This question appears in Mumbai University > Wireless Networks subject

Marks: 10 M

• Mate :

i. It is the Middleware for WSN which uses the Virtual Machine approach and it runs on top of the TinyOS.

ii. Mate has a high level user interface and the programs of this middleware are split into 24 instruction packets called capsules which can be fitted into a single TinyOS packet.

iii. Mate is basically a stack-based architecture and it has a concise instruction set that comprises three types of instructions performing different operations. This basically means that Mate works as a byte code interpreter. The program is injected into sensor networkfaster and easily by using Mate.

• Impala :

i. Impala uses modular programming approach. Impala basically provides mechanisms for network updates that are efficient enough to support dynamic applications.

ii. It also has autonomic behavior which increases its fault tolerance and network self-organization. However, the nature of its code instruction is that it doesn't allow hardware heterogeneity, which makes it unsuitable for devices with limited resources.

iii. Impala is energy efficient and its software is easily updatable but it does not provide QoS support.

• SINA :

i. SINA, Sensor Information and networking architecture is a middleware which allows sensor applications to issue queries and command tasks into the network, collect replies and results from the network, and monitor changes within these networks.

ii. SINA is mainly composed of three components: hierarchical clustering, which consists of grouping of the nodes based on their closeness or on their energy levels into clusters; attribute-based naming which replaces the standard id-based naming which cannot be used in data-centric networks like WSN; and location awareness: nodes should also know their physical location, and that is possible by using GPS.

• MiLAN :

i. MiLAN, Middleware Linking Applications and Networks is based on application driven approach providing QoS support to applications.

ii. MiLAN contains a network protocol stack which is used to configure and manage the network. It also uses a graph based approach to allow application to know how it performs using the collected data from different combinations of heterogeneous sensors and how to choose combination of sensors to satisfy its QoS requirements.

iii. MiLAN was originally designed for medical advising and monitoring.

• MidFusion :

MidFusion is also a middleware based on application driven approach trying to provide QoS support. But it is suitable only for networks which are Bayesian modeled.

• Mires :

i. Mires is an adaptation of the message-oriented middleware, used for traditional fixed distributed systems.

ii. It provides an asynchronous communication model suitable for WSN applications, which are event driven and has more advantages over the traditional request-reply model. It also adopts a component-based programming model using active messages to implement its publish-subscribe-based communication infrastructure.

iii. Mires' architecture basically includes a core component (a Publish-Subscribe service), a routing component, and some additional services, such as data aggregation. The Publish-Subscribe service manages the communication between middleware services. It also manages the topics list and the subscribed application to give the right topic to the related application.

iv. Mires sends only messages referring to subscribed topics, hence reducing the number of transmissions and energy consumption. Mires does not provide QoS support and security.

• TinyLIME :

i. This is also basically a database approach but unlike other database approaches it has a different programming paradigm.

ii. TinyLIME is a database middleware built over TinyOS which is based on LIME. TinyLIME extends

iii. LIME by adding features specialized for sensor networks which are not supported by LIME. But, TinyLIME is designed for environments in which clients only need to query data from local sensors. It does not provide multi-hop propagation of data through the sensor network.

• TinyDB :

i. It is designed and implemented as Acquisitional Query Processing (ACQP) system for collecting data from a sensor network.

ii. Comparing it to traditional technology, it has capabilities such as low power consumption and accurate query results.

iii. These are important advantages in a resource limited network environment. TinyDB is a distributed system. It runs on the top of TinyOS, with Structured Query Language (SQL) like interface to execute data from sensor nodes.

• Cougar :

i. Cougar applies database approach in sensor network. Basically, there are two types of data: stored data and sensor data.

ii. Signal processing functions in each node generate the required sensor data, and the data are communicated or stored in local storage facilities as relations in database system.Signal processing functions are modeled by using Abstract Data Type in Cougar.

iii. Cougar also uses SQL like language to implement queries.

• MagnetOS :

i. This middleware uses a VM. MagnetOS is a power-aware and an adaptive operating system which is specially designed for sensor and ad hoc networks.

ii. It constitutes a layer called a Single System Image, which provides higher abstraction for the heterogeneity of ad hoc networks' distributed nature. This abstraction lets the whole network appear as a single, unified Java VM.