Pub Date : 2015-05-04DOI: 10.1109/COMSNETS.2015.7098681
Rishabh Poddar, Anilkumar Vishnoi, V. Mann
Load balancing and auto scaling are important services in the cloud. Traditionally, load balancing is achieved through either hardware or software appliances. Hardware appliances perform well but have several drawbacks. They are fairly expensive and are typically bought for managing peaks even if average volumes are 10% of peak. Further, they lack flexibility in terms of adding custom load balancing algorithms. They also lack multi-tenancy support. To address these concerns, most public clouds have adopted software load balancers that typically also comprise an auto scaling service. However, software load balancers do not match the performance of hardware load balancers. In order to avoid a single point of failure, they also require complex clustering solutions which further drives their cost higher. In this context, we present HAVEN - a system for holistic load balancing and auto scaling in a multi-tenant cloud environment that is naturally distributed, and hence scalable. It supports multi-tenancy and takes into account the utilization levels of different resources as part of its load balancing and auto scaling algorithms. HAVEN leverages software-defined networking to ensure that while the load balancing algorithm (control plane) executes on a server running network controller software, the packets to be load balanced never leave the data plane. For this reason, HAVEN is able to provide performance at par with a hardware load balancer while still providing the flexibility and customizability of a software load balancer. We validate HAVEN on a hardware setup and our experiments confirm that it achieves high performance without any significant overheads.
{"title":"HAVEN: Holistic load balancing and auto scaling in the cloud","authors":"Rishabh Poddar, Anilkumar Vishnoi, V. Mann","doi":"10.1109/COMSNETS.2015.7098681","DOIUrl":"https://doi.org/10.1109/COMSNETS.2015.7098681","url":null,"abstract":"Load balancing and auto scaling are important services in the cloud. Traditionally, load balancing is achieved through either hardware or software appliances. Hardware appliances perform well but have several drawbacks. They are fairly expensive and are typically bought for managing peaks even if average volumes are 10% of peak. Further, they lack flexibility in terms of adding custom load balancing algorithms. They also lack multi-tenancy support. To address these concerns, most public clouds have adopted software load balancers that typically also comprise an auto scaling service. However, software load balancers do not match the performance of hardware load balancers. In order to avoid a single point of failure, they also require complex clustering solutions which further drives their cost higher. In this context, we present HAVEN - a system for holistic load balancing and auto scaling in a multi-tenant cloud environment that is naturally distributed, and hence scalable. It supports multi-tenancy and takes into account the utilization levels of different resources as part of its load balancing and auto scaling algorithms. HAVEN leverages software-defined networking to ensure that while the load balancing algorithm (control plane) executes on a server running network controller software, the packets to be load balanced never leave the data plane. For this reason, HAVEN is able to provide performance at par with a hardware load balancer while still providing the flexibility and customizability of a software load balancer. We validate HAVEN on a hardware setup and our experiments confirm that it achieves high performance without any significant overheads.","PeriodicalId":277593,"journal":{"name":"2015 7th International Conference on Communication Systems and Networks (COMSNETS)","volume":"212 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114596879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-05-04DOI: 10.1109/COMSNETS.2015.7098682
G. Mandyam, N. Ehsan
With the advent of HTML5, the next generation of browser-based technologies for web developers, new connectivity paradigms have arisen that will allow web developers to move beyond HTTP-based stateless transactions. However, such new features do not transition seamlessly from the desktop to mobile browsers. Specifically, handheld devices with limited battery life can be impacted by persistent connectivity. In this work, two recent technologies (WebSockets and WebRTC) are examined with respect to power consumption. Based on this analysis, recommendations for improving battery life when using these technologies are provided.
{"title":"Impact of HTML5 persistent connectivity to power consumption for the mobile web","authors":"G. Mandyam, N. Ehsan","doi":"10.1109/COMSNETS.2015.7098682","DOIUrl":"https://doi.org/10.1109/COMSNETS.2015.7098682","url":null,"abstract":"With the advent of HTML5, the next generation of browser-based technologies for web developers, new connectivity paradigms have arisen that will allow web developers to move beyond HTTP-based stateless transactions. However, such new features do not transition seamlessly from the desktop to mobile browsers. Specifically, handheld devices with limited battery life can be impacted by persistent connectivity. In this work, two recent technologies (WebSockets and WebRTC) are examined with respect to power consumption. Based on this analysis, recommendations for improving battery life when using these technologies are provided.","PeriodicalId":277593,"journal":{"name":"2015 7th International Conference on Communication Systems and Networks (COMSNETS)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117028918","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-05-04DOI: 10.1109/COMSNETS.2015.7098689
Sacha Trifunovic, Andreea Hossmann-Picu, T. Hossmann
By using device-to-device (D2D) communication, opportunistic networks promise to fill the gaps of the networking infrastructure in remote areas, to enable communication in emergency situations, and to inspire new applications. Yet, to become feasible in practice and accepted by users, it is crucial that the energy costs of D2D connections are small and shared fairly. Fairness, in particular, is a major issue with today's D2D technologies (Bluetooth, Wi-Fi Direct): since each connected peer must assume one of two different roles - access point/client, master/slave, the energy consumption inside a connected group is very asymmetric. While a large body of research exists on role assignment and topology control, the above issue of energy fairness is either not at all addressed (e.g. in the context of Bluetooth scatternets) or is addressed under fundamentally different conditions (e.g. in very dense and often static wireless sensor networks). In this paper, we tackle the fairness problem of the energy consumed in a group of D2D-connected nodes, by using role switching: the two types of roles are alternated among group members, thus producing a fairer cost sharing. First, we analyze contact traces for their group topologies and find that four simple motifs - clique, star, chain and NxM-clique - cover up to 94% of the aggregated lifetime of all connected groups. We then determine the optimal role switching strategies for these motifs by formulating the cycle of role assignments as an optimization problem. Since deriving the optimal cycle online, in a distributed manner is hardly possible in practice, we also propose two role switching heuristics for online use: a randomized switching scheme tunable for efficiency or fairness, and a deterministic scheme which additionally guarantees the group's connectivity. Finally, we evaluate our solutions on real contact traces and show that our heuristics find very good points of operation in the fairness-efficiency tradeoff.
{"title":"Passing the torch: Role alternation for fair energy usage in D2D group communication","authors":"Sacha Trifunovic, Andreea Hossmann-Picu, T. Hossmann","doi":"10.1109/COMSNETS.2015.7098689","DOIUrl":"https://doi.org/10.1109/COMSNETS.2015.7098689","url":null,"abstract":"By using device-to-device (D2D) communication, opportunistic networks promise to fill the gaps of the networking infrastructure in remote areas, to enable communication in emergency situations, and to inspire new applications. Yet, to become feasible in practice and accepted by users, it is crucial that the energy costs of D2D connections are small and shared fairly. Fairness, in particular, is a major issue with today's D2D technologies (Bluetooth, Wi-Fi Direct): since each connected peer must assume one of two different roles - access point/client, master/slave, the energy consumption inside a connected group is very asymmetric. While a large body of research exists on role assignment and topology control, the above issue of energy fairness is either not at all addressed (e.g. in the context of Bluetooth scatternets) or is addressed under fundamentally different conditions (e.g. in very dense and often static wireless sensor networks). In this paper, we tackle the fairness problem of the energy consumed in a group of D2D-connected nodes, by using role switching: the two types of roles are alternated among group members, thus producing a fairer cost sharing. First, we analyze contact traces for their group topologies and find that four simple motifs - clique, star, chain and NxM-clique - cover up to 94% of the aggregated lifetime of all connected groups. We then determine the optimal role switching strategies for these motifs by formulating the cycle of role assignments as an optimization problem. Since deriving the optimal cycle online, in a distributed manner is hardly possible in practice, we also propose two role switching heuristics for online use: a randomized switching scheme tunable for efficiency or fairness, and a deterministic scheme which additionally guarantees the group's connectivity. Finally, we evaluate our solutions on real contact traces and show that our heuristics find very good points of operation in the fairness-efficiency tradeoff.","PeriodicalId":277593,"journal":{"name":"2015 7th International Conference on Communication Systems and Networks (COMSNETS)","volume":"2 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114038075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-05-04DOI: 10.1109/COMSNETS.2015.7098670
Subhajit Majhi, Adrish Banerjee
In this paper, we consider a secondary incremental relaying (IR) network that operates simultaneously with multiple primary receivers in the same frequency band under spectrum sharing constraints. The IR network consists of multiple decode-and-forward (DF) relays and employs opportunistic relay selection for cooperation. We analyze the outage probability of the network where the selected relay opts to cooperate when the destination fails to decode the direct transmission from source.We derive the expression of asymptotic outage probability, diversity gain and coding gain. The diversity gain was found to be the same as that of a non-spectrum sharing system. Furthermore, we investigate the impact of number of relays and primary-to-secondary network distance on outage probability and coding gains, which is validated through numerical simulation.
{"title":"Asymptotic outage analysis of incremental decode and forward cognitive radio relay network","authors":"Subhajit Majhi, Adrish Banerjee","doi":"10.1109/COMSNETS.2015.7098670","DOIUrl":"https://doi.org/10.1109/COMSNETS.2015.7098670","url":null,"abstract":"In this paper, we consider a secondary incremental relaying (IR) network that operates simultaneously with multiple primary receivers in the same frequency band under spectrum sharing constraints. The IR network consists of multiple decode-and-forward (DF) relays and employs opportunistic relay selection for cooperation. We analyze the outage probability of the network where the selected relay opts to cooperate when the destination fails to decode the direct transmission from source.We derive the expression of asymptotic outage probability, diversity gain and coding gain. The diversity gain was found to be the same as that of a non-spectrum sharing system. Furthermore, we investigate the impact of number of relays and primary-to-secondary network distance on outage probability and coding gains, which is validated through numerical simulation.","PeriodicalId":277593,"journal":{"name":"2015 7th International Conference on Communication Systems and Networks (COMSNETS)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128968290","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-05-04DOI: 10.1109/COMSNETS.2015.7098718
Manish Chaturvedi, S. Srivastava
Intelligent Transportation Systems (ITS) play major role in generating fine grained vehicular traffic information for city wide or larger region. However, in developing countries like India, limited ITS infrastructure is available. On the other hand, cellular infrastructure is widely deployed in India with more than 867 million cellular connections and more than 70% cellular teledensity [1]. Also, on some major arterial roads, video cameras are deployed for surveillance purpose. The aim of this study is to assess feasibility of using these alternate sources to generate accurate traffic information for all the edges in a road network. The simulation results show that, even with large location error of 250-500 meters, edge level vehicle flow estimation with good accuracy (less than 10% error) is feasible using cellular network data. Using cellular network data alone, the edges can be classified as congested or uncongested. For edge level speed estimation, we propose a simple and novel approach for fusing widely available but erroneous flow data from cellular network with the spatially sparse but accurate flow-speed data from other sources (e.g. loop detectors or video cameras). The simulation results show that edge level speed estimation with good accuracy (less than 15% median error) is feasible using the proposed approach.
{"title":"Edge level vehicular traffic estimation using cellular infrastructure and other sources","authors":"Manish Chaturvedi, S. Srivastava","doi":"10.1109/COMSNETS.2015.7098718","DOIUrl":"https://doi.org/10.1109/COMSNETS.2015.7098718","url":null,"abstract":"Intelligent Transportation Systems (ITS) play major role in generating fine grained vehicular traffic information for city wide or larger region. However, in developing countries like India, limited ITS infrastructure is available. On the other hand, cellular infrastructure is widely deployed in India with more than 867 million cellular connections and more than 70% cellular teledensity [1]. Also, on some major arterial roads, video cameras are deployed for surveillance purpose. The aim of this study is to assess feasibility of using these alternate sources to generate accurate traffic information for all the edges in a road network. The simulation results show that, even with large location error of 250-500 meters, edge level vehicle flow estimation with good accuracy (less than 10% error) is feasible using cellular network data. Using cellular network data alone, the edges can be classified as congested or uncongested. For edge level speed estimation, we propose a simple and novel approach for fusing widely available but erroneous flow data from cellular network with the spatially sparse but accurate flow-speed data from other sources (e.g. loop detectors or video cameras). The simulation results show that edge level speed estimation with good accuracy (less than 15% median error) is feasible using the proposed approach.","PeriodicalId":277593,"journal":{"name":"2015 7th International Conference on Communication Systems and Networks (COMSNETS)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115417793","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-05-04DOI: 10.1109/COMSNETS.2015.7098697
Bernhard Distl, S. Neuhaus
Opportunistic networking poses a trade off. On the one hand, social links and the structure inherent in the opportunistic network are private information and should thus be protected. On the other hand, such information can be used to improve performance or security. We present an algorithm that can solve this dilemma: we can recognize pre-established social links among nodes without revealing private information, thus enabling privacy and performance and security. Our algorithm uses a hash-based approach for identity protection to reliably and efficiently detect existing social links in an anonymous environment. We implemented our algorithm in a smartphone application and evaluate its performance. Friendship detection time grows linearly with the number of friendships; for example, with 500 friendships, detection typically takes about 90 ms.
{"title":"Social power for privacy protected opportunistic networks","authors":"Bernhard Distl, S. Neuhaus","doi":"10.1109/COMSNETS.2015.7098697","DOIUrl":"https://doi.org/10.1109/COMSNETS.2015.7098697","url":null,"abstract":"Opportunistic networking poses a trade off. On the one hand, social links and the structure inherent in the opportunistic network are private information and should thus be protected. On the other hand, such information can be used to improve performance or security. We present an algorithm that can solve this dilemma: we can recognize pre-established social links among nodes without revealing private information, thus enabling privacy and performance and security. Our algorithm uses a hash-based approach for identity protection to reliably and efficiently detect existing social links in an anonymous environment. We implemented our algorithm in a smartphone application and evaluate its performance. Friendship detection time grows linearly with the number of friendships; for example, with 500 friendships, detection typically takes about 90 ms.","PeriodicalId":277593,"journal":{"name":"2015 7th International Conference on Communication Systems and Networks (COMSNETS)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115230688","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-05-04DOI: 10.1109/COMSNETS.2015.7098690
Antar S.H. Abdul-Qawy, A. Potluri
Networking devices in ISP networks and data centers have been deployed in an over-provisioning and redundant manner to meet the worst case traffic loads and to quickly recover from failures. These devices are idle or semi-idle most of the time with full power consumption. Today, there is widespread interest in techniques that help in reducing energy waste and achieving high levels of energy-efficiency. Solutions proposed in literature are either topology-oriented (ESTOP) or traffic-oriented (ESTA). In this paper, we propose a Hybrid Heuristic for Green Networking (HHGN) that exploits the parameters considered in both ESTOP and ESTA to identify a subset of network elements that can be switched off such that the power saving is maximized under network connectivity and edge bandwidth utilization constraints. We introduce two new heuristics, w-BFLP (weighted Betweenness, Flow, Links, Power) and w-BFP (weighted Betweenness, Flow, Power), that sort the nodes and the edges respectively in increasing order of their importance in the topology. Nodes and edges are then switched off from the least important until the connectivity threshold or maximum bandwidth utilization constraint on the edges as specified by the user is reached. We compare our approach with ESTOP and ESTA for edge optimization in terms of power gain, mean utilization of edge bandwidth, percentage of sleeping edges, fairness index, trade-off between power gain and edge utilization, and increase in the length of re-computed paths for real ISP and FatTree topologies using different power models. Experimental results show that HHGN gives the best performance independent of power models, traffic matrices and topologies tested.
{"title":"Performance analysis of parameters affecting power efficiency in networks","authors":"Antar S.H. Abdul-Qawy, A. Potluri","doi":"10.1109/COMSNETS.2015.7098690","DOIUrl":"https://doi.org/10.1109/COMSNETS.2015.7098690","url":null,"abstract":"Networking devices in ISP networks and data centers have been deployed in an over-provisioning and redundant manner to meet the worst case traffic loads and to quickly recover from failures. These devices are idle or semi-idle most of the time with full power consumption. Today, there is widespread interest in techniques that help in reducing energy waste and achieving high levels of energy-efficiency. Solutions proposed in literature are either topology-oriented (ESTOP) or traffic-oriented (ESTA). In this paper, we propose a Hybrid Heuristic for Green Networking (HHGN) that exploits the parameters considered in both ESTOP and ESTA to identify a subset of network elements that can be switched off such that the power saving is maximized under network connectivity and edge bandwidth utilization constraints. We introduce two new heuristics, w-BFLP (weighted Betweenness, Flow, Links, Power) and w-BFP (weighted Betweenness, Flow, Power), that sort the nodes and the edges respectively in increasing order of their importance in the topology. Nodes and edges are then switched off from the least important until the connectivity threshold or maximum bandwidth utilization constraint on the edges as specified by the user is reached. We compare our approach with ESTOP and ESTA for edge optimization in terms of power gain, mean utilization of edge bandwidth, percentage of sleeping edges, fairness index, trade-off between power gain and edge utilization, and increase in the length of re-computed paths for real ISP and FatTree topologies using different power models. Experimental results show that HHGN gives the best performance independent of power models, traffic matrices and topologies tested.","PeriodicalId":277593,"journal":{"name":"2015 7th International Conference on Communication Systems and Networks (COMSNETS)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114235737","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-05-04DOI: 10.1109/COMSNETS.2015.7098708
G. Peretti, V. Lakkundi, M. Zorzi
The emergence of Internet of Things and the availability of inexpensive sensor devices and platforms capable of wireless communications enable a wide range of applications such as intelligent home and building automation, mobile healthcare, smart logistics, distributed monitoring, smart grids, energy management, asset tracking to name a few. These devices are expected to employ Constrained Application Protocol for the integration of such applications with the Internet, which includes User Datagram Protocol binding with Datagram Transport Layer Security protocol to provide end-to-end security. This paper presents a framework called BlinkToSCoAP, obtained through the integration of three software libraries implementing lightweight versions of DTLS, CoAP and 6LoWPAN protocols over TinyOS. Furthermore, a detailed experimental campaign is presented that evaluates the performance of DTLS security blocks. The experiments analyze BlinkToSCoAP messages exchanged between two Zolertia Z1 devices, allowing evaluations in terms of memory footprint, energy consumption, latency and packet overhead. The results obtained indicate that securing CoAP with DTLS in Internet of Things is certainly feasible without incurring much overhead.
{"title":"BlinkToSCoAP: An end-to-end security framework for the Internet of Things","authors":"G. Peretti, V. Lakkundi, M. Zorzi","doi":"10.1109/COMSNETS.2015.7098708","DOIUrl":"https://doi.org/10.1109/COMSNETS.2015.7098708","url":null,"abstract":"The emergence of Internet of Things and the availability of inexpensive sensor devices and platforms capable of wireless communications enable a wide range of applications such as intelligent home and building automation, mobile healthcare, smart logistics, distributed monitoring, smart grids, energy management, asset tracking to name a few. These devices are expected to employ Constrained Application Protocol for the integration of such applications with the Internet, which includes User Datagram Protocol binding with Datagram Transport Layer Security protocol to provide end-to-end security. This paper presents a framework called BlinkToSCoAP, obtained through the integration of three software libraries implementing lightweight versions of DTLS, CoAP and 6LoWPAN protocols over TinyOS. Furthermore, a detailed experimental campaign is presented that evaluates the performance of DTLS security blocks. The experiments analyze BlinkToSCoAP messages exchanged between two Zolertia Z1 devices, allowing evaluations in terms of memory footprint, energy consumption, latency and packet overhead. The results obtained indicate that securing CoAP with DTLS in Internet of Things is certainly feasible without incurring much overhead.","PeriodicalId":277593,"journal":{"name":"2015 7th International Conference on Communication Systems and Networks (COMSNETS)","volume":"321 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114458516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-05-04DOI: 10.1109/COMSNETS.2015.7098694
A. Seetharam, Simon Heimlicher, J. Kurose, Wei Wei
In practice, Mobile Ad Hoc Networks (MANETs) must operate efficiently under a wide range of node mobility and link quality regimes. Stateful protocols such as OLSR are suitable for networks connected by stable paths, but are outperformed by stateless flooding in sparse and rapidly changing networks. Neither routing nor flooding alone perform well in heterogeneous networks comprised of both stable and highly dynamic components. Rather than design a new protocol for routing in heterogeneous mobile networks from scratch, we use an approach that leverages prior work by operating nodes individually as routers or flooders and switching mode in response to changing network conditions. We present two greedy algorithms, k-flood and adaptive-flood, that dynamically classify nodes as routers or flooders. Our simulations show that nodes can effectively adapt their individual operation as routers/flooders, achieving performance equivalent to, and in some cases significantly better than, that of network-wide routing or flooding alone.
{"title":"Routing with adaptive flooding in heterogeneous mobile networks","authors":"A. Seetharam, Simon Heimlicher, J. Kurose, Wei Wei","doi":"10.1109/COMSNETS.2015.7098694","DOIUrl":"https://doi.org/10.1109/COMSNETS.2015.7098694","url":null,"abstract":"In practice, Mobile Ad Hoc Networks (MANETs) must operate efficiently under a wide range of node mobility and link quality regimes. Stateful protocols such as OLSR are suitable for networks connected by stable paths, but are outperformed by stateless flooding in sparse and rapidly changing networks. Neither routing nor flooding alone perform well in heterogeneous networks comprised of both stable and highly dynamic components. Rather than design a new protocol for routing in heterogeneous mobile networks from scratch, we use an approach that leverages prior work by operating nodes individually as routers or flooders and switching mode in response to changing network conditions. We present two greedy algorithms, k-flood and adaptive-flood, that dynamically classify nodes as routers or flooders. Our simulations show that nodes can effectively adapt their individual operation as routers/flooders, achieving performance equivalent to, and in some cases significantly better than, that of network-wide routing or flooding alone.","PeriodicalId":277593,"journal":{"name":"2015 7th International Conference on Communication Systems and Networks (COMSNETS)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123800078","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-05-04DOI: 10.1109/COMSNETS.2015.7098684
Di Huang, Sanfeng Zhang, P. Hui, Zhou Chen
Opportunistic networks (OppNets) have emerged as prospective network architecture due to the popularization of mobile devices and little monetary cost. Routing is the major concern in designing OppNets. The main obstacle for routing protocol design in OppNets is little knowledge of future link patterns, which leads to blind and unpredictable packet forwarding behavior. To achieve better packet delivery rate, OppNets have to retain and deliver multiple copies of a message, which consumes more devices' energy and causes a lackluster OppNets service. To this end, we aim at the prediction of future link patterns to explore mobile connectivity. In this paper, we propose PreKR-the kernel regression based estimation framework for link pattern prediction. We initially extract best features that can represent the network evolution. Then we models historical structural features by kernel regression with the output of link probability. Experimental results show that our method outperforms state-of-the-art prediction methods up to 25%. We also find that both reachability prediction and high degree nodes prediction reach more than 90% accuracy. In the end, we propose heterogeneous architecture for PreKR deployment and investigate two prospective OppNets applications to show how PreKR improve system performance.
{"title":"Link pattern prediction in opportunistic networks with kernel regression","authors":"Di Huang, Sanfeng Zhang, P. Hui, Zhou Chen","doi":"10.1109/COMSNETS.2015.7098684","DOIUrl":"https://doi.org/10.1109/COMSNETS.2015.7098684","url":null,"abstract":"Opportunistic networks (OppNets) have emerged as prospective network architecture due to the popularization of mobile devices and little monetary cost. Routing is the major concern in designing OppNets. The main obstacle for routing protocol design in OppNets is little knowledge of future link patterns, which leads to blind and unpredictable packet forwarding behavior. To achieve better packet delivery rate, OppNets have to retain and deliver multiple copies of a message, which consumes more devices' energy and causes a lackluster OppNets service. To this end, we aim at the prediction of future link patterns to explore mobile connectivity. In this paper, we propose PreKR-the kernel regression based estimation framework for link pattern prediction. We initially extract best features that can represent the network evolution. Then we models historical structural features by kernel regression with the output of link probability. Experimental results show that our method outperforms state-of-the-art prediction methods up to 25%. We also find that both reachability prediction and high degree nodes prediction reach more than 90% accuracy. In the end, we propose heterogeneous architecture for PreKR deployment and investigate two prospective OppNets applications to show how PreKR improve system performance.","PeriodicalId":277593,"journal":{"name":"2015 7th International Conference on Communication Systems and Networks (COMSNETS)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130440844","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：