Analyzing the hidden-terminal effects and multimedia support for wireless LAN
Introduction
In recent years, there has been an increasing trend towards personal computers becoming “portable” and “mobile”. In order to offer broadband communication services and to provide universal connectivity to mobile users, it is important that a suitable standard for wireless local area networks (WLAN) be designed, and an approach to interconnect these WLANs to the existing wired local area networks (LANs) and broadband networks be developed.
To satisfy the above needs of wireless data networking, Study Group 802.11 was formed under IEEE Project 802 to recommend an international standard for wireless LAN [9]. The key objectives include: (1) A WLAN should appear at the logical link control layer (LLC) as just another 802.xLAN (e.g. Ethernet and token ring); (2) the response time should not be so large that the productivity of end users is compromised.
Meanwhile in Europe, ETSI (Europe Telecommunication Standardization Institute) also formed a group called HIPERLAN to study a standard for WLAN, which is different from what IEEE was doing. The HIPERLAN is targeted to a more ambitious goal to support much higher bandwidth than IEEE 802.11 LAN [14].
There has been several works reported in the literature about the performance of IEEE 802.11 and the HIPERLAN [1], [2], [5], [17], [6], [7], [13], [14]. Most of them were on the simulation study. Some of them include analysis of capacity [2] and stability [1]. None of them, however, had a detailed analysis of hidden-terminal effect. Hidden-terminal problem occur when two stations are within the receiving area of each other, yet they cannot hear each other due to signaling loss, interference, multi-path effect, or relay fading. Hidden terminal is a significant problem in wireless and mobile communications.
In this paper, we first give a formal analysis of hidden-terminal effect of HIPERLAN. We derive network throughput under the influence of hidden terminals, and show that throughput is reduced by more than the percentage of hidden probability. We believe that this is the first work in the literature that provides a formal analysis of hidden-terminal effect of WLAN.
In the second part of this paper, we present results of a rigorous set of simulation experiments comparing the two WLAN protocols. These include the following: (1) evaluation of hidden-terminal effects, partly to verify the analysis result; (2) performance study of real-time traffic supports. Both sets take into consideration of various parameters include frame length, traffic load, and measure network throughput, voice delay distribution, and bandwidth left for data traffic.
The paper is organized as follows. Section 1.1 gives a survey of related work. Section 2 presents an overview of the two WLAN protocols. The analysis of HIPERLAN is presented in Section 3. Section 4 describes the simulation model and settings. 5 Simulation results on hidden terminal effect, 6 Simulation results on real-time communication support study and compare the two protocols, by simulation, on their hidden-terminal effects and real-time traffic supports, respectively. Finally, Section 7 concludes the paper.
On the IEEE 802.11 MAC protocol, Chhaya and Gupta conducted performance study of the basic and RTS/CTS schemes, taking into account the decentralized nature of communications, and hidden-terminal effects [5], [17]. Crow et al. [6], [7] investigated through simulations, the DCF and PCF functions of the IEEE 802.11 MAC, in supporting of both time-sensitive packet voice and non-real-time packet data traffic. Cali et al. [2] analyzed the IEEE 802.11 MAC protocol capacity, and suggested a protocol enhancement by properly tuning the backoff algorithm.
On the HIPERLAN MAC protocol, Wilkinson et al. [14] reviewed the process of ETSI RES10 on standardizing the protocol. Anastasi et al. [1] formally analyze the stability and performance of HIPERLAN [1]. Nenonen and Mikkonen [16] compared the performance of HIPERLAN and the Modified MDR by simulation. Weinmiller [13] evaluated the performance of HIPERLAN and IEEE 802.11 by varying station number and packet size.
All of the above studies, except [1], [2], focused on simulation study. There has been no formal analysis of hidden terminal effect on either of the two protocols appeared in the literature. Moreover, none of the simulation study of HIPERLAN has explicitly showed the delay distribution of voice packets, or the data throughput achievable in the integrated service of voice and data communication. These and related results are reported in this work.
Section snippets
IEEE 802.11 protocol description
The IEEE 802.11 committee has been working on the establishment of a standard for wireless LANs. A detailed description of specifications on different layers was presented in [9]. The standard is concentrated on multiple physical layers, and the common medium access control sub-layer.
Analysis of hidden-terminal effect
In this section, we first formally derive network throughput of HIPERLAN when hidden probability is non-zero. Hidden-terminal effect of IEEE 802.11, including both basic and RTS/CTS scheme, are then discussed.
Settings and assumptions for IEEE 802.11
In our simulation, we accepted the parameter values and some assumptions used in Ref. [3]. Two different simulation models are presented. One is ad hoc network that offers only asynchronous data transmission service, and the other is an infrastructure network that offers time-bounded services. Both simulation models are implemented using the physical layer parameters specified in the standard for the DSSS (Direct sequence spread spectrum) implementation. To simplify the complexity of simulation
Simulation results on hidden terminal effect
We evaluate the hidden terminal effect on (A) throughput using basic scheme, (B) when varying frame size, only basic scheme, and (C) RTS/CTS.
Simulation of IEEE 802.11
We evaluate how well the MAC protocol can support real-time (mainly voice) communication, considering (A) effect of payload size, (B) effect of K, (C) number of stations supported, and (D) bandwidth left for data traffic.
(A) Effect of payload size
In Fig. 14, we see that over 99% of the voice packets’ delays are within 500 ms when payload size is equal 100bytes or above. For payload is 50 bytes, over 2% of packets have a delay longer than 500 ms, i.e., these packets do not meet the timing
Conclusion
We have formally analyzed the hidden-terminal effect on HIPERLAN, and have studied and compared two WLAN MAC protocols: IEEE 802.11 and HIPERLAN, including the effect of hidden-terminal and their support of real-time traffic. We explicitly formulated network throughput in terms of hidden probability and other network parameters. We found that, when hidden probability is greater than zero, the achievable throughput is reduced by more than the percentage of hidden probability.
By simulation, we
References (13)
- G. Anastasi, L. Lenzini, E. Mingozzi, Stability and performance analysis of HIPERLAN, in: Proceedings of IEEE Infocom,...
- F. Cali, M. Conti, E. Gregori, IEEE 802.11 Wireless LAN: capacity analysis and protocol enhancement, in: Proceedings of...
Medium access control of wireless LANs for mobile computing
IEEE Network
(1994)- H.S. Chhaya, S. Gupta, Performance of asynchronous data transfer methods of IEEE 802.11 MAC protocol, IEEE Personal...
- B.P. Crow, I. W, J.G. Kim, P. Sakai, Investigation of the IEEE 802.11 medium access control (MAC) sublayer functions,...
- B.P. Crow, I. Widjaja, J.G. Kim, P. Sakai, IEEE 802.11 wireless local area network, IEEE Communications Magazine,...
Cited by (13)
Performance of wireless networks with hidden nodes: A queuing-theoretic analysis
2005, Computer CommunicationsCitation Excerpt :The techniques presented in Ref. [16], therefore, are not applicable in the present context. Other related works include [19,20], which also do not take queuing into account. In the context of ALOHA and CSMA/CD, early works carried out packet delay analysis.
A dynamically adaptable polling scheme for voice support in IEEE802.11 networks
2003, Computer CommunicationsCitation Excerpt :Another important characteristic of the proposed polling scheme is the immediate removal of the silent stations from the polling list for a number of PCF rounds, based on the hangover duration of the used silence detectors. Refs. [5,6] considered the removal of silent stations from the polling list provided that these stations responded with NULL frames to a number of consecutive CF-Poll frames. In the CSSR polling scheme, the silent stations are immediately removed from the AP polling list and remain out of it for a number of PCF rounds.
Improving the Performance of Wireless LANs: A Practical Guide
2014, Improving the Performance of Wireless LANs: A Practical GuideIn-kernel passive measurement of the performance impact of hidden terminals in 802.11 wireless networks
2011, PE-WASUN'11 - Proceedings of the 8th ACM Symposium on Performance Evaluation of Wireless Ad Hoc, Sensor, and Ubiquitous NetworksPerformance modelling of IEEE 802.11 DCF using equilibrium point analysis
2009, International Journal of Wireless and Mobile ComputingI-Poll: Improved polling scheme for augmenting voice support in IEEE 802.11 PCF
2006, IEICE Transactions on Communications
- 1
Supported in part by NSF (grant NCR-9714700), 3Com Technology Development Center, and Lockheed Martin Missiles and Space; currently on sabbatical leave visiting 3Com TCD.