Methods of estimating signal-to-noise ratios

Pulse or digital communications – Testing – Signal noise

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C375S329000

Reexamination Certificate

active

06317456

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to methods for estimating a signal-to-noise ratio (SNR) of a signal. More specifically, the invention is directed to methods of estimating the SNR of a signal having a signal component and a noise component wherein the signal has been first modulated by a transmitter and then demodulated by a receiver so that the signal can be represented as a complex symbol having a real part and an imaginary part.
2. Description of the Related Art
Digital communications systems have emerged in the telecommunications field as important and efficient means for transmitting and receiving signals. These signals comprise noise components and signal components and must be filtered and transformed so that receivers in the system can adequately “hear” the signal components and output usable data to a user of the system. In digital communications systems, like orthogonal FDM (OFDM) systems, the noise components of the signals can overwhelm the desired components of the signal and so it has been desired to estimate or otherwise determine the SNR of the signal to thereby allow critical decisions like handoff and receiver shutdown to be made and to allow signal combining techniques to be undertaken.
OFDM systems, and many other communication systems, thus have an inherent need to estimate the SNR of the received signal. However, for conventional multiple and single carrier systems, methods have not been developed for estimating or calculating the SNR without a pilot signal being used. Most of these systems accordingly rely on determining the signal plus noise power as a figure of merit for the receivers in the systems. The SNR is typically used in diversity combining techniques, an optimal form of which is the “maximal ratio combining” (MRC) technique wherein different copies of the same information are combined after they are weighted by the ratio of the instantaneous signal voltage and the noise power of each copy.
SNR estimation techniques for calculating or otherwise estimating the SNR of a signal are describe in W. C. Jakes, Microwave Mobile Communications, IEEE Press, 1974, pp. 419-420. The Jakes reference teaches a technique wherein the SNR is calculated for a single carrier system by repeatedly turning the transmitter in the system on and off. During the off interval the receiver calculates the noise power and during the on interval calculates the signal plus noise power. These two power signals are then subtracted so that only the noise component remains. However, this technique takes valuable bandwidth away from the information transmission capability of the system. This technique and other personal communication system techniques are data aided techniques and are therefore complicated and computationally expensive. In cellular networks and PCS, MRC is oftentimes performed using a “Rake” receiver wherein the output of each of the “fingers” in the receiver is weighted by the output of the respective pilot fingers. Unfortunately, pilot aided coherent combining of the actual weights is not a function of the signal strength only, but of the signal strength plus the noise strength and, therefore, accurate SNR calculation or estimation cannot be accomplished with this technique. Thus these techniques, and others similar to it, do not adequately produce a calculation or estimation of the noise present in the signal.
There exists a long-felt need in the art, recognized by the inventor hereof, for methods of accurately estimating the SNR in a communication system. These methods should be simple to implement and effective across a wide range of communication systems such as asymmetric digital subscriber line (ADSL) systems, OFDM wireless and wireline systems, digital multitone (DMT) systems, digital radios, digital audio broadcasting systems, and other systems. It would further be desirable if such methods estimated the SNR of a system without taking up bandwidth in the system and provided accurate SNR results. Such needs have not heretofore been satisfied or achieved in the art.
SUMMARY OF THE INVENTION
The aforementioned long-felt needs are met, and problems solved, by methods provided in accordance with the present invention for estimating a SNR of a signal having a signal component and a noise component, wherein the signal has first been modulated and then demodulated by a receiver so that the signal can be represented as complex symbol having a real part and an imaginary part. The complex symbol is then preferably stored in a digital memory that is then accessed so that the complex symbol can be operated on such that the real and imaginary components of the complex symbol are rotated in a plane having a real and imaginary axis so to place the signal component substantially on the real axis and the noise component will lie substantially on the imaginary axis.
It is then desired to calculate a mean-squared value of the real part of the complex symbol and a variance value of the imaginary part of the complex symbol. By dividing these two values it is possible in accordance with the invention to obtain an estimation of the SNR which is a critical value for the system. The methods of the present invention efficiently and accurately estimate the SNR of a signal in a communication system. By rotating the complex symbol representation of the signal in a complex plane, it is possible to readily calculate the mean-squared value of the real part of the complex symbol and to calculate the variance from the imaginary part of the complex symbol which provides a simple and straightforward technique for estimating the SNR for many different types of modulated and demodulated signals. These results have not heretofore been attainable in the telecommunication and signal processing arts. These and other features of the present invention will become apparent from the following detailed description considered in conjunction with the accompanying drawings. It is to be understood, however, that the drawings are designed solely for purposes of illustration and not as a definition of the limits of the invention, for which reference should be made to the appended claims.


REFERENCES:
patent: 4542514 (1985-09-01), Watanabe
patent: 5768473 (1998-06-01), Eatwell et al.
patent: 5809065 (1998-09-01), Dapper et al.
patent: 6108610 (2000-08-01), Winn
patent: 6122015 (2000-09-01), Al-Dhahir et al.
patent: 6122610 (2000-09-01), Isabelle

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Methods of estimating signal-to-noise ratios does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Methods of estimating signal-to-noise ratios, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Methods of estimating signal-to-noise ratios will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2598970

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.