Pulse or digital communications – Receivers – Interference or noise reduction
Reexamination Certificate
2000-03-10
2004-03-09
Tran, Khai (Department: 2631)
Pulse or digital communications
Receivers
Interference or noise reduction
C375S362000, C375S375000, C375S355000
Reexamination Certificate
active
06704377
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to the field of wireless communications, more particularly to correcting frequency errors in coherently demodulated wireless communication systems.
2. Description Of The Related Art
Explosive growth in the market for internet and intranet related applications has provided the impetus for a greater demand for fixed wireless networking services and systems. A wireless internet access system (WIAS) illustrated in
FIG. 1
is composed of four major parts: (a) multiple data base stations (BS)
100
(
a
) and
100
(
b
) which provide wireless connectivity and gain coverage to subscriber units
102
(
a
)-(
d
) of a large geographical area (for example, residential and corporate terminal equipment as illustrated in FIG.
1
); (b) wireless modems
170
(
a
)-(
c
) (hereinafter “WM”) which are connected to BS
100
(
a
) or
100
(
b
) via wireless links
115
(
a
)-(
c
); (c) a data switching center (DSC)
125
with integrated management functions; and (d) a backbone transmission network
135
interconnecting (a)-(c) above.
As can be seen from
FIG. 1
, corporate terminals
102
(
c
) and
102
(
d
) can be, and many times are, connected to WM
170
(
c
) via a local area network (LAN) and a wireless router or firewall (not shown). Additionally, BS
100
(
a
) and
100
(
b
) may communicate with DSC
125
via frame relays (not shown). Further in conventional wireless internet access systems or networks, DSC
125
is interconnected with backbone transmission network
135
by a router and/or firewall (not shown for clarity).
FIG. 2
illustrates BS
100
(
a
) and
100
(
b
) of
FIG. 1
in an operational mode. Each BS
100
(
a
) and
100
(
b
) provides 360° RF coverage on the order of several gigahertz (preferably operating in the 3.5 GHz spectrum using approximately 5 MHz wide channels), sending and receiving signals over air lines
115
(
a
)-(
c
) between individual subscriber units
102
(
a
)-(
d
) served by BS
100
(
a
) and/or
102
(
b
). More particularly, the designated geographical area of subscribers served by each BS
100
(
a
) and
100
(
b
) is typically called a cell
150
, defined by its coverage area as shown in
FIG. 2
, where BS
100
(
a
) and
100
(
b
) are situated in designated cells
150
(
a
) and
150
(
b
). Within each cell
150
(
a
) or
150
(
b
) reside a plurality of subscribers
102
(
a
)-(
d
) served by the BS
100
(
a
) and/or
100
(
b
) includes a plurality of access points (hereinafter “AP”, not shown in
FIG. 1
) serving as an interface between individual subscribers
102
(
a
)-(
d
) of a cell
150
(
a
)-(
b
) served by BS
100
(
a
)-(
b
). Each access point includes receiver and transmitter circuitry of the base station for communicating with individual subscribers
102
(
a
)-(
d
) within a designated cell
150
(
a
)-(
b
).
A channel is the wireless link between an AP antenna and a WM antenna. There are a plurality of channels for receiving packets of information transmitted along various frequency bands, be it from an AP transmitter to WM receiver or vice versa. In either case, the WM and/or AP receiver can function in only one frequency band, and hence in only one channel, at a time. Further within the receiver circuitry of an AP and/or WM, there are several components used for synchronizing the time and frequency of an incoming packet with the receiver circuitry, so as to provide acceptable receiver performance.
As will be explained in more detail below, a synchronization as well as an equalization process is performed in the receiver for each incoming packet of information received by the receiver. Based on the synchronization process, a frequency offset estimation is calculated to account for the frequency offset which develops within the receiver during processing of the packet. As explained in co-pending U.S. patent application Ser. No. 09/XXX,XXX, entitled “FREQUENCY AND TIME SYNCHRONIZATION IN SEVERE DELAY SPREAD CHANNELS” and filed concurrently with the present application, a frequency offset exists because oscillators in both the transmitter and the receiver have different frequencies, although desirably they should have the same frequency. Accordingly, a frequency offset correction (hereinafter termed a stage
1
frequency offset correction algorithm) is performed in the synchronizer, and a frequency offset estimate (or phase drift (pd), as described in co-pending application 09/XXX,XXX) is generated by the synchronizer.
However this estimate is not perfectly accurate, i.e., there is some difference between the frequency offset and the frequency offset estimate (phase drift). This difference is termed a frequency offset estimate error, and should be corrected by application of a second correction, or a stage
2
frequency offset correction algorithm. Specifically, the frequency offset estimation errors affect the packet error rate (PER) performance, or probability that a transmitted packet cannot be received correctly by the receiver. For example, if the frequency offset estimate error causes a 1 degree per symbol phase drift, then after 180 symbols are processed by a receiver, all the symbol phases will be about 180 degrees off of the correct phases. This phase inaccuracy causes inaccurate and/or erroneous demodulation results, and thus packet errors.
To help understand the current implementation used for determining frequency offset estimation for channels in a wireless network and/or system, the following terms are defined. Each detected packet is divided into symbol segments allocated to various components within the receiver. For example, time symbol sequences can be allocated to a Barker detection unit, synchronizer and equalizer of a receiver. Typically each incoming packet includes in upwards of 2,000 time symbols that are processed in the various components of the receiver.
As previously discussed, to correct the frequency offset, the current implementation utilizes two frequency correction algorithms, a stage
1
frequency offset estimation algorithm and a stage
2
frequency correction algorithm. The function of the stage
1
frequency offset correction algorithm is to determine an initial frequency offset estimate (i.e., coarse frequency adjustment) for each incoming packet of information. The function of the stage
2
frequency correction algorithm is to compensate for any frequency error resulting from the stage
1
algorithm's determination.
FIG. 3
illustrates components comprising part of a receiver typically used in a WM and/or AP of a wireless system. The current stage
1
and stage
2
frequency correction algorithms are explained with reference to
FIG. 3
, and depicts a part of a receiver
200
comprising a frequency synchronizing unit
215
, frequency correction unit (FCU)
230
, equalizer
235
, adaptive frequency offset correction (AFOC) unit
240
and decision device
245
.
In
FIG. 3
, each incoming packet is processed in Barker detection circuitry (informing the receiver of an incoming packet), and the received signal is subject to frequency synchronization in a frequency synchronizing unit
215
. Any frequency offset developed between an oscillator of the receiver
200
(not shown) and the oscillator of the transmitter (AP) for example, can cause a constant phase drift between two time symbols. To account for this drift, frequency correction unit
230
receives inputs from frequency synchronizing unit
215
and an adaptive frequency offset correction unit (AFOC)
240
to determine a per-symbol phase drift. For example, frequency synchronizing unit
215
outputs an initial pd calculation which is then modified by an output from AFOC
240
for each sequentially received symbol of the packet.
Equalizer
235
processes the output of FCU
230
on a per-symbol basis to remove the effects of inter-symbol interference introduced by the channel. The output of equalizer
235
, a complex number representing a symbol of the incoming packet, is fed to decision device
245
. Decision device
245
maps this complex number to the closest QPSK symbol on t
Lucent Technologies - Inc.
Tran Khai
Tran Khanh Cong
LandOfFree
Method of correcting frequency errors for coherently... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of correcting frequency errors for coherently..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of correcting frequency errors for coherently... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3217101