Wireless Pers Commun

DOI 10.1007/s11277-017-4496-5

A 3-Parameter Routing Cost Function for Improving

Opportunistic Routing Performance in VANETs

Mohammad Naderi1 • Farzad Zargari2 • Vahid Sadatpour1 •

Mohammad Ghanbari3,4

 Springer Science+Business Media New York 2017

Abstract Dynamic topology and unstable wireless links make efficient data delivery in
vehicular ad-hoc networks (VANETs) a challenging issue. To tackle this issue, several
routing strategies for finding a route with high reliability and low latency in VANETs have
been proposed. One of theme is Opportunistic routing (OR) paradigm which can improve
the reliability of routing in vehicular ad-hoc networks (VANETs) by broadcasting trans-
mission features and employing additional backup links from other neighboring nodes. In
this paradigm a node sends the packet to a number of its neighbors called forwarding set.
The way transmission priority is assigned to the nodes in the forwarding set can highly
affect the performance of opportunistic routing. In this Paper a new measure composed of
three parameters is proposed for selecting the best forwarding node. Simulations on
highway environment indicate that the proposed method improves the routing performance
in all grounds of QoS compared to the other published methods in the literature.

Keywords VANET  Opportunistic routing  Link state  Link stability  Packet

advancement

& Farzad Zargari

[email protected]
Mohammad Naderi
[email protected]
Vahid Sadatpour
[email protected]
Mohammad Ghanbari
[email protected]
1
Department of Computer Engineering, Science and Research Branch, Islamic Azad University,
Tehran, Iran
2
Research Institute for ICT, aka Iran Telecom Research Center, End of North Kargar St., Tehran,
Iran
3
School of Computer Science and Electronic Engineering, University of Essex, Colchester, UK
4
School of Electrical and Computer Engineering, University of Tehran, Tehran, Iran

123
 M. Naderi et al.

1 Introduction

Vehicular ad-hoc networks (VANETs) are wireless networks which do not require any
fixed infrastructure. These networks suffer from several problems such as highly variable
network topology, variable communication conditions and road constrained mobility
patterns which make the routing protocol in the VANETs a challenging issue [1, 2]. Two
kinds of highly reported packet transmission approaches for VANETs in the literature are:
unicast transmissions, and broadcast transmission [1, 3, 4]. Broadcast transmission pro-
tocols exploit wireless transmissions’ characteristics to send each packet to a set of
neighboring nodes. In the early broadcast transmission protocols each node that receives a
packet, relays it to its neighboring nodes. This approach suffers from broadcast flooding
problem due to the large number of unnecessary packet retransmissions [5]. Opportunistic
routing paradigm is among the solutions to alleviate this problem [6]. In this approach only
one of the receiving nodes in the forwarding set broadcasts the received packet and if the
node fails to receive and forward the packet (to be out of the coverage area or does not
listen to the wireless channel), another receiving node among the nodes in the forwarding
set transmits it and this goes on in the same manner [7]. Through such forwarding set of
nodes, opportunistic routing improves the reliability of packet transmission in VANETs.
This is because the probability of delivering a packet to at least one neighboring node is
greater than the probability of delivering it to only one specified node [8]. In this way there
is not a predefined next hop for forwarding a packet (such as OLSR, DSR or Geographic
routing) [9] and the forwarding node is selected according to the characteristics and quality
of the transmission links [10]. Figure 1 depicts a basic model of opportunistic routing
paradigm for the case in which the destination node is out of the effective transmission
range of the source node. In this case each sender node should rank order the nodes in its
forwarding set according to specified metric(s). An important issue which highly affects
the performance of opportunistic routing paradigm is the way the nodes are prioritized in
the forwarding set for transmission of the received packet.
The opportunistic routing paradigm are classified as: Geographic, link state aware,
probabilistic, optimization based and cross-layer. Opportunistic routing is widely used in
wireless sensor networks and mesh networks [11–15]. Nevertheless, in its employment on
VANETs it suffers from VANET inherent characteristics of frequent link failures. In [16] a
geo-inspired opportunistic routing for VANETs is proposed in which the forwarding node
for routing uses the last gathered information about its neighboring nodes. Since the

Fig. 1 Packet forwarding using opportunistic routing

123
A 3-Parameter Routing Cost Function for Improving…

employed information may be out of date, it results in frequent failures and moreover the
wrong information may cause selection of unsuitable link. In [17] the next hop is selected
based on the angle between sender, receiver and the forwarding node. In this way the
vehicles moving in the opposite direction can be easily discarded but since this approach
does not consider other parameters such as link quality or packet advancement, it will
increase number of hops, end-to-end delay and link failure rate. In [18] the nearest hop to
the destination is selected as the next hop. Since this method does not consider link quality,
it suffers from high packet loss and low packet delivery ratio. LSGO [19] employs both
geographical and link state (ETX) parameters to propose a new opportunistic routing
method. This method does not consider the link failure and increased probability of packet
loss in distant communications. LLA [10] method employs link state (EATX) and link
stability parameters for opportunistic routing. Since both link state and link stability
parameters tend to select near hops as the next hop, as a result the total number of hops and
end-to-end delay are increased. Bidirectional stable communication (BSDC) method in
[20] employs link quality and distance metrics to select a relay node. It has two negative
points. First the distance metric is not normalized and second since the link stability is not
taken into account, it suffers from packet drop due to frequent failure in the selected relay
nodes. Moreover, since each node in this method stores a list of other nodes’ IDs and
broadcasts them to the beacons, this method suffers from increasing overhead and colli-
sions in the network. It is worth noting that most of the previous works have not employed
a realistic model for wireless communication to evaluate the performance of their proto-
cols. In this paper three metrics for opportunistic routing are employed. Link state and link
stability parameters are used to select more reliable nodes in the forwarding set, and packet
advancement parameter is used along with these parameters to increase the priority of
selecting distant hops among the reliable nodes in the forwarding set. We claim this will
lead to a more reliable opportunistic routing algorithm with lower hop count, end-to-end
delay and higher packet delivery ratio. The results of the performed simulations with inter
vehicle propagation model [21] confirm our claim. The rest of the Letter is organized as
follows. Section 2 presents our proposed scheme for opportunistic routing in VANETs.
Section 3 shows the simulation environment exploited to validate the proposal and presents
and discusses the simulation results and comparison with other approaches. Finally, Sect. 4
concludes the paper.

2 Proposed Method

One of the main factors which affects the quality of opportunistic routing is the employed
metrics for rank ordering of nodes in the forwarding set. Another important issue is the way
the metric in the developed protocol is modelled. For example, the distance metric is used
in [19, 20, 22] but it is modeled in three different ways which highly affect the perfor-
mance. In this section the proposed three metrics for OR along with their modeling
methods are explained. The employed three metrics (link state, link stability, and packet
advancement) are discussed in more detail in the following.

2.1 Link State

Link state indicates the quality of the wireless link between two nodes and is modeled by
packet delivery ratio pij . Packet delivery ratio between two neighboring nodes i and j is

123
 M. Naderi et al.

measured by the ratio of the number of received packets at the receiving node j to the total
number of transmitted packets by the sender node i, within a specified time interval [23].
dp
pij ¼ ð1Þ
sp
where dp and sp indicate the number of correctly delivered packets and sending beacons in
a time interval for a specified beacon transmission rate. It is worth noting that the beacon
message rate is fixed and both sender and receiver are aware of it.

2.2 Link Stability

Link stability metric is proposed in [10] as a measure for the lifetime of the wireless link
between two nodes and is defined as:
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi !
 
2 2 ffi
min sxi ðt0  DtÞ  sxj ðt0  DtÞ þ syi ðt0  DtÞ  syj ðt0  DtÞ ; r
sij ¼ 1  ð2Þ
r
 T
where si ðt0  DtÞ ¼ sxi ðt0  DtÞ; syi ðt0  DtÞ represents the movement vector of node i
in the previous interval ðt0  DtÞ and r is the maximum communication range. The sta-
bility index sij of link (i, j) at any time t0 is based on information on vehicles movement in
the previous interval ðt0  DtÞ, as the normalized increase  of distance between vehicles i
and j in the past ðt0  DtÞ interval [10]. Link stability sij is in the range of ½0; 1 and larger
sij implies lower variation in the position of two vehicles w.r.t. each other and hence higher
link stability. Link stability is based on the idea that when two vehicles i and j were
connected at interval ðt0  DtÞ, they will be most likely to be connected one more Dt time.
It is worth noting that the vehicles are assumed to use GPS system and can get their
geographical positions and inform their neighbors about the geographical positions via
beacon messages.

2.3 Packet Advancement

We propose packet advancement metric to indicate the achieved advancement towards

destination by selecting a node in the forwarding set. Distance metric is used in many
routing algorithms [19, 20, 22]. Rehman et al. [20] and Celimuge and Ohzahata [22]
employ the distance between sender and relay nodes without and with normalization,
respectively. The negative point in these methods is that they aim to select the relay node
such that to minimize the distance between the relay node and receiver, which is not
necessarily achieved by maximizing the distance between the sender and relay nodes. In
[19] packet advancement, reducing the distance to the receiver node, is selected as a metric
for routing but transmission range of the sender is not taken into account and hence packet
loss in the network increases. This metric should be used along with normalized metrics of
link state and link stability, and hence it should be normalized, as well. Sending a packet
form node i to node j the packet advancement metric is defined as:
dij ¼ Did  Djd ð3Þ
where Did and Djd are the distances from node i and node j to the destination node,
respectively. The packet advancement metric aij is:

123
A 3-Parameter Routing Cost Function for Improving…

(
0 dij \0; dij [ r
aij ¼ dij ð4Þ
0  dij  r
r
where r is the maximum communication range. Link state and link stability metrics give
higher rank to the closer nodes in the forwarding set. This will increase the total number of
hops and as a result the end-to-end delay and meanwhile the network throughput and
packet delivery ratio are reduced. On the other hand, packet advancement metric tends to
select far nodes in the forwarding set and considering this metric along with the other two
metrics it can reduce the total number of hops and minimize its negative side effects.
The following example provides an explanation about the idea behind using packet
advancement along with the other metrics. Suppose a packet is sent from source to des-
tination as shown in Fig. 2 and the resulted value of Link State and Link Stability for the
entire links in the path ABC to destination node is 0.9. For the links in the path from E to
destination the same metrics have the value of 0.85. With these metrics, the ABC path is
preferred over path from E to destination. Thus it is obvious that even though the indi-
vidual links in the upper path has higher reliability than the lower path, but the reliability of
the total lower path is higher than that of the upper path. This is because the lower path is
made up of only two links instead of four links in the upper path from ABC to destination.
Exploiting packet advancement along with the other two metrics can make longer
distant nodes with lower link quality (such as node E in the above Figure) to be selected as
the forwarding node, leading to smaller hop count and lower end-to-end delay. The three
metrics of the Link state, link Stability and packet Advancement (LSA) are combined to
define a rank ordering parameter of the nodes in the forwarding set:
LSAij ¼ pij  sij  aij ð5Þ

By using this parameter set, the selected forwarding node is the node which makes a
compromise amongst the highest link state, link stability and packet advancement.
Algorithm 1 indicates the proposed method in details. As shown in Algorithm 1, LSA
measure is computed using the received beacon messages by the sender node. The sender
node generates ordered forwarding list according to the computed LSA measure and
transmit it in the packet. The highest ranked node in the set will forward the packet
immediately upon receiving the packet and the other nodes start timers. If forwarding node
successfully transmits the packet before the timeout the other nodes stop their timers.

Fig. 2 A packet transmission example to indicate the importance of packet advancement

123
 M. Naderi et al.

Otherwise, the next highest ranked node will transmit the packet and the remaining nodes
restart the timers.

The proposed method, hereafter named LSA, will be contrasted against three other
methods in an environment representing real vehicle traffic information. In the next section
the simulation environment and the simulation results are described.

3 Simulation Results

Ns-2 version 2.35 [24] simulator was employed along with FleetNet [25] which is based on
real vehicle movement patterns captured from real traffic for 60 s in a highway of 15 km
long and 10 m wide in Germany. The minimum and maximum speed of vehicles were 77
and 168 km/h, respectively. One packet per second constant bit rate pattern is used for
packet transmission from a source node to the destination node. The reported simulation
results are averaged over 20 simulation runs.

123
A 3-Parameter Routing Cost Function for Improving…

To increase the accuracy of the simulation results the fading effect of wireless com-
munication should be taken into account as well. Nakagami channel fading model is used
to derive the channel parameter values [21]. This model has been used in the literatures,
such as [21, 26]. The parameter values are listed in Table 1. This simulation environment
was used to compare the performance of the proposed LSA method with that of the LLA
[10], LSGO [19] and BDSC [20] methods. The simulations are performed in two scenarios,
different nodes densities and different distances between source and destination nodes. In
the simulations, five network quality of service (QoS) characteristics were evaluated. They
were: throughput, end-to-end delay, average number of hops, packet delivery ratio and
jitter. Each QoS is defined in the following.

3.1 Packet Throughput

Packet throughput indicates the ratio of the correctly received bits in the destination within
the time interval of start of transmission to the end of simulation. Figure 3 shows the
achieved throughput of the three tested methods for various number of nodes in the
simulations. The throughput of the proposed method (LSA) outperforms the other methods
consistently. Moreover, as Fig. 3 shows, for smaller number of nodes the performance gain
of the proposed method over the other methods becomes more significant. This is because
in lower node densities selecting stable and far node in the forwarding set, which is a
distinct feature of the proposed method, becomes more reliable. Moreover, Fig. 4 depicts
resultant throughput for different distances between source and destination when there
were 236 vehicles in the highway. The results in Fig. 4 indicate higher throughput of the
proposed method compared to the other methods in the entire distances.

3.2 Average End-To-End Delay

In the simulations the average end-to-end delay is computed by averaging the end-to-end
delay time of correctly received packets. Average end-to-end delay characteristic of the
methods tested for various number of nodes is depicted in Fig. 5 Simulation results in this
Figure indicate that the proposed method achieves less average end-to-end delay compared

Table 1 Simulation Parameters

Parameter Value

Number of vehicles 105, 236, 340, 459

Simulated area 15 km 9 10 m
Mobility generator FleetNet
MAC/PHY IEEE 802.11 p
Maximum transmission range 450 m
Channel rate 6 Mbps
Packet size 512 Byte
Data rate 1 packets per sec
Propagation model Nakagam
Hello interval 1s
Number of runs 20
Simulation time 60 s
Number of source/destination pairs 10

123
 M. Naderi et al.

Fig. 3 Throughput for various numbers of node in the highway scenario

Fig. 4 Throughput for different source destination distances in 236 nodes case

to other methods in the entire tested node densities. Moreover, Fig. 6 indicates the sim-
ulation results for different source destination distances when there were 236 vehicles in
the highway. The results in Fig. 6 indicate lower end to end delay of the proposed method
compared to the other methods in the entire distances.

3.3 Average Number of Hops

It is the average number of hops which correctly received packets traverse from source to
destination. Figure 7 indicates the average number of hops for four tested methods and
Fig. 8 shows the performance of the proposed method on the same QoS parameter for
different distances among source and destinations. The proposed method achieves the least

123
A 3-Parameter Routing Cost Function for Improving…

Fig. 5 End-to-end delay for various numbers of nodes in a highway scenario

Fig. 6 End to end delay for different source destination distances in 236 nodes case

average number of hops among the tested methods in both scenarios. This emphasizes the
significance of advancement parameter in reducing the number of hops that a packet should
pass to reach its destination.

3.4 Packet Delivery Ratio

The next tested parameter in the simulations is packet delivery ratio (PDR) which is
defined as the number of received packets in the destination to the total number of the sent
packets. Figure 9 indicates PDR for the three resulting methods at four different number of
nodes and Fig. 10 depicts resultant PDR for different distances between source and des-
tination when there were 236 vehicles in the highway. The proposed method outperforms
the others in the entire tested node densities. Since by increasing the node density there will
be more chance to find stable nodes amongst the distant nodes, PDR increases for higher

123
 M. Naderi et al.

Fig. 7 Average hop count for various numbers of nodes in a highway scenario

Fig. 8 Average hop count for different source destination distances in 236 nodes case

node densities in all three methods. Nevertheless, the average PDR of the proposed method
outperforms the others.

3.5 Jitter

The last tested parameter is jitter which plays an important role in multimedia streaming.
Jitter is defined as the standard deviation of the delay of the received packets. Figure 11
shows the jitter results for the tested methods in the simulations. The average jitter of the
proposed method is consistently less than the others in the entire node densities. Moreover,
Fig. 12 indicates the simulation results for different source destination distances when
there were 236 vehicles in the highway. The results in Fig. 12 indicate lower jitter of the
proposed method compared to the other methods in the entire distances.

123
A 3-Parameter Routing Cost Function for Improving…

Fig. 9 Packet delivery ratio for various numbers of node in the highway scenario

Fig. 10 Packet delivery ratio for different source destination distances in 236 nodes case

4 Conclusion

In this paper a new measure (LSA) to improve the performance of the opportunistic routing
is proposed. LSA consists of three metrics: link state, link stability and packet advance-
ment. The link stability and link state metrics tend to choose near nodes in the forwarding
set while packet advancement metric increases the probability of selecting far nodes in the
forwarding set. The simulations were performed by NS-2.35 based on traffic patterns from
FleetNet for highways. The simulation results indicate that our proposed method outper-
forms the previously reported methods of LLA, LSGO and BDSC, in all aspects of network
QoS parameters, such as throughput, end-to-end delay, average hop numbers, packet
delivery ratio and jitter, in the entire testing conditions. This means, the proposed method

123
 M. Naderi et al.

Fig. 11 Jitter for various numbers of node in the highway scenario

Fig. 12 Jitter for different source destination distances in 236 nodes case

has superior performance in all grounds of QoS. In general, improving in one aspect of
QoS can be due to trading for the degradation of other QoS.

References
1. Li, F., & Wang, W. (2007). Routing in vehicular ad hoc networks: A survey. Vehicular Technology
Magazine, 2, 12–22.
2. Kumar, Neeraj, & Dave, Mayank. (2016). BIIR: A beacon information independent VANET routing
algorithm with low broadcast overhead. Wireless Personal Communications, 87, 869–895.
3. Rak, J. (2013). Providing differentia levels of service availability in VANET communications. IEEE
Communication Letter, 17, 1380–1383.

123
A 3-Parameter Routing Cost Function for Improving…

4. Saleet, H., Langar, R., Naik, K., & Boutaba, R. (2011). Intersection-based geographical routing protocol
for VANETs: A proposal and analysis. IEEE Vehicular Technology Transactions, 60, 4560–4574.
5. Reina, D. G., Toral, S. L., Johanson, P., & Barrero, F. (2015). A survey on probabilistic broadcast
schemes for wireless ad hoc networks. Ad Hoc Networks, 25, 263–292.
6. Chakchouk, N. (2015). A Survey on Opportunistic Routing in Wireless Communication Networks.
IEEE Communications Surveys and Tutorials, 17, 2214–2241.
7. Thorat, S. A., & Kulkarni, P. J. (2015). Opportunistic routing in presence of selfish nodes in MANET.
Wireless Personal Communication, 82, 689–708.
8. Laufer, R., Dubois-Ferriere, H., & Kleinrock, L. (2012). Polynomial-time algorithms for multirate
anypath routing in wireless multihop networks. IEEE/ACM Transactions on Networking, 20, 742–755.
9. Qureshi, K. N., AbdulAltameem, A. H., & Altameem, A. (2016). Road aware geographical routing
protocol coupled with distance, direction and traffic density metrics for urban vehicular ad hoc net-
works. Wireless Personal Communication, 92, 1–20.
10. Rak, J. (2014). LLA: A new anypath routing scheme providing long path lifetime in VANETs. IEEE
Communications Letters, 18, 281–284.
11. Biswas, S., & Morris, R. (2005). ExOR: Opportunistic multi-hop routing for wireless networks. ACM
SIGCOMM Computer Communication Review, 35, 133–144.
12. Sadeghi, B. Kanodia, V., Sabharwal, A., & Knightly, E. (2002). Opportunistic media access for multi
rate ad hoc networks. In ACM in proceedings of the 8th annual international conference on mobile
computing and networking (pp. 24–35).
13. Shin, W. Y., Chung, S. Y., & Lee, Y. H. (2013). Parallel opportunistic routing in wireless networks.
IEEE Transactions on Information Theory, 59, 6290–6300.
14. Liu, C., & Wu, J. (2013). On multicopy opportunistic forwarding protocols in nondeterministic delay
tolerant networks. IEEE Transactions on Parallel and Distributed Systems, 23, 1121–1128.
15. Rosario, D., Zhao, Z., Braun, T., Cerqueira, E., Santos, A., & Li, Z. (2013). Assessment of a robust
opportunistic routing for video transmission in dynamic topologies. IEEE in Wireless Days (WD), 2013,
1–6.
16. Leontiadis, I., & Mascolo, C. (2007). Geopps: Geographical opportunistic routing for vehicular net-
works. In IEEE international symposium on a world of wireless, mobile and multimedia networks (pp.
1–6).
17. Yu, D., Ikram, M., & Ko, Y. B. (2009). (VVOF) velocity vector-based opportunistic forwarding in
vehicular sensor network. Communications and Information Technology, ISCIT, 2009, 627–628.
18. Ting, Y., Li, Y., Ma, X., Shang, W., Sanadidi, M. Y., & Gerla, M. Y. (2013). Scalable opportunistic
VANET content routing with encounter information. In 21st IEEE international conference on network
protocols (ICNP), IEEE (pp. 1–6).
19. Cai, X., He, Y., Zhao, C., Zhu, L., & Li, C. (2014). LSGO: Link state aware geographic opportunistic
routing protocol for VANETs. EURASIP Journal on Wireless Communications and Networking, 2014,
1–10.
20. Rehman, O., Ould-Khaoua, M., & Bourdoucen, H. (2016). An adaptive relay nodes selection scheme for
multi-hop broadcast in VANETs. Computer Communications, 87, 76–90.
21. Cheng, L., Henty, B. E., Stancil, D. D., Bai, F., & Mudalige, P. (2007). Mobile vehicle-to-vehicle
narrow-band channel measurement and characterization of the 5.9 GHz dedicated short range com-
munication (DSRC) frequency band. IEEE Journal on Selected Areas in Communications, 25,
1501–1516.
22. Celimuge, C. W., & Ohzahata, S. (2012). VANET broadcast protocol based on fuzzy logic and
lightweight retransmission mechanism. IEICE Transaction on Communication, 95, 415–425.
23. Couto, D., Bicket, J., & Morris, R. (2005). A high-throughput path metric for multi-hop wireless
routing. Wireless Networks, 11, 419–434.
24. The Network Simulator—ns-2, The Network Simulator—ns-2. http://www.isiedu/nsnam/ns/. Accessed
23 May 2013.
25. Enkelmann, W. (2003). Fleet-net-applications for inter-vehicle communication. In IEEE proceedings
intelligent vehicles symposium (pp. 162–167).
26. Charash, U. (1979). Reception through Nakagami fading multipath channels with random delays. IEEE
Transactions on Communications, 27, 657–670.

123
 M. Naderi et al.

Mohammad Naderi received B.Sc. degree in Computer Engineering

from Hamedan University of Technology, Hamedan, Iran in 2013. He
is currently M.Sc. student in Computer Engineering at Science and
Research Branch IAU University, Tehran, Iran. His research interests
include wireless multihop networks, mobile ad-hoc networks and
vehicular ad hoc networks.

Farzad Zargari received B.Sc. degree in Electrical Engineering from

Sharif University of Technology and M.Sc. and Ph.D. degrees in
Electrical Engineering from University of Tehran, all in Tehran, Iran.
He is currently associate professor at the Information Technology
Department of Iran Telecom Research Center (ITRC), Ministry of
Telecommunications and Information Technology of Iran. He is also a
teaching academic staff in the Computer Engineering Department of
Science and Research branch of Islamic Azad University. His research
interests include multimedia systems, wireless networks, and hardware
implementation of image and video coding standards.

Vahid Sadatpour received B.Sc. degree in Computer Engineering

from Kashan University, Kashan, Iran in 2006 and M.Sc. degree in
Computer Engineering from Science and Research Branch IAU
University, Tehran, Iran, in 2009. He is currently Ph.D. candidate in
Computer Engineering at Science and Research Branch IAU Univer-
sity, Tehran, Iran. His research interests include wireless multihop
networks, wireless sensor networks, mobile ad-hoc networks and
vehicular ad hoc networks.

123
A 3-Parameter Routing Cost Function for Improving…

Mohammed Ghanbari is the professor of video networking at the

University of Essex, United Kingdom and Tehran University. He is
best known for his pioneering work on two-layer video coding for
ATM networks (which earned him an IEEE Fellowship in 2001), now
known as SNR scalability in the standard video codecs. He has reg-
istered for eleven international patents on various aspects of video
networking and was the co-recipient of A.H. Reeves prize for the best
paper published in the 1995 Proceedings of IET on the theme of digital
coding. He is the author of eight books and his book: An Introduction
to Standard Codecs, received the year 2000 best book award by IET
Prof. Ghanbari has authored or co-authored about 650 journal and
conference papers, many of which have had a fundamental influence in
the field of video networking. He has served as Associate Editor to
IEEE Transactions on Multimedia (IEEE-T-MM from 1998 to 2004).

123