US 6,618,452 B1 · Burst carrier frequency synchronization and iterative frequency-domain frame synchronization for OFDM

Abstract

Carrier frequency and frames are synchronized in bursty data transmissions over unknown channels that cause inter-symbol interference. The synchronization procedure comprises two stages. The first stage performs a time-domain processing of samples to exploit a periodic signal repetition and to extract the coarse timing, the frequency offset and also to resolve frequency ambiguities. The second stage estimates the fine time offset of a received modulation signal. A coarse estimate of a frame start position of a received sequence of desired data samples may be improved by using the coarse timing estimate to generate frequency-domain received samples. A frequency-domain correlation is then determined between the frequency-domain received samples and noiseless samples. When using a fixed number of training samples, a “sandwich” preamble (“sandamble”) is utilized to achieve greater efficiency than a conventional repetition preamble.

Claims64

Dependent claims

A method of transmitting a sequence of data samples, comprising the steps of:

initially transmitting a first preamble comprising a sequence of preamble samples;
transmitting the sequence of data samples;
and
subsequently transmitting a second preamble comprising the sequence of preamble samples, wherein the first preamble comprises a first guard region and a symbol sequence, the first guard region comprises a first subset of the symbol sequence, the second preamble comprises a second guard region and the symbol sequence;
and the second guard region comprises a second subset of the symbol sequence and whereby the sequence of data samples is transmitted after the step of initially transmitting the first preamble, and before the step of subsequently transmitting the second preamble.

2 more elements

A method of receiving a sequence of desired data samples, comprising the steps of:

receiving a signal sequence that includes a first preamble followed by the sequence of desired data samples followed by a second preamble, wherein the first preamble comprises a symbol sequence, and the second preamble comprises the symbol sequence;
hypothesizing a plurality of locations of the first preamble in the received plurality of data samples;
for each of the hypothesized locations, determining a hypothesized first preamble and a corresponding hypothesized second preamble;
for each of the hypothesized locations, determining a correlation between the hypothesized first preamble and the corresponding hypothesized second preamble;
and
using the correlations to determine locations of a most likely hypothesized first preamble and of a most likely second preamble.

3 more elements

A method of receiving a sequence of desired data samples, comprising the steps of:

receiving a signal sequence that includes a cyclic prefix guard interval followed by the sequence of desired data samples, followed by a cyclic postfix guard interval;
determining a coarse estimate of a starting position of the sequence of desired samples in the received signal sequence;
and
generating an improved estimate of the starting position of the sequence of desired samples in the received signal sequence by:
using the coarse estimate of the starting position of the sequence of desired samples in the received signal sequence to generate frequency-domain received samples;
determining a frequency-domain correlation between the frequency-domain received samples and noiseless samples;
and
generating the improved estimate of the starting position of the sequence of desired samples in the received signal sequence based on an argument of the frequency-domain correlation between the frequency-domain received samples and the noiseless samples.

5 more elements

An apparatus for receiving a sequence of desired data samples, comprising:

means for receiving a signal sequence that includes a first preamble followed by the sequence of desired data samples followed by a second preamble, wherein the first preamble comprises a symbol sequence, and the second preamble comprises the symbol sequence;
means for hypothesizing a plurality of locations of the first preamble in the received plurality of data samples;
means for determining a hypothesized first preamble and a corresponding hypothesized second preamble for each of the hypothesized locations;
means for determining a correlation between the hypothesized first preamble and the corresponding hypothesized second preamble for each of the hypothesized locations;
and
means for using the correlations to determine locations of a most likely hypothesized first preamble and of a most likely second preamble.