There are two distinct approaches for enabling wireless mobile computers to communicate with each other. The first is to utilize the existing cellular network infrastructure originally developed for voice communications. The other approach is to let users who wish to communicate with each other form an ad-hoc network network and collaborate among themselves to deliver data packets from a source to its destination possibly via one or more intermediate nodes. This form of networking, although limited in range by the individual node's transmission range, has several advantages when compared to traditional cellular systems including on-demand setup, fault tolerance, unconstrained connectivity. A mobile ad hoc network is an autonomous system of mobile routers connected by wireless links. The routers are free to move randomly and organize themselves arbitrarily. Thus, the network's wireless topology may change rapidly and unpredictably. Such a network may operate in a stand-alone fashion, or may be connected to the larger Internet. Studies conducted so far on protocol performance and other communications aspects in wireless ad-hoc networking have been based on analyses of simulation results. Our goal is to provide an actual implementation of a mobile ad-hoc network by making use of open wireless LAN standards like IEEE 802.11, TCP/IP protocol suite. This is done through a viability study of how an ad-hoc network can be implemented using several LAN segments each of which is interconnected using standard protocols for bridges and routers. This ad-hoc network should eventually be able to function as a virtual private network(VPN). By implementing a VPN, the performance of applications such as audio-video conferencing, multicast applications such as shared whiteboard, network audio and video conferencing tools will be tested.
Once we have a working ad-hoc network we can extend its functionalities to a MAN environment. The different LAN segments in the VPN will be tested for their interoperability over public networks including telephone modem-based, Internet, X.25(emulated over Ethernet).

Project Team: Amiya Bhattacharya, Nilanjan Banerjee, Wook Choi and Sundar Ramachandran

Over the last several years, the worldwide cellular communications market has undergone explosive growth. This can be attributed to several factors, including decreasing prices, improved radio coverage, and compact, lightweight terminals. As the number of users increases given a fixed ratio spectrum allocation, the size of radio coverage cells must decrease, in order to accommodate the higher subscriber densities. Personal communications systems (PCS) is a new generation of mobile communication network, which is expected to support services like seamless coverage and service, personal mobility, voice/data capabilities, low cost by a single identity number and a pocketable communication terminal. The number of users of PCS network will be much higher than today's cellular users. Cells will be smaller and smaller in size to support this increasing number of users and traffic- actually to reuse the limited available radio spectrum.

Location Management is an essential component in wireless cellular networks for supporting mobility. As cells are getting smaller in size day by day, to support increasing number of users, this topic has already got the attention of the researchers. Lots of location-tracking schemes have been proposed for location update and terminal paging. We believe that a good location management approach should use the mobility history of individual users to dynamically create individualized location areas. In my thesis I have studied the existing update and paging schemes and implemented the LeZi-update scheme. LeZi-update is a path-based location-tracking and paging scheme that dynamically builds and maintains a dictionary of individual user's path updates. The update cost is reduced by
variable-to-fixed length encoding of LZ algorithm, whereas the paging cost is reduced by its built-in prediction power.

Project Team: Abhishek Roy and Amiya Bhattacharya

Over the past few years, much discussions and debates haves fathomed among the
researchers and technologists, over the specifics of the next generation wireless communication technology as well as its role in the evolution of the information technology of the future. A careful look at these discussions reveals that there are essentially two common demands posed on the wireless service that are universally agreed upon. First, the wireless service of the next generation must evolve into an information service that is convergent of multiple media such as voice, video and loss-sensitive data. Second, the
subscribers need to be provided with ubiquitous seamless connectivity, so far as this service is concerned. To elaborate, today's voice-centric wireless service must retain and likely enhance the current quality of voice; and in addition, should at least provide standard Internet services (web, email, file transfer and remote login) and real-time/streaming video capability.

The fact that the so called ``anytime anywhere'' accessibility should apply to all the services offered by the provider, poses a new kind of demand on the location management system. The objective of the location management problem in the current voice-centric wireless network has been the optimization of the combined registration and paging costs, for a successfully call delivery. Knowledge of the subscriber's current cell or location area (a designated cluster of cells) is sufficient for that purpose. On the other hand, it would
be quite useful to have good capability of predicting individual subscriber's movement, on the part of a wireless infrastructure designed to support the mobile information technology of the future. One reason behind this conjecture is the need for bandwidth reservation to support end-to-end quality of service (QoS) guarantee for multimedia traffic. Irrespective of whether the reservation is hard or soft state based, it would be extremely wasteful to follow a conservative reservation scheme with highly mobile terminals. Considering the
scarcity of wireless bandwidth, this even may not be affordable at all. A highly predictive reservation scheme that learns subscriber's movement profile seems to be a more reasonable alternative. Another need for building movement profile may arise from some sort of locator services that can be augmented with wireless data service. The most convenient service access location for a mobile user is not necessarily the closest one, -- it could be the one on the way to current destination. The location management database can intelligently guess the possible direction and immediate destination for the ongoing movement.

We have earlier proposed a novel location update scheme for today's cellular architecture which possesses the capability of learning the movement profiles of individual subscribers. Naturally, it is also capable of using the learned profile to make good prediction about the subscriber's next move, -- at least to the extent possible from an information-theoretic perspective. Moreover, it is quite a simple matter to build group movement profiles by easy aggregation of such individual's movement database. In fact, this technique uses a well
known principle of computational learning that treats the learning problem as a data compression problem. Designed around the acclaimed Lempel-Ziv LZ78 universal compressor-decompressor duo, this scheme has been named as the LeZi-update algorithm. Simply said, it uses the learning capability of a data compression algorithm because a good compressor must always be a good learner and predictor, to be able to compress the data well. However, to achieve the power of a universal predictor, the LeZi-update scheme must use path-based update messages as opposed to the traditional position-based (cell or location area) messages.

Being the derivative of a universal predictor, LeZi-update possesses a highly desirable characteristic -- it imposes very weak assumption as far as the mobility model of individuals. Given that the movement profile of a user is stationary, the scheme is asymptotically optimal no matter what. The real-life interpretation to this is that the scheme would be able to track the movement profile of a subscriber, as long as he/she has a definitive stationary pattern of movement. This is true for most of us because our lifestyle and habitual behavior dictate these patterns. In reality, however, movement profile may
still change from one stationary pattern to another due to major change in lifestyle, such as a change of job or residential neighborhood. In this paper, we attempt to reflect this reality by making the mobility model even weaker -- from stationary to piecewise stationary. By analysis and simulation, we would establish that the LeZi-update technique performs very well under this almost model-independent scenario. As a result this scheme shows potential to be considered as a prime candidate for mobility support in next-generation wireless networks.

Project Team: Amiya Bhattacharya and Abhishek Roy

In a Client-Server system, data transmission from server to client can be performed in two ways: Pull-Based and Push-Based. In Pull-Based transmission, client sends requests to server to ask for data. In Push-Based transmission, server pushes data to client.
One kind of Client-Server systems is called Asymmetric Communication System, where the downstream (server to client) communication capacity is much greater than the upstream (client to server) capacity. In such system, it is believe that the broadcast is an efficient way to transmit data from server to client. Wireless network is such a system.
How to schedule data to broadcast is of importance in the overall system performance. Right now, most researches are focusing on either Pull-Based system or Push-Based system. However, fully pushed-based broadcast may not be useful for some clients, and fully pull-based broadcast may not be efficient. We are considering to combine scheduling for these two systems to make it to be of practicability.
There are some ideas for consideration:

1. Combining Pull-Based scheduling with Push-Based scheduling. In purely Push-Based system, the data items to be transmitted are fixed. We just need to set the order to broadcast. If the amount of data items is huge, it may not be good to transmit each of them. In such condition, Pull-Based model is appreciated to set which item need to be transmitted.

2. Arranging channels for Pull-Based broadcast and Push-Based broadcast separately. The Pull-Based broadcast is for most popular and regular data items. The Push-Based broadcast is for less popular and irregular items. 3. Need Long-term scheduling for Push-Based broadcast channel and Short-term scheduling for Pull-Based broadcast channel.

4. Combining several scheduling criteria, such as popularity, priority, waiting time, stretch, service time. Scheduling based on just one criteria, which is the way most proposals do right now, may not be good enough.

Project Team: Sourav Pal

Our main objective in this proposal is to provide effective wireless data services in 3G CDMA networks and accordingly develop robust protocols which can adapt to the changing traffic pattern, be it voice or multimedia traffic. We propose to investigate some fundamental aspects of wireless data networking which will not only be confined to the static optimization paradigm, where all the input parameters remain unchanged during the course of services. Due to the host mobility, variation in speed, mobility pattern and direction of movement, unpredictable fluctuations in the bit-rate and bit-error rate of wireless links, channel conditions and fluctuation of cumulative load on wireless bandwidth, there is a need for on-line decision making, thus leading to a dynamic optimization paradigm. Extensive performance modeling, analysis and simulation experiments will also be
conducted to test the efficiency and robustness of our proposed protocols.

Although, we will concentrate mainly on the CDMA networks, we hope to apply our methodologies for performance evaluation for other data networks also. Similar models can be used to evaluate the airlink performance of a GPRS/EDGE network, by computing the throughput of each user. In the process, we will also calculate the delay each user has to put up with to transmit his data. Using multi-slot reservation, the packet delay will be reduced and the bandwidth assigned to a user can be varied dynamically. To maximize throughput, more dynamic optimization at the GPRS MAC/RLC level will be required.
Finally, we will evaluate the impact of this layer-2 GPRS performance when mapped
onto CDMA networks (cdma2000 and W-CDMA).

Project Team: Mainak Chatterjee

An Approach Based on Random Graph Theory

We believe a multi-cluster, multi-hop packet radio network architecture for wireless system should be able to dynamically adapt itself with the changing network configurations.
Due to the dynamic nature of the mobile nodes, the association and dissociation to and from clusters perturb the stability of the system and th reconfiguration of the system is unavoidable. Choosing clusterheads optimally, which act as the mobile base station, is an important issue since the clusterheads decide the topology of the network and are responsible for resource allocation. Frequent clusterhead changes adversely affect the performance of other protocols such as scheduling and resource allocation that rely on it.
Most of the existing algorithms for choosing clusterheads, are greedy heuristics which do not include all the relevant optimization criteria into account.

We propose a graph theoretic approach and show how this problem can be tackled with the concepts of random graphs. Although, most related graph algorithms are NP-hard, it can be shown that with certain realistic assumptions the run time can be made polynomial.
We propose to find an analytical model and validate that model by performing exhaustive simulations. The parameters of our model will be node-mobility, node-degree, and transmission range.

Routing is another critical component in any multi-hop wireless network. Traditional routing protocols and single hop protocols are not suitable for multi-hop mobile wireless networks since there is no fixed home agents to maintain routing information. Due to the mobility of the hosts and the limitations of the wireless channels, the problem of routing is complex.
Inefficient routing protocols will degrade the throughput of channel access and increase the overhead as the number of hosts increase.

In addition to the conventional, cellular based applications, a number of new applications have evolved recently, which require wireless multi-hopping between remote users, without relying on the fixed wired network. These applications have stricter requirements than ordinary voice applications and hence, need to specify their expectations from the system as a whole. The applications show no degradation as long as they are in the wired domain.
The issue of interfacing the wired network to the wireless one demands more stringent guarantees when the wireless section is multi-hop. One must guarantee the quality of service (QoS) not only over a single hop but over an entire wireless multi-hop path.
The key component to such provisioning is QoS routing, which requires QoS information at each source to be propagated to the intermediate nodes. To address the issue of QoS routing, we propose to solve the following problems.

Project Team: Damla Turgut and Mainak Chatterjee

The main purpose of this project is to study the effect of video quality when it is transmitted over the wireless channel and design efficient coding schemes which will be best suited for that environment. Due to the dynamic nature of the channel and the uncertainty associated with the user's mobility, No constant bit-rate encoding will be able to survive the dynamism offered by the hostile environment. Hence, we propose an adaptive scheme which will be more robust to the varying channel conditions and will take into consideration the channel state and will adaptively change the encoding parameters without disrupting the on-going session.

We propose to use the most recent information available about the channel to encode the incoming video stream. The decoder at the receiver side, apart from decoding, will also monitor the signal strength and the bit error rate as experienced by the packets. These parameters will be captured and passed on to the encoder using a feed-back control mechanism.


The parameters for the feedback are the subject of study. The feedback from the network (channel) will help us make wise decisions and help us predict the channel in near future. The translation of the feedback parameters to the encoding parameters will be done
with the help of a transfer function. The design of the transfer function and its response time is an important issue. If the channel recovers before the response time, then it does not help. It will be useful if the duration of bad phase is large enough compared to the response time of the feedback system.

The idea is to employ dynamically changing network policies on the stream. We can imulate different traffic load patterns by artificially generating streams that are simple noise generators. The idea is to find out how the QoS changes in response to this noise. Obviously to demonstrate this fault-tolerance behavior, the QoS and path need to be pre-negotiated.The experiment will tell us the response time for the value adaptation.

Scalable coding schemes help in playing back a video sequence at different resolutions. We propose to use Discrete Wavelet transform (DWT) for the multi-resolution signal decomposition. The decomposition of the frames into frequency subbands will depend on the feedback control system. The encoder will decide the number of subbands to be encoded and transmitted. To our knowledge, there has been no real-time adaptive implementation of wavelet-based video compression algorithm on a DSP board. This proposal aims at opening up a new avenue for future work on DSP-based implementation
of wireless video-communication products in the future.

Project Team: Mainak Chatterjee and Sourav Pal

With the advent of widespread wireless networks, more and more applications are migrating from the wire line domain to the wireless domain. Previously deployed high bandwidth wire line applications are no longer suited because of the inherent limitation on the bandwidth and unreliable channel conditions. Users demand new protocols and new systems that can react to the varying network conditions including limited bandwidth and possible network disconnections. Few of the general application areas for mobile computing are described below. As an example of an application well suited for a mobile computing, consider a calendar-scheduling program. A common office activity is reserving conference rooms for presentations and meetings, known as the group calendar application. We would like managers to be able to reserve rooms even while away on a business trip. Traditional calendar programs assume that everyone updating the (shared) calendar has reliable network connections and thus the software cannot handle disconnected mobile hosts. Hence, mobile computing environments can offer improvements for this application area. Another mobile application is email using the POP protocol that allows a server to
hold a user's mail and allows the user to retrieve the email from any system. With only slight additional coding, a POP client can support mobile computing by pre-fetching all new messages during a (possibly) brief network connection and queuing the user's replies until the user reconnects. Emergency response is also well suited for mobile computing systems. Often emergency response personnel need specific information, such as geographic information in a flood region or a copy of a emergency call report for a medical or police emergency. Fixed networks are unavailable for an emergency response team, and often the only wireless communications available are satellite communications with very high latency.
Ideally, a mobile system supporting emergency response teams would allow two-way communication of both voice and graphical data. Thus, this application requires higher bandwidth than POP email or calendar management.

Much research literature has focused on systems designed to support distributed computing. However, mobile computing systems have some unique requirements and change their focus areas. We identify some crucial features of mobile-computing systems
and emphasize how these differ from traditional distributed systems.

One of the major tradeoffs in mobile computing systems is the tradeoff between offering strong consistency guarantees and robustly handling disconnected workgroups or unreliable network communication. Strong consistency guarantees required timely reliable transmission between client and server. A system, which allows for disconnected workgroups is limited in the consistency guarantees which it can make. Another important issue facing mobile computing systems is how much transparency to support. Existing mobile computing systems all require changes to the server software. By changing the software on client side, the mobile computing system can offer more flexible network architecture. Systems that use existing client/server software (high flexibility) are less able to support dynamically reconfiguring network topologies. We expect that as mobile computing systems become more flexible, they will support operation that is both transparent and flexible. As computing becomes ubiquitous the place of mobile computer systems requires
many features and characteristics not common or important on stationary systems. We
envision a future where many types of very different systems will be mobile, many with
temporary connections to some kind of networks for short periods of time. Such systems:
smart identification badges, medical records and status, personal planners and assistants,
and others such systems have many common problems (transient network connections,
data management, merging and updating shared information) and each has specific needs.
We believe an environment can be created to facilitate understanding, specifying, designing and programming such systems by using a graphical based visualization toolkit which supports the use of object components. In contrast to a programming language tool
which helps in creating an application program or a CASE tool which aids in design we
focus on the particular needs of mobile systems and sharing commonality through the
reuse of mobile objects. An important part of building mobile systems is a visualization
environment that aids in abstracting mobile issues and identifying features and functionality to create an object-component based system.

Project Team: Damla Turgut