Time Reversal Signal Processing in Communications - A Feasibility Study
A typical communications channel is subjected to a variety of signal distortions, including multipath, that corrupt the information being transmitted and reduce the effective channel capacity. The mitigation of the multipath interference component is an ongoing concern for communication systems operating in complex environments such as might be experienced inside buildings, urban environments, and hilly or heavily wooded areas. Communications between mobile units and distributed sensors, so important to national security, are dependent upon flawless conveyance of information in complex environments. The reduction of this multipath corruption necessitates better channel equalization, i.e., the removal of channel distortion to extract the transmitted information. But, the current state of the art in channel equalization either requires a priori knowledge of the channel or the use of a known training sequence and adaptive filtering. If the ''assumed'' model within the equalization processor does not at least capture the dominant characteristics of the channel, then the received information may still be highly distorted and possibly useless. Also, the processing required for classical equalization is demanding in computational resources. To remedy this situation, many techniques have been investigated to replace classical equalization. Such a technique, the subject of this feasibility study, is Time Reversal Signal Processing (TRSP). Multipath is particularly insidious and a major factor in the deterioration of communication channels. Unlike most other characteristics that corrupt a communications channel, the detrimental effects of multipath cannot be overcome by merely increasing the transmitted power. Although the power in a signal diminishes as a function of the distance between the transmitter and receiver, multipath further degrades a signal by creating destructive interference that results in a loss of received power in a very localized area, a loss often referred to as fading. Furthermore, multipath can reduce the effectiveness of a channel by increasing inter-symbol interference. Here, a symbol is the fundamental unit of information. Although a signal may have a sufficient signal-to-noise ratio (SNR) at a receiving site, the signal may not be interpretable because it is composed of time-delayed replicas of the original transmission due to the multiple paths between transmitter and receiver. Although not previously employed for communications systems, developments in Time Reversal Signal Processing (TRSP) indicate the potential for compensating the transmission channel while mitigating the need for detailed a priori knowledge of the channel characteristics. Furthermore its simplicity, viz. a viz. equalization, makes it particularly attractive. The successful use of TRSP can increase channel bandwidth, thereby enabling the proportional increase in the volume of information. It implicitly compensates for distortion by using the equivalent of an imbedded phase conjugation technique for the equalization. This is an astounding property of the TRSP than can be taken advantage of for this problem. This feasibility study is directed toward showing that TRSP is a viable method for mitigating the effects of multipath on the transmission of information through a communications channel. In this report, we first briefly describe the theory behind the use of TRSP in communications. The channel is composed of a single transmitter-receiver pair operating in an unknown, possibly inhomogeneous, medium that includes scatterers that can contribute to multiple paths between the transmitter and receiver. These multiple paths (multipath) express themselves as reverberation in the received signal and attendant distortion in the information. Once having established the theory behind the use of TRSP to mitigate multipath distortion, we will describe simulations of the performance of suggested systems using this approach. Finally, we will establish conclusions based on our observations in the simulations.
- Research Organization:
- Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- US Department of Energy (US)
- DOE Contract Number:
- W-7405-ENG-48
- OSTI ID:
- 15006881
- Report Number(s):
- UCRL-ID-147014; TRN: US200412%%276
- Resource Relation:
- Other Information: PBD: 30 Jan 2002
- Country of Publication:
- United States
- Language:
- English
Similar Records
Wireless Fading Channel Models: From Classical to Stochastic Differential Equations
Speech coding