Devices in wireless mesh networks can operate on multiple channels and automatically adjust their transmission rates for the occupied channels. This paper shows how to improve performance guaranteed multi casting transmission coverage for wireless multi-hop mesh networks by exploring the transmission opportunity offered by multiple rates (MR) and multiple channels (MC).
We investigate the characteristics and behavior of transmissions with different rates in wireless multi-hop mesh networks. We then propose parallel low-rate transmissions and alternative rate transmissions to explore the advantages of MRMC under the constraint of limited channel resources. A novel link-controlled multi-rate multi-channel multicast algorithm is also designed to extend wireless multicast coverage with high throughput. Our NS2 simulation results demonstrate the improved multicast quality of LC-MRMC in much larger wireless areas as compared to current studies.
Multicast in a wireless mesh network (WMN) is promising to efficiently utilize wireless sources in providing flexible and reliable wireless connections to a group of receivers. This paper proposes to improve transmission coverage with high through put for multicast in a large-scale WMN. Complicated wireless multi casting interference is a major obstacle to achieving this. Such interference is caused by 1) consecutive transmissions on the same multi-hop WMN paths, and 2) parallel delivery of multicast data on paths that have at least one interfering hop. We use the example in Fig.1 to illustrate the effect of this complicated interference.
Multiple Rates And Multiple Channels In WMNs:
A. About the Throughput-Coverage Tradeoff:
The tradeoff between network throughput and transmission coverage introduced in the literature is based on single-hop wireless transmissions. We conduct NS2 simulations to observe the transmission behavior of different rates on a multi-hop wireless path. The simulations use the parameter settings listed in TableI. To set 802.11 MAC and physical wireless parameters, we refer to the specifications of Lucent ORiNOCO11b cards. Also, a probabilistic Nakagami propagation model, representing channel fading characteristics of a wide-range urban settings, is employed in the simulations.
B. Parallel Low-rate Transmission (PLT):
We consider a simple wireless multicast in Fig.3 in which n1 is located in the 11Mbps transmission range of n0 and n2 is located out of the 11Mbps transmission range but within the 5.5Mbps transmission range of n0. For the sake of connectivity, it is not unusual for n0 to transmit at 5.5Mbps in the literature,however this limits the throughput that n1 can potentially achieve since n0 is capable of transmitting at 11Mbps.
C. Analysis of the Parallel Low-Rate Transmission:
We now generalize the achieved observations from the above simulations for other transmission rates.
Alternative Rate Transmission:
Although PLT requires a simple process to effectively improve transmission ranges with high throughput, its significant benefit relies on the availability of orthogonal channels.
A. Benchmark Rate:
To save orthogonal channels for PLT transmissions, regular transmissions employ single channels. Bearing the motivation of improving both throughput and coverage in mind, by referring to our observations in Section III A, in ART, regular transmissions use such a rate that provides the best balance between coverage and performance over multiple hops. We call it the benchmark rate, denoted as R . We now analyze how to achieve R among n available rates.
B. Alternative Rate Transmission:
The ART algorithm classifies wireless mesh nodes as regular nodes and PLT nodes . Regular nodes transmit at the benchmark rate (denoted as R ) via single channels. PLT nodes employ the PLT transmission to send packets at rate R5.
C. PLT Transmission Rate:
In order to achieve the best channel utilization to gain the largest throughput-guaranteed transmission coverage, the PLT rate R should have a transmission coverage satisfying.
Link Controlled Multi Rate Multi Channel Multicasting Tree (LC-MRMC):
The construction of an LC-MRMC tree is initiated by the registration procedure of receivers. Each multicast receiver broadcasts a REGISTRATION packet to the multicast sender s, including the fields of Group ID identifying the multicast group, Hop Count recording the hop number between a mesh node and s, and Forwarder list recording the IP address and the reliability of a REGISTRATION forwarder
In this section, we use NS2 simulations to evaluate the performance of the proposed ART and LC-MRMC. The simulations use the parameters listed in Table I.
A. Art Evaluation:
B. Performance Evaluation in a Random WMN:
For evaluating LC-MRMC, we compare the average multicast throughput, the average multicast delay and the multicast coverage of five different wireless multicast schemes in a wireless network with 100 mesh nodes.
In this paper, we investigated the transmission opportunity afforded by multiple transmission rates and multiple channels for wireless multicast in mesh networks. Our development was based on several interesting findings derived from observations of our simulation of the behaviour of MRMC transmissions in multi-hop WMNs.
Parallel low-rate transmission was proposed to improve high-throughput coverage by simple processes with light over heads. In order to address the challenge to PLT posed by limited channel resources, the alternative rate transmission scheme was designed to alternatively run regular and PLT transmissions in order to extend interference-free coverage with high throughput.
We then presented the new LC-MRMC algorithm that employs the minimum number of reliable on-tree forwarders to form ART paths to extend high-throughput coverage in a multicast environment. Our NS2 simulation results proved that LC-MRMC delivers multimedia flows with higher performance to a coverage area which is at least 85% larger than existing multi-rate multi-channel schemes.
Source: The Robert Gordon University
Author: Wanqing Tu