2015 13th International Conference on Frontiers of Information Technology
Content Centric Networking Approach in Cognitive Radio Ad hoc Networks Rana Asif Rehman
Byung-Seo Kim
Department of Electronics and Computer Engineering Hongik University Sejong City, 337-701, Republic of Korea Email:
[email protected]
Department of Computer and Info. Comm. Engineering Hongik University Sejong City, 337-701, Republic of Korea Email:
[email protected] propose two new models: called datagram and predicate models. Moreover, they also describe the relationship between traditional and content-based networking. For futuristic Internet paradigm perspective, Information-Centric Networking (ICN) [5] is a newly proposed scheme in which communication is receiver-driven and rely on content names instead of their locations. It provides the more efficient and faster routing mechanism specially in peer-to-peer communication without any need of overlay systems.
Abstract—Content Centric Networks (CCN) is recently proposed scheme for next generation Internet communication in which nodes do not need to communicate based on locations information instead they used contents names. In cognitive radio ad hoc networks (CRAHNs), nodes have dynamic network situation due to intermittent channels conditions as well as the presence of primary users activities. Based on IP-based communication approach, it is hard to maintain end-to-end connections among cognitive radio (CR) nodes in CRAHNs. In this paper, by leveraging the benefits of CCN, we propose a new cross layer architecture, named content centric networking for cognitive radio ad hoc networks (CCN-CRAHNs). Proposed scheme is evaluated by using simulation which shows the advantages of CCN-CRAHNs as compare to IP-based CRAHNs in term of average end-to-end delay in cognitive environment.
Many active ICN project like 4WARD [6], CCN [7], COAST [8], COMET [9], PURSUIT [10], CONVERGENCE [11] etc. are currently working in this arena, in which content centric networking (CCN) [7] more attracts the research community, thanks to the simple, due to its uncomplicated communication procedure. It provides built-in mobility support and nodes do not need to perform handover whenever some node travel from one position to another [12]. By using this approach, there is no need to assign IP address to each participating node. Furthermore, there is no any routing loops exists during communication. Early research on CCN is more investigated in wired networks. However, it also shows its fruitful benefits in wireless ad hoc networks [13], [14], [15], [16], [17], [18], [19].
Keywords—Content, Interest, Cognitive Radio, Named Data Networking, Ad Hoc Networks
I.
I NTRODUCTION
FCC report [1] brings a great revolution in the field of dynamic spectrum access (DSA). Now-a-days various intelligent and smart programmable devices have been used to share, sense and utilize the spectrum opportunities (SOPs) [2]. Cognitive Radio (CR) is one of the technology that can adjust their attributes based on operational environment. Two types of users named licensed or primary users (PUs) and unlicensed or secondary users (SUs) formalized the cognitive radio networks (CRNs), in which, SUs utilize the SOPs in the absence of PUs while PUs exploits the spectrum by using fixed spectrum allocation policies [2]. CRNs can also be distributed into further two famous networks named Infrastructure-based cognitive radio networks and Cognitive radio ad hoc networks (CRAHNs). Due to rapid advancement in technology, various devices such as tablets, smart phones etc. have been evolved. As a result, Internet traffic has increased exponentially, specially in the case of video traffics, will be around 80% of the total traffic in 2016 [3]. Now-a-days, users are more interested in fetching their required contents from different online resources like YouTube, Facebook etc. by experiencing less delay as well as decoupling from their locations. IP-based communication is also called location-based communication and has number of issues like data availability, security, location dependent etc., which is not suitable for future communication and applications requirements.
In CRAHNs, CR nodes need to preserved single end-to-end connection in order to communicate and do not fully exploit the broadcast nature of the wireless channels. Due to channel fluctuations, channel errors and unpredictable PU activity, it is hard for CR nodes to hold accurate routing state and machineto-machine connectivity. Therefore, in this paper, by leverages the benefits of simple CCN communication paradigm, we proposed a new fine grained cross layer architecture, named content centric networking for cognitive radio ad hoc networks (CCN-CRAHNs). In CCN-CRAHNs, CR nodes do not need to keep alive machine-to-machine path connectivity in order to communicate specially in the presence of unpredictable channel conditions and PUs activities. In addition, every CR nodes have capability to store the contents, which is further unique characteristic of proposed scheme to reduce the content retrieval latency in the network. The rest of the paper is organized as follows. In Section 2, we describe the proposed architecture in detail. Section 3 provides the simulation results and analysis, and finally the conclusions are presented in Section 4.
Initially, the fundamental concept of content based networking approach is described in [4]. In which, the authors 978-1-4673-9666-0/15 $31.00 © 2015 IEEE DOI 10.1109/FIT.2015.61
314 309
II.
are based on content names instead of their locations, so naming is an indispensable module of the proposed CCNCRAHNs architecture. Basically, proposed scheme uses hierarchy based naming scheme for the sake of simplicity. In this scheme, the content names are represented such as URI (i.e. Content/Broadcasting/Image/V1), which are highly expressive, human-readable, and customized. Other naming schemes like flat naming can also be used. However, it should be persistent and normally naming technique depends upon applications requirements.
P ROPOSED A RCHITECTURE
Fig. 1 shows the proposed cross-layer architecture of CCNCRAHNs. At the bottom, spectrum sensing, sharing and mobility are performed by conventional cognitive medium access control (MAC) and Physical (PHY) layers.
B. Security Mechanism
Fig. 1.
In proposed CCN-CRAHNs architecture, each CR-Data message has a signature over the actual data and its name which is produced by a key information that is included besides the actual data in a packet. In traditional IP-based CRAHNs, we need to secure data plus channels itself as well. On the other side, in proposed scheme, we only need to secure our contents. In CCN-CRAHNs, every CR nodes experience robust security procedures which are decoupling from their locations. Moreover, in CCN-CRAHNs, it is difficult for an adversary or attacker to find some victim based on its location. Therefore, well-known attack such as distributed denial of service (DDoS) is almost impossible in this environment, in which many nasty agent and handler nodes send their traffic to some specific victim nodes based on their location.
Proposed CCN-CRAHNs architecture
However, in this work, our focus is only on upper layers. We introduced a new layer named CCN layer in the stack, in which content names are used for communication instead of IP addresses. In addition, two types of messages named Cognitive Radio Interest (CR-Interest) and Cognitive Radio Data (CR-Data) are proposed for CCN-CRAHNs. Both CRInterest and CR-Data messages contain the following important fields as shown in table I and II. In CCN-CRAHNs, each CR node has two types of tables, named Content Store (CoS) and Unsatisfied Interest Table (UIT). CoS is used for the storage of data contents while UIT keeps tracking of the CR-Interest messages. Proposed CCN-CRAHNs further introduce four main components: CCN-Cognitive Strategy, Security, Caching, Naming. In next sections, each component is described in detail. TABLE I.
C. Caching Mechanism Unlike IP-based CRAHNs, proposed scheme provides innetworking caching mechanism which is one of the unique feature of it. Through this mechanism, the content availability in the network is also increased. In traditional approach, every content packet is fetched based on its source location information. However, in CCN-CRAHNs, if available, the relay CR nodes can also send the desired contents to the requesters. As a result, the content retrieval time is also reduced. Due to rapid proliferation in technology, each communication device has huge storage capacity. In CCN-CRAHNs, all CR nodes also have capability to store contents packets and least recently used (LRU) technique is used for contents replacement in CoS. In addition, different other cache replacement and decision techniques [20] can also be used in this regards.
CR-I NTEREST MESSAGE FORMAT
Field
Description
CR-Interest Name CR-Interest Lifetime Nonce
Requested content name It shows the lifetime of the CR-Interest and specified in seconds Random value use to avoid duplication of CR-Interests
Channel Information
TABLE II. Field
Description Requested content name
CR-Data Length
Length of the requested content
CR-Data Content
Actual data payload
Signature Type
In this section, the core processing of proposed CCNCRAHNs is described in detail. CCN-Cognitive strategy module deals with all the communication mechanisms of the proposed scheme in cognitive environment. CR-Interest and CRData messages utilize the underlying layers information (i.e. primary user activity, channel information etc.). In addition, this module provides the efficient transport and forwarding mechanisms for both CR-Interest and CR-Data messages. CCN-Cognitive strategy contains two further main components, named forwarding and transport which will be explained in detail in next sections.
CR-DATA MESSAGE FORMAT
CR-Data Name
Channel Information
D. CCN-Cognitive Strategy
Contains the PU activity free channels information
Contains the PU activity free channels information That represents the type of signature such as SHA256 etc.
Fig. 2 describes the basic communication process in CCNCRAHNs. In which, consumer node S wants to download data from provider node U. It sends a CR-Interest message towards node U which contains the name of the desired data. Node
A. Naming Mechanism All contents naming related activities are considered in this component. Since the communication in CCN-CRAHNs
310 315
information in its CR-Interest message and broadcast it to all neighbors nodes.
Fig. 2.
Algorithm 1 describes the CR-Interest processing at the relay node. On the reception of CR-Interest message, CR node checks either it is expired or some duplicate message. If it is not, then it further lookups the message in its CoS table. If it exists, then CR node sends data back to the requester which follows the UIT table entries toward the requester node. Otherwise, it further checks its UIT table for any corresponding entry exist in it. If there is any corresponding entry, then it discards CR-Interest message (i.e. CR-Interest already sent by other node). Otherwise, it updates its UIT table by adding incoming CR-Interest information and waits for channel availability. When some free channels are available, then it forwards the CR-Interest message towards neighbor nodes.
An example of communication in CCN-CRAHNs
T acts as relay node and after reception of CR-Interest, it checks its CoS and UIT tables. If there is no any information available, it forwards the CR-Interest message towards node U after adding the corresponding entry in its UIT table. Provider node U sends back the required data towards consumer node S which follows the same UIT table entries. Relay node T also saves the data in its CoS. Furthermore, in future, if some other node R requires the same data, then the node T will fulfill the request by using its CoS.
Algorithm 2 exemplifies the CR-Data processing at the relay node. When CR-node receives some CR-Data message, it lookups in its UIT table for any corresponding entry regarding this CR-Data message. If there is not, then the CR node considers it as unsolicited CR-Data message and discards it. Otherwise, it saves the incoming message in its CoS and looks for any channel availability. If there is some channels available, then it updates the channel information in its CRData message, discards the corresponding entry from UIT, and forwards it to next nodes. 2) Transport Mechanism: Proposed CCN-CRAHNs also follows some transport operations of conventional transmission control protocol (TCP) from IP-based CRAHNs. Due to mobility and dynamic condition of channels, it is possible that some CR-Interest or CR-Data messages may be lost during communication. Therefore, if some CR-Data messages are not received by consumer CR node within some specific time interval (RTO), it is responsibility of the CR node to send again the corresponding CR-Interest messages, if it still needs the CR-Data.
1) Forwarding Mechanism: This module describes the processing of both CR-Interest and CR-Data messages at relay CR node by taking into account the available channels information. Proposed CCN-CRAHNs scheme does not depend upon endto-end forwarding decisions. Instead, CR-Interest and CR-Data messages are forwarded hop-by-hop fashion according to the available SOP information. Reactive flooding-based approach is used in this forwarding process, in which, source node encapsulate the name of the desired content plus signature 311 316
III.
P ERFORMANCE E VALUATIONS
Network Simulator 3 (NS3) [21] is used for simulation purpose and some simulation parameters are shown in table III. Performance of the proposed scheme is compared with IPbased CRAHNs by taking an assumption that all the CR nodes can get information about primary users activities as well as channels by exploiting some centralized database [22]. TABLE III.
S IMULATION PARAMETERS
Parameter
Value
Propagation Delay Model Propagation Loss Model Technology Routing Protocol Area (m×m) Tx Power Content Size Simulation Time CoS Size Nodes Data Rate Replacement Policy Traffic Type Packet Size
Constant speed propagation Friis propagation IEEE802.11g AODV 1000×1000 18 dBm 100-500 packets 600 sec 10,000 packets 100 6 Mbps LRU CBR traffic 1024 bytes
Fig. 3. Average end-to-end delay as a function of the number of content size
IV.
C ONCLUSION
In this paper, we have proposed a new cross layer architecture, named content centric networking for cognitive radio ad hoc networks (CCN-CRAHNs). Proposed CCN-CRAHNs has two unique features as compared to conventional IP-based CRAHNs. First, in CCN-CRAHNs, CR nodes do not need to keep alive machine-to-machine path connectivity in order to communicate specially in the presence of unpredictable channel conditions and PU activities. Second, every CR nodes has capability to store the contents in its CoS which is further unique characteristic to reduce the content retrieval latency in the network. Initially, performance evaluation shows that CCNCRAHNs performs better than traditional IP-based CRAHNs in terms of end-to-end delay.
Initially, we take 10×10 grid topology. In IP-based CRAHNs, AODV is used as main routing protocol. In proactive or geolocation based approaches, nodes need to periodically exchange the information with other nodes. As a result, it makes the routing mechanism worst especially in the presence of PUs and dynamic channel condition in cognitive ad hoc environment. Therefore, we apply reactive message forwarding mechanism in our proposed architecture and also compare the performance with reactive-based routing protocol (i.e. AODV).
As a future work, we will more investigate the performance of proposed CCN-CRAHNs by using real time applications and topologies.
Content store (CoS) utilizes the least recently used (LRU) replacement policy and each CR-node can store up to 10000 data packets. Five different types of contents are used in simulations, in which each content consists of 100 to 500 data packets of size 1024 bytes. Total simulation time is 600 seconds and results are averaged value of 10 times simulation runs. Average end-to-end delay is used as a performance metric to evaluate the performance of both approaches (i.e. CCNCRAHNs and IP-based CRAHNs).
ACKNOWLEDGMENT This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2013R1A1A2005692).
According to Fig. 3, as we increase the number of packets, the average end-to-end delay increases as well in both cases. However, proposed CCN-CRAHNs outperforms as compare to IP-based CRAHNs because in CCN-CRAHNs, CR nodes do not require any end-to-end path establishment as well as maintenance mechanism for communication. On the other hand, in IP-based CRAHNs, before forwarding any data packets, CR nodes need to find and keep active some suitable path from source to destination. Moreover, in the presence of PU activities and channels fluctuations, this condition looks unrealistic and hard to find in dynamic cognitive environment.
R EFERENCES [1] [2]
[3] [4]
[5]
In addition, in IP-based CRAHNs, the delay is high because if some path is broken due to some channel errors, CR nodes again need to perform rerouting, which further exacerbates the network performance. However, in CCN-CRAHNs, communication is totally based on content names despite of their locations.
[6]
312 317
FCC, “Spectrum policy task force report,” ET Docket, pp. 02–155, 2002. I. F. Akyildiz, W.-Y. Lee, and K. R. Chowdhury, “Crahns: Cognitive radio ad hoc networks,” Ad Hoc Networks, vol. 7, no. 5, pp. 810–836, 2009. Cisco, “Cisco visual networking index: forecast and methodology: 2011-2015,” 2012. A. Carzaniga and A. L. Wolf, “Content-based networking: A new communication infrastructure,” in Developing an Infrastructure for Mobile and Wireless Systems. Springer, 2002, pp. 59–68. B. Ahlgren, C. Dannewitz, C. Imbrenda, D. Kutscher, and B. Ohlman, “A survey of information-centric networking,” Communications Magazine, IEEE, vol. 50, no. 7, pp. 26–36, 2012. M. Brunner, H. Abramowicz, N. Niebert, and L. M. Correia, “4ward: a european perspective towards the future internet,” IEICE transactions on communications, vol. 93, no. 3, pp. 442–445, 2010.
[7]
[8] [9] [10] [11] [12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21] [22]
V. Jacobson, D. K. Smetters, J. D. Thornton, M. F. Plass, N. H. Briggs, and R. L. Braynard, “Networking named content,” in Proceedings of the 5th international conference on Emerging networking experiments and technologies. ACM, 2009, pp. 1–12. COAST, “Coast: Content aware searching retrieval and streaming.” http://www.coast-fp7.eu. COMET, “Comet: Content mediator architecture for content-aware networks.” http://www.comet-project.org. PURSUIT, “Pursuit: Pursuing a pub/sub internet.” http://www.fp7pursuit.eu. CONVERGENCE, “Convergence: Host centric to content-centric.” http://www.ict-convergence.eu. D.-h. Kim, J.-h. Kim, Y.-s. Kim, H.-s. Yoon, and I. Yeom, “End-to-end mobility support in content centric networks,” International Journal of Communication Systems, 2014. S.-Y. Oh, D. Lau, and M. Gerla, “Content centric networking in tactical and emergency manets,” in Wireless Days (WD), 2010 IFIP. IEEE, 2010, pp. 1–5. M. Amadeo, C. Campolo, A. Molinaro, and G. Ruggeri, “Contentcentric wireless networking: A survey,” Computer Networks, vol. 72, pp. 1–13, 2014. L. Han, S.-S. Kang, H. Kim, and H. P. In, “Adaptive retransmission scheme for video streaming over content- centric wireless networks,” Communications Letters, IEEE, vol. 17, no. 6, pp. 1292–1295, 2013. M. Amadeo, C. Campolo, and A. Molinaro, “Design and analysis of a transport-level solution for content-centric vanets,” in Communications Workshops (ICC), 2013 IEEE International Conference on. IEEE, 2013, pp. 532–537. G.-H. Jung, W.-S. Jung, Y.-B. Ko, J. hoon Kim, and J.-H. Park, “An efficient name-based loss recovery for wireless content centric networking,” in Computers and Communication (ISCC), 2014 IEEE Symposium on, June 2014, pp. 1–6. M. Amadeo, A. Molinaro, and G. Ruggeri, “E-chanet: Routing, forwarding and transport in information-centric multihop wireless networks,” Computer Communications, vol. 36, no. 7, pp. 792–803, 2013. R. A. Rehman, J. Kim, and B.-S. Kim, “Ndn-crahns: Named data networking for cognitive radio ad hoc networks,” Mobile Information Systems, vol. 2015, 2015. G. Zhang, Y. Li, and T. Lin, “Caching in information centric networking: a survey,” Computer Networks, vol. 57, no. 16, pp. 3128–3141, 2013. NS3, “Network simulator 3,” http://www.nsnam.org. A. Al-Ali, Y. Sun, M. Di Felice, J. Paavola, and K. Chowdhury, “Accessing spectrum databases using interference alignment in vehicular cognitive radio networks,” Vehicular Technology, IEEE Transactions on, vol. 64, no. 1, pp. 263–272, Jan 2015.
313 318
