Shinji Motegi

DAG based in-network aggregation for sensor network monitoring

Wireless sensor network monitoring is important for network maintenance, since it keeps the observer aware of node failures, resource depletion etc. Since communication overheads increase if the sink collects data individually from all sensor nodes, in-network data aggregation methods have been proposed which reduce the overheads. They form a routing tree and data follows up from the edge of the tree to the sink. However, in the event of heavy packet loss, the error margin of the collected data received by the sink grows. Furthermore, when the assumed hop count of the edge of the tree is smaller than the actual count, data can not be followed up from the edge. For the reasons mentioned above, observers find it problematic to assess the state of the network, since the error margin increases as the accuracy of the collected data falls. In this paper, we propose a new in-network aggregation method for sensor network monitoring. The method provides fault tolerance for packet loss by forming a directed acyclic graph (DAG), which allows a node to have multiple parent nodes. In addition, the method can ensure correct data transmission timing, according to the actual hop count of the edge of the DAG. Furthermore, we evaluated the proposed method in comparison with the existing methods, from the perspective of the error margin of the collected data

Implementation and evaluation of on-demand address allocation for event-driven sensor network

In this paper, we propose an energy-efficient address allocation method for the event-driven sensor network, and implement and evaluate the proposed method. The proposed method allocates a temporary address only to a sensor node which detects an event, on an on-demand basis. By performing simulation studies, we evaluated the proposed method and compared it with one of the existing methods from the perspective of the number of control messages for the address allocation. The results show that the number of control messages of the proposed method is small compared to that of the existing methods. We also evaluated the processing time overhead of the proposed method using the implemented system. Although the proposed method has little extra overhead, the results show the processing time is short enough for practical use.

Relay Control for Data Dissemination of Spontaneous Vehicular Networks

Spontaneous vehicular networks formed by vehicles contribute to safe and comfortable driving by disseminating vehicular movement information. Inter-vehicle communication protocols have been proposed that can provide indirect communications by relaying packets. Although the relay expands the area over which packets can deliver, it generates new communication traffic, hence existing protocols have some mechanism to reduce this traffic, for example the selection of a minimum set of relaying vehicles. However, communication delays increase due to the overhead of the mechanism when the numbers of vehicles increase. In this paper, we propose a relay control protocol for the data dissemination of spontaneous vehicular networks. The protocol controls the relay of each packet, on an as-required basis, in order to adapt to various vehicular communication environments. In addition, the protocol reduces communication traffic by avoiding duplicate relays through a duplicate relay detection algorithm. Furthermore, we evaluate the proposed protocol through experiments and compare it with one of the existing protocols

Performance Evaluation of Protocols for Inter-Vehicle Communications

Inter-vehicle communication (IVC) protocols have been studied to provide safe and comfortable driving. Each vehicle using the IVC protocols periodically broadcasts its current information (location, moving direction, speed, etc.), and other vehicles must receive such information exactly and in time. Other vehicles use received information to alert, advise, and navigate drivers, helping them become aware of the existence of other vehicles. This kind of application is helpful in various situations. For example, when a driver is approaching an intersection, the driver is not aware of vehicles that are also approaching the same intersection from another direction. To realize such inter-vehicle communication, we need to consider many issues such as protocol design, wireless standard, evaluation methodology, etc. Previous works on both mobile ad hoc networks and inter-vehicle communications used conventional metrics, e.g., packet delivery ratio, average delay, and path optimality, to study the performance of protocols. However, such metrics cannot be applied to inter-vehicle communications directly because the identities of the prospective receivers are a priori unknown. Moreover, many requirements must be considered to judge whether a IVC protocol satisfies the objectives of IVC applications. Although a vehicle get information properly, it is useless if that information arrives too late. This paper proposes new performance metrics to evaluate IVC protocols by the means of reliable and timely communications. We also introduce a methodology to perform realistic evaluation through simulation. Realistic vehicular traces and simulation models are required to get correct evaluation results. We then use the defined metrics to evaluate the previously proposed protocol [1]. According to the simulation results, the proposed protocol is a good candidate for real implementation because it passes all requirements of inter-vehicle communications.

Adaptive channel and time allocation for body area networks

Personal health care is a promising application of a body area network (BAN). The BAN for health care application consists of a data collector and multiple wearable sensors that send data to the collector according to a sampling interval determined by the application. The data collector may be any hand-held or portable device such as a mobile phone, a PDA or a wristwatch. An individual user needs a BAN to detect, track or manage his/her health. Multiple BANs for such purpose are likely to appear in a nearby area, especially a place where population density is high. Since multiple BANs must share wireless medium in this scenario, a problem of packet collisions among BANs is unavoidable. To deal with such problem, this study proposes an adaptive scheme to allocate channel and time for coexisting BANs. The proposed scheme distinguishes inter-BAN and intra-BAN communications, and adaptively allocates channel and time according to the current number of BANs that exist in a nearby area. As a salient feature, the proposed scheme allocates time by attempting to satisfy the requirement of sampling interval determined by the application. The evaluation results demonstrate that our scheme achieves much higher packet delivery rate than the existing scheme.

Resource Allocation for Coexisting ZigBee-Based Personal Area Networks

Personal healthcare is a promising market of a personal area network (PAN). The PAN consists of a data collector and multiple wearable sensors which must send data to the collector according to the frequency specified by each PAN. The data collector may be any hand-held or portable device, e.g., a mobile phone, a PDA, or a wristwatch. An individual user needs a PAN in order to detect, track, or manage his/her health. Multiple PANs for such purposes are likely to coexist in a same area, especially a place where a large number of users exist. Because multiple PANs must share resources (e.g., wireless channels, transmitting time) in this scenario, a problem of packet collisions among such PANs is unavoidable. As a result, the personal healthcare system using PAN cannot be provided as a service for the users because the data collector cannot collect sensing data from each sensor correctly. This paper proposes an adaptive scheme to allocate resources for coexisting PANs. The scheme distinguishes a PAN and a node when allocating resources such as wireless channel and transmitting time. Resources are dynamically allocated according to the current number of PANs that exist in the same area. In particular, the proposed scheme automatically reallocates resources when any PANs come into or leave from the coexisting area. As one of salient features, the scheme allocates transmitting time to meet a requirement on reporting interval determined by an application. Thus applications that are sensitive to delay can collect sensing data correctly and in time. The evaluation results show that our scheme achieves higher packet delivery rate than the standard ZigBee.

Address Autoconfiguration for Event-Driven Sensor Network

An event-driven sensor network composed of a large number of sensor nodes has been widely studied. A sensor node sends packets to a sink when the node detects an event. For the sink to receive packets it fails to acquire, the sink must send re-transmission requests to the sensor node. To send the requests to the sensor node using unicast, the network address of the sensor node is required to distinguish the sensor node from others. Since it is difficult to allocate the address manually to a number of nodes, a reasonable option is to use existing address autoconfiguration methods. However, the methods waste the limited energy of the sensor nodes due to using a number of control messages to allocate a permanent address to every node. In this paper, we propose an energy-efficient address autoconfiguration method for the event-driven sensor network. The proposed method allocates a temporary address only to a sensor node which detects an event, on an on-demand basis. By performing simulation studies, we evaluated the proposed method and compared it with one of the existing methods based on the number of control messages for the address allocation. The results show that the number of control messages of the proposed method is small compared to that of the existing method. We also evaluated the process time overhead of the proposed method using the implemented system. Although the proposed method has little extra overhead, the results show the processing time is short enough for practical use.

Design and implementation of "kubit" for sensing and control ubiquitous applications

Towards the wide acceptance of ubiquitous networking, practical and deployable ubiquitous applications must be provided in a concrete and easily understandable form to end-users. We designed and implemented ubiquitous nodes, KDDI Ubiquitous Bit (kubit). The highest priority was given to application-oriented design and ease of deployment in a typical home network environment. The kubit quickly detects anything suspicious with built-in sensors, and sends notification of the event via e-mail containing an image to be displayed on a PC or mobile phone, as well as controlling the existing home appliances via a PC or mobile phone or led by the event.

Poster: Node placement adviser for sensor network deployment

In this poster, we propose a new tool, i.e. a node placement adviser, which supports deployment workers of sensor networks. The adviser tool is realized in the form of mobile terminal software and acts as a proxy of an additional node, as well as communicating with those already placed. Based on the communication results, it indicates how favorable the current location is to a worker.

Communication Algorithm for Statistic Monitoring in People-Centric Sensing Networks

Mobile phones have great potential to act as global mobile sensing devices due to already being equipped with sensors, e.g., sound, image and acceleration sensors. People as individuals or special interest groups can apply the new sensing devices to form sensing networks called people-centric sensing networks that sense what we are doing and support our daily activities. Most of the sensing applications depend on the ability to monitor statistics including max, average, ranking, rather than raw sensor readings. Such an integration of sensing-capable mobile phones into the networking infrastructure shifts the network’s main utility from data communication to information filtering. Thus, data reduction is one of the major networking challenges for people-centric sensing networks. In this paper, we propose a communication algorithm for statistic monitoring in people-centric sensing networks. The main feature of the proposed algorithm is the ability to reduce communication traffic by eliminating redundant sensing data transmission without increasing the error of the statistics during monitoring. We show how the proposed algorithm is applied to a ranking monitoring application. We also include a simulation study demonstrating the advantages of the proposed algorithm. Furthermore, we show a prototype system where the proposed algorithm is implemented into a mobile phone.