RESEARCH OVERVIEW

• Capacity of Wireless Channels and Networks One of the fundamental issues in the design of a wireless communication s ystem is how fast data can be transmitted reliably. The theoretical upper bound for speed of data transmission over a channel is called the channel capacity. The capacity of time-varying channels depends on what is known about the channel at the trans mitter and/or receiver. It also depends on whether rate can be adaptive or constant, as well as whether or not service outage can be tolerated. Our recent results in this area include investigating capacity or capacity regions of fading broadcast channels, fading multiple access channels, broadcast channels with intersymbol interference and colored noise, and multiantenna channels with covariance feedback. One of our most recent results shows the duality between broadcast and multiple access channels, such that the capacity region of one channel can be used to find the capacity region of the other, dual channel. This result is quite powerful and we are currently applying it to multiantenna MAC and broadcast channels, and trying to extend it to nonGaussian and nondegraded broadcast channels. We are also investigating capacity limits of ad-hoc networks with both a finite and asymptotically large number of nodes.
  
  
• Adaptive modulation and coding: Wireless channels vary over time due to fading and changing interference conditions. Adaptive modulation and coding exploits these variations to maximize the data rate that can be transmitted over such channels. Such schemes can also be designed in a prioritized manner so that high-priority (or low resolution) bits get through in all channel conditions, and as the channel improves the lower-priority (high resolution) bits can be transmitted. We are currently looking at adaptive multimedia modulation techniques, adaptive turbo coded modulation, and adaptive rate, power, and BER techniques for cellular and ad-hoc networks. The goal of these adaptive schemes is to optimize performance in the face of randomly changing channel and interference conditions.
  
  
• Adaptive CDMA for Multimedia We have investigated adaptive CDMA techniques for voice and data users. The data users adaptively vary their rate using multiple spreading gain, multiple codes, or variable signal constellations. Results indicate that the variable spreading gain approach achieves the highest throughput. We are currently investigating extensions of this work to incorporate power control and multiple classes of users.
  
  
• Turbo Codes and Iterative Decoding Turbo codes have created much excitement in the research community due to their exceptional performance in AWGN channels. Our work in this area is investigating adaptive turbo coded modulation, turbo codes for multiantenna systems, and turbo coded CPM modulation.
  
  
• Multiantenna systems Multiple antennas are known to significantly increase the capacity of wireless systems. We are investigating multiple facets of using multiple antennas in a wireless system design. Specifically, we are researching the capacity limits of such systems when the channel is not perfectly known at either the transmitter or receiver. We are also investigating optimal space-time coding techniques for such systems.
  
  
• Multiple Access and Dynamic Resource Allocation Sharing the limited spectrum within and between different multiuser systems can be done through a variety of techniques: the most common are frequency division, time division (or multiplexing), spread spectrum coding, and hybrid combinations of these techniques. In addition, the spectrum can be more efficiently utilized by reusing frequencies at spatially separated locations - this is the idea behind cellular communication systems, but can also be used in ad-hoc networks. Optimal ways to share the spectrum and reduce interference through power control, dynamic rate adaptation, and dynamic channel allocation in both cellular and ad-hoc networks are also being looked at.
  
  
• Joint Source/Channel Coding Another ongoing area of research is joint source/channel coding. Shannon's fundamental theorem showing that source coding and channel coding can be separated without any loss of optimality does not apply to general time- varying channels, or to systems with a complexity or delay constraint (i.e. any real system). Since distortion by the source encoder decreases with data rate, while channel errors increase with data rate, the joint source/channel coding problem reduces to allocating bits in an optimal way between the source and channel encoders. This optimal allocation will depend on the source data (voice, video, images, etc.) as well as on the channel characteristics. Joint source/channel coding techniques for multimedia data and distributed source coding for multiantenna systems are our current focus in this area. We are also looking at developing a generalized source/channel separation theorem for time-varying channels.
  
  
• Wireless Networks The nature of wireless channels suggests that the infrastructure to support cellular, personal, and indoor wireless communication systems might differ significantly from the Internet and other hard-wired systems. Issues of access, resource allocation, routing, mobility management, network security, and network control may borrow from the standards of wireline networks, but must also take into account the unique character of terminal mobility and the randomly-varying channel.
  
  
• Energy-Constrained Networks Many types of wireless networks, like sensor networks, have nodes with a finite battery life. Once these batteries die away the node is gone, and can no longer perform its intended function (e.g. data collection) nor participate in forwarding information from other nodes. This provides an interesting challenge in the optimization of such energy-constrained networks, since energy-conservation is not typically included in network protocol design. We are investigating the capacity limits of energy-constrained networks as well as optimal routing protocols that maintain full connectivity within the network for as long as possible.
  
  
• Wireless Communications for Control Applications Many automated systems today require distributed control where the control commands and observations are transmitted via wireless links (e.g. automated highway systems). However, most distributed controllers can experience catastrophic failure due to delayed or lost packets, which are common on wireless links. In this research we are investigating optimal joint designs of distributed controllers and the wireless networks that connect the controllers. The goal is to insure stability under all operating conditions and maintain the tightest control parameters given the operating constraints of the communications network.