QoS Enabled Cross-Layer Multicast Routing over ... - Science Direct

Available online at www.sciencedirect.com Available online at www.sciencedirect.com

ScienceDirect

Procedia Computer Science 00 (2009) 000–000 Procedia Computer Science 125 (2018) 215–227

Procedia Computer Science www.elsevier.com/locate/procedia

International Conference on Smart Computing & Communication (ICSCC) 2017

QoS Enabled Cross-Layer Multicast Routing over Mobile Ad Hoc Networks Dinesh Chandera, Rajneesh Kumarb

a

Research Scholar, CSE Department, M. M. Engineering College, M. M. University, Ambala, India Professor, CSE Department, M. M. Engineering College, M. M. University, Ambala, India, India

b

Abstract Quality of group communication using mobile ad-hoc networks depends on various factors like channel fading, signal quality, path loss, transmission and reception power of mobile nodes, mobility, link life and battery backup. Different layers are involved in communication, but the behavior of these layers is isolated with each other and overall network performance may effect from their operations. A cross-layer approach can extract the critical information from multiple layers which can be further utilized to enhance the overall network performance and Quality of Service (QoS). In this paper, Cross-layer Multicast Routing (CLMR) is introduced to enhance the QoS using a tree-based multicast routing protocol. In order to achieve QoS, optimization of the tree operations and tree management cost has been done. CLMR exploits the functionality of PHY layer, Application layer and Routing Layer for QoS oriented communication. Performance of CLMR is analyzed using Multicast Ad-Hoc On-Demand Distance Vector (MAODV) routing protocol under various parameters, i.e. Throughput, Delay, Packet Delivery Ratio, Link Cost and Energy Consumption. © 2018 The Authors. Published by Elsevier B.V. Peer-review under responsibility of the scientific committee of the 6th International Conference on Smart Computing and Communications. Keywords: MANET; Cross-layer; Multicast; Group Communication; MAODV

1. Introduction Mobile ad-hoc network (MANET) uses low-frequency wireless links for group communication. Due to the low frequency of wireless link data transmission suffers from the behavior of the layers being used for network

a

Corresponding author. Tel.: +91-9896024340; E-mail address: [email protected]

1877-0509 © 2018 The Authors. Published by Elsevier B.V.

Peer-review under responsibility of the scientific committee of the 6th International Conference on Smart Computing and Communications. 1877-0509 © 2018 The Authors. Published by Elsevier B.V. Peer-review under responsibility of the scientific committee of the 6th International Conference on Smart Computing and Communications 10.1016/j.procs.2017.12.030

Dinesh Chander et al. / Procedia Computer Science 125 (2018) 215–227 Dinesh Chander/ Procedia Computer Science 00 (2017) 000–000

216

operations. The traditional layered approach is not supportive enough to achieve QoS due to the inherent nature of the MANET. In the cross-layer technique, information sharing takes place between nonadjacent layers to optimize the overall performance of the network. Following are some issues related to these layers: i) PHY Layer: Nodes consume energy in transmission and reception of the data related to routing information, the user’s data and data related other network operations. In case of excessive energy consumption, the battery may be exhausted during data exchange and thus may result in link breakage. A low battery may affect the wireless transmission range also and drop the ongoing communication. Therefore, it’s required to optimize the energy consumption w.r.t above discussed factors for better and uninterrupted communication [1][2]. ii) MAC Layer: This layer manages the access to the wireless medium and it’s fair utilization. It is also responsible to control the contention and collision level over the shared wireless channel. If the MAC layer fails to manage the designated operations, then a lot of packet retransmission may take place and consume more energy than required. If this fact can be passed on to PHY layer, energy consumption can be optimized. Therefore, a cross-layer solution may be used to optimize the operations for each layer. [3-11]. iii) Network Layer: This layer keeps the track of each link and the data rate required for communication. Due to node’s mobility, the network topology changed frequently that cause the frequent updates in routing information. This frequent updates in routing information may cause frequent link breaks. Route reconfiguring may consume unwanted energy, results in depletion of node’s battery. In order to send their data after route reconfiguring multiple nodes can try to access the channel at the same time, and cause collision over the wireless channel [16] [17]. iv) Transport Layer: It controls the congestion over a network. The congested network may bring down the overall network performance. A cross-layer solution may be used for MAC and transport layer for performance optimization [18][19][20]. By using the cross-layer interaction between layers many QoS parameters like energy, security, tree management cost and various control overhead can be optimized for improved performance. A typical hypothetical cross-layer design can be shown as in figure 1, in which PHY, MAC, and Network layer are exchanging their information to form upper layer information, similarly, Application and Transport layer form a lower layer information. And further these two cross-layer exchange their information to form cross-layer interaction. Application Layer Transport Layer

Cross layer Upper layer information Cross layer information exchange

Network Layer MAC Layer

Cross layer Lower layer information

PHY Layer Fig. 1. A typical cross-layer design

1.1 Common cross-layer solutions  For PHY layer, a cross-layer solution can be used to optimize the power for wireless links in such a way that maximum transmission range can be ensured [1][2].  At the network layer, optimal route selection schemes may be deployed to adopt the dynamic topology [14][15].  Congestion control at end terminals may be used for reliable communication at the transport layer [12] [13]. 1.2 Why cross-layer multicast? Multicasting can reduce the communication cost, but still, there are some issues which may degrade the performance of routing protocol. Depends upon the Routing operations (Tree-based/Mesh-based), the protocol may consume the


217

excessive resources thus cause the performance degradation. Cost of Tree-based routing operations is more expensive as compared to mesh-based operations due to complex multicast tree management iterations which consume multiple resources at the same time. Operations of the other layers are also affected by frequent tree updates. Simple layered architecture cannot support tree construction cost due to the dynamic nature of the MANET. Therefore, a cross-layer solution which can support the optimal tree operations w.r.t network resources as well as performance to meet QoS. For example, if the SNR ratio received from the PHY layer and the interference level from the link layer can be used for controlling transmission at the transport layer and the route selection at network layer can enhance the QoS. Unnecessary tree operations cause extra control overhead, collisions, congestion, routing load and delay over the network [18]. A MapReduce-based technique can also handle the huge volume of data to reduce network congestion and improves performance [36]. Thus the main aim of this paper is to propose a design to improve the QoS by reducing tree management cost and optimizing tree operations. The organization of the paper is the following: Section 2 gives literature review about proposals to achieve crosslayer design; section 3 illustrates the cross-layer proposed scheme; section 4 represents the proposed scheme in pictorial form; section 5 discusses the performance analysis of the proposed system and finally the conclusion and future scope discussed in section 6. 2. Literature review The growing demand for MANET’s applications and its unique characteristics have attracted the attention of many researchers to improve the QoS. Many cross-layer schemes have been proposed so far to optimize the network performance. Few of them have been surveyed here to propose the new design. Medium access control (MAC) and routing enabled cross-layer cooperative transmission (MACR-CCT) [21] scheme uses the MAC and routing layer data for interference management. On the basis of intermediate distance and gain, relays are selected to forward the packets to multiple receivers. As compared to IEEE-802.11, CoopMAC, Reco-MAC, it supports delayed transmission with less error over a given range. Cross-layer caching [22] is a hybrid method which offers variable storage buffer for multiple receivers. The joint coaching method is used to determine the cache data over multiple layers. Acceptable level of cache gain rate is maintained for multiple users. Reliable and Adaptive Multicast (RAMCAST) [23] uses PHY and MAC layer for reliable multicast transmission. Access Points (AP) are selected dynamically by verifying the channel conditions and end users send the feedback of AP and as per the collected feedback, selected data is retransmitted only to avoid the extra control overhead. Network-based Linked Controlled algorithm [24] can reduce the dependency over intermediate nodes by calculating the residual energy, data rate, and transmission power. The network topology is maintained only for the single hop neighbor, to reduce the control overhead. Results show that it can enhance the network performance, lifetime and throughput etc. Fuzzy logic based self-organized multicast routing [25] can manage the network resources with collecting the traffic statistics. Each end terminal adjusts its traffic rate according to the delay and packet delivery ratio is altered accordingly. The cross-layer mechanism is used to cope with congestion, thus results in enhanced network performance and less packet retransmission. Experiments show the compatibility of the proposed scheme with Treebased and Mesh-based routing methods. Multicast communication suffers from the limited energy resources and the optimal utilization. The energy-aware Multicast tree can be constructed using a genetic algorithm (GA) [26], in order to provide reliable transmission over wireless networks. A short interval of the multicast session may cause packet retransmission and it can also increase the multicast tree construction cost in terms of resource utilization. GA selects the nodes having enough energy with QoS enabled links to ensure the maximum lifetime for multicast sessions. Variations in multicast traffic may have an impact on resource utilization and can degrade the performance of the transmitter and receiver by consuming the excessive resources. It can also interrupt the routing decisions. Light tree base logical topology [27] can be used to resolve this issue. The light tree uses direct links for multicast communication and a heuristic method is used to predict the data flow dynamics and prevents the traffic blocking. It offers scalable, reliable and resources efficient multicast communication as compared to light-tree single hop logical tree method.

218


Video transmission [28] over wireless links suffers from the various factors i.e. channel fading, unfair resource allocation, contention and packet loss etc. IEEE 802.11e (EDCA) was developed to resolve these issues, but still, it cannot manage the resource requirements for end users. A cross-layer multicast data mapping and transmission scheduling can ensure the QoS requirements of end users. Experiments show its compatibility with the H.264 library as compared to traditional EDCA. It can be extended to support the QoS for multiple layers over multi-hop networks. P2P based multicast tree construction framework [29] supports real-time video streaming over a scalable network environment. The multicast tree is constructed dynamically and dedicated links are used for data transmission. Backup slots are used to reestablish the broken links, but packet retransmission consumes extra bandwidth. Quality of Experience (QoE) is delivered by managing the level of PSNR for end users. Tree-based SVC video coding reduces the extra control overhead and delay also. This framework can be extended to provide the support for content delivery networks. Each layer consumes network resources as per requirements and layerwise resource consumption cost varies. A cross-layer multivariable cost method [30] can be used to optimize the resources at application/network layer. Cost function deals with different parameters, i.e., Packet Errors, Bit Error, PSNR and Jitter etc. Cross-layer solution based on the combination of PHY layer link adoption and Network Coding can reduce the overall Frame Loss Ratio for real-time video multicast [31]. Feedback is used for channel access and resource management, thus results in the reduction of packet retransmission. SNR ratio is used to determine the packet loss at individual video layer to ensure the QoE for end users.The proposed scheme can be adopted for the Wireless Mobile IPTV operators. Reliable transmission can be achieved through cross-layer based Raptor codes [32]. Forward error correction is used to estimate the channel conditions and MIMO switching is further utilized to maintain the acceptable level of reliable communication. Cross-layer protocol stack [33] offers multi-functions (congestion management, energy optimization, and contention control) for multi-flow ad hoc networks. Proposed solution subdivides the constraint of each layer and distributed computations are used for individual layer optimization. It can be extended to adapt the channel fading and mobility constraints. The quality of real-time transmission can be improved by enhancing the functionality of individual layer [34]. For a MAC layer, an extended modulation method is used, whereas error correction is performed at the application layer and resource optimization is performed at PHY layer. Experiments show that it can adopt the dynamic channel fading environment. Security attack over tree-based routing can interrupt the multicast group communication and it is difficult to detect the minor changes in tree maintenance phase. The attack may be spread over the network through multiple groups. Traditional IDS is not suitable for these attacks, but the combination of machine learning approach [35] can extend the capabilities of IDS. Multiple algorithms can be used with machine learning method which can identify the various symptoms related to the attack. It builds a classification metric for decision making. Results show its performance in terms of low false rejection ratio and less resource consumption. N. Chilamkurti et al. [37] analyzed that DSR has no in-built capability to determine whether a packet loss occurs due to congestion in a network or node failure, cause inefficient utilization of energy. To overcome this issue they extend the DSR with the help of cross-layer design. With this cross-layer design, a route is calculated only if the packet loss occurs due node failure. Simulation results presented that the proposed scheme has 50% lower route calculation as compared to DSR. V. Thilagavathe et al. (96) presented a cross-layer congestion control scheme to reduce the congestion problem in the network. They extract the congestion information from the transport layer and MAC layer to inform the source node about the congestion on selected routes. This scheme is implemented on Ad hoc On-demand Multipath Reliable and Energy-Aware QoS Routing Protocol (AOMP-REQR) and simulation result proved better in terms of PDR, delay and packet loss. 3. Cross-layer Multicast Link aware Routing (CLMR) Cross-layer Multicast Link aware Routing (CLMR) exploits the information from multiple layers, i.e. PHY layer, Application, and Routing layer, to form a multicast group with the consideration of the following parameters:


219

 Signal Strength  Path Loss Factor  Link Life Time  Cost of tree updates The value of signal strength, path loss (PL) factor is collected from PHY layer and it is used by another layer for decision making. Additionally, the cost of tree updates is also considered in this scheme, which incurs extra control overhead at the time of construction of the multicast tree. A node which satisfies over mentioned constraints will be allowed to join the multicast group otherwise, a JOIN request for that particular node will be ignored. Following are the membership criteria. i) Signal strength: If a non-member node wants to join the group, then it should have a better quality of signal strength which can be derived as. Equation: I Ss = (St * G)/dn * Pl Where Ss : Signal Strength St : Packet transmitted with t signal strength dn : Distance between receiver node and the sender node In figure 2 variations in signal strength for an individual mobile node have been shown. Nodes having low signal strength are not eligible for group communication.

Fig. 2. Variations in the signal strength of sender and receiver node(s).

ii) Path loss: Path loss (PL) or path attenuation in wireless communication is a reduction in the power of a link. A link with less power may not be able to deliver a data packet without error within a desired interval of time. The path loss factor of the link between two nodes in MANET is calculated with the help of PHY layer [11]. The received signal strength of route request (RREQ) packet is subtracted from the transmitted power to get the path loss. Equation: II Path Loss (PL) factor measured DB can be derived as PL = 20Log10(d)  20Log10(f)  32.44  G(Tx, Rx) Where, G: Gain (Transmitter, Receiver) d : Distance from the transmitter f: Signal Frequency Tx: Gain by transmitter antenna Rx: Gain by receiver antenna 3.1 Link Life Time It can be defined as the maximum lifespan of a wireless link until a link break occurs, for a particular node (over a specific time interval). It is directly proportional to the quality of wireless link being used. For weak signal strength,


220

it may be of the short interval. It is also affected by different factors also such as Mobility, battery life, antenna direction. Link Stability Factor (LSF) can be defined as: Equation: III LSF   Link  failure /  Ti Where Ti is time interval 3.2 Tree Link cost (TLC) Tree Link cost (TLC) is the cost which is required to maintain a tree link and it can be derived as, Equation: IV TLC  (rB  rq B) / LC Where: • rB : the amount of bandwidth currently in use by existing connections, • rq B : the amount of bandwidth requested by the newly arriving group of participants, • Link Capacity ( LC ): total bandwidth of the link.

3.3 Multicast Tree Dynamics Multicast Tree is constructed to manage the data of a specific group and its member nodes. Multicast Tree may grow and shrink as per the group join/leave operations and it may cause extra control overhead thus results in degraded network performance. Due to dynamic topology, routes may be updated frequently and routing layer does not share this information with the application layer. This communication gap may cause a huge packet drop. i) Tree Update Ratio (Tur): It can be defined as the number of tree updates over a specific interval due to the following Tree management Operations: GROUP JOIN, GROUP LEAVE, TREE MERGE, TREE PURNE. Equation: V  Tur   N tu /  Ti Where Tur : tree updates ratio, N tu : no. of tree updates, Ti : time interval ii) Control overhead (CTR): Ratio of control overhead can be defined as, Equation: VI  CTR   Tur /  Ti

Where, CTR is control overhead, Tur : tree updates ratio, Ti

: time interval

4. Representation of proposed cross-layer scheme

Quality of service in MANET is an essential to achieve for uninterrupted group communication in crucial situations. However, the traditional layered approach is not supportive enough to achieve QoS due to the inherent nature MANET [38]. The proposed scheme is presented here through the flowchart in figure 2. In this scheme PHY layer, MAC layer and Network layer are interacting with each other to form the cost-effective tree and related operation to reduce the control overhead. All the required parameters of consideration are retrieved from the PHY layer to be


221

processed by the MAC layer and the network layer to achieve the QoS oriented service in terms of reduced control overhead and overall optimized group formation. A detailed flowchart of the proposed scheme is given in figure 3.

Start

1 Get Signal Strength using Equation:I

Calculate Node‐>Residual Energy ei

Calculate PL using Equation:II

Calculate LSF using Equation:III

Calculate TLC using Equation:IV

NO

If (Sr,e, PL, LSF, TLC) ==True

Group(Join, False)

YES

If (Ntu> Th) && CTR >Th)

YES

NO

Group(Join, True)

Tree Update==TRUE, Ntu++

Calculate Tur using Equation:V

Calculate CTR using Equation:VI

0 NO

0

If (another Group Join Request==True)

YES

Dinesh Chander/ Procedia Computer Science 00 (2017) 000–000

222

Dinesh Chander et al. / Procedia Computer Science 125 (2018) 215–227

0 NO

If (Group (Join, True))

NO

1

Data Transmission==True

Stop

4.1 Flowchart description

Fig. 3. Proposed cross-layer scheme

Multicast group is formed using Equations I, II, II and IV. The equation I describe that how to calculate the signal strength followed by path loss factor calculations. Link stability factor and tree link cost are calculated using Equations III and IV. The residual energy of each node is also calculated using PHY layer. In figure 4 residual energies of nodes has been extracted from the PHY layer to consider the nodes for further participation of the multicast group.

Fig. 4. Group Membership constraints

After fulfilling the above constraints, a current ratio of tree updates w.r.t. control overhead is calculated using equation V and VI respectively. If these are less than the threshold value, only than group communication is initiated by activating the multicast routes. In figure 4, it has been depicted that node 4, 5 and 6 has not enough battery and signal strength to participate the group formation. In figure 5 we have shown that the nodes 1, 2 and 3 qualified to participate in multicast group formation and route activation due to enough signal strength and battery power.

Dinesh Chander et al. / Procedia Computer Science 125 (2018) 215–227

223

Fig. 5. Multicast route activation for data transmission

If the multicast route is activated successfully, then the application starts the data distribution for current group members.Group joining process is repeated for the individual JOIN request. The multicast route is activated only for the eligible candidates and extra control overhead is controlled by avoiding frequent tree updates. The number of tree updates and its impact over control overhead is also monitored and minimize the operational cost of a tree link. 5. Simulation and performance analysis

The NS-2 simulator is used for performance analysis of the proposed cross-layer scheme CLMR. We have used the MAODV multicast routing protocol as a base to implement the CLMR. Following are the different simulation scenarios: Table 1. Simulation configuration Simulation Parameters Node(s) 30 MAC Protocol 802.11 Antenna Type Omni directional Terrain 1200x1200 Ad Hoc Multicast Routing MAODV Protocol Simulation Time 50 Seconds Group Size 1 Propagation Model TwoRayGround Simulator NS-2 Node’s Speed 120ms Queue Type DropTrail/Priority Queue Initial Energy 10.0j Traffic Type CBR Packet Size 1024 Bytes IFQ Length 50 Simulation Scenario(s) CLMR: Cross-layer Multicast Routing nCLMR: No Cross-layer Multicast Routing

a) nCLMR: No cross-layer Multicast Routing: Performance analysis of MAODV without using CLMR scheme b) CLMR: Cross-layer Multicast Routing: Performance analysis of MAODV using CLMR scheme.


224

51.6 51.4 51.2 51 50.8 50.6 50.4 50.2

nCLMR CLMR

nCLMR

Packet Deliery Ratio

Throughput

In order to analyze the performance of proposed scheme w.r.t various QoS parameter like Throughput, PDR, Routing Load, End-to-End Delay, Mean Data Delay, Tree Update Ratio, Tree management Cost, Ideal Tree Cost and Number of Tree Links are observed.

92 91 90 89

nCLMR

88

CLMR

87 86 nCLMR

CLMR

CLMR

Figure 6 shows the throughput of MAODV with and without CLMR. MAODV without CLMR is less efficient in terms of throughput, while with CLMR scheme its throughput has increased up to a significant level.

Figure 7 shows the PDR of MAODV with and without CLMR. Without CLMR, MAODV has a very less PDR, but using CLMR scheme PDR has shown significant improvement.

2.15 2.14 2.13 2.12 2.11 2.1 2.09 2.08 2.07

nCLMR CLMR

nCLMR

CLMR

End‐to‐End Delay

Fig. 7. Packet Delivery Ratio

Routing Load

Fig. 6. Throughput analysis

14.9 14.8 14.7 14.6 14.5 14.4 14.3

nCLMR CLMR

nCLMR

CLMR

Fig. 8. Routing load

Fig. 9. End-to-End Delay

Figure 8 shows that without using CLMR, MAODV suffered from extra control overhead. While CLMR scheme offers less Routing Load for MAODV.

Figure 9 above shows the end-to-end Delay of MAODV. In case of nCLMR, end-to-end Delay is higher as compared to CLMR.

0.35 0.3 0.25 0.2 0.15 0.1 0.05 0

nCLMR CLMR

nCLMR

nCLMR

1.50E+08 1.00E+08

CLMR

5.00E+07 0.00E+00 CLMR

Fig. 11. Energy Consumption

Figure 11 shows the Energy Consumption of nCLMR is much higher as compared to CLMR.

0.8 0.6 0.4

nCLMR

0.2

CLMR

Ideal‐Tree‐Cost

8

1

6 4

Ideal‐Tee‐ Cost‐nCLMR

2 0

0

0 10 20 30 40 49 nCLMR

CLMR

Fig. 12. Tree Update Ratio

Figure 12 shows the Tree update ratio for nCLMR and CLMR. In case of nCLMR, more tree updations were performed as compared to CLMR. 10 8 6

Tree‐mgmt‐ Cost‐nCLMR

4 2

Tree‐mgmt‐ Cost‐CLMR

0

Fig. 13. Ideal-Tree-Cost

Figure 13 represents variations in Ideal Tree cost for nCLMR and CLMR. In case of nCLMR, there are some variations but after some time interval, it is almost constant. 5 4 3 2 1 0

Tree_Links ‐CLMR Tree_Links ‐nCLMR

0 10 20 30 40 49

0 10 20 30 40 49

Time Interval

Time Interval

Fig. 14. Tree-management cost

Ideal‐Tee‐ Cost‐CLMR

Time Interval

No. of Tree Links

Tree Update Ratio

2.50E+08 2.00E+08

nCLMR

Fig. 10. Mean Data Delay

225

3.00E+08

CLMR

Figure 10 shows the Mean Data Delay of MAODV. In case of nCLMR, it is higher as compared to CLMR.

Tree‐mgmt‐Cost

Energy Consumption


Mean Data Delay

Figure 14 represents Tree-management-Cost of nCLMR and CLMR. Due to excessive tree updates, nCLMR has a high tree management cost and it is also varying after some interval and reaches up to its peak but after that, it becomes constant until the end of the simulation. In case

Fig. 15. No.of Tree Links

Figure 15 represents the number of tree links used by MOADV using nCLMR and CLMR. It can be observed that nCLMR used more tree links as compared to CLMR. It can be observed that in starting of simulation, more tree links are required and in mid-interval,

226


of CLMR, it is constantly increasing and it also reaches to its peak, which is less than nCLMR, and it starts decreasing till the end of the simulation.

nCLMR used excessive tree links as compared to CLMR till the end of the simulation.

6. Conclusion and future scope In this paper, cross-layer multicast link aware routing (CLMR) was introduced. Tree-based multicast routing protocols suffer from the extra control overhead due to unnecessarily tree management operations. Tree management cost also increases due to different factors i.e. mobility, link errors, dynamic topology, channel fading, etc. There is a need to establish highly stable links by considering the above mentioned constraints. CLMR builds the group using member nodes which have higher signal strength and energy levels. A node cannot join the group without satisfying the group membership constraints. NS-2 is used for simulation purpose and simulation results show that without using CLMR, MAODV could not perform well and it has a less throughput and PDR due to extra control overhead. It also suffers from higher end-toend delay and mean data delay. Its energy consumption and tree management costs are also higher. While CLMR reduces the extra control overhead as well as the tree management cost thus results in improved QoS. CLMR show reduced delay, energy consumption with higher throughput and PDR. It can also be observed that the CLMR used fewer links as compared to nCLMR. Finally, it can be concluded that MAODV performed well using CLMR scheme which can be extended for other multicast routing protocols. References [1] P. S. Kalshetti, Sanjay Mahadev Koli, MAC-PHY Layer Optimization for Digital Video Transmission in Wireless Network, International Conference on Advances in Communication and Computing Technologies, IEEE-2014, pp.1-5 [2] J. Park, Yusik Yang, JaekwonKim, APP-MAC –PHY Cross Layer Technique for Robust Video Streaming over Wireless Channels, ICTC2013, pp.197 – 198. [3] Xueyuan Su, Sammy Chan, Masaki Bandai, A Cross-layer MAC Protocol for Underwater Acoustic Sensor Networks, EEE Sensors Journal, 2016, pp.4083 – 4091 [4] D. Singhal, Rama Murthy Garimella, Energy Efficient Cognitive Cross-layer MAC Protocol, ICACCI, IEEE-2015, pp.987 - 992 [5] K. A. Rahman and Kemal E. Tepe, Towards a Cross-Layer Based MAC for Smooth V2V and V2I Communications for Safety Applications in DSRC/WAVE Based Systems, IEEE Intelligent Vehicles Symposium, 2014, pp.969 – 973 [6] Q. T. Hoang, Xuan Nam Tran, Improved Cross-Layer Cooperative MAC Protocol for Wireless Ad hoc Networks, APSIPA, IEEE, pp.1-7 [7] P. S. Kalshetti, Sanjay Mahadev Koli, MAC-PHY Layer Optimization for Digital Video Transmission in Wireless Network, International Conference on Advances in Communication and Computing Technologies, IEEE-2014, pp.1-5 [8] U. F. Abbasi, Azlan Awang, Nor Hisham Hamid, A Cross-Layer Opportunistic MAC/Routing Protocol to Improve Reliability in WBAN, APCC, IEEE-2014, pp.36 – 41 [9] S. H. Lee, Lynn choi, Cross-Layer Route Optimization using MAC Overhearing for Reactive Routing Protocols in MANETs, ICTC, IEEE2013, pp.550 – 555 [10] S. H. Lee, Lynn choi, Cross-Layer Route Optimization using MAC Overhearing for Reactive Routing Protocols in MANETs, ICTC, IEEE2013, pp.550 – 555 [11] Belani S, Kumar P, Gupta H., A Path loss Sensitive Stable Routing Protocol for MANET, International Journal of Computer Applications. 2013 Jan 1;72(8). [12] M. A. Gawas, Lucy J.Gudino, K. R. Anupama, Cross-layer Congestion Aware Multi Rate Multi Path Routing Protocol for Ad hoc Network, ICSC, IEEE, pp.88 - 93 [13] D. Sunithat, А. Nagaraju, G. Narsimha, А Cross-Layer АрргоасЬ Сог Congestion Control in Multihор Mobile Аd Нос Networks, INDIACom, IEEE, 2014, pp.54-60. [14] Samuel Baugh, Gruia Calinescu, David Rincon-Cruz, Kan Qiao, Improved Algorithms for Two Energy-Optimal Routing Problems in AdHoc Wireless Networks, Big Data and Cloud Computing (BDCloud), Social Computing and Networking (SocialCom), Sustainable Computing and Communications (SustainCom) (BDCloud-Social Com-SustainCom), 2016 IEEE International Conference, 2016, pp.509 – 516 [15] Jianhui Lv, Xingwei Wang, Kexin Ren, Min Huang, Keqin Li, ACO-Inspired Information-Centric Networking Routing Mechanism, Computer Networks, Elsevier-2017,In Press [16] P. T. Kalaivaani, A. Rajeswari, Partial Correlation Based Cross Layer Approach with Routing in Wireless Sensor Networks, Wireless Personal Communications, Vol.94(4), Springer-2017, pp 2125–2148 [17] AmrMohamed, Hussein Alnuweiri, Cross-Layer Optimal Rate Allocation for Heterogeneous Wireless Multicast, EURASIP, Journal onWireless Communications and Networking, Vol.2009, pp.16 pages. [18] Ridhima Mehta, D. K. Lobiyal, Cross-layer optimization using two-level dual decomposition in multi-flow ad-hoc networks, Telecommunication Systems, Springer-2017, pp.1-17 [19] S. Hasanpour, M. Hoorfar, A.B. Phillion, Characterization of transport phenomena in porous transport layers using X-ray microtomography , Journal of Power Sources, Vol.353, Elsvier-2017, pp.221-229


227

[20] Mutalifu Abulikemu, Jérémy Barbé, Abdulrahman El Labban, Jessica Eid, Silvano Del Gobb, Planar hetero junction perovskite solar cell based on CdS election on transport layer, Thin Solid Films, Vol.636, Elsvier-2017, pp.512-518. [21] Qilin Wu, Xianzhong Zhou, Fangzhen Ge, A cross-layer protocol for exploiting cooperative diversity in multi-hop wireless ad hoc networks, Wireless Networks, Springer-2016, pp 1–20. [22] Lin Zhang; Zhao Wang; Ming Xiao; Gang Wu; Shaoqian Li, Centralized caching in two-layer networks: Algorithms and limits, IEEE 12th International Conference on Wireless and Mobile Computing, Networking and Communications (WiMob),2016, pp.1-5. [23] Young Deok Park; Seokseong Jeon; Kyungjun Kim; Young-Joo Suh, RAMCAST: Reliable and Adaptive Multicast Over IEEE 802.11n WLANs, IEEE Communications Letters, 2016, Vol.20 (7), pp.1441– 1444. [24] Quoc-Tuan Vien, Wanqing Tu, Huan X. Nguyen, Ramona Trestian, Cross-layer optimisation for topology design of wireless multicast networks via network coding, 39th Annual IEEE Conference on Local Computer Networks, 2014, pp.466-469. [25] Alireza Shams Shafigh, Beatriz Lorenzo Veiga, Savo Glisic, Cross layer scheme for quality of service aware multicast routing in mobile ad hoc networks, Wireless Networks, Springer-2016, pp 1–15. [26] Ting Lu, Jie Zhu, Shan Chang, Longfei Zhu, Maximizing multicast lifetime in unreliable wireless ad hoc network, Springer, Wireless Networks, 2016, pp 1–11. [27] Ashok Kumar Pradhan, Saurabh Keshri Kunal DasTanmay De, A heuristic approach based on dynamic multicast traffic grooming in WDM mesh networks, Journal of Optics, Springer-2017, Vol. 46 (1), pp 51–61. [28] Olfa Ben Rhaiem; Lamia Chaari Fourati, QoS improvement for multicast video streaming: A new cross-layer scheme-based mapping, scheduling and dynamic Arbitration Algorithms for Multicast Scalable Video Coding (CMSAA-MSVC), International Conference on Wireless Networks and Mobile Communications (WINCOM), IEEE-2015, pp.1-6. [29] Muge Sayit, Sercan Demirci, Yagiz KaymakE. Turhan Tunali, Adaptive, incentive and scalable dynamic tree overlay for P2P live video streaming, Peer-to-Peer Networking and Applications, Springer-2016, Vol.9 (6), pp.1074–1088. [30] Tien Anh L, Hang Nguyen Manh Cuong Nguyen, Application-network cross layer multi-variable cost function for application layer multicast of multimedia delivery over convergent networks, Wireless Networks, Springer-2015, Vol.21 (8), pp.2677–2692. [31] Jinwoo KimKyung A KimJin S JeongChung G Kang, Cross-layer design of adaptive modulation and coding for multicast system with random network coding, EURASIP Journal on Wireless Communications and Networking, Springer-2014, pp.1-10. [32] Berna Bulut, Evangelos Mellios, Denys Berkovskyy, Farah Abdul Rahman, Angela Doufexi, Andrew Nix, Cross-layer design of raptor codes for video multicast over 802.11n MIMO channels, IEEE 26th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), 2015, pp.1403-1408. [33] Ridhima Mehta, D. K. Lobiyal, Cross-layer optimization using two-level dual decomposition in multi-flow ad-hoc networks, Telecommunication Systems, Springer-2017, pp.1-17. [34] Jaeyoung Park; Yusik Yang; Jaekwon Kim, APP-MAC-PHY cross layer technique for robust video streaming over wireless channels, International Conference on ICT Convergence (ICTC), IEEE-2013, pp.197 - 198 [35] Menaka Pushpa Arthur, Kathiravan Kannan, Cross-layer based multiclass intrusion detection system for secure multicast communication of MANET in military networks, Wireless Networks, Springer-2016, Vol.22 (3), pp.1035–1059 [36] Singh H, Bawa S. A MapReduce-based scalable discovery and indexing of structured big data. Future Generation Computer Systems. 2017 Aug 31,73:32-43. [37] Chilamkurti N, Zeadally S, Vasilakos A, Sharma V. Cross-layer support for energy efficient routing in wireless sensor networks. Journal of Sensors. 2009 Apr 29;2009. [38] Conti M, Maselli G, Turi G, Giordano S. Cross-layering in mobile ad hoc network design. Computer. 2004 Feb;37(2), pp48-51. [39] Zhang Q, Zhang YQ. Cross-layer design for QoS support in multihop wireless networks. Proceedings of the IEEE. 2008 Jan;96(1), pp64-76.