Smart Computing Review A Review of Techniques to Resolve the Hidden Node Problem in Wireless Networks

Abstract

Recent developments of wireless communication devices have increased the interest in wireless networks. The hidden node problem is one of the major problems which leads to packet dropping and transfer delays via blind collisions. In this paper, we discuss the design factors of some existing mechanisms to deal with hidden node avoidance, and present a timeline of the development of these mechanisms. We classify and characterize the existing mechanisms into three categories, which are handshaking, busy tone multiple accesses, and routing management mechanisms. This classification and characterization provides a better qualitative comparison and presents a clear picture of the strengths and weaknesses of these mechanisms. Finally, we highlight the open issues that still need to be addressed.

Full text

Smart Computing Review, vol. 2, no. 2, April 2012
DOI: 10.6029/smartcr.2012.02.001
95
Smart Computing Review
A Review of Techniques to
Resolve the Hidden Node
Problem in Wireless
Networks
L. Boroumand 1, R. H. Khokhar 1 , L. A. Bakhtiar2 and M. Pourvahab3
1 Faculty of Computer Science and Information Technology University of Malaya 50603 Lembah Pantai,
Kuala Lumpur, Malaysia / laleh.b127@gmail.com, rashid@um.edu.my
2 Faculty of Computer, Islamic Azad University / Arak branch, Iran / leily.b@gmail.com
3 Faculty of Computer, International Pardis University / Guilan, Iran / mehrannet@gmail.com
* Corresponding Author: L. Boroumand
Received January 15, 2012; Revised February 27, 2012; Accepted March 13, 2012; Published April 30, 2012
Abstract: Recent developments of wireless communication devices have increased the interest in
wireless networks. The hidden node problem is one of the major problems which leads to packet
dropping and transfer delays via blind collisions. In this paper, we discuss the design factors of
some existing mechanisms to deal with hidden node avoidance, and present a timeline of the
development of these mechanisms. We classify and characterize the existing mechanisms into three
categories, which are handshaking, busy tone multiple accesses, and routing management
mechanisms. This classification and characterization provides a better qualitative comparison and
presents a clear picture of the strengths and weaknesses of these mechanisms. Finally, we highlight
the open issues that still need to be addressed.
Keywords: IEEE 802.11 wireless networks, collision avoidance, hidden nodes, degradation, quality
of service
Introduction
tilizing wireless devices for various applications increases their sensitivity. Such wireless devices are important
components of many applications. For example, these devices are used for communication in military operations;
the monitoring, tracking, and surveillance of borders; joining or leaving a network at any time in a conference
U

Boroumand et al.: A Review of Techniques to Resolve the Hidden Node Problem in Wireless Networks
96
room; weather forecasting; and ocean monitoring. Recent developments in wireless networks have highlighted the needs of
availability and reliability in this type of network.
The main concerns that lead to negative effects on these types of networks are a low packet delivery ratio and high end-
to-end delays. There are two main factors that cause low packet delivery ratios and high delay, but the one with a hugely
destructive effect on the performance of ad-hoc networks is the hidden node problem. Based on the topology and location
of nodes in networks, nodes can be categorized into two types including hidden nodes (blind nodes) and non-hidden nodes
(visible nodes). The second factor that degrades the performance of wireless networks is the overloading of buffers. It
refers to a situation in which node buffers are full, and the nodes cannot receive any new packets. However, buffer
overloading is beyond the scope of this paper. Our main focus is on the existing mechanisms of hidden node problems in
wireless networks.
Hidden nodes are a fundamental problem that can potentially affect any wireless network where nodes cannot hear each
other whereas they have short distance. In this situation, blind nodes cannot receive any control packets, so packets are sent
to the visible node regardless of any other nodes sending packets, which leads to collisions and packet loss [1-2]. There are
three situations where a node can be a hidden node. In the first situation, which has the worst throughput, all nodes are
hidden. In the second situation, all nodes are visible and contending with each other for resources like Access Points (AP),
so the network has the best throughput. The last situation, called hybrid schema, is one where hidden nodes and contending
nodes appear together [3]. The traffic in each network is classified by Wang et al. [40] into downstream and upstream
traffic. When the AP or sink (in the wireless sensor networks) sends a packet to the terminals, it is considered downstream
traffic, while packets sent in a reverse direction are upstream traffic generated in the network. In downstream traffic there is
no hidden node, because all nodes can hear the access point. But in upstream traffic, the network deals with hidden nodes
because of their different transmission ranges, which is explained with the help of the following example.
Figure 1 illustrates the classic hidden node scenario that consists of three nodes, A, B, and C. In this case, Node A can
hear Node B, but not Node C, as it is not in the same transmission range. Each node that is located in the other node’s
transmission range can receive its packet easily. Therefore, in Figure1 (a), Node B can receive packets from both Nodes A
and C. However, there will be a collision at Node B if these two nodes send their packets at the same time, and Node B
cannot successfully receive any packets. Although, there is another range which is defined for the wireless nodes called
carrier sense range. Each node that is located in the carrier sense range of other nodes can sense node transactions despite
not being able to receive them. Therefore, this range allows for concurrent transmissions that are collision-free [4-6].
Figure1 (b) shows the carrier sense range for Node C. The transmission range and carrier sense range can be similar, but it
is not necessary.
The immense effect of hidden nodes on efficiency, power consumption, transmission delay, and QOS can be considered
the main reason for the effort invested to eliminate them [7]. Recent research [8-11] has shown that more than 40% of
packets are being lost due to the hidden node problem. This percentage increases with an increase in the number of nodes.
Therefore, the network has mutable and unfair throughput, which should be unacceptable, especially for real-time systems
[12-13].
In addition, there is another type of node that causes problems, which is known as the exposed node problem. In this
type of problem, two nodes want to transmit to the same receiver, but one incorrectly concludes that it will interfere with
the other’s transmission, and does not transmit. This problem leads to throughput non-scalability [14]. Consequently, Jiang
and Liew [15] suggested that MAC-designing researchers should consider a trade-off between the exposed node and hidden
node problems, and try to reduce the effect of collision between them. Collisions force the source nodes to resend their

Smart Computing Review, vol. 2, no. 2, April 2012
97
packets repeatedly, which leads to more power consumption in both sender and receiver nodes. Transmission delay refers
to a period of time in which a packet is generated and delivered to the destination. However, no matter what the reason, it
diminishes the performance of the network which causes an increase of power consumption and delay [16].
This paper discusses the design factors of some existing mechanisms that try to deal with hidden node avoidance in
wireless networks. Particularly, the mechanisms which are used to reduce the effect of the hidden nodes are categorized
into three groups: 1) handshaking mechanisms in which the sender and receiver send some control packets through another
channel to each other before transferring information, 2) busy tone mechanism in which the node that has information about
the status of the line tries to avoid collisions by sending control packets to its neighbor nodes, and 3) the routing
management mechanism tackles with the hidden node by choosing a suitable routing path that contains any hidden node.
Section 2 discusses some recently proposed hidden node avoidance mechanisms in wireless networks. This section
includes three main sub-sections, which are handshaking, busy tone, and routing management mechanisms. We have
presented a timeline and comparisons of different hidden node avoidance mechanisms in Section 3. Finally, in Section 4,
the paper is concluded with highlights of open issues.
Hidden Node Avoidance Mechanisms in Wireless Networks
Most of the wireless networks include hidden nodes based on their topology and the node’s transmission range, especially
wireless networks in which nodes are distributed randomly. This kind of node has a negative effect on the throughput and
performance. Therefore, researchers try to find mechanisms for preventing collision that occur due to hidden nodes. Some
of these collision avoidance mechanisms utilize control packets and monitor the status of the channel (busy or free), while
others work based on hidden node detection and store information of the node via a coordinator. In this section, we
investigate the recently proposed solutions which are used for alleviating hidden node avoidance. As stated in Section 1,
these mechanisms are classified into three categories. In the following sections, these three categories are discussed in
detail.
■ Handshaking Mechanisms
Handshaking is known as a common mechanism in wireless communication for avoiding collision in a channel. Generally,
wireless communication uses two types of channels, a data channel, which is used to send and receive the actual data, and a
control channel, which is used by the control packet for managing the connection. The channel division makes
communication more reliable, because the probability of collision is eliminated between the control packet and the data
packet. In the following, the characteristics of some existing mechanisms are discussed which utilize the Request to Send
(RTS) and Clear to Send (CTS) packets used to eliminate the hidden node problem.
■ Multiple Accesses with Collision Avoidance
Multiple Accesses with Collision Avoidance (MACA) [17] uses two fixed-size controlling packets for avoiding collision
occurrences. The complete exchange process involves four packets: RTS, CTS, DATA, and ACK. RTS and CTS are
swapped before data transmission over the control channel. For reserving the medium, a sender transmits an RTS packet.
After receiving RTS, the destination node sends back a CTS packet. The CTS also contains a time value that alerts other
nodes to hold off from accessing the medium while the station initiates the RTS to transmit its data. At the end of the
transfer process, the receiver should return an ACK packet, which is a confirmation that the full amount of data was
received. This mechanism enhances communication by introducing two new parameters such as Inter Frame Space (IFS)
and Network Allocation Vector (NAV), which are used for determining the channel access priority and reserving the
medium. An IFS is a time interval in which nodes cannot transmit any packet. It is used to provide prioritized access to the
channel. Short Inter-Frame Space (SIFS) and Distributed Inter-Frame Space (DIFS) are two types of inter-frame spaces that
are supported by the RTS/CTS mechanism. SIFS is the shortest interval that is used for CTS, DATA, and ACK packets
[18]. An RTS packet can be transferred after waiting for the Distributed Coordination Function/Inter Frame Space (DIFS)
interval time that is longer than SIFS [19]. DCF is based on a CSMA/CA mechanism and a back-off procedure [3, 20].
In the CSMA/CA mechanism before any data transmission, a node broadcasts a signal to the network to tell others not to
broadcast. By broadcasting this signal, the node that wants send data can reserve the channel. In the back-off procedure, the
back-off counter would be decremented one by one in every idle time slot. If the channel becomes busy again in the back-
off procedure, the station would freeze its back-off countdown process and resume it if the channel becomes idle for the
DIFS time. When the back-off counter reaches zero, the station would transmit the frame immediately [21-22]. Network
Allocation Vector (NAV) is a time value that indicates the remaining time of the ongoing transmission that is stored by
each node. In fact, NAV provides some information about the future network traffic [18, 23].
Figure 2 illustrates the complete RTS/CTS mechanism with the help of a simple scenario. Node A as a transmitter starts
broadcasting an RTS packet, and the NAV timer of Node A starts counting. When Node B receives the RTS packet it
should wait for the SIFS time interval then send a CTS packet to Node A. In this step, the NAV timer for Node B should be

Boroumand et al.: A Review of Techniques to Resolve the Hidden Node Problem in Wireless Networks
98
initialized. In the final phase, when Node A transmits the DATA, Node B should send an ACK packet. This packet can
transmit separately or a payload on the other packet for reducing the traffic. The usage of RTS/CTS is helpful, especially
when a data frame is much longer than an RTS frame, because the network faces overhead due to using control packets
[24]. The assumption of getting RTS/CTS by all nodes which are in the same transmission range with the sender or receiver
is not logically correct in the RTS/CTS mechanism. Therefore, the network faces another problem, is called a masked node,
which is discussed next.
■ Multiple Accesses with Collision Avoidance for Wireless
Bharghavan et al., [25] proposed Multiple Access with Collision Avoidance for Wireless (MACAW), which is based on
MACA. MACAW presents a new control packet to alleviate the effects of hidden and exposed nodes. This mechanism uses
an RTS/CTS/Data Sending (DS)/DATA/ACK pattern for data transmission. The new control packet in the MACAW uses
the same mechanism as explained in Figure1. In this scenario, when Node B wants to transmit data to Node A, Node C can
hear the RTS. Node C has no more information about the connection between Nodes A and B because it cannot hear the
CTS from Node A. Therefore, if Node C wants to send data to Node B, it must defer this process. For avoiding an
unnecessary deferral situation, a DS packet is generated by Node B as an announcement for a successful connection.
In addition, another new control packet is proposed in [25], called Request for Request To Send (RRTS). According to
the proposed RRTS mechanism, when Node B cannot reply to the RTS as a receiver due to being in deferral situation, then
in the next connection period, it generates RRTS packet. Upon receiving this packet from Node A, it sends back an RTS
packet. After that the common control packet will be used to establish collision-free communication. In the meantime, each
node that receives the RRTS packet defers its transmission for two time slots. With these two new control packets, the
probability of a collision is reduced. On the other hand, unnecessary deferring is removed due to using a DS. Although
MACAW solves both the hidden and exposed node problems, it’s the number of its control packets increase in comparison
with MACA, and it may create control packet collisions. Also, MACAW, just like MACA, does not consider the masked
node problem.
■ Multiple Accesses with Reduced Handshake
Toh et al., [26] proposed another handshaking mechanism called Multiple Accesses with Reduced Handshake (MARCH),
which can face hidden terminals while exchanging fewer control packets. Figure 3 illustrates a routing scenario using
MARCH, in which Node A wants to send data to Node D while both Nodes B and C are located as forwarder nodes.
MARCH uses an RTS/CTS packet in the first hop, and after that, nodes send only a CTS packet to the next forwarder nodes.
Node A sends RTS to Node B; if Node B is ready to receive data from Node A, it sends back a CTSB (packet sent by Node
B). Meanwhile Node C overhears this packet. A timer is sets with Node C when it receives the CTSB, then after the timeout
of the CTSC timer, it sends to Node B for confirming that the channel is free. Same as before, Node D also overhears the
CTSC (packet sent by Node B), and the previous steps are repeated. All CTS packets in this mechanism involve the MAC
address of the source node, and the route ID is applied for preventing unnecessary CTS propagation via a node which
overhears the CTS message from these nodes, but it does not participate on the routing. Each node compares its RouteID
with the arrival packet. If they are similar, then after the timeout, the node should send CTS; otherwise it continues the

Smart Computing Review, vol. 2, no. 2, April 2012
99
listening channel for having knowledge about the channel status [26]. This mechanism cannot handle a masked node, hence
RTS/CTS/TTM created by Al-Mahdi et al. [12] for combating this problem.
■ RTS/CTS/TTM Mechanism for Overcoming both Hidden and Mask Node
Problems
As discussed above, one of the limitations of the RTS/CTS mechanism is that it cannot handle the mask node issues. The
main question addressed in this portion is the difference between the hidden node and the masked node, and how it can be
used to solve the masked node problem. Unlike the hidden node, the masked node is located in the same transmission range
as the source or destination nodes. It can hear the packets but cannot decode their signal or monitor its channel successfully,
due to the constant transmission of mobile nodes near it [27-29]. Mobile nodes interfere with ongoing transmissions when
they are out of range of the corresponding RTS or CTS. Therefore, these types of nodes can make collisions in the network
and all the concerns appear again that exist in dealing with collisions.
So the optimal solution is to determine the node that causes this problem and prohibit it from communicating with other
nodes in a critical situation. In [12], the authors purposed an RTS/CTS-based mechanism to combat the masked node
problem. In this mechanism, a new control packet is added to the RTS/CTS protocol which is known as Time To Masked
(TTM), that keeps the duration in which the node will be masked. Therefore, three states such as idle, busy, and masked are
considered for each node. When a node finds itself in a masked node situation, it cannot receive any packet from its
neighbors. Furthermore, each node has two buffers, and one of them has an infinite capacity used for data packets. The
other buffer that is used for storing the masked time with finite capacity is called the Stop and Masked (SM) buffer. In
addition, there are two types of TTM packets: Set TTM (S-TTM) and Remove TTM (R-TTM). This mechanism is
illustrated with the help of the following scenario.
As shown in Figure 4, the network has four nodes in the initialization step which are in the idle state. Suppose Node C
wants to send a packet to Node D; C as a sender node should check its SM. If the buffer is empty, it sends an RTS packet;
otherwise it should send RTS with Resume (RTS-R). RTS-R contains both the time to temporary stop reception and DATA
transmission time. The stop time is the period after which Node D stops its reception for interval time temporarily, which
equals the transmission time of a TTM packet, and then resumes its transmission. This will happen in the middle of the
network’s transaction, not during the initialization step, because this is the first communication in the network, and all
buffers are empty. Both Nodes B and D receive the RTS packet because they are in Node C’s transmission range. Only D
should send back the CTS because the RTS message belongs to it. Node D, as a destination, checks its SM and, based on
the buffer’s status, generates either a CTS or CTS with Resume (CTS-R) packet.

Boroumand et al.: A Review of Techniques to Resolve the Hidden Node Problem in Wireless Networks
100
In the following we discuss the trend of Node B when it received the RTS packet from Node C. Node B depicts
transmission time from this packet and sends S- TTM to all neighbors who contain a masked time that is equal to the
transmission time of Node C. Then Node B changes its status to masked, so it cannot hear any signals until it is in this
situation. Neighbors, who gave S-TTM from Node B, put it in the SM buffer as long as receiving R-TTM from Node B.
Considering the previous scenario if the RTS/CTS/TTM mechanism is not used, Node B as a masked node could create
some problem when it receives an RTS from Node C and Node A, while it cannot successfully decode these packets.
■Busy Tone Mechanism
All mechanisms which are part of this section use control packets which are known as Busy Tones (BTs). These type of
control packets are sent by the sender or receiver to their neighbors to reserve the channel during the transmission. Also,
some of these mechanisms such as FPDBT, VPDBT, and ADBT work on the power that should be used to transmit these
control packets. This part starts with Busy Tone Multiple Access (BTMA) as a basic busy tone mechanism that is continued
by other improved mechanisms.
■ Busy Tone Multiple Access
Busy Tone Multiple Access (BTMA) is a protocol that is based on the RTS/CTS mechanism [30-31]. In BTMA, a base
station broadcasts a signal to the other terminals to eliminate the effect of hidden nodes when one station uses a channel. In
fact, with this message, the base station reserves the channel for the sender until the end of transmission. This mechanism
should be changed in ad hoc networks without any central nodes. To optimize the BTMA mechanism, some recently-
proposed protocols are described as follows.
■ Enhanced Busy Tone Multiple Access
In Enhanced Busy Tone Multiple Access (EBTMA) [10], BTMA and MACA are combined to get better performance in
terms of throughput, packet delivery ratio, and average delay. In fact, two channels are used in this mechanism. The first
one is used for transmitting RTS/CTS/DATA/ACK packets, while the second one is used for transferring the busy tone
signals. Furthermore, it uses a back-off counter similar to that used in MACA. The simulation results show that EBTMA
has 0.4 Mbps better throughput rate compared to BTMA. Although this mechanism gains better throughput, combining
BTMA and MACA imposes a cost in hardware.
■ Dual Busy Tone Multiple Access
In this mechanism, Haas et al., [32] proposed two narrow bandwidth tones known as sender’s Busy Tone (BTs) and Busy
Tone which generates via receiver (BTr) which uses a single shared channel. The first one is generated by a transmitter
terminal after sending RTS, and it provides protection for an RTS packet. When a destination node receives an RTS, it
generates BTr that is used for two important goals: acknowledging the RTS and supplying protection for the transmitted
data packet. All nodes sensing any busy tones are not allowed to send RTS requests. In this mechanism, RTS and BTr solve
the hidden node problem.

Smart Computing Review, vol. 2, no. 2, April 2012
101
In [24], Uemura et al. introduced seven states for the nodes in DBTMA mechanism that are: IDLE, CONTEND, S-RTS,
S-DATA, WF-BTR, WF-DATA, and WAIT. The first thing that should be considered in the discussion about the state is
whether the node is the sender or receiver, which is illustrated in Figure 5. In analyzing the sender node, if it is in the IDLE
state at first, then while it senses or receives some data, the state can be changed to CONTEND or S-RTS. In the
CONTEND state, the node is not allowed to send RTS packets due to the channel being reserved by the other node.
Therefore, in this situation the timer plays an important role. If the node face timeout before the channel becomes free, it
should be back to the IDLE state, otherwise the state would change to S-RTS. In this state, the node sends its RTS packet
and selects a new state that is WF-BTR, which indicates the sender is waiting for the BTr packet from the receiver.
When BTr is received by the source node, it modifies its state to WAIT, during which time the node stays for a
mandatory time before sending data. After terminating this time, the data packet will be sent, and the state will be changed
to S-DATA. When the sender gets an acknowledgement from the destination, it goes back to the initial state. Node as a
receiver should wait in the IDLE state until it receives an RTS from the source. Then it should respond with BTr and
change the state to WF-DATA, which means that it is waiting for data. While DBTMA solves both the hidden node and
exposed node problems, it does not have a proper trend in the environment in which nodes have a large interference range.
Mechanisms that can support this type of node are described in the following subsections 2.2.4 to 2.2.7.
■ Fixed Power Dual Busy Tones
In the previous mechanisms, when interference range is larger than the transmission range, then the expected results are
changed. Fixed Power Dual Busy Tones (FPDBT) [33] is proposed that covers the networks which include nodes with large
Interference Ranges (IRs). In FPDBT, when Node A sends RTS to Node B, it compares the power of the signal of the
received packet with the Pthreshold. If this signal power is larger than Pthreshold, it means that the IR is less than the
transmission range. Hence Nodes A and B can communicate with each other based on the normal handshaking protocol.
Otherwise, this mechanism uses dual busy tones and virtual carrier sense (VCS) [36] to combat collisions. In this situation,
Node B sends BTr to Node A after its CTS packet, and Node A replies to it by transmitting BTt until it receives an ACK
from Node B. Both BTr and BTt are sent with maximum power (Pmax ), and all neighbors who sense this signal refrain
their transmission [33]. Utilizing FPDBT can solve the hidden node problem completely, but it makes the exposed node
problem worse. BTt and BTr with Pmax reserve an area which is larger than IR, and some nodes defer their transmissions
without any acceptable reason.

Boroumand et al.: A Review of Techniques to Resolve the Hidden Node Problem in Wireless Networks
102
■ Variable Power Dual Busy Tones
Variable Power Dual Busy Tones (VPDBT) [33] is a scheme proposed to solve the deferment of transmissions without any
acceptable reason in FPDBT. VPDBT begins its busy tones procedure with maximum power, but it modifies this power
during the transmission. Also, the VCS function is replaced with a busy tone sensing mechanism. When Node A wants to
start a transfer process, it checks the channel if it doesn’t sense the busy tones, then transmits the RTS packet. Instantly, it
sends BTt with a maximum power. Now Node A should wait to receive CTS from Node B in the CTS timeout interval. On
the receiver side, after receiving RTS, Node B checks the line and sends both CTS and BTr with PBT power if the line is
free, otherwise it rejects RTS. The PBT is a signal power with destination node that can receive packet successfully. Then
Node B is ready to receive data in the SIFS interval.
Node A sends data and changes the power of the BTt into PBT and waits for an ACK. When it receives the ACK, it
turns off the BTt. During the transmission process, neither sender nor receiver can receive packets, meanwhile the
specification time runs back off procedures. When they assure that the connection has failed, it turns off the busy tone
signal immediately.
■ Asymmetrical Dual Busy Tones
VPDBT improves on the FPDBT mechanism but it does not solve the exposed node problem completely, which has caused
some researchers to propose the Asymmetrical Dual Busy Tones (ADBT) mechanism [34]. CTS and ACK packets are the
main reasons for the exposed node problem, and ADBT modulates these packets on the BTr. Moreover, covering range is
changeable based on the number of interference sources. When Node A does not hear any busy tone signal, it sends an RTS
packet to Node B. Also BTt with the PBTt transmits in the control channel until RTS is received by Node B, then it sends
BTr with PBTt . After finishing the SIFS duration, Node B transmits BT-CTS modulated signal on the BTr and waits to
receive data. Upon receiving this signal via Node A, it starts sending DATA to the destination. A BT-ACK modulated
signal transmits on the BTr once the transfer is successful. Hence, the exposed node problem is eliminated. Additionally,
the number of collisions which happen between different ACK and CTS packets are removed. In the VPDBT, BTt and BTr
cover the IR of the single interference source. In the scenario in which there are multiple interference sources, the
VPDBT’s mechanism cannot combat collisions. ADBT utilizes a new mechanism to solve this problem in which the
Coverage Range (CR) is the same as VPDBT at first. But it recalculates the CR when multiple sources are detected in the
network, and their signal accumulation leads to collision in case of concurrent data transmission. This feature causes ADBT
to be more flexible than VPDBT.
■ Low Energy Wireless Medium Access through Record Keeping
The Low Energy Wireless Medium Access through Record Keeping (LE-WiMARK) solution that is introduced by Kalfas
et al. in [35] alleviates hidden and exposed node effects by recording special information about nodes. Each node sends a
Status Message (STmess) to others which are located in its range. The details of the STmess content is shown in Figure 6.
The preamble field can determine whether or not the node is critical. A node has three statuses in this mechanism, which
are IDLE, Receiving, and Transmitting. If the status of the node changes to receiving it means that the node is in a critical
situation and busy tones should avoid collisions. The second field keeps the MAC address of the node which sends STmess.
The two next fields can be null if the node is IDLE and propagates this message only for announcing its activeness to its
neighbors. If the node plays a role as a receiver, the MAC AddSource field stores the MAC address of the transmitter, while
if it is a sender, the destination’s MAC address is located on the MAC Add Destination. Pf lag will set when this message is
transmitted with maximum signal power. Therefore, when a node receives a STmess, if a power flag is set, then it stores the
max power in its table for the next transmission. By storing and updating the information periodically, nodes can make the
best decision to create a collision-free transmission.
■Routing Management Mechanism
Besides RTS/CTS and BTMA, a network can handle the hidden node problem with another mechanism that manages
routing paths. Some such mechanisms work based on the node grouping while others use beacons to find the best free-

Smart Computing Review, vol. 2, no. 2, April 2012
103
collision path. These mechanisms are investigated in this section. There is a major difference in the grouping strategies due
to the hidden node phase detection based on the type of wireless topology. Some of the mechanisms, which work based on
the star (infrastructure) topology, have a detection phase. The other mechanisms, which are used in peer to peer (ad hoc)
networks, don’t need the hidden node detection phase. Both topologies are discussed in detail in the following subsection.
■ Routing Management with Coordinator
The routing management with coordinator mechanism is proposed in [36]. This mechanism is used for grouping nodes into
four phases. The first one is hidden node situation discovery. The next phase is hidden relationship collection or polling.
Third phase works on the node grouping, and final phase is bandwidth allocation for each group.
Three types of devices - an end device, router, and coordinator - are defined by IEEE standards. A coordinator is
categorized as a fully-functional device (FFDS) that shares the same functionality as a router [37]. It is also responsible for
initializing and recovering network formation, and it helps node grouping via periodically sending a broadcast beacon. A
coordinator also monitors all channels to determine collisions based on their event’s time. After detecting the collision, the
coordinator wants to collect the information of all nodes, then, according to this information, it can classify nodes in
different categories. The information gathering, known as a polling phase, is the phase in which the coordinator starts
transmission by an aware message. This message forces nodes to enter the active mode so the coordinator sends the polling
message for them in turn and waits for an ACK in response. This process is continued by each node, and finally, all nodes
have information about their hidden node. They send their information to the coordinator after receiving a report message
from it, and the coordinator sends them to the hidden node graph. Now, the coordinator should start a grouping process
based on the node’s reports. For ensuring no hidden node exists in each group, nodes that are connected in the hidden node
graph cannot stand in the same category. After grouping, the coordinators allocate a time interval to each group, and in each
interval, CSMA/CA is used as a MAC protocol. By allocating the time interval, it can manage the routing process. After all,
based on the group size, a coordinator should assign specific bandwidths to each group. Whenever the coordinator
determines a collision, the grouping process performs repeatedly.
■ A Hidden Node Avoidance Mechanism
There is another mechanism for grouping nodes in an ad hoc network. It is called the Hidden Node Avoidance Mechanism
(H-NAMe) [3]. This mechanism does not require hidden node detection, and the joining process of nodes to the group has
less complexity compared to the above-mentioned polling mechanisms. Besides all of these advantages, the new
mechanism scales to support multiple clusters of Wireless Sensor Networks (WSN). In multiple cluster networks, each
cluster has bidirectional radio connectivity with other nodes in the cluster. In the initial step, all nodes which belong to the
same cluster share the same contention access period (CAP), therefore hidden collision probability is increased.
H-NAME divides its grouping process into two main phases, which are intercluster grouping and intracluster grouping.
In the first one, nodes are grouped to avoid collisions in the routing process of each cluster. The intercluster step consists of
four phases including join requests, neighbor notifications, neighbor notification reports, and group assignment procedures.
In the first phase, Node A wants to avoid collisions, so it sends a broadcast group join request (GJR) message to the head
cluster. The ACK that is generated by the head cluster is received by all nodes in the cluster, based on the assumption that
all nodes have bidirectional connections. This ACK message has two responsibilities. The first one responds to Node A and
confirms the joining process. Also, it helps the neighbors who receive this packet to add Node A to their table. In the
second phase, all neighbors who are in Node A’s range should send a neighbor notification message (NNM) to Node A
within the timer, which is set via Node A after receiving an ACK from a head cluster. The timer should be long enough to
permute the objective. At the end of this step, Node A has the information of bidirectional neighbors who have already
joined the group. In the third phase, Node A should send a report message about all neighbors to the head cluster. The head
cluster works as a central manager of the cluster and should have detailed knowledge about all groups in the cluster and
their members. This knowledge is collected by all report messages from the nodes. Finally, the head cluster should manage
the number of groups and is responsible for load balancing and bandwidth allocation. Therefore, with intercluster grouping,
the nodes which are in the cluster choose a routing path that does not involve the hidden node. There is no assurance that
data collisions can be avoided when transmitting data between various clusters, so intracluster grouping is necessary. H-
NAME follows a proactive mechanism that avoids hidden node collisions and leads to better performance and throughput.
Besides, against the previous mechanism for grouping, it does not require collision detection, so it is not necessary to
regroup the nodes after each collision occurrs.
In [38], the authors propose an algorithm to divide up the communication area, rather than grouping the nodes. After
partitioning, the nodes are located in places in which collisions never occur. In the partitioning process, the distances
between the nodes, the carrier sense range, and the interference range of each node are considered. This mechanism is not
suitable for WSNs, in which the nodes are distributed randomly.
■Multi-Point Relay -Tree Mechanism

Boroumand et al.: A Review of Techniques to Resolve the Hidden Node Problem in Wireless Networks
104
In [39], a Multi Point Relay (MPR) mechanism is proposed that combines both MACA and grouping mechanisms to avoid
hidden node collisions. In an MPR-Tree mechanism, before data transmission, the source node creates a tree which
includes a set of MPR and normal nodes. Some of the routing protocols which work the base on flooding use MPR nodes as
described in [40]. The MPR-Tree has two pre-phases that it must go through before transmitting data. The first pre-phase
creates a source root tree, and the other uses a handshaking mechanism. A node who wants to transmit data should
broadcast a control message to its one-hop neighbors that contained the ID of the multicast group. This message is sent for
two reasons: to announce new data delivery, and to start the joining tree process. MPR nodes should flood the multicast
control message then neighbors who want to join the tree send back a join message to the MPR parent. Two types of join
messages are declared in this mechanism. One of them is called the MPR join message, which is generated by MPR nodes,
and the other is the normal join message that is produced by non-MPR neighbors. After tree creation, the nodes that
participate to construct the tree can be updated based on the hello packets which are transmitted via neighbors. After all, the
source node should start the data delivery process, but it is not sure about the hidden node problem, because it uses a
handshaking process. The MACA handshaking is used in this mechanism, which has been explained in Section 2.1.1.
■Graph and Beacon Signal Combination:
This scheme [41] works based on the combination of sensing functions and beacon signaling to prevent the participation of
potential hidden nodes in the routing area. Two graphs are used in this mechanism that illustrates the nodes and the
existence links between them. The first graph G(V, E ) shows the nodes which are in the same transmission range and their
links, while the second graph G (V* , E*) shows the nodes and their links which are in the carrier sense range. Besides, a
source node checks its routing table if there is no entry that goes towards the destination. If that is the case, it broadcasts a
Route Request (RREQ) message to its neighbors. RREQ broadcasts by the neighbors continue until one of them reaches the
destination node. Then, this node checks every possible path and selects the best routing path based on the metric. The
destination node sends back a Route Reply (RREP) message in the selection path after Node A transmits data on this link.
In this mechanism, when a forwarded node receives two RREQ which are transmitted by the same node, it drops the
duplicate one. If the sender nodes are different, it forwards the first RREQ message and drops the other messages. Also,
when a node receives the same RREQ in the next time frame, after the transmission of the RREQ packet, it should transmit
a beacon signal to its neighbor. This signal helps to detect the hidden node. The third and last one is the time intervals
which are introduced in MACA used in all transmitting control packets in this mechanism.
■ Distributed Point Coordination Function
The Distributed Point Coordination Function mechanism [42] is a combination of DCF and Point Coordination Function
(PCF), which improves the performance of wireless networks under a heavy traffic load. In PCF, the access point polls the
nodes because there is no contention to use the channel. Despite the collision avoidance in PCF, it is an infrastructure-based
mechanism, and this is the main reason that this mechanism combines it with DCF. In this mechanism, a node can operate
in three modes which are Idle, Master, and Slave. In the initial stage, all nodes are in the idle mode. As shown in Figure7,
when Node B as a destination node receives RTS transmitted by Node A, it changes its mode into a Master and replies to
this packet with a beacon (Be). Each node that receives the beacon should be synchronous with the Master as a Slave.
Therefore, a dynamic cluster is generated in which member nodes can be updated based on the beacon reception. In the
next step, Node B polls a source node, and then Node A should transmit the data packet or a null packet to all Slaves
through a peer-to-peer link. The polled node sends a null packet when it has no data to transmit. Based on the type of
packet which is transmitted by the polled node, the next operation will trigger. If it sends a data packet, a Master Time Out
(MTO) counter should decrement. Otherwise, an inactivity counter should increment. The MTO counter shows the duration
in which a Master node can access a channel without interruption, while the inactivity counter shows when a cluster should
be demolished. In two cases, the Master transmits Cluster End (CE) to the Slaves, such as when the MTO equals zero or the
inactivity counter reaches its threshold value. Each node that senses the CE changes its mode to Idle again. Figure 7
illustrates the DPCF packet exchanging process. Details about all timers that are used in this mechanism are discussed in
[42].
Comparisons of Hidden Node Avoidance Mechanisms
Figure 8 shows a timeline of some recently-proposed hidden node avoidance mechanisms. This timeline makes
mechanisms tracking easier for readers who are interested in investigating these mechanisms in detail. This figure reveals
that most of the mechanisms come from the MACA and use handshaking as a part of their solutions. Busy Tone Multiple
Accesses (BTMA) mechanisms were proposed between 2005 and 2010, and during that time many researchers tried to

Smart Computing Review, vol. 2, no. 2, April 2012
105
improve throughput performance with some modifications. As it can be observed, most of the Routing Management
Mechanisms have been proposed recently.
Table 1 summarizes the characteristics of solutions proposed for the hidden node problem. Particularly, this table shows
which mechanisms offer solutions for hidden, exposed, and masked node problems, whether and how they support large
interference, number of control packets, effect on delay, throughput, and quality of service.

Boroumand et al.: A Review of Techniques to Resolve the Hidden Node Problem in Wireless Networks
106
Handshaking Mechanism: In MACA, the delay is increased in comparison with a scenario that does not take into
account a hidden node. In MACAW, the problem of exposed nodes is eliminated. Therefore, the delay which is produced
by this problem will decrease. On the other hand, the number of control packets is more than the MACA mechanism, and
imposes the delay time to this protocol. Therefore, the amount of delay time depends on density and network topology. The
results of the simulation show that in the MARCH mechanism, when the network is under a high load of traffic, end-to-end
throughput improves more than 60 percent in comparison with MACA. Also, because of the reduction of handshaking
packets in MARCH, there is a lower possibility of collisions between control packets. Therefore, the MACA control
overhead is more than MARCH for all traffic loads. The delay of the MARCH mechanism in the light traffic is more than
the delay of MACA, while if traffic load increases, then the delay has a significant growth in MACA toward MARCH.
In both the RTS/CST/TTM mechanism and RTS/CTS, the collision probability increases when the traffic load increases.
But the probability of this happening in RTS/CTS/TTM is about 0.02 % less than RTS/CTS mechanism over the 0.08%
traffic load. The difference between the probability reaches to 0.03% over 0.24% traffic load. This gain happens when the
size of the RTS packet is equal to 20 bytes and the data rate is 20 packets /second.

Smart Computing Review, vol. 2, no. 2, April 2012
107
Busy Tone Mechanism: BTMA’s effect on throughput is better than MACA’s effect, but a collision still exists in this
mechanism because of the delay in busy tone transmission. In simulation analysis, the throughput of EBTMA is higher than
BTMA. In both mechanisms, by increasing the average number of nodes per hop, the throughput will be decreased. The
delay of EBTMA is less than a conventional RTS/CTS mechanism and by increasing the number of nodes, this parameter
grows. In a simulated environment, when the channel rate and control packet lengths are constant, the utilization of the
network will be increased by adding the nodes in the network in DBTMA mechanism. In other words, in a dense
environment DBTMA has a better utilization toward a MACA mechanism. In a MACA mechanism, by increasing the
number of nodes in the network, the performance will be somewhat permanent. The throughput rate with FPDBT is higher
than VPDBT, and this difference is more intuitive especially in a saturated area, while in the high traffic load, VPDBT’s
performance is better than FPDBT. The throughput of ADBT is about two times VPDBT when the number of transmitted
packets increases. Also, the ADBT’s delay falls remarkably toward VPDBT’s. FPDBT has a negative effect on the exposed
node problem, therefore VPDBT tries to diminish that effect, but it cannot completely remove the exposed node problem.
In simulation results, LE-WiMARK shows better throughput under a high traffic load against the DBTMA mechanism. But
the simulation graph and in low traffic loads, it shows that both mechanisms have similar behavior. The analysis of the
delay versus throughput indicates that, when the throughput increases, the parameter of delay will be incremented in the
DBTMA mechanism. In LE-WiMARK, the growth of the delay parameter is smaller than DBTMA.
Routing Management Mechanism: In a routing management mechanism that works based on a coordinator with a
detection phase, a delay will increase, but throughput also increases also because of the management of the coordinator. A
graph and beacon combination mechanism has a better throughput in comparison with RTS/CTS. However, when the
network becomes dense, the throughput of the network will drop under both mechanisms, because the chance of a collision
will have increased. In the graph and beacon combination, due to creation of a graph model, the delay of route
establishment is greater than a conventional mechanism. But the lower data transmission delay can compensate for the
delay in route establishment. Hence, the graph and beacon mechanism has a smaller delay time in total. In PDCF, both
throughput and delay in the light traffic is the same as the PCF and DCF, but under a high traffic load, DPCF throughput
has an outstanding increment. Finally, delay has a great reduction in comparison with both PCF and DCF.
Conclusions and Open Issues
In this paper, we have presented hidden node avoidance mechanisms in wireless networks. We have classified and
characterized them into three main categorizes: handshaking, busy tone, and routing management mechanisms. We have
also provided a timeline of these mechanisms and summarized them in Table 1. In fact, this Table evaluates these
mechanisms based on their influence on hidden, exposed, or masked node problems, and affect on efficiency and
performance. Most of these mechanisms imposed overhead to the network and we have compared the overhead to the
number of control packets.
This paper has opened the following issues which could create a favorable context to enhance the existing mechanisms
or to create new ones:
1. Routing management mechanisms, which appear more prominent in recent years, try to select the best path which
does not have hidden nodes. Therefore, future research could investigate the probability of the combination of these
mechanisms and handshaking mechanisms.
2. In fact, the routing management mechanisms detect hidden node overlaps by creating routing tables, and it can
increase the speed of data transmission without collisions. The previous mechanisms for grouping nodes work on
clustering and creating trees. One research question that could be asked focuses on the mechanism which can be used
to classify the nodes in order to get the best results. It would be better if future work could focus on static or dynamic
backbones to better combat the hidden node problem. Nodes which are members of the backbone store the node’s
information and forward data through a free-collision path.
3. Additionally, in wireless networks in which nodes can move, hidden nodes cannot be successfully predicted, because
hidden nodes change for mobile nodes. Therefore, researchers should work on dynamic hidden node detection,
especially for wireless sensor networks. Because of the node characteristics in the WSNs, the whole network faces
some limitations which can affect routing protocols. Hence, routing management mechanisms that are used for hidden
node collision avoidance cannot be similar to previous mechanisms.
4. Timers have a key role in all mechanisms. Some of these mechanisms focus on the details of timers but others do not.
So for future work, a researcher can concentrate on timers to reduce overhead and latency.
References
[1] M. Luo, J. Hsu, S. Mo, R. Ghanadan, “Distributed medium access control for multiple hop tactical networks,” in Proc.
of IEEE Military Communications Conference, pp. 107, Nov. 2008.

Boroumand et al.: A Review of Techniques to Resolve the Hidden Node Problem in Wireless Networks
108
[2] H. Khalife, N. Malouch, “Interaction between hidden node collisions and congestions in multihop wireless ad-hoc
networks,” in Proc. of IEEE International Conference on Communications, pp. 3947-3952, 2006. Article
(CrossRefLink)
[3] O. Ekici, A. Yongacoglu, “Fairness and throughput performance of infrastructure IEEE 802.11 networks with hidden-
nodes,” Physical Communication, vol. 1, no. 4, pp. 255-265, Dec 2008. Article (CrossRefLink)
[4] A. Rahman, P. Gburzynski, “Hidden Problems with the Hidden Node Problem,” in Proc. of 23rd Biennial Symposium
on Communication, pp. 270-273, 2006. Article (CrossRefLink)
[5] L.B. Jiang, S. C. Liew, “Removing hidden nodes in ieee 802.11 wireless networks,” in Proc. of Vehicular Technology
Conference, 2005.
[6] F. Liu, J. Lin, Z. Tao, T. Korakis, E. Erkip, S. Panwar, “The hidden cost of hidden terminals,” in Proc. of IEEE
International Conference on Communications (ICC), pp. 1-6, May 2010 .
[7] A Koubaa, R. Severino, M. Alves, E. Tovar, “H-name: A hidden-node avoidance mechanism for wireless sensor
networks,” Network, 2009.
[8] C. Lim, H. Luo, C-o. Choi, “Rain: A reliable wireless network architecture,” in Proc. of 14th IEEE International
Conference on Network Protocol, pp. 228-237, Nov. 2006. Article (CrossRefLink)
[9] A. Koubaa, R. Severino, M. Alves, E. Tovar, “Improving quality-of-service in wireless sensor networks by mitigating
hidden-node collisions,” IEEE Transactions on Industrial Informatics, vol. 5, no. 3, pp. 299-313, Aug. 2009. Article
(CrossRefLink)
[10] Z.J. Haas, J. Deng, F. R. Hall, “Dual busy tone multiple access DBTMA - performance results,” in Proc. of Wireless
Communication and Networking conference (WCNC), pp. 1328-1332, 1998.
[11] F.Y. Hung, I. Marsic, “Performance analysis of the IEEE 802.11 DCF in the presence of the hidden stations,”
Computer Networks, vol. 54, no. 15, pp. 2674-2687, Oct. 2010. Article (CrossRefLink)
[12] A. Iyer, C. Rosenberg, “Understanding the key performance issues with MAC proto- cols for multi-hop wireless
networks,” Wireless Communications and Mobile Computing, vol. 6, no. 6, pp. 745-760, Sep. 2006. Article
(CrossRefLink)
[13] Y. Wang, N. Yan, T. Li, “Throughput analysis of IEEE 802.11 in multi-hop ad hoc networks,” in Proc. of Wireless
Communications, Networking and Mobile Computing (WiCOM), pp. 1-4, Sep. 2006.
[14] Z.J. Haas, J. Deng, “Dual busy tone multiple access (DBTMA)-a multiple access control scheme for ad hoc
networks,” IEEE Transactions on Communications, vol. 50, no. 6, pp. 975-985, Jun 2002. Article (CrossRefLink)
[15] L.B. Jiang, S. C. Liew, “Improving throughput and fairness by reducing exposed and hidden nodes in 802.11
networks,” IEEE Transactions on Mobile Computing, vol. 7, no. 1, pp. 34-49, Jan. 2008. Article (CrossRefLink)
[16] S. Ray, D. Starobinski, J. Carruthers, “Performance of wireless networks with hidden nodes: a queuing-theoretic
analysis,” Computer Communications, vol. 28(10):1179–1192, June 2005. Article (CrossRefLink)
[17] P. Karn. “Macaa new channel access method for packet,” in Proc. of Computer Networking Conference on
ARRL/CRRL Amateur Radio, pages 134140, 1990.
[18] M. Fatima, M. Maaza, K. Said, “Techniques of detection of the hidden node in wireless ad hoc network,” Lecture note
in Engineering and computer science, vol. 2191, pp. 1802-1805, 2011.
[19] C. Yu, S. Hall, B. Lee, O. Hall, “Medium access control mechanisms in mobile ad hoc networks,” Mobile Computing
Handbook CRC Press, 2004.
[20] J. Peng, L. Cheng, B. Sikdar, “A wireless MAC protocol with collision detection,” IEEE Transactions on Mobile
Computing, vol. 6, no. 12, pp. 1357-1369, Dec. 2007. Article (CrossRefLink)
[21] A. T. Barry, M. Veres, A. Campbell, L. Sun, “Supporting differentiation in wireless packet networks using distributed
control,” IEEE J-SAC, vol. 19, no. 10, pp. 2081-2093, 2001.
[22] S. Leng, L. Zhang, Y. Chen, “IEEE 802.11 MAC protocol enhanced by busy tones,” in Proc. of IEEE International
Conference on Communications, vol. 5, pp. 2969 -2973, May 2005.
[23] M Najimi, “A hidden node aware network allocation vector management system for multi-hop wireless ad hoc
networks,” in Proc. of International Conference in Communication Engineering, pp. 32-36, 2010.
[24] W. Uemura, M. Murata, “A cooperative broadcasting method for a sensor network,” International Journal of Ad hoc,
Sensor & Ubiquitous Computing (IJASUC), vol. 2, no. 2, pp. 1-10, 2011.
[25] K. Liu, S. Leng, H. Fu, L. Li, “A novel dual busy tone aided mac protocol for multi-hop wireless networks,” in Proc.
of IEEE International Conference on Dependable, Autonomic and Secure Computing (DASC ’09), pp. 373-378, Dec.
2009.
[26] C.-K. Toh, V. Vassiliou, G. Guichal, C.-H. Shih, “March: a medium access control protocol for multihop wireless ad
hoc networks,” in Proc. of 21st Century Military Communications Conference, pp. 512-516, 2000.

Smart Computing Review, vol. 2, no. 2, April 2012
109
[27] S. Ray, J.B. Carruthers, D. Starobinski, “Evaluation of the masked node problem in ad-hoc wireless LANs,” IEEE
Transactions on Mobile Computing, vol. 4, pp. 430-442, 2005. Article (CrossRefLink)
[28] J.L. Sobrinho, R. de Haan, J.M. Brazio, “Why RTS-CTS is not your ideal wireless LAN multiple access protocol,” in
Proc. of IEEE Wireless Communications and Networking Conference, pp. 81-87, 2005. Article (CrossRefLink)
[29] S. Parvin, T. Fujii, “Hidden node aware routing method using high- sensitive sensing device for multi-hop wireless
mesh network,” EURASIP Journal on Wireless Communications and Networking, Sep. 2011. Article (CrossRefLink)
[30] H. Al-Mahdi, M.A. Kalil, F. Liers, A. Mitschele-Thiel, “Collision reduction mechanism for masked node problem in
ad hoc networks,” AEU - International Journal of Electronics and Communications, vol. 63, no. 9, pp. 754-776, Sep.
2009. Article (CrossRefLink)
[31] C. Wu, V. Li, “Receiver-initiated busy-tone multiple access in packet radio networks,” in Proc. of the ACM Workshop
on Frontiers in Computer Communications Technology, vol. 17, no. 5, 1988. Article (CrossRefLink)
[32] A.A. Abdullah, Lin Cai, F. Gebali, “Enhanced busy-tone-assisted mac protocol for wireless ad hoc networks,” in Proc.
of 72nd IEEE Vehicular Technology Conference, pp. 1-5, Sep. 2010. Article (CrossRefLink)
[33] L. Aceves, J.J. Garcia, C.L. Fullmer, “Performance of floor acquisition multiple access in ad-hoc networks,” in Proc.
of Computers and Communications, pp. 63-68, Jul. 1998.
[34] B. Shrestha, E. Hossain, S. Camorlinga, “IEEE 802.15.4 MAC with GTS transmission for heterogeneous devices with
Application to wheelchair body-area sensor networks,” IEEE transactions on Information Technology in Biomedicine,
vol. 15, no. 5, pp. 767-77, Sep. 2011. Article (CrossRefLink)
[35] G. Kalfas, G.I. Papadimitriou, P. Nicopolitidis, A.S. Pomportsis, “Le-wimark: An intelligent power-efficient ad hoc
mac protocol with busy tone and power control,” in Proc. of Wireless Pervasive Computing, 2008.
[36] L.J. Hwang, S.T. Sheu, Y.Y. Shih, Y.C. Cheng, “Grouping strategy for solving hidden node problem in IEEE
802.15.4 lr-wpan,” in Proc. of IEEE international conference on Wireless Internet, pp. 26-32, Jul 2005.
[37] V. Bharghavan, A. Demers, S. Shenker, L. Zhang, “MACAW: a media access protocol for wireless LAN’s,” in Proc.
of the Conference on Communications Architectures, Protocols and Applications, pp. 212-225, 1994. Article
(CrossRefLink)
[38] Z. Yu, W. Feng, W. Zhang, L. Zhang, Y. Ren, “Virtual sieve: A distributed channel assignment algorithm resolving
the hidden node problem,” in Proc. of International Conference on Communications and Mobile Computing (CMC),
pp. 445-448, Apr. 2011. Article (CrossRefLink)
[39] C-Y. Oh, J. Park, J. Ahn, M. Lee, T-J. Lee, W. Cha, P. S. Mah, “Tree-based multicast protocol using multi-point
relays for mobile ad hoc networks,” in Proc. of International Conference on Ubiquitous and Future Networks
(ICUFN), pp. 174 -178, June 2011.
[40] C. Wang, B. Li, K. Sohraby, M. Daneshmand, Y. Hu, “Upstream congestion control in wireless sensor networks
through cross-layer optimization,” IEEE Journal on Selected Areas in Communications, vol. 25, no. 4, pp. 786-795,
May 2007. Article (CrossRefLink)
[41] R. Bhakthavathsalam, “A new paradigm of NAV alteration to overcome masked node problem in 802.11 MAC based
MANETs,” in Proc. of International conference on Wireless Network, pp. 319-327, July 2008.
[42] J. Alonso-Zrate, C. Crespo, Ch. Skianis, L. Alonso, Ch. Verikoukis, “Distributed point coordination function for IEEE
802.11 wireless ad hoc networks,” Ad Hoc Networks, vol. 10, no. 3, pp. 536-551, May 2011. Article (CrossRefLink)
Laleh Boroumand Jazi received her B.S. in hardware computer engineering from Azad
University of Najafabad in Iran. She received her DNIIT in network engineering from NIIT,
Iran. She works at ACECR in Isfahan University of Technology (IUT) as an NIIT consoler and
lecturer. She works on different subjects like mobile cloud computing, sensor networks, and
computer architecture.

Boroumand et al.: A Review of Techniques to Resolve the Hidden Node Problem in Wireless Networks
110
Rashid Hafeez Khokhar is working as a Senior Lecturer in the Faculty of Computer Science
and Information Technology, University of Malaya in Malaysia. He received his Ph. D and M.S
(by research) in Computer Science from Universiti Teknologi Malaysia in Malaysia. He did his
M.S. (Information Technology) and M.S. (Statistics) from Preston University in Pakistan and
University of the Punjab in Pakistan, respectively. He is serving as a member of the editorial
board of 2 international journals, and as a reviewer for 8 international journals and 7
international conferences. His current research interests include Mobile Cloud Computing,
Vehicular & Mobile Ad Hoc Networks, Artificial Intelligence, and Real-Time Communications.
Leily A. Bakhtiar received her B.S. in hardware computer engineering from Azad University
of Najaf Abad and her M.S. in Computer Architecture from Azad University of Arak, Iran in
2009 and 2011, respectively. Her research interests include computer architecture and wireless
ad hoc and sensor networks.
Mehran Pourvahab received a bachelor’s degree in Computer Software Engineering from the
Islamic Azad University of Bafgh Branch in Iran, and he is now a Master’s student in
Information Technology Networking at the International Pardis University of Guilan in Iran.
His current research interests include Network Security, Wireless Networks, Web
Programming, Network Communications, and Databases.
Copyrights © 2012 KAIS