Token-Based MAC Protocols: Wireless Networks Essay

Abstract

Wireless networks have higher flexibility as compared to wired networks due to the lack of trunking and cabling in the networks. Both desktop computers and laptops can be fitted with wireless cards which enable the machines to be connected to the net. To connect the machines, various protocols are applied to enhance the proper connection of various standalone machines. Wireless networks have their problems at the same time when compared to wired networks.

When wires are used, some protocols like the Ethernet are allowed to use dedicated wires which go hand in hand with the dedicated medium that can avail the transmission and consequent reception of data. This allows the network to be able to transfer data at a higher rate than the wireless networks. In the case of the wireless networks, the radio frequency in which the network is tuned is the only medium. This limits the mechanisms which the network is using for access control (MAC) (Butala, Tong 2005).

In all networks, CSMA (“Collision Sense Multiple Access”) is used to access the network medium. Wired networks use collision sense multiple access with “collision detection (CSMA/CA)” whilst the wireless networks use “collision sense multiple access with collision avoidance” (CSMA/CA) (Ergen, Lee, Sengupta, 2003). The key difficulties of wireless data transmission are the encoding of the data, speed of transmission, access of the hidden nodes through stations (hubs), overcoming of the cyclic transmissions caused by transmission errors and failures. The offered technologies and protocols involve the key solutions of the stated problems, however, some of them originate from other problems. Hence, the paper aims to assess the data transmission protocols and define the weaknesses and strengths of each protocol.

Token-based protocols

“Token-based access control protocols” have been there for some time now. Some protocols that put into practice a comparable access control system are the Frottle which is open source, as well as Wireless Cyclic Token Protocol” (WiCTP), Wireless Token Ring Protocol (WRTP). These protocols generate a master/main node that is used to control the access to the network medium by bringing into play a token that is switched over between the nodes on the particular network.

Client nodes queue the data packets that they are obliged to send. At the time that they receive the tokens, they send the queued packets by the data that is contained within that token. This technique is very straightforward and successfully eliminates the concealed node crisis at the same time as increasing the stability of the network since two nodes cannot move at any one point in time.

System models of the data transmission networks may vary, however, the key aim of the network is to perform the reliable and stable data-transmission process. Because MAC, Data Base Management, Admission Control and Information Access Management are the key aspects of the wireless network creation, most of the existing protocols involve these aspects as the basic features for reliable work. Nevertheless, some of them are not suitable for wide application and data coordination, due to various reasons. These reasons are closely linked with the multitude of parameters that need to be adjusted for any particular data-transmission instance, which is not suitable for everyday use.

Frottle

As in the case of Frottle, expansion of WiCCP seems to have come to an end. The latest release accessible as of 2004 is version 0.5, accessible as both a Linux kernel module and also as a Windows XP driver. Thus, it is further attractive to wide-ranging users seeing that Linux is not required to utilize the protocol. Unfortunately, the Windows driver only sustains the client mode, but this is still a bit more preferable to Frottle, which has many supplementary dependencies.

This is a Linux-only execution of a token-based access control mechanism. It at present relies on the Linux kernel’s IPtables abilities of packet filtering to control the access to the network, and it uses the “Transmission Control Protocol and Internet Protocol” (TCP/IP) stack to be in touch between the master and also the clients. It runs as a “userspace application” and utilizes the IPtables QUEUE rule which is set to queue some packets for transfer (Luo, Lu 2005).

The Frottle wrap-up communicates amid the master and the client nodes using TCP/IP port 999. This is a key disadvantage, in view of the fact that its reliance on both TCP and IPtables confines its portability. Without IPtables, Frottle has no efficient way of controlling the packet queuing, and also without a TCP/IP stack, it cannot correspond between the nodes. For any embedded structure wishing to run the Frottle protocol, they must have the TCP/IP stack and also the IPtables modules which should be compiled into the kernel, which adds to the size of that kernel spectacularly (Aad & Castelluccia 2001).

While TCP/IP would approximately certainly be amassed into any networked structure, IPtables can boost the required sizes by many KBs, a serious distress for almost all embedded systems. The Frottle wrap-up also lets unrestricted right of entry on port 999, a noteworthy weakness. If one more server runs a tune-up on port 999, a client who is part of the Frottle ring can be able to circumvent the “token access control and hence directly access that service.

This can realize a spectacular drop in routine and in cases where that service is heavily in use, allows the “hidden node problem” to reappear. While port 999 is not in use by any typical network service, a client could have the capability to circumvent the “access control” on that network to be able to achieve a higher feat at the expense of the other users on that network. Frottle does not include auto-detection of the master nodes. Thus, each of the nodes must be set individually, and thus have at least one single network crossing point of the same subnet as in the master sequentially to swap over control packets.

This relentlessly limits the flexibility of that protocol, as any protocol with auto-detection of master nodes and no dependence on TCP/IP would efficiently allow numerous subnets on the identical access point. This would add to the number of probable network setups many times over. While Throttle has been on hand since 2003 in the month of August, development of the protocol seems to have come to an end, and at the moment, no new edition has been out (Yi-Sheng and Takawira 2004).

The results of a Frottle execution on a network are also for the most part subjective, with no firm results existing for assessment against the standard CSMA/CA “access control” mechanism. This, shared with its numerous execution disadvantages, leads users to try to find alternatives other than Frottle (Ergen et al 2001).

Because the key problem of Frottle technology is the solution of hidden node parts of the network, it should be emphasized that it is mainly used in widespread WLAN setups with numerous nodes, that are connected by the means of directional antennas and have high upload (the radius of such networks is 50M and wider). Hence, as it is emphasized by Yi-Sheng and Takawira (2008, p. 451): “IEEE 802.11 is suited for bridging the last mile for broadband access only to a very limited extent. Newer standards such as WiMAX assign time slots to individual stations, thus preventing multiple nodes from sending simultaneously and ensuring fairness even in over-subscription scenarios.”

In the light of this statement, it should be emphasized that IEEE 802.11 uses 802.11 RTS/CTS, and this principle is applied in the Frottle protocol as the technology of handshake packets. Actually, the commonly used RTS/CTS protocol is not the perfect solution to the hidden node problem, nevertheless, it is capable to decrease throughput even further. Hence, the only solution is the use of adaptive adjustments and regimes that are not available for the average users for their complexity and absence of a user-friendly interface.

Wireless Cyclic Token Protocol

The projected protocol implements a token-passing access control apparatus, analogous to the mechanisms implemented by Frottle and also WiCCP, called the

“Wireless Cyclic Token Protocol” (WiCTP). WiCTP works by use of passing a token to each node on the particular network. Only the node which is in possession of the token may be able to transfer the data, thus allowing only one of the nodes to transfer at any one point in time, thus getting rid of collisions completely. Each of the nodes on that network is grouped as either master or slave. The master nodes have knowledge about all other nodes on the particular network, but each of the slaves only knows about the masters. The master node reins all access to the entire network, in view of the fact that it handles all the allocation of the tokens to each of the nodes.

This allows the main master node to have a vast suppleness in the control of that network. It can then re-order the process of passing of the tokens to better utilize that network, or provide preferential handling to definite nodes, thus allowing the master node to make sure that all the nodes that in general would be downgraded to lower performance obtain a fair share of the network’s bandwidth (Metropolis et al 2000).

The key aspect of WiCTP network communication is the Carrier Sense Multiple Access with Collision Avoidance technology. This helps to define whether the collision-free transmission is possible on the definite part of the root. Assuming that the network is taken as the distance between A and B points with some intermediary Access Point, the nodes of the network may be hidden or visible. If the nodes are hidden, it is impossible to define whether the node is transferring or not. Hence, if two hidden nodes are transferring to an open node, the data is colliding, causing loss of the information, and collision paradox, when A and B have to retransmit, while neither A nor B know whether the other point is transferring.

WiCTP is aimed at solving this problem by using the Request to Send/Clear to Send (RTS/CTS) protocol. Hence, information is sent in RTS packets with special marking, which points out the required transmission delay. Hence, any station that hears such a signal will prevent transmission for the duration pointed.

While not yet put into practice, WiCTP uses the integration of dissimilar “Quality of Service” (QoS) algorithms. When any node is connecting to a wireless network, it sends a packet broadcasting its presence to the particular network. When the master node receives this packet, it goes ahead and replies to that node with information about the master node with which it has come to be associated with, and also makes that node a slave. The master node adds the new slave’s data to its list of existing slaves and then places it in the lineup for receiving the tokens. When the slaves receive the token, it confirms to see if that token is from the master node to which it has been directly associated.

If so, it will verify for any data that necessitates being sent, and then examines the token to observe how many packets it may propel as allocated by the main the master. The slave then sends its queue of data packets to all the other lined-up recipients.

When the slave has no supplementary data packets to launch, or it has sent the highest figure of packets, it drives the tokens back to the master node. Since only the slaves with the tokens may broadcast the data, the concealed node problem is hence eliminated. No, any two nodes can be able to transfer at the same instant, thus the collisions are eliminated completely (Spyropoulos and Raghavendra 2010).

WTRP

Key features of this protocol are as follows:

• It is regarded as a medium access control protocol for wireless networks.
• It maintains guaranteed quality of the services by controlling the bandwidth and latency values.
• The token defines the transmission order, and each may be abandoned after the specified amount of time
• Each network ring has a unique MAC address
• If owner (the central station) leaves the ring, another station occupies its place.

The Wireless Token Ring Protocol is regarded as the innovative medium access control tool for wireless networks of local area scale. In most researches, it is opposed to IEEE 802.11 networks, as in comparison with this standard, the WTRP provides the reliable connection tools in the terms of latency and bandwidth. This protocol is based on the principle of reducing the retransmissions caused by collisions. By the statement by Ergen et al (2009), the key principles of this protocol are closely linked with the statement that stations prevent cycled transmission if these are filed:

WTRP is a distributed protocol that supports many topologies, as not all stations need to be connected or to a central station. WTRP is robust against single node failures, and recovers gracefully from multiple simultaneous faults. WTRP is suitable for inter-access point coordination in ITS DSRC, safety-critical vehicle-to-vehicle communications, and home networking, and provides extensions to other networks and Mobile IP. (Ergen et.al. 2010, p. 1870)

In the light of this statement, it should be emphasized that the actual benefit of using this protocol is the opportunity to create partial connectivity. Furthermore, the existing connectivity principle is able to create the unique overall architecture of the network system that is featured with the data link layer. Hence, the Mobility Manager and Channel Allocator may be used for Admission Control, as well as Information Base Management. (Metropolis 2000)

Conclusion

From the above methods, it is evident that there are various diverse ways in which Token-based MAC protocols for Supporting Timely Transmission of Real-Time Traffic in Wireless Networks can be enhanced. This includes the use of Wireless Central Coordinated Protocol” (WiCCP), Wireless Cyclic Token Protocol and throttle among others. The methods have been seen to have various advantages and disadvantages which determine the most applicable process.

One of the critical conditions that the implementers of the protocol should focus on is the use of protocols is avoidance of collisions within the data packets in transit which makes the transmission of information a problem. Efficiency of information transfer is the most important aspect in communication between different nodes in any network and thus the protocol used should be of great essence (Yi-Sheng and Takawira 2004).

According to the discussion presented in this paper, networks which use WiCTP are more stable in networks as compared to CSMA/CA protocol. In WiCTP, the error rates in data transfer are predominantly decreased whilst the error rates in CSMA/CA are a bit less than the latter. WiCTP has better throughput and is more stable and it also shares the bandwidth between the nodes in a more stable manner.

References

Aad, I & Castelluccia, C 2001, “Differentiation mechanisms for IEEE 802.11”, in Proc. IEEE INFOCOM’01, 209–218.

Butala, A., Tong, L. 2005 ‘Cross-Layer Design for Medium Access Control in CDMA Ad Hoc Networks’, EURASIP J. Applied Signal Processing, vol. no. 2, pp. 129-143, 2005.

Ergen, D. et al 2001,“WTRP—wireless token ring protocol,” IEEE Trans. Veh.Technol., 53(6):1863–1881.

Ergen, M., Lee, D. Sengupta, R., 2009 ‘WTRP-Wireless Token Ring Protocol’ WTRP Journal. Vol 4; No 6.

Ergen, M., Lee, D. Sengupta, R. 2003 ‘Wireless Token Ring Protocolperformance comparison with IEEE 802.11 Eighth IEEE International Symposium on Computers and Communication’, Antalya, Turkey.

Luo, H., Lu, S. 2005 ‘A Topology-Independent Wireless Fair Queueing Model in Ad Hoc Networks’, IEEE J. Selected Areas in Comm., vol. 23, no. 3, pp. 585-597.

Metropolis, A. W. 2000 “Equations of state calculations by fast computing machines,” J. Gem. Phys., vol. 21(6):1087–1092.

Spyropoulos, A., Raghavendra, C. 2010 ‘A Token-based Greedy Chain Scheduling Algorithm (T-GCSA) for Situation Aware Wireless LANs’ University of Southern California, Los Angeles.

Yi-Sheng, L., Takawira, F. 2004 ‘Token Based Medium Access Control in Wireless Networks’ School of Electrical and Electronic Engineering, University of Natal.

Yi-Sheng, L., Takawira, F., Hong-Jun X. 2008 ‘A Hybrid Token-CDMA MAC Protocol for Wireless Ad Hoc Networks’, IEEE Transactions on Mobile Computing, Vol. 7, No. 5.