Interference Minimization and Dynamic Sub-Carrier Allocation in Broadband Wireless Networks

Interference Minimization and Dynamic Sub-Carrier Allocation in Broadband Wireless Networks

Ju Wang (Computer Science Department, Virginia State University, Petersburg, VA, USA) and Jonathan Liu (CISE Department, University of Florida, Gainesville, FL, USA)
DOI: 10.4018/ijwnbt.2014070101


Efficient channel allocation is the key to fully exploit the signal diversity presented in the multi-carrier physical link in today's broadband wireless access networks. There are mounting evidences that the 4G and future-generation systems will take advantage of two opposite types of access methods, one using centralized control method and the other using a distributed approach. The authors study the distributed channel allocation problem in this article, formulated as a non-linear optimization problem, in broadband wireless networks. The signal properties of the multi-carrier radio interface in 3G and 4G networks are discussed to comprehend the complexity of the channel allocation problem. The authors propose a novel distributed heuristic algorithm based on the particle-swarm searching method. The distributed approach allows user stations to quickly switch sub-carriers with minimum intervention from the base station. The work presented in this paper shows an effective method to allocate a large number of channels while minimizing the possible interference. Extensive numerical experiments are conducted to evaluate several versions of distributed channel allocation algorithm for these new problem settings. The authors' results show that PSO-based method converges quickly in all our numerical experiments despite the high-dimensional searching space. The proposed technologies will eventually allow the true mobile steaming video/audio experience anywhere and anytime, which will have the huge impact to business, entertainment and people's quality of life.
Article Preview

1. Introduction

We have witnessed the rapid growth of wireless access networks during the recent years. It has been driven by the increasing numbers of smartphone and Wireless Sensor Networks (WSNs) applications (Brar, 2001). The debate between two types of access methods, one using centralized control method and the other using a distributed approach, remains largely unsettled as new evidences arrives from both sides.

Broadband wireless access networks have historically favoured centralized access methods due to the method’s ability to provide guaranteed Quality of Service (QoS). However, hybrid access schemes are commonly used in the third/fourth generation systems, which allow the access providers to support a seamless integration of different types of data services.

For both broadband access network and wireless sensor networks, a fundamental challenge is to mitigate the co-channel interference, which is the limiting cause to the throughput of the networks beyond certain scale. To increase the system capacity, multiple frequency channels/carriers are often deployed to achieve maximum signal isolation in space, time and frequency domains (Jordan, 1996). In such a network, wireless devices could transmit at different frequency channels based on specific channel allocation strategy. Channel access control is particularly important in the third- and fourth-generation broadband networks, due to the aggressive physical layer techniques such as OFDM modulation technique. Although not mandatory in the 4G technical specification, dynamic channel allocation is one of the key mechanisms to maximize the spectrum efficiency and achieve high bandwidth. Such is our motivation to study a novel distributed channel allocation method from the perspective of minimizing interference violation.

On the other side, most infrastructure-less wireless networks (such as conventional 802.11 networks and WSNs) use a shared-channel model within the cell boundary. Channel arbitration falls to the users in a distributed manner. Such assumption should attribute partially to the physical complexity of multi-channel transceiver, but more on the fact that channel allocation problem is NP-hard by nature (Krumke, 2000) especially the distributed ones.

Figure 1 shows a graphic representation of a conventional channel allocation model in a multi-cell configuration, subject to the different interference rules. For illustration, a cell is represented by a vertex, and the network has a regular hexagon structure. On the left side, the interference range is defined as one hop of a uniform radius, which results in a 6-node neighborhood for all nodes. On the right-hand graph, the interference range is two hop of a uniform radius, and the neighborhood from any node now contains 19 nodes. For a collision-free channel allocation, the former case will require 7 channels and the later need 19 channels. One centralized algorithm to do this involves fan-rotation (Misra, 1992) and cd-path inverting.

Figure 1.

Cells in a regular hexagon topology (a) interference radius = one hop, requires 7 channels for collision free allocation. (b) interference radius= two hops requires 19 channels


Complete Article List

Search this Journal:
Open Access Articles
Volume 9: 2 Issues (2020)
Volume 8: 2 Issues (2019)
Volume 7: 2 Issues (2018)
Volume 6: 2 Issues (2017)
Volume 5: 1 Issue (2016)
Volume 4: 3 Issues (2015)
Volume 3: 4 Issues (2014)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing