding could increase throughput of network, while
intra-flow network coding could provide better reliability. It is easy for us to think of such a question:
whether we could integrate these two kinds network coding in routing algorithm.
I
2
MIX (Intra-flow and Inter-flow MIXing) [25] is the first routing that addresses the integration of
inter-flow and intra-flow network coding in order to benefit from both of them. To facilitate the
introduction of I
2
MIX, [25] first proposed IMIX which only exploited intra-flow network coding and
OSPR (Opportunistic Single-Path Routing), and find routes with more hops to increase the number of
overhearing transmissions. In I
2
MIX, packets from all flows are coded together at each node. Besides
the benefits in IMIX and COPE, I
2
MIX also discussed three unique benefits due to integration of two
kinds of network coding: linear coding gain, early-forwarder gain and gossip gain.

5.2 Hybrid of Inter-Flow NC and OR

Opportunistic routing (OR) was first proposed in [6] which takes advantage of the broadcast nature
of wireless medium and explores forwarding capacity of all intermediate nodes that overhear packets.
In recent years, researches have shown that OR could also improve the throughput of WMNs
significantly. In contrast to traditional routing, OR broadcasts packet first and then selects the one that
received the packet and closest to the destination as the next hop. Thus the routes in OR is decided
dynamically during the packet forwarding and the node closest to destination is selected as next hop,
which guarantee the performance improvement.
There is a lot of attempt to combine NC and OR. In this section, the hybrid routing that combines
inter-flow NC and OR is discussed, while the hybrid routing that integrates intra-flow NC and OR is
investigated in next section.

5.2.1 XCOR

XCOR (NC with Opportunistic Routing)[26] is based on SOAR[27] which is an opportunistic
routing and designed to facilitate multiple flows, thus easy to integrate with inter-flow network coding.
As other OR, in XCOR packets are broadcasted at each hop and the node closest to destination in ETX
is given the priority to forward the received packets immediately, which solve the problem of duplicate
transmissions in OR. To depict the coding gain, XCOR proposed utility gain.
In [25], I
2
MIX claims that combining opportunistic routing with inter-flow NC is not feasible,
because the routes in OR is not predetermined, while inter-flow NC need fixed path routing. To solve
this problem, XCOR uses reception reports and detects the coding opportunity dynamically.
Each intermediate node checks packets from different flows and codes them if the utility
obtained after coding is larger than that of without coding. Simulation results show that
equipped with both two techniques, XCOR could exploit their individual potential and synergy.

5.2.2 CORE

Review of Network Coding based Routing Algorithm for Wireless Mesh Networks
Xing Shao, Ru-chuan Wang
149
An observation that coding gain is obtainable if the node with most coding opportunity in forwarder
candidates is selected as next hop in opportunistic routing, motivates the design of CORE (Coding-
aware Opportunistic Routing Mechanism) [28], which combines hop-by-hop opportunistic forwarding
and localized inter-flow network coding for improving the throughput performance of WMNs.
In CORE, when a node has a packet to send, it simply broadcasts the packet. The forwarder
candidates receiving the packet collaborate to select the next forwarder with the most coding
opportunities among them in a localized manner. This hop-by-hop packet forwarding process is
repeated until the packet reaches its destination.
On one hand, CORE selects the node with the most coding gain in forwarder candidates to forward
packets through opportunistic forwarding; on the other hand, CORE attempts to maximize the coding
gain in a single transmission through localized network coding. Thus, CORE could benefit from both
opportunistic routing and inter-flow network coding.

5.3 Hybrid of Intra-Flow NC and OR

In opportunistic routing, the coordination among intermediate forwarders to avoid duplicated packet
transmission, is a complex problem and influences the resource consumption and performance of OR.
However, the complex negotiation among forwarders is not feasible in lossy networks. Besides, the
coordination among forwarder candidates in OR prevents spatial reuse and thus underutilize the
wireless medium. The solution to this problem is not satisfied until the hybrid routing that combines
intra-flow random linear network coding and OR was proposed, which eliminates the coordination
problem in OR.

5.3.1 MORE

To address the coordination in OR, MORE (MAC-independent Opportunistic Routing and
Encoding) [29] exploits random linear network coding in packet transmitting, which could eliminate
the possibility of useless duplicated transmission and inter-candidate coordination.
MORE uses a file transmission example to explain its routing process. At the source, the file to be
transmitted is broken up into batches of K native packets. Then the source creates a random linear
combination of the K native packets in the current batch and broadcast the coded packet. In MORE, the
transmitted packets are all coded and the coefficients of coding are transmitted with packets. When an
intermediate node receives a packet, it first checks the linearly independence between the coding vector
of the received packet and that of packets stored in its buffers. If their relationship is linearly
independence, i.e. the received packet is innovative to this node, it will accept the packet. Otherwise, it
will discard it. When the destination has received K linearly independence packets of the same batch, it
will decode them and send an acknowledgement to the source to inform it move onto next batch.
The checking of linearly independence between packets avoids the forwarding of duplicated and
useless transmission. Thus, MORE needs no special coordination among forwarder candidates. Besides,
for the existence of spatial reuse, MORE could achieve significant performance improvement
compared with traditional OR. However, there still exist some uninnovative packet transmissions in
MORE which consumed much bandwidth resource.

5.3.2 CodeOR

In inter-flow network coding, node encodes the packets received from other nodes currently.
However, in the intra-flow network coding of MORE, the source needs to split the file into multiple
batches and encode only packets of the same batch. The source only transmits the packets of the same
batch until receiving the acknowledgment for the batch, which degrades performance of MORE as the
network size scales up.
Only one batch of packets transmitting degrades the routing performance, while too much batches
transmitting concurrently may lead to congestion. To solve this problem, Lin etc. introduced the sliding
window mechanism of TCP flow control into MORE and proposed a novel routing, CodeOR(Coding
in Opportunistic Routing)[30]. In CodeOR, each node maintains a sending window to limit that only
packets of batches in the window could be transmitted in the entire network. Besides, CodeOR uses
end-to-end acknowledgement (E-ACK) and hop-by-hop acknowledgement (H-ACK). E-ACK is
Review of Network Coding based Routing Algorithm for Wireless Mesh Networks
Xing Shao, Ru-chuan Wang
150
transmitted to the source only when batch i and all batches before i have been decoded at the
destination to inform that source node could move the sending window to the i+1. A node uses H-ACK
to notify its upstream nodes that a sufficient number of coded packets has been received in a batch, so
that the upstream node can start to transmit new batches and avoid redundant transmission. By
transmitting multiple batches of a window simultaneously, CodeOR substantially improves the
performance of MORE and is particular suitable for large scale network.

5.3.3 CCACK

MORE has a problem that transmission of unneeded packets leads to waste of bandwidth which
makes it impossible for MORE to realize the maximum possible gains. The prior solutions leverage
credit mechanism using measurements of offline loss rates to control the transmission of coded packets,
which leads to their performance heavily depend on the accuracy and freshness of the loss rate
measurements, while CCACK (Cumulative Coded Acknowledgement) [31], an improved version of
MORE, proposed an approach oblivious to loss rates.
The approach is called Cumulative Coded Acknowledgement scheme and allows nodes to
acknowledge network coded traffic to their upstream nodes in a simple way, oblivious to loss rates, and
with practically zero overhead. Knowing which packet has been received by forwarder candidates,
upstream node will send only those packets innovative to forwarder candidates and avoid unnecessary
transmission, thus save bandwidth. In addition, the scheme enables an efficient credit-based rate
control algorithm.

5. Comparison of Typical NC based Routing Algorithm

Table 1. Comparison of Typical Network Coding based Routings
Routing
NC Type
Metric
Overhear
Traffic Pattern
Characteristic
COPE[19]
ROXC[20]
CAMP[21]
RCR[22]
DCAR[23]
MMSR[24]
I
2
MIX[25]
XCOR[26]
CORE[28]
MORE[29]
CodeOR[30]
CCACK[31]
Inter-Flow
Inter-Flow
Inter-Flow
Inter-Flow
Inter-Flow
Inter-Flow
Inter&Intra-Flow
Inter-Flow & OR
Inter-Flow & OR
Inter-Flow & OR
Inter-Flow & OR
Inter-Flow & OR
ETX
ECX
ETX
RTN
CRM
ERC
NTX
ETX
G-Dist
ETX
ETX
ETX
Yes
Yes
No
Yes
Yes
No
Yes
Yes
Yes
No
No
No
Chain &Cross
Chain &Cross
Chain
Chain &Cross
Chain& Extend of Cross
Chain
Chain& Extend of Cross
Chain &Cross
Chain &Cross
No
No
No
First practical
Coding-aware
Coding-aware Multi-path
Coding-aware Rate control
Coding-aware on entire path
Markovian Metric
Integration of Inter&Intra-Flow
Integration of Inter-Flow & OR
Integration of Inter-Flow & OR
Avoid Coordination
Sliding window
Avoid uninovative transmission

In this section, the representative aspects of previous discussed various solutions are summarized in
Table 1. The table further clarifies the main features of these solutions and provides a noticeable
comparison between these approaches.

6. Conclusion

In this paper, the taxonomy of existing network coding based routing for WMNs is presented. Then
typical network coding based routing algorithms for WMNs are reviewed in order of category by
analyzing their representative features, relative strength and weakness. Although significant
performance improvement in throughput and reliability has been obtained, several research problems in
network coding based routing are still open, such as the combination of coding gain and traditional
routing metric, impact of traffic patterns, rate adaption, tradeoff between coding opportunity increasing
and load balancing, and so on. Solutions to these problems will motivate the further progress of
network coding based routing for WMNs.

7. Acknowledgement

Review of Network Coding based Routing Algorithm for Wireless Mesh Networks
Xing Shao, Ru-chuan Wang
151
The subject is sponsored by the National Natural Science Foundation of P. R. China (No.60973139,
60773041, 61003039, 61003236), Scientific & Technological Support Project (Industry) of Jiangsu
Province (No.BE2010197, BE2010198), The Special Foundation for Development of Modern Service
Industry of Jiangsu Province, Project sponsored by Jiangsu provincial research scheme of natural
science for higher education institutions (10KJB520013, 10KJB520014), Scientific Research &
Industry Promotion Project for Higher Education Institutions(JH10-14), Science & Technology
Innovation Fund for higher education institutions of Jiangsu Province (CXZZ11_0406, CX10B-196Z,
CX10B-197Z, CX10B-200Z), The Six kinds of Top Talent of Jiangsu Province (2008118), Doctoral
Fund of Ministry of Education of China (20103223-120007), Key Laboratory Foundation of
Information Technology Processing of Jiangsu Province (KJS1022). Thank anonymous reviewers for
their helpful suggestions on the quality improvement of this paper.

8. References

[1] Ian F.Akyildiz, Xudong Wang, Weilin Wang, “Wireless mesh networks: a survey”, Computer
Networks, Elsevier, vol.47, no.4, pp.445-487, 2005.
[2] Kai Yang, Jianfeng Ma, "A Hierarchical Hybrid Routing Protocol for Wireless Mesh Network",
JCIT: Journal of Convergence Information Technology, AICIT, vol.5, no.8, pp.146-156, 2010.
[3] Nori Mohammed AL-Kharasani, Zuriati Ahmad Zukarnain, "Performance Evaluation of Routing
with Load-Balancing in Multi-Radio Wireless Mesh Networks", JDCTA: International Journal of
Digital Content Technology and its Applications, AICIT, vol.5, no.2, pp.64-71, 2011.
[4] Zhanmao Cao, Limin Peng, "Destination-oriented Routing and Maximum Capacity Scheduling
Algorithms in Cayley Graph Model for Wireless Mesh Network", JCIT: Journal of Convergence
Information Technology, AICIT, vol.5, no.10, pp.82-90, 2010.
[5] Daniel Aguayo, John Bicket, Sanjit Biswas, Glenn Judd, R. Morris, “Link-level Measurements
from an 802.11b mesh network”, In Proceeding of ACM SIGCOMM 2004, pp.121-132, 2004.
[6] Sanjit Biswas, Robert Morris, “ExOR: opportunistic multi-hop routing for wireless networks”, In
Proceedings of ACM SIGCOMM 2005, pp.133-144, 2005.
[7] Raffaele Bruno, Maddalena Nurchis, “Survey on diversity-based routing in wireless mesh
networks: Challenges and solutions”, Computer Communications, Elsevier, vol.33, no.3, pp.269-
282, 2010.
[8] Nunez-Martinez, Mangues-Bafalluy, “A survey on routing protocols that really exploit wireless
Mesh network features”, Journal of Communications, Springer, vol.5, no.3, pp.211-231, 2010.
[9] Douglas S.J. De Couto, Daniel Aguayo, John Bicket, Robert Morris, “A high-throughput path
metric for multi-hop wireless routing”, In Proceeding of ACM MobiCom, pp.134-146, 2003.
[10] Richard Draves, Jitendra Padhye, Brian Zill, “Routing in multi-radio, multi-hop wireless mesh
networks”, In Proceedings of the 10th annual international conference on Mobile computing and
networking, pp.114-128, 2004.
[11] Anand Prabhu Subramanian, Milind M. Buddhikot, Scott Miller, “Interference aware routing in
multi-radio wireless mesh networks”, In Proceedings of IEEE WiMesh 2006, pp.55-63, 2006.
[12] Rudolf Ahlswede, Ning Cai, Shuo-Yen Li, Raymond W. Yeung, “Network Information Flow”,
IEEE/ACM Transactions on Information Theory, IEEE/ACM, vol.46, no.4, pp.1204-1216, 2000.
[13] Tracey Ho, Ralf Koetter, Muriel Médard, David R. Karger, Michelle Effros, “The benefits of
coding over routing in a randomized setting
”, In Proceedings of IEEE International Symposium
on Information Theory, pp.442-443, 2003.
[14] Shuo-Yen Robert Li, Raymond W. Yeung, Ning Cai, “Linear network coding”, IEEE/ACM
Transactions on Information Theory, IEEE/ACM, vol.49, no.2, pp.371-381, 2003.
[15] Koetter Ralf, Muriel Médard, “An algebraic approach to network coding”, IEEE/ACM
Transactions on Networking, IEEE/ACM, vol.11, no.5, pp.782-795, 2003.
[16] Sidharth Jaggi, Michael Langberg, Sachin Katti, Tracey Ho, Dina Katabi, Muriel Médard,
“Resilient Network Coding In The Presence of Byzantine Adversaries”, In Proceedings of IEEE
INFOCOM 2007, pp.616-624, 2007.
[17] Sidharth Jaggi, Peter Sanders, Philip A. Chou, Michelle Effros, Sebastian Egner, Kamal Jain,
Ludo M.G.M. Tolhuizen, “Polynomial time algorithms for multicast network code construction”,
IEEE Transactions on Information Theory, IEEE, vol.51, no.6, pp.1973-1982, 2005.
Review of Network Coding based Routing Algorithm for Wireless Mesh Networks
Xing Shao, Ru-chuan Wang
152
[18] Christina Fragouli, Dina Katabi, Athina Markopoulou, Muriel Médard, Rahul Hariharan,
“Wireless network coding: opportunities & challenges”, In Proceedings of IEEE Military
Communications Conference, pp.8-11, 2007.
[19] Sachin Katti, Hariharan Rahul, Wenjun Hu, Dina Katabi, Muriel Médard, Jon Crowcroft, “XORs
in the air: practical wireless network coding”, IEEE/ACM Transactions on Networking,
IEEE/ACM, vol.16, no.3, pp.497-510, 2008.
[20] Bin Ni, Naveen Santhapuri, Zifei Zhong, Sarihari Nelakuditi, “Routing with opportunistically
coded exchanges in wireless mesh networks”, In Proceeding of WiMesh 2006, pp.157-159, 2006.
[21] Song Han, Zifei Zhong, Guihai Chen, Edward Chan, Aloysius K.Mok, “Coding-aware multi-path
routing in multi-hop wireless networks”, In Proceedings of the IEEE 27th International
Performance, Computing, and Communications Conference, pp.93-100, 2008.
[22] Yan Yan, Zhuang Zhao, Baoxian Zhang, Hussein T. Mouftah, Jian Ma, “Rate-Adaptive Coding-
Aware Multiple Path Routing for Wireless Mesh Networks”, In Proceedings of 2008 IEEE Global
Telecommunications Conference, pp.543-547, 2008.
[23] Jilin Le, John C.S.Lui, Dah-Ming Chiu, “DCAR: Distributed coding-aware routing in wireless
networks”, IEEE Transactions on Mobile Computing, IEEE, vol.9, no.4, pp.596-608, 2010.
[24] Yunnan Wu, Saumitra M. Das, Ranveer Chandra, “Routing with a Markovian Metric to Promote
Local Mixing”, In Proceedings of IEEE INFOCOM 2007, pp.2381-2385, 2007.
[25] Chuan Qin, Yi Xian, Chase Gray, Naveen Santhapuri, Srihari Nelakuditi, “I
2
MIX: integration of
intraflow and inter-flow wireless network coding”, In Proceedings of 5th Annual IEEE
Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and
Networks Workshops, pp.1-6, 2008.
[26] Dimitrios Koutsonikolas, Y.Charlie Hu, Chih-Chun Wang, “XCOR: synergistic interflow network
coding and opportunistic routing”, In Proceedings of the ACM MobiCom 2008, pp.1-3, 2008.
[27] Eric Rozner, Jayesh Seshadri, Yogita Mehta, Lili Qiu, “Simple opportunistic routing protocol for
wireless mesh networks”, In Proceedings of Wi-MESH 2006, pp.48-54, 2006.
[28] Yan Yan, Baoxian Zhang, Jun Zheng, Jian Ma, “CORE:A coding-aware opportunistic routing
mechanism for wireless mesh networks”, IEEE Wireless Communications, IEEE, vol.17, no.3,
pp.96-103, 2010.
[29] Szymon Chachulski, Michael Jennings, Sachin Katti, Dina Katabi, “Trading structure for
randomness in wireless opportunistic routing”, In Proceedings of ACM SIGCOMM 2007:
Conference on Computer Communications, pp.169- 180, 2007.
[30] Yunfeng Lin, Baochun Li, Ben Liang, “CodeOR: opportunistic routing in wireless mesh networks
with segmented network coding” In Proceedings of 16th IEEE International Conference on
Network Protocol, pp.1092-1648, 2008.
[31] Dimitrios Koutsonikolas, Chih-chun Wang, Y.Charlie Hu, “CCACK: Efficient network coding
based opportunistic routing through cumulative coded acknowledgments”, In Proceedings of
IEEE INFOCOM 2010, pp.216-224, 2010.

Review of Network Coding based Routing Algorithm for Wireless Mesh Networks
Xing Shao, Ru-chuan Wang
153