Interference suppression for OFDM-based UWB communication

• US 20050265218A1
• Filed: 05/28/2004
• Published: 12/01/2005
• Est. Priority Date: 05/28/2004
• Status: Active Grant

• Current Assignee: Mitsubishi Electric Research Laboratories, Inc. (Mitsubishi Electric Corporation)
• Original Assignee: Mitsubishi Electric Research Laboratories, Inc. (Mitsubishi Electric Corporation)
• Inventors: Zhang, Jinyun, Molisch, Andreas F., Li, Ye
• Granted Patent: US 7,355,959 B2
• Time in Patent Office: Days
• US Class Current: 370/203
• CPC Class Codes: H04B 1/719 Interference-related aspects
• CPC Class Codes: H04L 25/022 of frequency response
• CPC Class Codes: H04L 25/0224 using sounding signals
• CPC Class Codes: H04L 27/261 Details of reference signals

Abstract

A method suppresses interference in a time-frequency hopped, ultra wide bandwidth system. Signals corresponding to a transmitted symbol are received. The received signals are frequency demodulated, and a coefficient vector w that minimizes a cost function $C\left(w_n\right)=\frac{1}{\sum _k{\lambda }^{n-k}}\sum _k{\lambda }^{n-k}{w_n^H{x}_k-c_k}^2,$
is estimated, where λ is a weighting factor between 0, x is a vector representing the received signals, H is a frequency response of a channel used to transmit the symbol, and c is a known training symbol. A coefficient vector, ŵ_n={circumflex over (R)}_n⁻¹{circumflex over (d)}_n, is determined, where a correlation matrix for the received signals is ${\hat{R}}_n=\frac{1}{\sum _k{\lambda }^{n-k}}\sum _k{\lambda }^{n-k}{x}_k{x}_k^H,$
and an estimate of a crosscorrelation is ${\hat{d}}_n=\frac{1}{\sum _k{\lambda }^{n-k}}\sum _k{\lambda }^{n-k}{x}_k{c}_k^{*}.$
The received signals are then combined according to ${\hat{c}}_n=\frac{1}{\sqrt{C\left({\hat{w}}_n\right)}}{\hat{w}}_n^H{x}_n=\frac{1}{\sqrt{1-{\hat{w}}_n^H{\hat{d}}_n}}{\hat{w}}_n^H{x}_n.$

Claims

1. A method for suppressing interference in a time-frequency hopped, ultra wide bandwidth system, comprising:

receiving signals corresponding to a transmitted symbol;

frequency demodulating the received signals;

estimating a coefficient vector w that minimizes a cost function

$C<br><br>\left(w_n\right)=\frac{1}{\underset{k}{\sum <br><br>}<br><br>{\mathrm{\lambda <br><br>}}^{<br><br>n-k<br><br>}}<br><br>\underset{k}{\sum <br><br>}<br><br>{\mathrm{\lambda <br><br>}}^{<br><br>n-k<br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{<br><br>w_n^H<br><br>\text{\hspace{1em}<br><br>}<br><br>x_k-c_k<br><br>}^2,$

where λ

is a weighting factor between 0, x is a vector representing the received signals, H is a frequency response of a channel used to transmit the symbol, and c is a known training symbol;

determining a coefficient vector ŵ

_n={circumflex over (R)}_n<sup>−

1</sup>{circumflex over (d)}_n,

where a correlation matrix for the received signals is

${\hat{R}}_n=\frac{1}{\underset{k}{\sum <br><br>}<br><br>{\mathrm{\lambda <br><br>}}^{<br><br>n-k<br><br>}}<br><br>\underset{k}{\sum <br><br>}<br><br>{\mathrm{\lambda <br><br>}}^{<br><br>n-k<br><br>}<br><br>x_k<br><br>x_k^H,$

and an estimate of a crosscorrelation is

${\hat{d}}_n=\frac{1}{\underset{k}{\sum <br><br>}<br><br>{\mathrm{\lambda <br><br>}}^{<br><br>n-k<br><br>}}<br><br>\underset{k}{\sum <br><br>}<br><br>{\mathrm{\lambda <br><br>}}^{<br><br>n-k<br><br>}<br><br>x_k<br><br>c_k^{*};<br><br>$

and

combining the received signals according to

${\hat{c}}_n=\frac{1}{\sqrt{C<br><br>\left({\hat{w}}_n\right)}}<br><br>{\hat{w}}_n^H<br><br>x_n=\frac{1}{\sqrt{1-{\hat{w}}_n^H<br><br>{\hat{d}}_n}}<br><br>{\hat{w}}_n^H<br><br>x_n.$

2. The method of claim 1, in which the received signals include multiple copies of the transmitted symbol, the multiple copies being transmitted at different instances in time.

3. The method of claim 1, in which each of the multiple copies of the transmitted symbol is obtained by a different receive antenna.

4. The method of claim 1, further comprising:

transmitting pilot tones having a fixed frequency for a particular symbol, and different frequencies for different symbols.

5. The method of claim 4, in which the different frequencies for the different symbols vary periodically.

6. The method of claim 4, in which a frequency response of the channel is estimated from the pilot tones.

7. The method of claim 1, further comprising:

phase shift demodulating the combined signal; and

decoding the phase shift demodulated signal to recover the transmitted symbol with suppressed interference.