Condition and location-aware channel switching scheme for multi-hop multi-band WLANs
Introduction
Mult-hop relay networks are deployed in a situation where communication infrastructure is either completely demolished or does not exist a priori. These networks are easily and quickly deployed, however, performance such as throughput and end-to-end delay is drastically effected as the data is relayed over multiple hops. Similarly, for traditional networks, there has been tremendous growth in the network traffic for last couple of decades, thanks to bandwidth hungry applications like HD video, tele-conferencing, telemedicine, e-gaming etc. This growth resulted in the explosion of the wireless devices to make up for the capacity required [1]. Many advance technique have been designed to make up for the inexorable rise in performance demand such as beamforming, massive MIMO and spatial multiplexing etc. [2], [3]. However, these techniques along with dense deployment of devices cannot help much as long as the channel quality does not support the capacity required. The channel quality always deteriorates due to number of reasons such as path losses, wireless interference, Medium Access Control protocols (MAC) design, the topology of the network, the mobility of the end-user and the environment in which the wireless network operates [4], [5], [6], [7]. Therefore, we need to keep the channel quality intact to meet the performance demands of the end-users. In this vein, switching the channel whenever there is a decline in its health in order to improve the performance of the networks has been proven to be a promising avenue to meet the ever-changing demand in the traffic [8], [9]. However, given the similar characteristics of the new channel, the gain in performance may be short-lived and inconsiderable.
In recent years, the problem has received a lot of attention, as deteriorating channel condition can lead to decreased throughput and disconnection between the devices. Channel switching within the same band has been carried out effectively in past to improve system performance [8], [9], [10], [11], [12]. The improvement results from spatial, temporal, congestion and spectral diversity of the channels. However, given the diverse environment in which the network operates, a single band does not suffice. For example, a node in direct contact with access point (AP) may be behind an obstacle the next moment. A node in high node density area may be in a sparse area the moments later. A node closer to AP may move far away from it a while after. Therefore, using single band would not add much to the performance much.
We assert that there does not exist a universal single most suitable band. Sometimes we need to cater for the distance, sometimes we want a node behind obstacles to have better channel quality and sometimes we are interested in eradicating the interference. A single band does not accomplish all the requirements. We need to consider the physical conditions and the inherent characteristics of the frequency bands and then assign a channel from the band that offers best performance in terms of throughput and delay. For example, 2.4 GHz band has only three orthogonal channels and is already overcrowded with electronic appliances found everywhere [13], [14]. The meagre number of channels along with over-crowding of the band lead to increased interference from other devices and intermittent connectivity issues which result in lowered throughput. However, the upside of 2.4 GHz band is its ability to penetrate solid objects more efficiently than 5 GHz band does. Similarly, a signal from 2.4 GHz suffers less path losses as compared to 5 GHz enabling it travel farther than 5 GHz. On the other hand, 5 GHz has 12 orthogonal channel and is not populated much [14], [15]. Additionally, as compared to 2.4 GHz, 5 GHz offers more space for channel bonding where we can increase channel width from 20 MHz up to 80 MHz, if required [14], [16]. Therefore, for a node in direct contact with AP and if SNR allows, assigning a channel from 5 GHz is more advisable. Furthermore, there is a need to improve the spectral efficiency, as the popularity of wireless communication is pushing operators to the limit of a spectrum shortage [17]. We believe using multiple bands will also impede the spectral resource scarcity.
In this paper, we consider a multi-hop WLAN system that works over multiple frequency bands where the sender and the receiver are not in direct contact with each other that is, there is only one source-destination pair while all other nodes assist the transmission. Multi-hop relay networks are mostly deployed to extend the coverage area of the networks in exhibitions, conferences or areas where connection to a network is not available a priori. Scenarios where relay networks are deployed include disaster hit area where rehabilitation services are needed but network infrastructure is completely demolished post-catastrophe and in military tactical communication for fast establishment of communication infrastructure during deployment of forces in a foreign territory [18]. In these types of multi-hop networks, the channel condition and traffic load on the source-relay link tend to be different than on the relay-destination link. This difference in channel condition and traffic result in either buffer overflow or channel under-utilization depending upon whether the traffic load on a particular link is less or more than the traffic load it can accommodate [19]. The buffer overflow and the channel under-utilization drastically impede the throughput and end-to-end delay. To address this problem, we propose receiver’s feedback based bitrate adaption scheme that controls the bitrate of the sender considering the channel condition on either side of the relay.
We consider 2.4 GHz and 5 GHz bands. We develop dynamic switching technique where the relevant characteristics of 2.4 GHz and 5 GHz are taken into account. Where most of the channel switching techniques require an extra channel for negotiating the channel switch [20], [21], [22], the proposed technique does not require any extra channel for channel negotiation. Condition based inter-band or intra-band channel switching is carried out by an adaptive threshold. By estimating SNR of the channels, we predict exactly when packet will fail to reach the next hop and destination. We assign a channel that gives best throughput for the given condition, load in the network and location of the node. Whereas the relay nature of the networks degrades the throughput significantly [23], [24], the propose scheme considerably makes up for the dilapidation in the performance.
The main contributions of this paper are summarized as follows:
- •
We develop an analytical model that guides the channel switching of the system in multi-band environment. The model is validated using network simulator (ns-2).
- •
Ad Hoc Traffic Indication Messages (ATIM) based, a single rendezvous condition-based, location based channel scheme is designed that seamlessly reassign a new channel to the links on the basis of SNR threshold. The SNR threshold is adaptive and is varied according to the channel conditions in a certain link in order to give best performance.
- •
A mobility management model is developed that predicts the location of the station after certain time T seconds. Based on the location, SNR after T seconds is estimated. If throughput is expected to drop, channel is switched beforehand. The proposed mobility management model is also validated using ns-2 simulations.
- •
We develop a technique that handles the different bitrates on the receiving and the sending side of a relay node.
The rest of the paper is organized as follows. In Section 2, we have given a brief literature survey. Section 3 consists of channel assessment and switching model. We use adaptive threshold for channel switching which is given in Section 4. In Section 5, we present our proposed mobility management model. The exception of performance mismatch is given in Section 7. Line of Sight determination is given in Section 7. Section 8, contains results and discussion. Finally, conclusion and future work is given in Section 9.
Section snippets
Related work
Capacity improvement through channel switching within a band has been extensively studied. Researchers have proposed different techniques and protocols that exploit available channels. A review of multi-channel MAC protocols is given in [4], [11] and comparative analysis through simulation is carried out in[22]. By letting nodes dynamically choose which channel to switch to, multi-channel MAC protocols attempt to improve the network performance. These protocols differ in how devices come to an
Channel switching in multi-band problem
In this section, we describe the considered multi-band WLAN system model and the channel-switching problem in multi-band scenario. We also present the formal problem formulation. Fig. 1 depicts the considered relay-based, multi-hop WLAN system capable of operating over multiple band such as 2.4 GHz and 5 GHz. In the figure, the source, relay and the destination are indicated by S, R and D respectively whereby S and R are stationery nodes while D is a mobile station. The source-relay link is
Channel assessment and switching
let be mean SNR and let ϑT be threshold SNR. Considering Rayleigh Fading channels, probability (P) of SNR getting below threshold is given by;
For current SNR below the threshold SNR, the packet will most likely fail. Therefore, probability of packet getting failed is given by;
Channel is switched when probability of packet getting failed is above certain threshold Pt.
For channel switching, we exploit the Ad-hoc Traffic Indication Message (ATIM)
Adaptive threshold
In order to avoid frequent channel switching and improve the performance, we propose an adaptive threshold. To begin with, SNR threshold is initialized with 25 dB. For any given bitrate, SNR is calculated, and packet reception is recorded. SNR threshold will remain the same if current SNR is above the threshold SNR and packet was successfully received. However, current SNR being above the threshold and packet still getting failed, is the indication of a worsening channel condition, which is the
Mobility management model
We are interested to find where will the node move after T seconds and what will be the SNR at that location? For this purpose, we develop a mobility management model that exploits the concept of Angle of Arrival (AoA) [34] and Time of Flight (ToF) [35]. Based on AoA, we can estimate direction of the mobile node depending upon where the signal is coming from and whether the angle is getting larger or smaller.
An illustration is shown in Fig. 8a and b where a mobile station D (assuming it to be
Handling capacity and traffic overloading constraints
Based on the conditions explained in previous sections, we assume both sender and the receiver choose the best channel. However, channel condition at relay-destination link tends to be different. Furthermore, additional relay nodes may be attached to the relay node in question as shown in Fig. 12. If R2 is transmitting data via R1, the load at R1 will be different than at S. For these two reasons, the constraint 2 above is violated that leads to packets being dropped at relay nodes, which in
Performance evaluation
In this section, we study the performance of the proposed technique through simulation. We call our proposed technique as channel switching in multi-band (CSMB) and compare it with 802.11b and 802.11a that operates on 2.4 GHz and 5 GHz respectively. We also compare the results with greedy approach that instantly switches to the better channel, if it finds one with better SNR in the list.
We have used throughput and delay for comparison. Throughput is taken as the number of bits per second
Conclusion and future work
We presented an efficient channel-condition and station-location aware channel switching technique for multi-hop multi-band wireless LAN networks. We developed a model that act as guideline to switch channel to a better one from a suitable band based on the channel conditions and location of the stations. The proposed technique assess the channel conditions and predicts exactly when will packet fail to reach the next hop. Additionally, based on the incoming signal, we estimate the speed and
Declaration of Competing Interest
The Authors declare that there is no conflict of interest.
Acknowledgment
The authors wish to thank Professor Nei Kato and Associate Professor Zubair Md. Fadlullah of Graduate School of Information Sciences, Tohoku University, Japan for their valuable suggestions during our discussion. The authors would also like to thank Ministry of Education, Culture, Sports, Science and Technology for financial support of this research.
Asad Alireceived M.S. degree in Information Technology from School of Electrical Engineering and Computer Science, National University of Science and Technology, (SEECS, NUST), Pakistan in 2011 and M.S degree in Applied Information Sciences from Graduate School of Information Sciences, Tohoku University, Japan in 2019. He has been serving as lecturer at Faculty of Information and Communication Technology (FICT), Balochistan University of Information Technology, Engineering and Management
References (40)
- et al.
Channel, capacity, and flow assignment in wireless mesh networks
Comput. Netw.
(2011) - et al.
Filtering noisy 802.11 time-of-flight ranging measurements from commoditized WiFi radios
IEEE/ACM Trans. Netw.
(2017) - et al.
A survey on low latency towards 5G: RAN, core network and caching solutions
IEEE Commun. Surv. Tutor.
(2018) - et al.
Millimeter wave cellular wireless networks: potentials and challenges
CoRR
(2014) - et al.
Full-duplex wireless communications: challenges, solutions, and future research directions
Proc. IEEE
(2016) - et al.
Channel assignment techniques for multi-radio wireless mesh networks: a survey
IEEE Commun. Surv. Tutor.
(2016) - et al.
Channel assignment strategies for multiradio wireless mesh networks: issues and solutions
IEEE Commun. Mag.
(2007) - S. Uludag, T. Imboden, K. Akkaya, A taxonomy and evaluation for developing 802.11-based wireless mesh network testbeds,...
- et al.
Dynamic switching with heterogeneous channels in multichannel 802.11 WLANs
2015 12th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON)
(2015) - et al.
Throughput optimal switching in multichannel WLANs
IEEE Trans. Mobile Comput.
(2013)
Traffic-aware channel assignment for multi-transceiver wireless networks
Multi-channel medium access control protocols: review and comparison
J. Inf. Optim. Sci.
Routing and interface assignment in multi-channel multi-interface wireless networks
IEEE Wireless Communications and Networking Conference, 2005
Intelligent channel bonding in 802.11n WLANs
IEEE Trans. Mobile Comput.
The impact of channel bonding on 802.11n network management
Proc. ACM 7th Conf. Emerg. Netw. Exp. Technol, CoNEXT
On the performance of a medium access control scheme for the reconfigurable wireless networks
MILCOM 97 Proceedings
Multi-hop wireless transmission in multi-band WLAN systems: proposal and future perspective
IEEE Wirel. Commun.
Cited by (5)
Multiradio Parallel Offloading in Multiaccess Edge Computing: Optimizing Load Shares, Scheduling, and Capacity
2024, IEEE Internet of Things JournalPartially overlapping channel assignment using bonded and non-bonded channels in ieee 802.11n wlan
2022, Computers, Materials and ContinuaMulti-Band Multi-Hop WLANs for Disaster Relief and Public Safety Applications
2020, 2020 IEEE Wireless Communications and Networking Conference Workshops, WCNCW 2020 - Proceedings
Asad Alireceived M.S. degree in Information Technology from School of Electrical Engineering and Computer Science, National University of Science and Technology, (SEECS, NUST), Pakistan in 2011 and M.S degree in Applied Information Sciences from Graduate School of Information Sciences, Tohoku University, Japan in 2019. He has been serving as lecturer at Faculty of Information and Communication Technology (FICT), Balochistan University of Information Technology, Engineering and Management Science (BUITEMS), Quetta, Pakistan since 2012. He was awarded Higher Education Commission (HEC) scholarship by Government of Pakistan for studies at SEECS-NUST and MEXT Scholarship by Government of Japan to perform his studies in Japan. He has over 30 research papers published in International Journals of repute and refereed conferences. He has received number of awards, research and travel grant for his research work. His research interests are in the area of multi-band transmission and network optimization.
Faisal A. Khan received BS degree in Software Engineering and MS degree in Telecommunications in 2004 and 2007, respectively, from Bahria University Karachi. He received MS and Ph.D. degrees in Electrical and Computer Engineering from the Georgia Institute of Technology, USA in 2012. He has been working with the Faculty of Information and Communication Technology at BUITEMS since October 2007. He has worked as Dean of the faculty from August 2013 to Nov 2018. Currently, he is working as the Pro-Vice Chancellor, BUITEMS since Dec 2018. Dr. Khan has worked with the Communications Systems Center (CSC) at Georgia Institute of Technology from 2009 to 2012. His research at CSC focused on Vehicular Ad hoc Networks under the title Safety-Message Routing in VANETs. His research work at Bahria University focused on Radio-Wave Propagation into Buildings and Client-Server Architecture for Cellular Handheld devices. Awards and distinctions to his credit include Fulbright Ph.D. scholarship award 2009–2012, British Council Momentum award 2018, outstanding leadership award IEEE Karachi section 2015, excellence in leadership award by World Confederation of Business, Houston, Texas, 2015; best teacher award BUITEMS 2008, Bahria University scholarship award 2006, and undergraduate distinction Bahria University 2004. Dr. Khan is Senior Member (SM) of the Institute of Electrical and Electronics Engineers (IEEE), Piscatway, NJ, USA. He has been invited speaker and session chair at a number of international IEEE conferences. He has also been the conference chair of the IEEE ICE Cube Conference 2016 and 2018. Besides professional undertakings, he is the founder of Quetta Going Green, a citizens’ movement for the promotion of environmental awareness in the city of Quetta. He is also the founder of Honesty Mart, an initiative for the promotion of trust and honesty in the society.
- 1
Pro-Vice Chancellor.