In this talk I'll present an overview of joint equalization and decoding schemes for space-time-coded transmissions over broadband wireless channels. Promising schemes that exploit multipath and spatial diversity for both space-time trellis and block codes investigated at AT&T Shannon Laboratory over the past two years will be described. Practical training-based channel estimation schemes will be presented. Furthermore, I'll describe a frequency-domain approach for combining joint equalization/decoding with multi-user detection to further enhance system capacity. A main theme in my talk is exploiting the rich structure of space-time codes to enhance the performance and reduce the complexity of receiver signal processing. The third generation TDMA system EDGE will be taken as a case study. .
In this talk we highlight the salient properties of ad-hoc wireless networks and focus on energy efficiency. We consider the case of session-based multicasting and describe some algorithms that are motivated by the need to save energy. In the process, we obtain key properties of energy efficient networking and project these into overall wireless network design and operation.
We consider a wireless communication link, with multiple antennas at both transmit and receive sites. The link, which operates in a rich scattering environment, is impaired by additive white Gaussian noise. The transmitter, which is subject to a power constraint, may not know the outcome of the random matrix channel. When the transmitter only knows the channel statistics, the link is operated at a limit on the probability of outage. Guided by information theory, we discuss the advantages and disadvantages of various means of expressing messages in spacetime. Space here has one complex dimension per transmit antenna. Spacetime is viewed as a long duration rectangle into which one can write a message in a variety of different ways. The goal is to find the way that enables the receiver to reduce the self-interference of the received waves to effect the best tradeoff between performance (Shannon capacity) and complexity.
A wireless link demonstrating enormous bit-rates is in operation at Bell Laboratories in Holmdel, New Jersey. It is part of the BLAST (Bell Labs Layered Spacetime) project for exploring the effectiveness of multiple antenna communication architectures.
A review will be presented on the paradigm of radio on chip vs. radio on module for next generation communication applications. The continued demand for ubiquitous wireless connections at higher data rates puts tremendous pressure on technical requirements for advanced radio implementation. In this talk we look at several integrated solutions for OFDM, OSCM and Bluetooth to help answer the question: when will single chip radios be available?
As the communication shifts its momentum from wired to wireless in the past decade, the demand of key RF components, such as power amplifiers, and filters increases tremendously. This drives the industry to develop new technologies and products to meet these commercial market demands.
Among these technologies, there are InGaP HBT, SiGe HBT, CMOS, GaN, SAW, etc. In this paper, we intend to give an overview of these new technology development and compare their technical advantages in different applications, which include power amplifier modules, RFIC chip set, and passive elements used in the cellular handsets, broadband amplifiers used in the infrastructure, front-end circuits used in the optical communications.
IIn a multiple access wireless communication system, one objective for a receiver is to equalize a designated channel and discriminate all other signals. The performance of such receivers, with or without using the training signals, always degrades when noise is increased. With the support of a prefilter system, the Noise Independent Autocorrelation (NIA) Matched Filter can achieve, in theory, consistent performance when SNR varies. A simulation shows that its performance remains constant when the SNR varies from -10db to +10db, and it is constantly better than the LMS method using training signals, especially in the low SNR area from 0db to -10db. In addition, it is near-far resistance, multiple access interference (MAI) resistance, and intersymbol interference (ISI) resistance.
In this talk I will present an overview of asynchronous VLSI and its application to the design of wireless communication systems. I will present the design of a new architecture for modeling networks using asynchronous VLSI, and our results that show that we will be able to simulate large scale wireless networks much faster than real-time thereby enabling a new class of adaptive protocols for wireless communication systems.
This talk will discuss scalability issues which arise in mature deployment of broadband wireless networks. As a growing percentage of users move to portable/mobile computing devices, wireless access networks will be required to carry ~Gbps/sq-Km in the not-too-distant future. We examine issues related to scaling of users, traffic and services in broadband wireless networks, in terms of the impact on: (1) wireless modem speed; (2) cellular capacity and spectral efficiency; (3) MAC/link layer throughput; (4) mobility protocols for handoff and location management; and (5) application-level information. Technical challenges for each are discussed along with potential solutions where applicable.
Energy efficient wireless communication system design is an important and challenging problem. It is important because mobile units operate on batteries with limited energy supply. It is challenging because there are many different issues that must be dealt with when trying to design a wireless communication system for low energy. These issues include amplifier design, coding and modulation design, as well as network protocol design. This design would not be too complicated if each design could be decoupled from other designs. Unfortunately, while this has been the approach to designing complex systems, it is not optimum. The challenge is to integrate the design of various aspects of a wireless communications problem into a single design that achieves low power. In this talk we discuss a methodology for designing wireless communication networks that jointly optimizes an network wide objective function with an energy constraint yet does so in a computationally tractable manner.
Recently, there is considerable interest in the idea of cross-layer design of wireless networks. This is motivated by the need of providing a greater level of adaptivity to variations of wireless channels. In this talk, we first consider roles of signal processing in the design of medium access control (MAC) protocols for wireless networks. We then examine the problem of optimal packetization for intersymbol interference fading channels.
Public Internet access is developing in wireless LAN (WLAN) environments in order to meet the ubiquitous access, high data rate, and local services needs of current and future users. This talk describes a business model of franchises for multiple services providers, and the technical infrastructure to support it in relatively large-scale IP-based WLAN environments such as airports, shopping centers, hotels, and campuses. By relying on IP-level services mechanisms and pre-configured MPLS paths, the WLAN can simultaneously support different air interfaces, franchises for multiple services providers, and a multi-segment LAN environment with local IP mobility capabilities. A router associated with each access point helps realize this highly distributed IP networking environment, and a QoS-enabled switched Ethernet core supports virtual networks and QoS services.
* Work done jointly with J. Zhang and N. Tu (NEC USA C&C Res. Labs) and J. Li (Thomson Multimedia USA)