With the spread of IoT technology, many applications using sensor data have emerged. In these applications, the accuracy management of sensors is crucial. The measurement accuracy of sensors decreases over time due to a systematic error known as bias. Because the bias can grow significantly over time, it is essential to perform sensor calibration periodically, such as annually, to correct for the bias and the maintain accuracy. Calibration is typically performed manually for each sensor, but as the number of sensors increases, it becomes challenging. Therefore, several automatic calibration methods have been proposed that estimate the correction values of each sensor based on the correlation of the measured values of sensors located in close proximity. However, the existing automatic calibration methods focus solely on estimating optimal correction values, and the accuracy of bias estimation, i.e., the amount of error included, is unclear. We propose a novel automatic calibration method that aims to minimize the bias in all sensors and maintain the bias errors of all sensors within the required accuracy threshold for a long time. The proposed method estimates a probability distribution of each sensor's bias and periodically calibrates a part of the sensors based on the estimated distributions. We show that our method can maintain the bias of each sensor within a required range for a long time through simulations.

The value of personal health data continues to be a debated topic in HCI and society more broadly. We investigate the monetary value people attach to their health data. Using a custom mobile app for 14 days with 55 participants, we collected health data (sleep duration, sleep quality, pain intensity, wake-up times) and a daily monetary data valuation using a reverse second-price auction. Participants bid to sell their data to a for-profit company, the government, or academia. Our findings indicate that people value their data differently based on who is buying. We also show that people are interested in monetizing their personal health data despite privacy and data protection concerns. The presented study helps us understand the data value landscape and paves way to a healthier data-driven future where people may benefit more from their own contributions, either in monetary or other forms.

We propose a new route guidance system (RGS) to mitigate traffic congestion called Flow-based Capacity-aware Rerouting (FCR), which achieves better global optimality by considering the capacity of detour routes. FCR introduces a flow-based traffic management method that enables us to compute detour paths in units of flow rather than per vehicle processing, which reduces the load to calculate detour paths and makes the centralized system practically feasible. With the flow-based traffic management, FCR also enables us to consider traffic volume that should bypass the congested roads to mitigate the congestion and suggest detour paths to the corresponding traffic volume of vehicles. By prioritizing travel time loss when choosing detour routes to apply, FCR leads to better global optimality with the limited volume of rerouting traffic while considering each rerouted vehicle’s benefit, such as travel time loss. FCR offers a new centralized system architecture for actual real-time traffic control, which improves both global optimality and each vehicle’s benefit within a feasible computational load.

Vehicular Ad-hoc NETworks (VANETs) in sparse vehicles scenarios can be regarded as a kind of Delay Tolerant Networks (DTNs), and how to provide reliable and efficient communications over them has been extensively studied. Past studies such as SADV and RDV showed that the assistance of low-cost unwired static nodes located at intersections, which work as routers to provide distance-vector or link-state routing functions, significantly improves the communication performance such as delivery ratio and delivery delay. Especially, RDV provides any previously configured value of expected packet delivery ratio by creating the required number of duplicated packet copies on the shortest paths. Despite of the high delivery ratio achievement, RDV still has problems, that is, traffic concentration in the shortest paths and its large delivery delay. In this paper, we extend RDV by using multiple paths to avoid packet concentration as well as to improve delivery delay while preserving the function to provide the preconfigured expected delivery ratio. Evaluation results show that the proposed method MP-RDV (Multi-Path RDV) achieves high load-balancing performance to provide better network capacity, lower delivery delay, and higher fault tolerance against topology changes.

OLSR is one of the representative proactive routing protocol for MANETs. However, link-state routing scheme in general poses high load with control messages so that reducing the control load is highly required. On the other hand, table-driven distance-vector routing is known as a low-load routing protocol. However, since it requires large time for path convergence in face of topology changes, it is generally not suitable for MANETs. There are several hybrid routing protocols of link-state and distance-vector schemes in the literature to take advantages of both schemes. By applying link-state schemes only for destinations within k-hop distance, they can reduce control load significantly while keeping fast convergence for destinations within k-hop distance. However, they fundamentally have a drawback of slow convergence for destinations outside the k-hop area even if topology change occurs as close as within k-hop distance. In this paper, we propose a new fundamental mechanism of hybrid routing protocol that also apply link-state routing within k-hop area, but achieves fast convergence for all destinations. Our new proposal takes both fast convergence and low-load advantages for every destination, which namely means to achieve a new design of scalable table-driven routing protocols that have both advantages of link-state and distance-vector routing schemes in the fundamental mean.

We propose a new joint MAC and Routing Approach for duty-cycled sensor networks and designed a protocol called JBS (Joint MAC and routing protocol for Beacon-Saving), which achieves far larger network lifetime than conventional ones. Our proposal improves energy performance of receiver-initiated MAC protocols by combining it with routing functionality. In receiver-initiated MAC, leaf nodes of the delivery tree periodically transmit wasteful beacons even though no data frame will be received. So, we propose to omit transmitting beacons of leaf nodes to save the transmission power as well as the accompanying idle listening. As revealed in the past studies, reduction of idle listening has large effect on the energy performance. In our protocol, we build a delivery tree in a distributed manner such that the number of relay nodes are minimized, meaning that we maximize the number of leaf node. When residual battery power of some relay node becomes low, the delivery tree is recomposed so that every relay node in it is with sufficient residual power. Evaluation results show that the effect of our approach to combine routing functionality is far larger than the power saving effect of MAC only.

To support various practical applications that are expected to work over vehicular networks, it is important to develop a network infrastructure that provides quality communications on which applications can rely. However, since a vehicular network is essentially a delay tolerant network (DTN) in which connected paths to the destinations do not always exist, providing reliable communications on such a network is a challenging research goal. In this paper, we propose a new table-driven distance-vector routing protocol called RDV (Reliable Distance-Vector routing) for static-node-assisted vehicular networks. RDV provides the guarantee that the expected packet delivery ratio exceeds the preconfigured value by creating the necessary number of packet copies sent on the shortest path. Evaluation results show that RDV provides a stably high packet delivery ratio and outperforms other conventional routing schemes in both packet delivery ratio and overhead of packet copies.

Wireless Sensor Networks (WSNs) should collect sensing data quickly without increasing the energy consumption in order to deal with various applications, particularly delay-sensitive applications. This paper proposes a method to schedule beacon transmissions for receiver-initiated MAC protocols, which are promising energy efficient protocols, to improve data collection delay. The key idea of the proposed method is scheduling beacon timings in the sequence of distance from leaf node to the root (i.e., sink) node so that data frames are forwarded to the sink node with small duration staying at each node. Each sensor node selects a time slot to send beacon in distributed manner based on their distances (i.e. hop counts) from the sink node. In addition, each node autonomously adjusts its beacon timing to avoid frame collisions. Computer simulations show that the proposed method can collect data in a shorter time with less energy consumption than conventional beacon scheduling methods for receiver-initiated MAC protocols.

MANET (Mobile Ad-hoc NETworks) is useful in many practical scenarios since it provides multi-hop communication without wired infrastructure. However, there is a problem that the communication performance of a flow may be easily degraded by even a single local congestion on the whole path. A solution for the problem is to use a detour path that avoids the local congestion. However, to this end, the detour paths should not use the nodes in the congested area, which is in fact relatively large due to the nature of radio waves. In the current state of the art, we do not have such alternative-path computation algorithms. In this paper, we propose an algorithm and a routing scheme to compute and utilize detour paths adaptively according to the network traffic conditions. Through evaluation, we show that the proposed scheme improve the communication performance by using the detour paths in practical network scenarios.

IEEE802.11-based Wireless Mesh Networks (WMNs) has been deeply studied as a low-cost wireless mesh networks. However, they still suffer from severe interference due to hidden terminals, which significantly degrades the performance to an impractical level. Several studies tried to use multiple frequency channels to avoid collisions among wireless links. However, static channel assignment requires so many orthogonal channels to achieve collision-freedom, resulting in that we still suffer from severe collisions as long as we use the well-populated IEEE802.11 standards. In this paper, we propose a new static channel assignment scheme CASCA (CSMA-aware Static Channel Assignment) that achieves collision-free channel assignment under 3-4 available channels by incorporating a CSMA-aware interference model as well as partially a routing functionality in it. CASCA enables us to realize collision-free WMNs using the license-free IEEE802.11 devices.

Although Wireless Mesh Networks (WMNs) over CSMA and TDMA MAC have been extensively studied, CSMA still suffers from heavy interference due to hidden terminals, while TDMA has a practical difficulty in accurate time synchronization. Slotted CSMA, in which we assume that CSMA runs within relatively large time-divided slots, could be one of the solutions since it works on loosely synchronized time slots and simultaneously avoids major collision under a properly designed feasible schedule. However, if we apply a schedule generally used in TDMA, there are several inconveniences: First, we need heavy overhead of centralized computation or negotiation among nodes in distributed scheduling. Second, end-to-end packet delivery delay grows large since TDMA requires a large number of slots to achieve collision-free scheduling. In this paper, we propose a new scheduling algorithm for slotted CSMA called CATBS (CSMA-Aware Time-Boundable Scheduling) that requires a smaller number of slots for feasible scheduling, and runs within a bounded time. Incorporating the scheduling, we design a routing protocol that enables flexible paths recomputation with small communication overhead. We consequently achieve a new fully distributed and robust network architecture of WMNs that provides practically high throughput and low delivery delay, which can be implemented on the off-the-shelf IEEE802.11 interfaces.

We propose a new static-node-assisted routing scheme RDV (Reliable Distance Vector routing) for vehicular networks. In RDV, static nodes located at intersections behave as distance-vector routers to assist packet forwarding, i.e., they exchange messages, compute paths, and help packets travel along the computed forwarding paths to their destinations. To support various practical applications over vehicular networks, RDV provides reliable communications such that the average packet delivery ratio exceeds the preconfigured delivery ratio by creating the necessary number of packet copies sent on the shortest path. Evaluation results show that the proposed scheme provides stably high packet delivery ratio, and outperforms other existing routing schemes in both packet delivery ratio and the number of generated packet copies.

Several approaches for IP fast rerouting have been proposed to augment resiliency of IP networks. For example, NotVia uses IP tunnels to avoid failure component, and FIFR prepares two routing tables to switch packets according to incoming interfaces. In this paper, for two-table approach such as FIFR, we propose to use 2-bit marking on packet header to overcome one of the problems called route bouncing. By preventing route bouncing, more efficient use of network resources are possible. We further show that our method not only prevents route bouncing but also prevents packet loops in case of multiple failure. Note that, although multiple failure occurs with low probability, once it occurs network performance significantly degrades since size of loops is far larger than microloops.

It is important to predict interaction among proteins to understand the system of life. We have proposed a method to predict threeway interactions among proteins from expression data based on conditional probability in our previous study. However, the proposed method has a problem of significant computational time, because it requires computation of a probability distribution as a preprocess of the algorithm. In this study, we propose a method to significantly reduce the computational time by computing the probability distribution with analytical approach, instead of simulation approach used in our previous study. We evaluated the computational time and accuracy of the proposed analytical computation method.

Several load balancing techniques for IP routing scheme have appeared in the literature. However, they require optimization process to compute optimal paths to meet traffic demand so that it requires a mechanism to measure traffic demand and to share them among all routes in order to follow dynamics of traffic. It naturally results in communication overhead and losing sensitivity to follow traffic dynamics. In this paper, we investigate a load balancing mechanism from another approach, i.e., based on IP fast reroute mechanisms. The main idea is simply to forward packets into detour paths supplied by IP fast reroute mechanisms only when packets meet congestion. This strategy enables us to use vacant resources adaptively as soon as they are required to avoid and dissolve the congestion. Through traffic simulation we show that IP fast reroute based load balancing mechanisms improve the capacity of networks.

Many dynamic metrics for wireless mesh networks (WMNs) to improve communication quality are proposed so far. However, all of them have possibility to create temporary routing loops which may cause severe congestion and disruption of communications. In order that WMNs are regarded as a reliable infrastructure of the next generation, such routing loops are strongly desired to be liminated since they bring the possibility of ervice disruption. In this paper, we propose a mechanism called Loop-free Metric Range (LMR) to make existing dynamic metrics loop-free by restricting the range of metric values to change. LMR is applicable to a major part of existing etrics such as ETX, ETT, MIC, and so on without any message overhead. We show the mechanisms of LMR, theoretical results on the range of loop-free etric values, and simulation results on the effectiveness of LMR.

To improve communication quality in wireless ad hoc networks, it is definitely important to select stable links as communication paths. To this end, various mobility metrics are presented in the literature. However, the routing loop problem, which is inevitably involved in link-state proactive routing scheme, has not been discussed well. In this paper, we propose a simple dynamic metrics in respect of link stability, which is loop-free throughout dynamic metric transition. Through a theoretical analysis of loop-freeness, we present our new strategy to achieve loop-freeness with sufficiently low cost for ad hoc networks.