Kenji Yoshigoe

The CICQ Switch with Virtual Crosspoint Queues for Large RTT

The memory size required for a combined input and crosspoint queued (CICQ) switch with an existing credit-based flow control is proportional to the flow control latency between the line cards and switch fabric. Recently, scalable packet switches were implemented with distributed units introducing a large RTT latency between the line cards and switch fabric, making the CP buffer requirement impractical. In this paper, a CICQ switch with virtual crosspoint queues (VCQs) is investigated. The VCQs unit resides inside the switch fabric and is dynamically shared among virtual output queues (VOQ) from the same source port and is operated at the line rate, making the implementation practical. The CICQ switch with VCQ can achieve high throughput for unbalanced traffic while reducing over 87% of the total buffer size requirement in the switch fabric, thus making the scalable implementation of a distributed CICQ switch practical.

Software-Defined Fog Network Architecture for IoT

Rapid increase in number and diversity of Internet-connected devices raises many challenges for the traditional network architecture, which is not designed to support a high level of scalability, real-time data delivery and mobility. To address these issues, in this paper we present a new model of Internet of Things architecture which combines benefits of two emerging technologies: software-defined networking and Fog computing. Software-defined networking implies a logically centralized network control plane, which allows implementation of sophisticated mechanisms for traffic control and resource management. On the other hand, Fog computing enables some data to be analysed and managed at the network edge, thus providing support for applications that require very low and predictable latency. In the paper, we give detailed insight into the system structure and functionality of its main components. We also discuss the benefits of the proposed architecture and its potential services.

Overcoming invasion of privacy in smart home environment with synthetic packet injection

We live in an increasingly connected world where not only can we open our garage doors with sensors, but from a remote location we can also lock our doors, or turn off our lights. Such convenience and flexibility come with great concerns for privacy and security. In this paper, we study privacy of smart home devices in home residence settings and show how homeowner’s privacy could be compromised via simple network traffic analysis. We first measure normal traffic patterns generated on commercial off-the-shelf (COTS) smart home devices, and identify possible privacy vulnerabilities. We designed a smart home hub-integrated solution to mitigate such risk by obscuring real network traffic with synthetic traffic. We suggest that the smart home industry consider incorporating this approach into their products to improve privacy in the smart home environment.

Threshold-based Exhaustive Round-Robin for the CICQ Switch with Virtual Crosspoint Queues

A multi-cabinet implementation of a combined input and crosspoint queued (CICQ) switch introduces a large RTT latency between the line cards and switch fabric, requiring a large crosspoint (CP) buffer. A virtual crosspoint queues (VCQs), proposed in literature are shared among a set of virtual output queues (VOQs) and CP buffers for the same input port, reducing minimal memory size required inside the switch fabric. In this paper, a threshold-based exhaustive round-robin (T-ERR) arbitration is employed to improve the throughput of the CICQ switch with VCQs. The T-ERR at VCQ and CP arbiters serves packets residing in a longer queue more frequently than packet residing in a shorter queue. The T-ERR is simple yet drastically increases the throughput of the CICQ switch with small CP buffers. Simulation evaluations with unbalanced traffic show that the throughput of the CICQ switch with VCQs improves from 80% to 94% for CP buffer size of 4 cells and 73% to 83% for CP buffer size of 2 cells for RTT = 64 cell time. Furthermore, its throughput is independent of switch size and RTT. Thus, the proposed scheme makes the scalable implementation of a distributed CICQ switch practical.

RC4-2S: RC4 Stream Cipher with Two State Tables

One of the most important symmetric cryptographic algorithms is Rivest Cipher 4 (RC4) stream cipher which can be applied to many security applications in real time security. However, RC4 cipher shows some weaknesses including a correlation problem between the public known outputs of the internal state. We propose RC4 stream cipher with two state tables (RC4-2S) as an enhancement to RC4. RC4-2S stream cipher system solves the correlation problem between the public known outputs of the internal state using permutation between state 1 (S 1 ) and state 2 (S 2 ). Furthermore, key generation time of the RC4-2S is faster than that of the original RC4 due to less number of operations per a key generation required by the former. The experimental results confirm that the output streams generated by the RC4-2S are more random than that generated by RC4 while requiring less time than RC4. Moreover, RC4-2S’s high resistivity protects against many attacks vulnerable to RC4 and solves several weaknesses of RC4 such as distinguishing attack.

Mining Twitter to Assess the Public Perception of the "Internet of Things".

Social media analysis has shown tremendous potential to understand publics opinion on a wide variety of topics. In this paper, we have mined Twitter to understand the publics perception of the Internet of Things (IoT). We first generated the discussion trends of the IoT from multiple Twitter data sources and validated these trends with Google Trends. We then performed sentiment analysis to gain insights of the public’s attitude towards the IoT. As anticipated, our analysis indicates that the publics perception of the IoT is predominantly positive. Further, through topic modeling, we learned that public tweets discussing the IoT were often focused on business and technology. However, the public has great concerns about privacy and security issues toward the IoT based on the frequent appearance of related terms. Nevertheless, no unexpected perceptions were identified through our analysis. Our analysis was challenged by the limited fraction of tweets relevant to our study. Also, the user demographics of Twitter users may not be strongly representative of the population of the general public.

Rate-based flow-control for the CICQ switch

A combined input and crosspoint queued (CICQ) switch with a flow control latency of round-trip time (RTT) packets requires each crosspoint (CP) buffer to hold the RTT packets in order to support any traffic pattern without blocking. In this paper, a distributed rate-based flow control internal to a CICQ switch is proposed. The new rate-based flow control allocates bandwidth to each virtual output queue (VOQ) using simple counters, and a minimal CP buffer requirement can be achieved without packet loss. The proposed rate-based flow control is shown to be more stable than the existing credit-based flow control previously proposed in the literature for realistic traffic while significantly reducing the CP memory size, thus making the implementation of a distributed CICQ switch practical

SDN-based concept of QoS aware heterogeneous wireless network operation

Software defined networking (SDN) is a new paradigm for wired networks which cleanly decouples control and the date plane, leaving all control functionality to logically centralized entity. By centralizing network intelligence it provides better visibility of the network conditions, and hence allows more sophisticated traffic management. In this paper, we investigate how the adoption of SDN concept can improve efficiency of heterogeneous wireless environments. We present architectural solution for SDN-based wireless networks, and elaborate on opportunities it provides from various perspectives. In order to exploit benefits of scalable cloud infrastructure, integration of Network Function Virtualization (NFV) is proposed. In particular, the proposed architectural solution allows fast deployment of new services for network operators, gives flexible content delivery options for application providers, and provides improved quality of experience (QoE) for end users.

LightGraph: Lighten Communication in Distributed Graph-Parallel Processing

A number of graph-structured computing abstractions have been proposed to address the needs of solving complex and large-scale graph algorithms. Distributed Graphlab and its successor, PowerGraph, are two such frameworks that have demonstrated excellent performance with high scalability and fault tolerance. However, excessive communication and state sharing among nodes in these frameworks not only reduce network efficiency but may also cause a decrease in runtime performance. In this paper, we first propose a mechanism that identifies and eliminates the avoidable communication during synchronization in existing distributed graph structured computing abstractions. We have implemented our method on PowerGraph and created LightGraph to reduce communication overhead in distributed graph-parallel computation systems. Furthermore, to minimize the required intra-graph synchronizations for PageRank-like applications, LightGraph also employs an edge direction-aware graph partitioning strategy, which optimally isolates the outgoing edges from the incoming edges of a vertex when creating and distributing replicas among different machines. We have conducted extensive experiments using real-world data, and our results verified the effectiveness of LightGraph. For example, when compared with the best existing graph placement method in PowerGraph, LightGraph can not only reduce up to 27.6% of synchronizing communication overhead for intra-graph synchronizations but also cut up to 17.1% runtime for PageRank.

Consumed-energy-type-aware routing for wireless sensor networks

This study proposes and investigates consumed- energy-type-aware routing (CETAR), a new metric to preserve the energy of active source nodes in the WSNs. CETAR uses statistics of the energy consumed for each type of node activities to select a node which seldom plays a role of source node as a routing node. By preserving the energy of active source nodes, CETAR can prolong the life-span of target specific operations in the WSN in which targets are not precisely known at the deployment of the network. Simulation model shows that the geographical and energy aware routing (GEAR) with CETAR can send significantly more packets than GEAR without CETAR for both uniform and non-uniform traffic regardless of network size. Most importantly, the new metric proposed in this paper can be widely deployed to existing energy aware routing protocols.