The objective of this research program is to devise and experiment techniques with mobile network and wireless technologies, protocols and applications. The research activities of the group include WPAN technologies (Bluetooth and UWB), Mobile ad hoc network (MANET), multihop cellular networks and the current new interests in mobile game development.

WPAN Technologies (Bluetooth and UWB)

Bluetooth

Development of Bluetooth applications is an area which NTRC has been actively pursuing. With keen technical and development support from leading Bluetooth solution providers such as Ericsson and CSR (Cambridge Silicon Radio), an application development team under the supervision of Assoc Prof Tong Yit Chow has developed a number of working prototypes for such devices as Bluetooth-enabled GSM/GPSR smart phones, Bluetooth LAN access point, wireless lab equipment, wireless location/product directory service, automated utility (e.g. electricity, gas, fuel, water) meter reading, and surveillance cams.

UWB Ad-Hoc Network

Recently, work has also been initiated to develop a generic wireless system based on UWB to support pervasive computing and communication in a home/office environment. At present, efforts are focused on protocol stack and embedded OS development for the generic system.  Nomadic WPAN applications that exploit the key advantages of UWB, including high data rate, robust multipath performance as well as high-accuracy indoor localization, will be developed to enable intercommunication and sharing of multimedia content and broadband services among a variety of UWB-enabled devices anywhere in and around the home/office environment.

In a UWB ad hoc network, several piconets (a piconet is a collection of several wireless (personal) devices that can communicate among devices) are formed in an ad hoc way without any infrastructure assistance. All the piconets are allowed to communicate with one another. The key issues to deploy a UWB ad hoc network includes:

• Piconet formation

• Packet routing

• Power-saving strategy

• Multiple access development

• Radio resource sharing

In the development of UWB protocol stack related to network layer in this project, we are focusing on the technique to form a piconet more efficiently and develop an effective routing protocol to forward packets from the source to the destination.

The formation of piconets will be based on a cluster-based ad hoc technique because clusterbased ad hoc is able to i) facilitate the wireless transmission management among devices; ii) provide a natural virtual backbone for inter-cluster routing; and iii) extract network information and require less information exchange among devices. The other research focus is to develop a routing protocol for use in UWB ad hoc network. We concentrate on the performance of the routing protocol to support pure IP ‘best effort’ traffic or data services with different quality-of-service requirement. It is possible that with multi-hop routing, UWB transmitters could reduce their transmit power emissions and thus allow more devices to be operated at the same time.

Currently, we are implementing some basic functions of ad hoc network using UWB radio technology as shown in the Figure 3.2. The source of the UWB transmitter will send the message to the destination via the one of the closest intermediate nodes. Some of the techniques, like packet forwarding and path selection strategies, are awaited to be realised. The idea of this project could be considered to apply for the future UWB ad hoc network.

Multihop Cellular Networks

Multihop cellular networks have been proposed as an extension to the conventional single-hop cellular network by combining the infrastructure that is used in ad hoc networks as shown in the Figure 3.3. Due to the potential of the multihop relaying to enhance coverage, capacity and flexibility, multihop cellular networks have been attracting considerable attention for future 3G and beyond systems. This network architecture provides future cellular systems with the possibility of peer-to-peer (mobile to mobile) communication as well as communication relayed through other fixed and/or mobile terminals, unlike present-day centralized cellular networks where every communication has to go through the central base station.

Due to high power level difference between transmit and receive signals, simultaneous transmission and reception in the same frequency band is not practical. Thus, WCDMA-TDD mode transmission is a suitable duplexing mode for the multihop cellular network. Moreover, the multihop transmission means that it will cost more radio resources for just one transmission. And different selection algorithm of the relay station could give different performance to the overall system. Thus, in order to meet the flexible requirement for the next generation cellular services, our project consists of the following tasks:

1. Dynamically allocate the WCDMA-TDD radio resource among mobile users in the service area. The relationship among coverage area, position of the relay station, different power control schemes and the system capacity are studied to achieve the best compromise

2. New multihop routing protocol that considers the specifics of the cellular environment, including the presence of base stations, asymmetric traffic (uplink versa downlink traffic), hop-point area and heterogeneous traffic types in cellular networks, will be investigated. Currently, the system model for the first task has already been built. Based on this model, both simulation and analysis results are accomplished. We are now studying different power control algorithms to achieve the best performance.

Mobile ad hoc network (MANET)

Introduction of MANET

The proliferation of wireless communication and mobile devices in recent years has opened the door to research on self-organizing networks that do not require a pre-established infrastructure as shown in Figure 3.4. Those spontaneous networks, normally called ad hoc networks, provide mobile users with ubiquitous communication capacity and information access regardless of the location.

The most important characteristic of such network is the independence of any fixed infrastructure or centralized administration. An ad hoc network is capable of operating autonomously and is completely self-organizing and self-configuring. Therefore, it can be rapid and easily deployed.

Another important property of an ad hoc network is the multihop capability. Unlike the cellular networks, which are single-hop wireless networks, an ad hoc network does not guarantee that a mobile node can directly communicate with its destinations all the time. A mobile node, which lies outside the transmission range of its specific destination, would need to relay its information flow through other mobile nodes. This implies that mobile nodes in ad hoc networks bear routing functionality so that they can act both as routers and hosts.

An ad hoc network can be deployed in an area where supports for mobile communication is not available, probably due to high deployment costs or disaster destruction. The typical application of ad hoc networks includes battle field communication, emergency relief and extension of the coverage area of cellular networks.

Project Description

It is a big research challenge to design suitable protocols especially at the MAC and the network layer that can cope well with high mobility, limited bandwidth and power scarcity of a typical ad hoc network. Our research focuses on designing an energy-efficient routing protocol. Dynamic routing is an important behavior of an ad hoc network so that energy metrics should be taken into consideration for the guarantee of the routing performance. The early “death” of some mobile nodes due to energy depletion may cause severe problems, such as network partition and communication interruption. The main objective of our current research topic is to maximally balance the energy consumption of mobile nodes in a network in order to prolong the whole network lifetime.

Current Progress

We have designed a routing algorithm named Energy Balancing Hybrid Routing (EBHR) and tested its performance with network simulator (ns2). In comparison with the typical dynamic routing protocol, such as dynamic source routing (DSR) and ad-hoc on-demand distance vector routing (AODV), and other energy balancing routing protocols, such as local energy-aware routing (LEAR) and time delay on-demand routing (TDOR). It was found that EBHR can actually achieve better performance in terms of energy balancing. At the same time, EBHR has satisfactory packet delivery ratio and end-to-end delay, which are important parameters for a routing protocol and are therefore comparable to those of DSR.

Mobile Game Development

Mobile gaming is the next in-thing of the future. According to market research reports by market research firms, the global games industry was worth US$31 billion for year 2003. For handheld and mobile platforms markets alone, it formed about 15.5% of the revenue in the global gaming industry. Last year, the Asia-Pacific region (excluding Japan) contributed 22 per cent of the global revenue from mobile gaming and predicted that in 2006, the figure will rise to 41 per cent.

In Singapore, established mobile gaming platforms such as handheld Nintendo games and Java enabled mobile phones have a loyal following. According to a Nokia spokesman, some of the more popular downloaded Java games for mobile phones include Puzzle Bobble, Mahjong 13 and Big 2. The spokesman also added that Java game downloads are increasing, especially when phones with colour screens are becoming more popular. These phones allow more fanciful games to be developed. More and more developers are creating Java games too.

In the light of these developments, Professor Tony Woo, Vice-President of Research, has initiated the mobile game projects. These projects are funded by the research office and supported by the GameLab of SCE. The objective is to provide students with a common interest in mobile gaming to work together and create mobile games. The initiative had selected 6 undergraduate students from school of EEE and SCE to work on this one-year project. In addition, some undergraduate students under Dr Peter H J Chong had also participated in the development work. The students designed the game with their own ideas and implemented through Java (J2ME) and C++ programming language. Following describes some of the games being developed and implement on the Bluetooth-enabled phone for multi-play gaming.