Rene L. Cruz - US grants

The principal investigators (PI) will develop and study new non- probabilistic models to analyze the behavior of control algorithms for resource allocation in packet switched data communications networks. Their approach to dealing with uncertainty differs from traditional approaches. Rather than specifying statistical properties of traffic they plan to employ and extend a non- probabilistic model for demand that specifies hard constraints satisfied by the traffic. The model facilitates the rigorous computation of bounds for parameters such as maximum delay, average delay, and buffering requirements in networks. They will focus on dynamic routing and flow control which involve the study of coupled queues. The use of models such as the nonprobabilistic model is likely to overcome the analytical problems traditionally encountered. They provide examples of problems that address these issues in networks using point to point and interacting broadcast links. As the size and demand for data networks grow, the economic impact of developing useful tools for engineering them will be quite significant. This approach develops some of these tools. ***//

The research area that this principal investigator (PI) will focus on during his Presidential Young Investigator award is the control and management of information flow in communication networks. The focus will be on overcoming the bottlenecks of switching and network processing that result from using optical transmission and electronic devices. Processing bottlenecks may arise in network protocols. The PI plans to develop analysis techniques for evaluating performance of higher level protocols using, for example, a non-statistical approach to predict worst case performance. To address the switching bottleneck, the PI plans to synthesize and investigate the performance of switching architectures which utilize massively parallel processing and the benefits of spatial statistical multiplexing and deflection routing.

9318416 Cruz The principal investigators will use their expertise in the areas of communications networks, optical devices, fiber optics, and systems engineering in a cohesive research effort aimed at demonstrating the feasibility of time-based optical multiplexors. Time-based multiplexing is a fundamental operation that must be performed in large-scale wide- area networks. The research will examine architecture alternatives and demonstrate feasibility by building and testing scalable prototype multiplexors. A major difficulty in the design of multiplexors is dealing with contention. The conceptual prototype of a time-based optical multiplexor, in this study, is based on the detection of optical packets at the packet rate, as opposed to the detection of the optical data at the much higher data rate. For example, the optical packets can be detected by splitting off a small part of the optical signal using an optical-fiber directional coupler, with these signals detected by a high-sensitivity detector. For a two-to-one multiplexor, the packet detection will be used to configure a 2 x 2 guided-wave directional coupler switch. If only one input has a packet, the packet is routed to the output. If both inputs have packets, then one packet is routed to the output while the other packet is sent through an optical fiber delay line, and subsequently fed back into another switch. Several packet detectors and switches can be used to implement various scheduling algorithms, and to offer a higher degree of system reliability. The multiplexors will be developed using commercially available fiber optic and optoelectronic components. We will iteraltively examine architecture alternatives and refine our prototypes. This will maximize the interaction between the device and network researchers. ***

9415684 Cruz The goal of this project is to develop new methods to efficiently predict and control the quality of service seen by heterogeneous users in an integrated services network. This is critical to the development of efficient and economically viable large scale high speed networks. Statistical sharing can be used to gain efficiency, but it introduces uncertainties in the quality of service seen by users of a network. Since the uncertainty that arises with statistical sharing is primarily due to queueing phenomena, this will be our main focus. Thus, one may view our project as a study on the behavior and control of queueing networks. Viewed in this larger context, our work will have an impact in other vitally important research areas, such as flexible manufacturing networks and transportation networks. We will, however, be primarily motivated by issues which arise in a communication context. We will develop techniques for examining end-to-end performance issues, and examine mechanisms for the allocation and scheduling of resources. One emphasis will be on uncovering mathematical structure of sample paths. This will have an impact on the utility of performance analyses involving both stochastic and non-stochastic models. The issues being examined are critical to the design of economical large scale high speed networks. ***

Traditionally, in order to address the many complex issues that arise in communication networks, a ``layered" approach has been taken. The different layers are typically assigned different ``responsibilities," so that the general network design problem can be decomposed into simpler, more manageable problems. However, the independent design of the different layers - which ignores the detailed nature of their mutual interactions, shared constraints, and cumulative impact on the network's overall performance - can lead to inefficiencies. Future wireless communication networks will be required to provide wide coverage and high capacity to mobile users generating bursty multi-media information. This heterogeneous traffic will impose upon the network time-varying quality of service (QoS) constraints. Since many multi-media applications have delay-sensitive information with varying reliability requirements, such as numerical data, voice, and video, the project will take end-to-end delay and data integrity as its dominant system constraints. The goal of this project is, in the broadest sense, to take a more global view of end-to-end performance, to better understand the interactions among the layers, to develop techniques that improve system performance through joint optimization over the various layers and to do so in the context of a end-to-end delay constraint. One key theme of the proposal involves the optimization of system performance in the context of multiple users, particularly in a system which does not employ a cellular-type architecture. This requires the development of techniques that, on the one hand, address the deleterious effects of multiple-user interference at the physical layer, yet also incorporate end-to-end QoS objectives, especially delay, just as with the development of routing and scheduling algorithms. Another major theme will be the determination of the optimal distribution and the real-time dynamics of error control functions implemented across t he layers of the network, subject to specified end-user requirements. This requires investigation of inherent tradeoffs between error-rate and decoding delay of FEC employed at the physical layer, as well as the interactions among error control functions, including retransmission strategies, invoked at higher levels.

In this project, three researchers, and a team of graduate students, from
the University of California, San Diego (UCSD), shall undertake
theoretical and experimental investigations of data modulation schemes for
efficient information transmission in conjunction with CDMA encoded
ultrashort pulses in an optical fiber. Efficient modulation formats will
result in aggregate transmission rates exceeding one terabit/second, with
individual user rates on the order of 100-1000 megabit/second. The specific
objectives of this project include modeling of the CDMA, statistical
analysis of the transmitted waveforms, investigation of various CDMA codes
that support thousands of users with minimal interference, bit error rate
analysis of received signals for various modulation schemes, means to
provide QOS levels, modeling and characterization of the distortions
induced by the fiber channel, adaptive equalization techniques for reducing
fiber distortions, computer simulations of the modulation schemes, and
experimental evaluation of the modulation schemes, transmitter, optical
channel, and receiver. The goal of this proposal is to demonstrate a
prototype network with several users employing a modulation format which
when scaled up to the full number of users will carry over one terabit per
second of information.

The potential impact of the work will be in the proof that CDMA encoding of
ultrashort pulses is a realizable and desirable alternative to wavelength
division multiplexing (WDM).

The need for more bandwidth and capacity in wireless systems currently is the main culprit for the
great interest in the development of wireless communications systems operating at millimeter wave
frequencies and higher. The future needs of broad-band interactive services (1Gb/s) demand the
application of optical fiber feed networks for distribution of the radio signals to and from the antennas at
the various base stations. Fiber-optic technologies have reached the stage where insertions into various
commercial RF systems are being considered. Today, there are three main steps in the evolution of
RF/Photonics systems for wireless communications. The first step has been in the direction of using
photonics to slowly replace conventional RF components, such as, the coax that is used to interconnect
the antenna to the electronics. Optical fibers, in contrast to coaxial cable, provide a more ideal medium for
broadband RF communication systems. The light weight property of fibers, and its immunity from other
signal interference make them very critical in the development of future RF distribution systems. The
second, and more challenging step, is in the seamless integration of photonics and RF wireless circuits.
The challenge in this step is to use photonics and RF circuits as complementary systems and blend them
together. Finally, the third step is towards the development of optically coupled antennas. In this step the
aim is to eliminate the need of local oscillators, mixers, amplifiers and a host of other parts by directly
feeding an antenna through a fiber at millimeter wave frequencies.
Here, it is proposed that an array of RF modulator/photodetectors be integrated directly to an array
of antennas. This new RF/photonic antenna array system, with the appropriate space-time processing and
coding, will form a iosmart antennaln that can enhance network coverage, capacity, and quality. It is
envisioned that a large number of such RF/Photonic antenna elements could be networked together into a
star configuration, feeding in and out of a radio hub.
As a transmitter, the proposed optoelectronic device operates as a photodiode, while as a receiver
the device operates as an optical modulator. It has already been demonstrated that this dual function of a
semiconductor electroabsorption modulator and photodiode in the same device for duplex operation, can
occur, using bias control as a transmit/receive mode control. For full duplex operation, two
modulator/photodiode devices need to be incorporated in the each transceiver element.
We propose to directly drive a coplanar waveguide (CPW)-fed slot antenna by converting optical
power into microwave power and vice versa using these RF modulator/photodetectors. As a transmitter,
the CPW line is connected to the active surface of the photodetector, from which the microwave power
propagates to feed the radiating slot. The photodetector is fed via an optical fiber from beneath. When the
device functions as an optical modulator, the receive function can also be achieved. Preliminary results
for a single antenna show that a very good bandwidth and radiation patterns can be achieved.
It should be noted that these elements can be interconnected via the fiber to achieve summation,
mixing and other signal processing functions, at the antenna site or at a remote site. Some preliminary
results have been achieved in the area of multiple functionality for the optoelectronic components, such as
modulation, photodetection, self-biasing and RF frequency mixing. They have shown properties, such as
high bandwidth and high power, that are desirable for the antenna applications. A main emphasis here is
to further investigate the material and device designs for the optoelectronic component that can
incorporate into the smart antenna architecture.
The proposed approach will have significant impacts on wireless communication systems by
providing higher system bandwidth capacity and enhancing their reliability. It may lead to a new type of
long distance, broadband network infrastructure that supports transparent transport of optical signals.
Our team is formed to provide the expertise in the four key elements for this proposed research.
Our project will provide a good opportunity to train graduate and undergraduate students in one of the
most exciting interdisciplinary areas in science (RF, photonics, signal processing and communications).
The interactions between the researchers at the different institutions will be aided by the close
collaboration that exists between the members of the group.

In this proposal, three researchers from the University of California, San Diego (UCSD), specializing in the fields of optics, communications, and computer networks, are collaborating on the Ultra-High-Capacity Optical Communications and Networking initiative. It is felt by many researchers that the most efficient and economical way to utilize optical transmission technology for large scale networking is to use wavelength division multiplexing (WDM) in a circuit switched mode, overlaid with packet switching implemented with electronics. While this may indeed be the case, it is important to investigate alternative approaches that have great potential. The UCSD team has been investigating novel techniques of information transmission via optical fiber, where code division multiple access (CDMA) using ultrashort laser pulses is employed. Compact, low cost fiber-based ultrashort pulse sources are currently being developed, making the technology suitable for future practical networks. When an ultrashort pulse is encoded for CDMA, the pulse spreads out in time and resembles a noise burst that is transmitted on the optical fiber. At the receiving node, a decoder is applied to the received signals from multiple users, which matches only the encoding of the desired transmitter. The matching signal component is transformed back to an ultrashort pulse form that can be detected over the remaining interference from other users with nonlinear optical techniques. A novel high resolution pulse synthesis and detection technique for ultrashort pulses developed at UCSD enable various data transmission formats to be considered, such as ultrafast packet transmission with on/off keying, pulse position modulation, and amplitude modulation. The CDMA scheme enables large scale, asynchronous, concurrent access to the transmission resources. With a suitable architecture, this can be exploited to simplify network control, and increase reliability and flexibility.

The objective of this proposal is to conduct basic research by investigating theoretically and verifying
experimentally data modulation schemes for efficient information transmission in conjunction with CDMA encoded ultrashort pulses in an optical fiber network. Efficient modulation formats will result in aggregate transmission rates exceeding 10's of terabits/second, with individual user rates on the order of 1-10 gigabits/second. The specific objectives of this proposal include modeling of the optical CDMA for ultrashort Gaussian pulses, complete statistical analysis of the transmitted waveforms, investigation of various optical CDMA codes that support thousands of users with minimal interference, bit error rate analysis of received optical signals for various modulation schemes, modeling and characterization of the distortions induced by the fiber channel, adaptive equalization techniques for reducing dispersion and other fiber distortions, computer simulations of the modulation schemes, and experimental evaluation of the communication system: transmitter, optical channel, and receiver. The various phases of the proposed project complement each other. Combined together, they provide for in-depth knowledge of the theoretical and experimental issues of communicating with CDMA encoded ultrashort pulses. These findings will be shared with the scientific community, enhancing not only the knowledge base of other researchers in the field, but also of the students conducting the research. We shall demonstrate a prototype optical network with several users employing the modulation format that will carry over 10 terabits per second of information, when scaled up to the full number of users.

The potential impact of the work will be in the proof that optical CDMA encoding of ultrashort pulses is a
realizable and desirable alternative to WDM. Currently, WDM is the preferred multiplexing method due to its
simplicity and low cost. While WDM does increase the transmitted bandwidth significantly, it still does not fully utilize the available optical bandwidth due to both the need for guard bands between channels and the under utilization of channels. In contrast, CDMA encoded ultrashort pulses share the entire bandwidth without the need for guard bands, leading to efficient utilization of transmission resources. Using CDMA can also provide a highly flexible and robust infrastructure, upon which packet switching can be overlaid. The CDMA format also provides a degree of security, as no data can be extracted without knowledge of the codes employed.

The proposed project will investigate a broad class of packet-based network architectures, protocols, and services where the core switch/route fabric has limited or no buffering capability. These networks - called mbuffered networks - have two distinguishing features. First, packet loss due to contention is treated as an erasure that can be corrected via coding techniques. Information is encoded into code words and each codeword is divided into fragments. The redundancy built into the codeword acts like a "virtual buffer" that mitigates contention and packet loss so that if several of the codeword fragments are erased as they pass through the network, there is still a high probability that the information within the codeword can be decoded correctly at the destination. Adaptive flow control can be implemented by adjusting the coding overhead (code rate) as well as the fragment generation rate. The second distinguishing feature is the robustness with respect to hardware and routing failures. In particular, different codeword fragments belonging to the same codeword can be sent using different routes within the network to increase resilience. Route diversity also provides unique security and authentication features.

The intellectual merit of the proposed project is the exploration of new architectural approaches that use little or no buffering in high-speed networks where buffers are becoming increasingly difficult to implement. Results of this project will impact research directions in optical systems technology, and increase the base of knowledge in communication systems theory. The project will provide unique training of both undergraduate and graduate students in a systems-oriented multi-disciplinary effort.

In this project, various radio resource allocation problems that arise in multi-hop wireless communication networks are proposed, where the main focus is on broadband wireless infrastructure networks. The aim is to help facilitate the realization of such networks by developing system level approaches to reducing cost and maximizing performance. The resource allocation problems considered include transmission scheduling, routing, power control, and topology configuration, including node placement, and node density planning. In this project, rigorous mathematical analysis techniques will be developed that have not previously been applied to the design of wireless infrastructure networks. These techniques will be combined with practical engineering approaches to yield novel network architectures and improved insight into the design space for multi-hop wireless networks. The results of the project could help to enable low cost, ubiquitous wireless networks that support very high data rates. Development of resource allocation technology underlying broadband wireless infrastructure networks, as presented here, may enable other related applications, including high bandwidth sensor networks and ad-hoc networks. For example, cost effective high bandwidth wireless infrastructure network technology may lead to "grass root" development of highly robust communications infrastructures whose control and ownership is distributed among end-users. The improvement of the efficiency of wireless information transport may enable large-scale video surveillance networks.