Comparative Study of Hidden Node Problem and Solution Using Different Techniques and Protocols

JOURNAL OF COMPUTING, VOLUME 2, ISSUE 3, MARCH 2010, ISSN 2151-9617 HTTPS://SITES.GOOG LE.COM/SITE/JOURNALOFCOMP UTING/ 65 COMPARATIVE STUDY OF HIDDEN NODE PROBLEM AND SOLUTION USING DIFFERENT TECHNIQUES AND PROTOCOLS Viral V. Kapadia, Sudarshan N. Patel and Rutvij H. Jhaveri Abs t ra ct — Hidden nodes in a wireless network refer to nodes that are out of ran ge of other no des or a collection of nodes. We will discuss a few problem s introduced by the RTS /CTS mechanism of collision avoidance and f ocus on the virtual jamming problem, wh i ch allows a malicious no de to effectively jam a large fragment of a wireless network at min imum expense of p ower. We have also discussed WiCCP ( Wireless Central Coordinated P rotocol) which is a proto col booster that also p rovides good solu tion to hidden nodes. Index Terms — Hi dden Term inal Problem , CSMA, Hidd en Termin al, Exposed Terminal, MAC A ——————————  —————————— 1 H IDDEN TERMINAL PROBLEM IDDEN no des are the node s that are no t in the ran ge of other nodes or a group of nodes. Take a physical star topology with an access point with many nodes s u r r o u n d i n g i t i n a c i r c u l a r f a s h i o n : E a c h n o d e i s w i t h i n communica tion range of the access point, bu t the nodes cannot com municate with e ach other as they do not have physical co nnecti on to each ot her. In a w ireless net work, it i s possible that the node at the far edge of the a ccess point's range, known as r, can see the a ccess point, but it is unlikely that the same node can see a node on the opposite end of the access point's range, r2. Th es e nodes ar e known as h idden. The problem is when nodes r and r2 start to send packets simultaneously to the access p oint. Since no de r and r2 cannot sense the carri er, Carrie r Sense Multiple Access with Collision Avoid ance (CSMA/CA) does not work. To overcome this problem, handshaking i s implemented in conjun ction w ith t he CSMA /CA sch eme. The sa me pro blem exists in a MANET [2]. The hidden node pro blem can be observed easily in widespread (>50m radius) WLAN setups with many nodes that use directio nal antennas an d have high upload. This is why IEEE 802.11 is suited for bridging the last mile, for broadband access, only to a very limited e xtent. Newer standards such as Wi-MAX assi gn time slots to individ ual stations, thus preve nting multiple nodes from sending simultaneously and ensuring fairness, even in over- subscription scenarios [2]. IEEE 802.11 uses 802.11 RTS/CTS ackn owledgment and handshake packets to partly overcome the hi dden node problem. RTS /CTS is not a complet e solution and may decrease th roughput even further, but a daptive acknowledgment s from the ba se station can help too. Figure . 1 Hidden Node [3] 2 CARR IER SENSE MULTIPLE ACCE SS (CSMA) In Carrier Sens e Multiple Acce ss: 1. If the channel is idl e then transmit. 2. If the channel for communica tion is free then it is going to transmit wi thout any precaution tha t there might be collision. 3. If the ch annel is busy , wait fo r a ra ndom tim e. 4. Waiting time is calculat ed using Truncated Binary Exponential Ba ckoff (BEB) algor ithm. Fig. 2 Exposed Nodes [ 3] H ————————————————  Viral V. Kapadia 1 i s w i t h t h e D e p artment o f Com p uter En g ineerin g , Birla Vishvakarma Maha vidyalaya, Vallabh Vidya nagar,, Gujara t, India.  Sudarshan N. Patel 2 is with the Department of Computer Engin eering, A .D. Patel Institute of Techno logy, New Vallabh Vidyanagar, Gujarat, India.  Rutvij H. Jhaveri 3 is with the Department of Computer Engineering and Information Techno logy, Shri S’ad Vid y a Man da l In st it ute o f Technolo gy , Bharuch, Gujarat, India. JOURNAL OF COMPUTING, VOLUME 2, ISSUE 3, MARCH 2010, ISSN 2151-9617 HTTPS://SITES.GOOG LE.COM/SITE/JOURNALOFCOMP UTING/ 66 3 H IDDEN TERMINALS The notorious hidden node probl em deals with a configuration of t hree nodes, like A , B , and C in Figure 1, whereb y B is within the transm ission range of A and C , while C is outside the range of A . In a situ ation like this, C will not be able to detect the on going transmission of A to B by carrier sensing and, consequently, it can inadvertently i nterfere with B ’s reception of A ’s packet [1]. The transmiss ion range of a node A is defined as t he area inside which othe r nodes are able to corre ctly receive A’s packets. On the othe r hand, the carrier sen se range of A is t he area encompassing those node s whose transmission A can perceive (carrier sense) while not necessarily being able to receive the transmitted packets [1] . Generally, it is unreason able to assume that the t wo areas are alw ays the sam e, e.g., th e carr ier sense r ange ca n be twice the transmission range [7]. Suppose th at ever y node in Figure 1 ha s the same transmission range (represented by a solid circle). Node C is out of the transm ission range of node A and thus w ould appear as a hidde n node to A . However, if the carrier sense range of C is larger than the transm ission rang e of A (see the dashed circle), C is no mor e hidden beca use it can se nse the transmission of A and thus avoid interfering with it. This mechanism fo r elimi nating the hi dden node prob lem has been described in [7]. 4 E XPOSED TERMINALS In wireless networks, the Exposed Node Problem occurs when a node is prevented fr om sending pa ckets to othe r nodes due to a ne ighboring t ransmitt er. Consider a n example of 4 nodes labeled R1, S1, S2, and R2, where the two receivers are out of range of each other, yet the two tra nsmitters in the middle are i n range of eac h other as shown in Figure 3. Here , if a transmission betw een node S1 a nd node R1 is taking place, node S 2 is prevented f rom transmitting to node R2 as it concludes after carrier sense t hat it will interfere with the transmission by its neighbor node S1. However note that node R2 could st ill receive the transmis sion from no de S2 without interferen ce because it is out of range fr om node S1 [1]. IEEE 802.11 RTS/ CTS mechanism help s to solve this problem only if the nodes ar e synchronized. When a node hears an RTS fr om a neighbor ing node, but not the c o r r e s p o n d i n g C T S , t h a t n o d e c a n d e d u c e t h a t i t i s a n exposed node and is permi tted to tran smit to other neighboring n odes [1]. If the nodes are not synch ronized, the problem may occur that the sender wil l not hear the CTS or the ACK during th e transmission of data of the second sender Figure 4. Fig. 3 Expose d Terminal Problem [3] Fig. 4 RTS/CTS HANDSHAKE with A CK [6] [7] 5 R TS-CTS HANDS HAKE IN ACT ION  A is the source which is in the range of B, D and C.  B is the destina tion which is in the ra nge of A, D an d E.  A is the source which is in the range of B, D and C.  B is the destina tion which is in the ra nge of A, D an d E.  B sends ACK after receiving one data packet.  Improves link reliability u sing ACK Figure 4. 6 M ULTIPLE ACCES S COLLISION AVOIDANCE (MACA)  Uses Request-To-Send (RTS) and Clear To-Send (CTS) handshake to reduce the effects of hidden terminals.  Data transfer dur ation is included in RTS and CTS, which helps other nodes to be silent for this duration.  If a RTS/CTS packet collides, nodes wait for a random tim e which is calcu lated using BE B algorith m. Draw back : Cannot avoid RTS/C TS co ntrol packet collisions . 7 S OLUTIONS The other methods that can be employed to solve hi dden node problem are:  Increase transmittin g power from the nodes.  Use Omni-directional antennas.  Remove obstacles.  Move the node.  Use protocol enhancement softwa re.  Use antenna diversity.  Wireless Central C oordinated Protocol. 7.1 I ncrease Transmitting Power from the Nodes Increasing the power (measur ed in mWatts) of the nodes can solve the hidden node problem by all owing the ce ll around each node to increase in size, encompassing all of the other nodes. This conf iguration e nables the non-h idden nodes to detect, or hear, the hidden n ode . If the non-hidden nodes can hear the hidde n node, the hidde n node is no longe r hidden. Because wireless LANs use the CSMA/CA protocol, nodes will wait for the ir turn befo re communicating with the access point. 7.2 U se Omni-directional Antennas Since nodes using di rectional a ntennas are nearly inv isible to nodes that ar e not positioned in the di rection the antenna is JOURNAL OF COMPUTING, VOLUME 2, ISSUE 3, MARCH 2010, ISSN 2151-9617 HTTPS://SITES.GOOG LE.COM/SITE/JOURNALOFCOMP UTING/ 67 aimed at, directional antennas shoul d be used only for Fig. 5 RTS/CTS HANDSHAKE withou t ACK [6] [7] very sm all networ ks (e.g., dedicated point-to-point connec tions). Use Omni-dir ectional antennas f or widesprea d networks consisting of more than two nodes [2]. 7.3 R emove Obstacles Increasing the power on mobile nodes may not work, if for example, the node that is hi dden is that hiding behind a cement or steel wall pre venti ng communication with other n o d e s . I t i s d o u b t f u l t h a t o n e w o u l d b e a b l e t o r e m o v e s u c h an obstacle, but rem oval of the obstacle is another method of remedy for the hidden node p roblem. Keep th ese types of obstacles in mind when performing a site survey [2]. 7.4 M ove the Node Another method of solvi ng the hidden node problem is moving the nodes so that they can all hea r each other. If it is found that th e hidden node prob lem is the result of a user mo vin g hi s com put er t o an are a th at i s hi dden fro m th e ot her wireless nodes, it may be necessa ry to have that user move again. The alternat ive to forcing users to move is ex tending the wireles s LAN to add proper cover age to the hidden area , perhaps using addi tional access poi nts. 7.5 W iCCP (Wireless Central Coordinated Protocol) WiCCP is a protocol booster for 802.11b DCF based wireless networks that prov ides cyclic token-passing medium access, and scheduled allocat ion of the available network resources, elimin ating the "Hi dden Node" pr oblem. It is a pure kernel implementat ion resulting in high efficiency traffic control. Its not requir ed extra configur ation e.g. static ARP tables or dedica ted routi ng contexts. WiCCP can be use d in fixed wireless network deployments [9]. It is interesting to note under what conditions WiCCP will work, and when it will not work - at least opti mally. WiCCP will outperform systems that do not run it when th e utilization of the b andwidth increases above some high percentage. If we are running standard Ethernet utilization would be about 80%. Ab ove this percentage of utilization , whatever that is , the channel assign ment ability of Wi CCP will allow the utilization t o increase almost to 100% or at least as close as is hum anly possible [9]. Looking on the other end of the scale , standard 802.1 1b will work best w hen the utili zation is low, and the levels are set correctly so that at the a ccess point all power level are the same. Under lo w utilization it is likely that the powe r levels do not affect things too much. The main quest ion is regarding heavy traff ic. WiCCP allows a guarantee of bandwidth for a p articular user and this solution appears to be the co rrect solution for this case to solve this p roblem. The ability to offer a guarantee and then offer mo re on top of t hat where availa ble is worth while. The problem is the overhea d of polling. As the num ber of user s increases, WiCCP wi ll tend to have issues with assigning timeslots to each, ensuring latency. Standa rd 802.11b will ha ve a defin ite advantage when ther e are a lot of statio ns, and very few want to transmit most of the time - Such as 50 lap tops who only check the ir mail onc e every 15 minu tes (without reading), as opposed to 50 users attempting to sur f the web. 8. C ONCLUSION Hidden node problem can be solved b y many means but each solution is for particul ar scenario. Using different techniques like Incr ease Transmitting Power From the Nodes, Use Omni-di rectional antenna s, Remove obstacles, Move the nodes , Use protoco l enhancemen t software, Use antenna diversity, Wireles s Central Coordina ted Protocol etcetera would increase the performance of ad-hoc networks a lot. 9. R EFERENCES [1] Ashikur Rahman , Pawel Gburz ynski, “ Hidden P roblems wi th the Hidden Node Problem”, 2 3rd Biennial Symposiu m on Commun ications [2] Pommer, Herman n,“Roaming zwischen Wi reless Local Are a Networks”, VDM Verl ag, Saarbrücken 2008, ISBN 978-3-836 4-8708- 5 [3] C. Rama Krishna, “STTP on Wireless Commu nication “, 2009. [4] D. Chen, J. Deng, P. K. Varshn ey, “Protecting wirele ss networks against a denial of service attac k based on virtu al jamming”, 9th ACM Annual International Conferenc e on Mobile Computing and Networking (Mobi Com) Poster, 2003. [5] P. Karn, “MACA–a new ch annel access method for packet radio”, 9th Co mputer Netwo rking Conf erenc e on ARRL/CR RL Amat eur Radio, pp. 134–140, 1990. [6] S. Ray, J. B. Carruthers, and D. Starobi nski, “RTS/CTS-induced congestion in ad hoc wirel ess LANs”, WCNC, 2003. [7] K. Xu, M. Gerla, S. Bae, “How effective is the IEEE 802.11 RTS/CTS handshake in ad hoc networks?”, IEEE GlOBECOM, vol ume 1, pp. 17–21, 2002. [8] J. Bellardo, S. Savage,”802.11 deni al of service attack s: Real vulnerabilit ies and pra ctical solu tions”, Proc-S ecurity Symp osium, 2003. [9] “Patras wire less Software solutions”, available at http:/ / patraswireless.net/software.html" Vira l V. K ap adia – Ph.D. candid ate, Lec turer, Departm ent of Com puter Engineering, B irla Vishvakarma Mahavidyalaya , Vallabh Vidyanaga r. He is author of 6 papers, wi th 3 papers in international confe rences and 3 in nationa l confer ences . Sudarsha n N. Patel – Ph.D. candidate, Lecturer , Department of Computer Engineering, A. D. Patel Institute of T e chnology, New V allabh Vidy anagar. The research area of interest a r e Mobile Ad-hoc Netwo rks and Dist ribute d Operatin g Syste m. Rutvij H. Jhaveri – Member of ISTE, Sr . Lecturer , Department of Computer Engineering, Shri S’ad Vidya Mandal Institut e of T echnology , He is author of 5 pa pers, with 1 paper in interna tional conference and 4 in national confe r ences. The research area of inte rest is: “Mobile Ad-hoc Networks”.