Ioannis Pefkianakis

WiFi-Assisted 60 GHz Wireless Networks

Despite years of innovative research and development, gigabit-speed 60 GHz wireless networks are still not mainstream. The main concern for network operators and vendors is the unfavorable propagation characteristics due to short wavelength and high directionality, which renders the 60 GHz links highly vulnerable to blockage and mobility. However, the advent of multi-band chipsets opens the possibility of leveraging the more robust WiFi technology to assist 60 GHz in order to provide seamless, Gbps connectivity. In this paper, we design and implement MUST, an IEEE 802.11-compliant system that provides seamless, high-speed connectivity over multi-band 60 GHz and WiFi devices. MUST has two key design components: (1) a WiFi-assisted 60 GHz link adaptation algorithm, which can instantaneously predict the best beam and PHY rate setting, with zero probing overhead; and (2) a proactive blockage detection and switching algorithm which can re-direct ongoing user traffic to the robust interface within sub-10 ms latency. Our experiments with off-the-shelf 802.11 hardware show that MUST can achieve 25-60% throughput gain over state-of-the-art solutions, while bringing almost 2 orders of magnitude cross-band switching latency improvement.

Mobile Data Offloading Through Caching in Residential 802.11 Wireless Networks

As the ever growing mobile data traffic challenges the economic viability and performance of cellular networks, innovative solutions that harvest idle user-owned network resources are gaining increasing interest. In this work, we propose leasing wireless bandwidth and cache space of residential 802.11 (WiFi) access points (APs) for offloading mobile data. This solution not only reduces cellular network congestion, but, due to caching, improves also the user-perceived network performance without overloading the backhaul links of the APs. To encourage residential users to contribute their bandwidth and cache resources, we design monetary incentive (reimbursement) schemes. The offered reimbursements directly determine the amounts of available bandwidth and cache space in every AP, which in turn affect the caching policy (where to cache each content file) and the routing policy (where to route each mobile data request). In order to reduce operators total cost for serving mobile data requests and leasing resources, we introduce a framework for the joint optimization of incentive, caching, and routing policies. Using a novel WiFi usage dataset collected from 167 residences, we show that in densely populated areas with relatively costly network capacity upgrades, our proposal can halve operators total cost, while reimbursing up to 9€ per month each residential user.

Characterizing home wireless performance: The gateway view

In this paper, we analyze a large dataset of passive wireless measurements and obtain insights about wireless performance. We monitor 167 homes continuously for 4 months from the vantage point of the gateway, which allows us to capture all the activity on the home wireless network. We report on the makeup of the home wireless network, traffic activity, and performance characteristics. We find that in most homes, a small number of devices account for most of the observed traffic volume and the bulk of this traffic activity occurs in the evenings. Studying link performance, we find that overall, the vast majority of transmissions are carried out at high data rates and the wireless networks have good coverage. We find a small number of episodes where performance is poor; a few homes have a disproportionate number of poor performance reports. Investigating further, we observe that most of these are not caused by poor coverage (pointing to network interference). Our results significantly add to the understanding of home wireless networks and will help ISPs to understand their subscriber networks.

Practical MU-MIMO user selection on 802.11ac commodity networks

Multi-User MIMO, the hallmark of IEEE 802.11ac and the upcoming 802.11ax, promises significant throughput gains by supporting multiple concurrent data streams to a group of users. However, identifying the best-throughput MU-MIMO groups in commodity 802.11ac networks poses three major challenges: a) Commodity 802.11ac users do not provide full CSI feedback, which has been widely used for MU-MIMO grouping. b) Heterogeneous channel bandwidth users limit grouping opportunities. c) Limited-resource on APs cannot support computationally and memory expensive operations, required by existing algorithms. Hence, state-of-the-art designs are either not portable in 802.11ac APs, or perform poorly, as shown by our testbed experiments. In this paper, we design and implement MUSE, a lightweight user grouping algorithm, which addresses the above challenges. Our experiments with commodity 802.11ac testbeds show MUSE can achieve high throughput gains over existing designs.

Mining Usage Patterns in Residential Intranet of Things

Ubiquitous smart technologies gradually transform modern homes into Intranet of Things, where a multitude of connected devices allow for novel home automation services (e.g., energy or bandwidth savings, comfort enhancement, etc.). Optimizing and enriching the Quality of Experience (QoE) of residential users emerges as a critical differentiator for Internet and Communication Service providers (ISPs and CSPs, respectively) and heavily relies on the analysis of various kinds of data (connectivity, performance, usage) gathered from home networks. In this paper, we are interested in new Machine-to-Machine data analysis techniques that go beyond binary association rule mining for traditional market basket analysis considered by previous works, to analyze individual device logs of home gateways. Based on multidimensional patterns mining framework, we extract complex device co-usage patterns of 201 residential broadband users of an ISP, subscribed to a triple-play service. Such fine-grained device usage patterns provide valuable insights for emerging use cases such as an adaptive usage of home devices, and also “things” recommendation.

Characterizing mobile user habits: The case for energy budgeting

In this paper, we collect and analyze data from 85 smartphone users over a 9 month period. Different from existing work, we study device usage patterns in concert with network performance in space and time. Our results uncover predictable mobility patterns, where users are moving between hubs (i.e., home or workplace) and transit locations. In hubs, users are typically connected using Wi-Fi, while in transit locations cellular connectivity dominates with highly varying performance (from EDGE to HSPA+). Interestingly, there are set of apps over time running on user devices, independent of the location, network conditions, and device resources (e.g., battery level). These apps can aggressively use the network, which leads to significant device resource consumption (e.g., energy), as shown by our controlled experiments. We discuss how our findings can be used to budget mobile device available resources and improve user experience.

LTERadar: Towards LTE-Aware Wi-Fi Access Points

Major LTE hardware vendors (e.g. Qualcomm, Ericsson), mobile service providers (e.g. Verizon, T-Mobile), and standardization bodies (e.g. LTE-U forum, 3GPP) are extending LTE networks into unlicensed spectrum bands to boost the speeds and coverage of mobile networks. However, the deployment of LTE in unlicensed has raised serious concerns regarding their adverse impact on Wi-Fi networks in the same bands. We design LTERadar, a lightweight interference detector, that runs on Wi-Fi devices and accurately detects LTE interference in real time. LTERadar is a purely software-based solution that is independent of specific hardware or technology of the LTE interferer (e.g. LTE-U, LAA, the dominant LTE unlicensed protocols). Our implementation and evaluation with off-the-shelf Wi-Fi APs show that LTERadar achieves more than 90% of the interference detection accuracy in operational networks.

Accurate 3D Localization for 60 GHz Networks

RF-based (mainly Wi-Fi and BLE) localization systems have been gaining the significant amount of attention from industry, since they do not require the deployment of any special and costly infrastructure such as cameras and depth sensors but leverage existing wireless Access Points (APs) and mobile clients. However, extensive field trials have shown that such solutions can be inaccurate (e.g., meter-level localization error) and/or impractical (e.g., requiring centrally-controlled, dense AP deployments). In this paper, we focus on millimeter-wave wireless networks, which are becoming increasingly popular for both indoors and outdoors connectivity, to overcome the aforementioned limitations and to provide accurate 3D localization. Specifically, we propose mWaveLoc, a system that exploits the small wavelength (millimeter) and directional communication of millimeter-wave (e.g., in 60 GHz) networks for accurate and practical 3D localization. mWaveLoc relies only on a single AP to track a devices position and works with existing off-the-shelf 802.11ad 60 GHz devices. Our implementation and experimental results on commodity 802.11ad testbeds show that mWaveLoc can achieve centimeter-level AP-client distance estimation accuracy and decimeter-level 3D localization accuracy in Line-Of-Sight settings.

Extracting usage patterns of home IoT devices

Ubiquitous connectivity and smart technologies gradually transform homes into Intranet of Things, where a multitude of connected, intelligent devices allow for novel home automation services. Providing new services for home users (e.g., energy saving automations) and Internet Service Providers (e.g., network management and troubleshooting) requires an in-depth analysis of various kinds of data (connectivity, performance, usage) collected from home networks. In this paper, we explore new Machine-to-Machine data analysis techniques that go beyond binary association rule mining for traditional market basket analysis considered by previous studies, to analyze individual device logs of home gateways. We introduce a multidimensional patterns mining framework, to extract complex device co-usage patterns of 201 residential broadband users of an ISP, subscribed to a triple-play service. Our results show that our analytics engine provides valuable insights for emerging use cases such as monitoring for energy efficiency, and “things” recommendation.

Demo: WiFi-Assisted 60 GHz Wireless Networks

Despite years of innovative research and development, multi-Gbps 60 GHz wireless networks are still not mainstream. The unfavorable propagation characteristics due to short wavelength and high directionality, makes the 60 GHz links highly vulnerable to blockage and mobility. However, the advent of multi-band chipsets opens the possibility of leveraging the more robust WiFi technology to assist 60 GHz in order to provide seamless, Gbps connectivity. In this demonstration, we will present MUST, an 802.11-compliant real-time system that provides seamless, high-speed connectivity over multi-band 60 GHz and WiFi devices. MUST has two key design components: (1) a WiFi-assisted 60 GHz link adaptation algorithm, which can instantaneously predict the best beam and PHY rate setting, with zero probing overhead at 60 GHz; and (2) a proactive blockage detection and switching algorithm which can re-direct ongoing user traffic to the robust interface within sub-10 ms latency. We have implemented MUST on off-the-shelf devices where our experiments show high throughput gain and almost 2 orders of magnitude cross-band switching latency improvement over state-of-the-art solutions.