U.S. patent application number 15/701412 was filed with the patent office on 2018-03-29 for array-to-array beamforming and iterative time reversal techniques.
This patent application is currently assigned to Ziva Corporation. The applicant listed for this patent is Ziva Corporation. Invention is credited to Maha Achour, Kris Gregorian, Mark Hsu, Anis Husain, Jeremy Rode, David P. Smith, Jeremy M. Ward.
Application Number | 20180091205 15/701412 |
Document ID | / |
Family ID | 58406969 |
Filed Date | 2018-03-29 |
United States Patent
Application |
20180091205 |
Kind Code |
A1 |
Smith; David P. ; et
al. |
March 29, 2018 |
ARRAY-TO-ARRAY BEAMFORMING AND ITERATIVE TIME REVERSAL
TECHNIQUES
Abstract
In examples, two arrays of Radio Frequency nodes achieve
enhanced beamforming for communications between the arrays by
successively sending sounding signals from one array to the other
array. Each sounding signal sent by the first of the two arrays is
beamformed through time reversal of an immediately preceding
sounding signal received by the first array from the second array,
and each sounding signal (except the initial sounding signal) sent
by the second array is beamformed through time reversal of an
immediately preceding sounding signal received by the second array
from the first array. The initial sounding signal sent by the
second array may be omnidirectional, beamformed through a
guesstimate, random, predetermined, or determined through a search
of the area where the arrays are located. With sufficient
beamfocusing, the arrays may communicate by sending and receiving
data from one array to the other array.
Inventors: |
Smith; David P.; (Ellicott
City, MD) ; Rode; Jeremy; (San Diego, CA) ;
Husain; Anis; (San Diego, CA) ; Achour; Maha;
(Encinitas, CA) ; Gregorian; Kris; (Encinitas,
CA) ; Ward; Jeremy M.; (Solana Beach, CA) ;
Hsu; Mark; (Richmond, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Ziva Corporation |
San Diego |
CA |
US |
|
|
Assignee: |
Ziva Corporation
San Diego
CA
|
Family ID: |
58406969 |
Appl. No.: |
15/701412 |
Filed: |
September 11, 2017 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
15277934 |
Sep 27, 2016 |
9793969 |
|
|
15701412 |
|
|
|
|
62234557 |
Sep 29, 2015 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H04B 7/022 20130101;
H04B 7/0617 20130101; H04B 7/024 20130101 |
International
Class: |
H04B 7/06 20060101
H04B007/06; H04B 7/022 20060101 H04B007/022; H04B 7/024 20060101
H04B007/024 |
Claims
1. A method of radio frequency (RF) communication between arrays of
nodes, the method comprising steps of: aligning/synchronizing nodes
of a first array in time, frequency, and phase, the first array
comprising a first plurality of nodes, the first plurality of nodes
comprising at least two first array ad hoc nodes, each node of the
first array comprising an RF receiver and an RF transmitter,
whereby the first array is enabled to operate as a coherent array
with coherent RF transmission properties and coherent RF reception
properties; aligning/synchronizing nodes of a second array in time,
frequency, and phase, the second array comprising a second
plurality of nodes, the second plurality of nodes comprising at
least two second array ad hoc nodes, each node of the second array
comprising an RF receiver and an RF transmitter, whereby the second
array is enabled to operate as a coherent array with coherent RF
transmission properties and coherent RF reception properties; and
successively sending RF sounding signals from the first plurality
of nodes to the second plurality of nodes and from the second
plurality of nodes to the first plurality of nodes, the RF sounding
signals comprising a first RF sounding signal from the first
plurality of nodes to the second plurality of nodes, each of the RF
sounding signals sent by the second plurality of nodes being
beamformed through time reversal of an immediately preceding RF
sounding signal of the RF sounding signals received by the second
plurality of nodes from the first plurality of nodes, and each of
the RF sounding signals sent by the first plurality of nodes except
the first RF sounding signal being beamformed through time reversal
of an immediately preceding RF sounding signal of the RF sounding
signals received by the first plurality of nodes from the second
plurality of nodes.
2. The method of claim 1, wherein: each of the at least two first
array ad hoc nodes is free to move in absolute terms and with
respect to each other node of the at least two first array ad hoc
nodes in at least two dimensions and each of the at least two
second array ad hoc nodes is free to move in absolute terms and
with respect to each other node of the at least two second array ad
hoc nodes in at least two dimensions.
3. The method of claim 2, further comprising: determining whether
sufficient focusing between the first plurality of nodes and the
second plurality of nodes has been achieved, thereby obtaining a
determination; terminating the step of successively sending in
response to the determination indicating that sufficient focusing
between the first plurality of nodes and the second plurality of
nodes has been achieved; and transmitting RF data communications in
at least one direction selected from the group consisting of from
the first plurality of nodes to the second plurality of nodes, and
from the second plurality of nodes to the first plurality of nodes,
the step of transmitting being performed after the determination
indicates that sufficient focusing between the first plurality of
nodes and the second plurality of nodes has been achieved.
4. The method of claim 3, wherein the step of determining comprises
measuring convergence of Signal-to-Noise Ratio (SNR) in at least
one of the first plurality of nodes and the second plurality of
nodes.
5. The method of claim 3, wherein the step of determining comprises
comparing mean square of aggregated differences of beamforming
weights metric in successive sounding iterations to a predetermined
limit.
6. The method of claim 3, wherein the step of determining comprises
computing a communication performance metric.
7. The method of claim 6, wherein the step of determining further
comprises computing a change in the communication performance
metric.
8. The method of claim 6, further comprising scatterer nulling by
the first plurality of nodes.
9. The method of claim 6, further comprising at least one step
selected from the group consisting of beam sweeping and sidelobe
sweeping.
10. The method of claim 6, wherein the step of transmitting the RF
data communications is performed with beamforming transmit/receive
gain.
11. The method of claim 9, further comprising: repeating at least
one of the steps of aligning/synchronizing the RF nodes of the
first array, aligning/synchronizing the RF nodes of the first
array, and successively sending in response to a need to maintain
link performance between the first plurality of nodes and the
second plurality of nodes determined based on an error rate or
Signal-to-Noise Ratio.
12. The method of claim 6, wherein first inter-node distances of
the first array and second inter-node distances of the second array
are greater than inter-array distances between the first array and
the second array.
13. The method of claim 6, wherein the at least two first array ad
hoc nodes comprise at least three first array ad hoc nodes, and the
at least two second array ad hoc nodes comprise at least three
second array ad hoc nodes.
14. A Radio Frequency (RF) communication apparatus comprising: a
first array of communication nodes comprising a first plurality of
nodes, the first plurality of nodes comprising at least two first
array ad hoc nodes, each node of the first plurality of nodes
comprising an RF receiver and an RF transmitter; a second array of
communication nodes comprising a second plurality of nodes, the
second plurality of nodes comprising at least two second array ad
hoc nodes, each node of the second plurality comprising an RF
receiver and an RF transmitter; wherein: the first array is
configured to align/synchronize the first plurality of nodes in
time, frequency, and phase, whereby the first array is enabled to
operate as a coherent array with coherent RF transmission
properties and coherent RF reception properties; the second array
is configured to align/synchronize the second plurality of nodes in
time, frequency, and phase, whereby the second array is enabled to
operate as a coherent array with coherent RF transmission
properties and coherent RF reception properties; and the first
array and the second array are further configured to successively
send RF sounding signals from the first plurality of nodes to the
second plurality of nodes and from the second plurality of nodes to
the first plurality of nodes, the RF sounding signals comprising a
first RF sounding signal from the first plurality of nodes to the
second plurality of nodes, each of the RF sounding signals sent by
the second plurality of nodes being beamformed through time
reversal of an immediately preceding RF sounding signal of the RF
sounding, signals received by the second plurality of nodes from
the first plurality of nodes, and each of the RF sounding signals
sent by the first plurality of nodes except the first RF sounding
signal being beamformed through time reversal of an immediately
preceding RF sounding signal of the RF sounding signals received by
the first plurality of nodes from the second plurality of
nodes.
15. The RF communication apparatus of claim 14, wherein: each of
the at least two first array ad hoc nodes is free to move in
absolute terms and with respect to each other node of the at least
two first array ad hoc nodes in at least two dimensions; and each
of the at least two second array ad hoc nodes is free to move in
absolute terms and with respect to each other node of the at least
two second array ad hoc nodes in at least two dimensions.
16. The RF communication apparatus of claim 15, wherein at least
one of the first array and the second array is further configured
to: determine whether sufficient focusing has been achieved between
the first plurality of nodes and the second plurality of nodes,
thereby obtaining a determination; stop successively sending the RF
sounding signals in response to the determination indicating to
that sufficient focusing; between the first plurality of nodes and
the second plurality of nodes has been achieved; and transmit RF
data communications in at least one direction selected from the
croup consisting of: from the first plurality of nodes to the
second plurality of nodes, and from the second plurality of nodes
to the first plurality of nodes, after the determination indicates
that sufficient focusing between the first plurality of nodes and
the second plurality of nodes has been achieved.
17. The RF communication apparatus of claim 16, wherein said at
least one of the first array and the second array is further
configured, in the course of determining whether sufficient
focusing, has been achieved between the first plurality of nodes
and the second plurality of nodes, to measure convergence of
Signal-to-Noise Ratio (SNR) in at least one of (1) the first
plurality of nodes and (2) the second plurality of nodes.
18. The RF communication apparatus of claim 16, wherein'said at
least one of the first array and the second array is further
configured, in the course of determining whether sufficient
focusing has been achieved between the first plurality of nodes and
the second plurality of nodes, to compare mean square of aggregated
differences of beamforming weights metric in successive sounding
iterations to a predetermined limit.
19. The RF communication apparatus of claim 16, wherein said at
least one of the first array and the second array is further
configured, in the course of determining whether sufficient
focusing has been achieved between the first plurality of nodes and
the second plurality of nodes, to compute a communication
performance metric.
20. The RF communication apparatus of claim 19, wherein said at
least one of the first array and the second array is further
configured, in the course of determining whether sufficient
focusing has been achieved between the first plurality of nodes and
the second plurality of nodes, to compute a change in the
communication performance metric.
21. The RF communication apparatus of claim 19, wherein the first
array is further configured to null scatterers.
22. The RF communication apparatus of claim 19, wherein the first
array is further configured to perform at least one step selected
from the group consisting of beam sweeping and sidelobe
sweeping.
23. The RF communication apparatus of claim 19, wherein said at
least one of the first array and the second array is further
configured to transmit the RF data communications with beamforming
transmit/receive gain.
24. The RF communication apparatus of claim 23, wherein said at
least one of the first array and the second array is further
configured, in response to a need to maintain link performance
between the first plurality of nodes and the second plurality of
nodes determined based on an error rate or Signal-to-Noise Ratio,
to repeat at least one of: (1) align/synchronize the RF nodes of
the first array, (2) align/synchronize the RF nodes of the first
array, and (3) successively send.
25. The RF communication apparatus of claim 19, wherein each
inter-node distance of the first array is greater than all
inter-array distances between the first array and the second
array.
26. The method of claim 19, wherein the at least two first array ad
hoc nodes comprise at least three first array ad hoc nodes, and the
at least two second array ad hoc nodes comprise at least three
second array ad hoc nodes.
27. An article of manufacture comprising at least one non-volatile
machine-readable storage medium with program code stored in the at
least one non-volatile machine-readable storage medium, the program
code, when executed by processors of nodes of a first array and of
nodes of a second array, each node of the first array and each node
of the second array comprising an antenna, a radio frequency (RF)
receiver coupled to the antenna, an RF transmitter coupled to the
antenna, and a processor coupled to the RF receiver and the RF
transmitter to control operation of the RF receiver and the RF
transmitter, the first array comprising a first plurality of nodes
that comprises at least two first array ad hoc nodes, the second
array comprising a second plurality of nodes comprising at least
two second array ad hoc nodes, configures the nodes of the first
array and the nodes of the second array to: align synchronize nodes
of the first array in time, frequency, and phase, enabling the
first array to operate as a coherent array with coherent RF
transmission properties and coherent RF reception properties;
align/synchronize nodes of a second array in time, frequency, and
phase, enabling the second array to operate as a coherent array
with coherent RF transmission properties and coherent RF reception
properties; and successively send RF sounding signals from the
first plurality of nodes to the second plurality of nodes and from
the second plurality of nodes to the first plurality of nodes, the
RF sounding signals comprising a first RF sounding signal from the
first plurality of nodes to the second plurality of nodes, each of
the RF sounding signals sent by the second plurality of nodes being
beamformed through time reversal of an immediately preceding RF
sounding signal of the RF sounding signals received by the second
plurality of nodes from the first plurality of nodes, and each of
the RF sounding signals sent by the first plurality of nodes except
the first RF sounding signal being beamformed through time reversal
of an immediately preceding RF sounding signal of the RF sounding
signals received by the first plurality of nodes from the second
plurality of nodes.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] The present application claims priority from U.S. patent
application Ser. No. 15/277,934, entitled ARRAY-TO-ARRAY
BEAMFORMING AND ITERATIVE TIME REVERSAL TECHNIQUES, tiled on 27
Sep. 2016, now allowed; which claims priority from U.S. Provisional
Patent Application Ser. No. 62/234,557, entitled ARRAY-TO-ARRAY
BEAMFORMING USING ITERATIVE TIME. REVERSAL TECHNIQUES, tiled on 29
Sep. 2015, which is hereby incorporated by reference in its
entirety as if fully set forth herein, including text, figures,
claims, tables, and computer program listing appendix (if
present).
FIELD OF THE INVENTION
[0002] This document relates generally to the field of wireless
communications. Selected examples relate to distributed/cooperative
communication arrays of nodes and/or antennas using Time-Reversal
(TR) techniques for communications between different arrays of
antennas/nodes where the receiver array and transmitter array are
located in different locations.
BACKGROUND
[0003] Distributed coherent communications are Radio Frequency (RF)
communications where coherent transmissions are made from a
transmit antenna array, and/or RF transmissions are received by a
synchronized receive antenna array. Each of the antenna arrays may
be formed by an array of nodes, with each of the nodes having one
or more of the antennas. Some or all of the nodes of either
transmit (Tx) node array or receive (Rx) node array may be ad hoc
nodes (as is described below). The "coherent" property of the
coherent communications refers to synchronization of the nodes, so
that (1) each of the nodes in the transmit array can transmit
synchronously, and/or (2) each of the nodes of the receive array
are synchronized and the received signals may be combined using a
common time reference, with resulting transmit and/or receive array
gain. Distributed coherent communications may offer significant
link budget gains and increased performance over those available
with single-antenna-to-single-antenna communications. Compared to
the use of single antenna transceivers, the use of multiple
antennas in wireless networks offers the promise of increased data
rates, reach distance, battery life, anti-jam capabilities,
spectral reuse, as well as reduced latency. Distributed coherence
can be leveraged into transmit beamforming (with, e.g.,
N.sup.2-fold increase in power for N transmit antennas), receive
beamforming (with, e.g., M-fold increase in power for M receive
antennas), and simultaneous transmit/receive beamforming gains
(with a theoretically-possible N.sup.2M-fold increase in power for
N transmit to M receive antennas).
[0004] Distributed communication and networking approaches,
however, may suffer from is one or more of a number of
disadvantages, including these:
[0005] 1. Some systems may improve established point-to-point
links, but cannot create links which would otherwise have not
existed;
[0006] 2. Some systems rely on complex weights and pre-coding
matrices calculations derived from link measurements, which may not
always be available or easily obtainable;
[0007] 3. Some systems employ long latency feedback techniques,
which may inject extra delays and may not converge in time
sufficiently short for operation in fast-changing dynamic and
otherwise challenging environments;
[0008] 4. Some systems rely on channel training in the forward and
return directions to estimate weights and pre-coding matrices
(i.e., "closed loop" or "explicit" approach), with associated
complexity and delays;
[0009] 5. Some systems require brute force, system-wide, long-term
and short-term synchronization of carrier frequency, modulation,
data, and time.
[0010] At a given array, collaborative beamforming weights may be
discovered using a "sounding" signal, an opportunistic or an
intentional signal emitted by the object of the beamforming, as
discussed in more detail below. The underlying assumption of such
collaborative beamforming, communications is that a single target
node is capable of emitting a sounding signal with sufficient power
to overcome link losses to reach the beamforming array, which can
then retrodirect the signal back to the target node as a
many-to-one (N:1) beamformer. In cases of one-to-many (1:M) receive
beamforming where a single node emits a signal and an array of
nodes on the receiving side uses beamforming weights to receive the
signal, the underlying assumption is that the emitting node
transmits with sufficient power for the signal to traverse the
channel and be detected by the array of nodes on the receiving
side.
[0011] It is not always the case, however, that a single
transmitting node can transmit with to power sufficient to be
detected on the other side of the link; in other words, a single
node may not be able to "close" the link on its own, particularly
where the transmitted signal is a sounding signal and initially is
not beamformed. For a many-to-many (N:M) array-to-array
beamforming, where a pair of arrays communicate with one another,
simultaneous transmit and receive gains can be used to enhance the
link budget. Link budget gains may be applied, for example, to link
improvement or reduction in transmit power. Table 1 below is a
summary of possible benefits from simultaneous transmit and receive
beamforming gains in exemplary systems:
TABLE-US-00001 TABLE 1 LPI/LPS/S Battery pectral (PA) Data Rate
Reach Anti-jam Latency Reuse Life Link .uparw.Log.sub.2(N.sup.2 M)
.uparw.N SQRT(M) .uparw.N.sup.2 M .dwnarw.Hops .dwnarw.N --
Improvement (Full TX Power) Stealth (1/N.sup.2 TX
.uparw.Log.sub.2(M) .uparw.SQRT(M) .uparw.M -- .uparw.N
.uparw.N.sup.2 Power)
[0012] Such gains may be achieved when transmit and receive
beamforming weights are derived using optimal algorithms and
applied at the Tx and Rx arrays. The correct channel information
between the array members on each side of the link and a mechanism
for the nodes to transmit/receive a corrected and weighted signal
at each of the array nodes are needed, so that beamforming is
achieved to within an accuracy required by the system. If the array
nodes are distributed (i.e., untethered physically separated radios
such as ad hoc nodes described below), each array may implement a
distributed algorithm across the nodes of the array, enabling the
array to operate in a coherent manner, providing frequency, and
phase/time synchronization/alignment of the clocks and oscillators
of the different nodes of the array. Certain array
synchronization/alignment methods for local synchronization of a
distributed set of array nodes are described in some of the related
and commonly-owned patent documents listed below.
[0013] The problem of initially detecting a signal such as a
sounding signal without a priori knowledge of the target, the
environment, and the beamforming weights, however, remains.
[0014] There is a need fix techniques for improving radio frequency
communications, and in particular for techniques to improve
beamforming in circumstances where sounding to transmissions from a
single node of a target array may not be detectable by the
transmitting array. There is also a need for techniques for
establishing and maintaining RF communications between arrays with
improved beamforming weights, in both static and
dynamically-changing environments, and in both Line-of-Sight (LoS)
and Non-Line-of-Sight situations (NLoS).
SUMMARY
[0015] Embodiments, variants, and examples described in this
document are directed to methods, apparatus, and articles of
manufacture that may satisfy one or more of the above described
and/or other needs.
[0016] Selected examples of communication architecture described in
this document enable improved communications by allowing a first
array of RF transmit-receive (TX/RX) nodes to communicate
cooperatively with a second array of R TX/RX nodes. In each of the
arrays, the TX/RX nodes may be tethered with a common clock
reference, operating as a coherent array; the nodes of a particular
"array" may be simply different (spatially diverse) antennas of the
same transceiver system. In each of the arrays, some or all of the
nodes may alternatively be untethered, ad hoc (as this term is
defined below) nodes with independent clock references which use
certain processes to achieve phase, frequency, and time
synchronization/alignment of the clocks of the individual nodes of
the array, enabling the nodes of a particular array to operate as a
coherent array and effect distributed coherent communications. As
has already been mentioned, various synchronization/alignment
processes are described in the related and commonly-owned patent
documents listed below.
[0017] Selected examples described in this document perform
iterative time reversal processes between two arrays of nodes, with
each array transmitting its sounding signals and refining its
beamforming weights and its following sounding signals iteratively,
as the array receives the sounding signals from the other array. In
examples, the largest node-to-node distance of either or both of
the arrays may be substantially smaller (by a factor of ten) than
the smallest distance between any of the nodes of one of the arrays
and any of the nodes of the other array. In examples, the largest
node-to-node distance of either or both of the arrays may be to
comparable to (not substantially smaller than) the smallest
distance between any of the nodes of one of the arrays and any of
the nodes of the other array, resulting in a bistatic
configuration.
[0018] In an embodiment, a method of radio frequency (RF)
communication between arrays of nodes includes
aligning/synchronizing a plurality of ad hoc nodes of a first array
in time and frequency; aligning/synchronizing a plurality of ad hoc
nodes of a second array in time and frequency; transmitting a first
initial sounding signal from the plurality of ad hoc nodes of the
first array to the plurality of ad hoc nodes of the second array;
and successively sending sounding signals from the plurality of ad
hoc nodes of the first array to the plurality of ad hoc nodes of
the second array and from the plurality of ad hoc nodes of the
second array to the plurality of ad hoc nodes of the first array,
wherein each sounding signal sent by the plurality of ad hoc nodes
of the second array is beamformed through time reversal of an
immediately preceding sounding signal received by the plurality of
ad hoc nodes of the second array from the plurality of ad hoc nodes
of the first array, and each sounding signal except the first
initial sounding signal sent by the plurality of ad hoc nodes of
the first array is beamformed through time reversal of an
immediately preceding sounding signal received by the plurality of
ad hoc nodes of the first array from the plurality of ad hoc nodes
of the second array.
[0019] In an embodiment, an apparatus includes a first array
comprising a plurality of first ad hoc radio frequency (RF) nodes,
and a second array comprising a plurality of second ad hoc RF
nodes. In the apparatus, the first array and the second array are
configured to: align/synchronize the plurality of first ad hoc RF
nodes in time and frequency; align/synchronize the plurality of
second ad hoc RF nodes in time and frequency; transmit an initial
sounding signal from the plurality of first ad hoc RF nodes of the
first array to the plurality of second ad hoc radio frequency (RET)
nodes of the second array and successively send sounding signals
from the plurality of first ad hoc RF nodes of the first array to
the plurality of second ad hoc RF nodes of the second array and
from the plurality of second ad hoc RF nodes of the second array to
the plurality of first ad hoc RF nodes of the first array, wherein
each sounding signal sent by the second array is beamformed through
time reversal of an immediately preceding sounding signal received
by the second array from the first array, and each sounding signal
except the initial sounding signal sent by the first array is
beamformed through time reversal of an immediately preceding
sounding signal received by the first array from the second
array.
[0020] In an embodiment, an article of manufacture includes at
least one non-volatile machine-readable storage medium with program
code stored in the at least one non-volatile machine-readable
storage medium. The program code may be executed by processors of
nodes of a first array and of nodes of a second array, each node of
the first array and each node of the second array comprising an
antenna, a radio frequency transceiver coupled to the antenna., a
local oscillator, and a processor coupled to the radio frequency
transceiver to control operation of the transceiver. When the code
is thus executed, it configures the nodes of the first array and
the nodes of the second array to align/synchronize the nodes of the
first array in time and frequency; align/synchronize the nodes of
the second array in time and frequency, transmit a initial sounding
signal from the first array to the second array; and successively
send sounding signals from the first array to the second array and
from the second array to the first array, wherein each sounding
signal sent by the second array is beamformed through time reversal
of an immediately preceding sounding signal received by the second
array from the first array, and each sounding signal except the
initial sounding signal sent by the first array is beamformed
through time reversal of an immediately preceding sounding signal
received by the first array from the second array.
[0021] In an embodiment, a method of radio frequency (RF)
communication between arrays of nodes includes
aligning/synchronizing a plurality of ad hoc nodes of a first array
in time and frequency; aligning/synchronizing a plurality of ad hoc
nodes of a second array in time and frequency; step for Reciprocal
Convergence process between the first array and the second array to
enable beamformed transmissions from the first array to the second
array and from the second array to the first array; and sending RF
data communications from the first array to the second array and/or
from the second array to the first array, the step of sending being
performed after the step for Reciprocal Convergence process.
[0022] In an embodiment, a method of radio frequency (RF)
communication includes aligning/synchronizing a plurality of ad hoc
nodes of a first array in time and frequency; transmitting an
initial sounding signal from the first array to a second array of
nodes successively sending sounding signals from the first array to
the second array, wherein each sounding signal except the initial
sounding signal is beamformed through time reversal of an
immediately preceding sounding signal received by the first array
from the second array; determining whether sufficient focusing
between the first array and the second array has been achieved,
thereby obtaining a determination; terminating the step of
successively sending in response to the determination indicating
that sufficient focusing has been achieved; and sending RF data
communications from the first array to the second array in response
to the determination indicating sufficient focusing.
[0023] These and other features and aspects of selected embodiments
not inconsistent with is the present invention(s) will he better
understood with reference to the following description, drawings,
and appended claims.
BRIEF DESCRIPTION OF THE FIGURES
[0024] FIG. 1 illustrates selected nodes and links of two arrays
configured in accordance with one or more features described in
this document;
[0025] FIG. 2 illustrates selected elements of a node of an array
configured w accordance with one or more features described in this
document;
[0026] FIG. 3 illustrates selected features of an example of an
emitted sounding signal;
[0027] FIG. 4 and FIG. 5 illustrate selected aspects of simulations
of signals on receivers of ad hoc node arrays;
[0028] FIG. 6A and FIG. 6B illustrate selected aspects of
simulations of convergence of exemplary Iterative Time-Reversal or
Reciprocal Convergence Algorithm processes;
[0029] FIG. 7 illustrates selected steps and decision blocks of an
exemplary Reciprocal Convergence Algorithm;
[0030] FIG. 8 illustrates selected aspects of sweeping by an array
of a main beam with sidelobes;
[0031] FIG. 9A and FIG. 9B illustrate selected aspects of
transmissions from one end array to another end-array through
intermediate array (s) in a cascade manner;
[0032] FIG. 10 illustrates selected steps and decision blocks of an
exemplary Reciprocal Convergence Algorithm process with scatterer
nulling performed between two ad hoc arrays; and
[0033] FIG. 11 illustrates selected steps of another example of a
Reciprocal Convergence Algorithm process with scatterer nulling,
performed between two ad hoc arrays.
DETAILED DESCRIPTION
[0034] The words "embodiment," "variant," "example," and similar
words and expressions as used here refer to a particular apparatus,
process, or article of manufacture, and not necessarily to the same
apparatus, process, or article of manufacture. Thus, "one
embodiment" (or a similar expression) used in one place or context
may refer to a particular apparatus, process, or article of
manufacture; the same or a similar expression in a different place
or context may refer to a different apparatus, process, or article
of manufacture. The expression "alternative embodiment" and similar
words and phrases are used to indicate one of a number of different
possible embodiments, variants, or examples. The number of possible
embodiments, variants, or examples is not necessarily limited to
two or any other quantity. Characterization of an item as
"exemplary" means that the item is used as an example. Such
characterization does not necessarily mean that the embodiment,
variant, or example is a preferred one; the embodiment, variant, or
example may but need not be a currently preferred embodiment,
variant, or example. All embodiments, variants, and examples are
described for illustration purposes and are not necessarily
strictly limiting.
[0035] The words "couple," "connect" and similar words with their
inflectional morphemes, as well as similar words and phrases, do
not necessarily import an immediate or direct connection, but
include within their meaning connections through mediate
elements.
[0036] The expression "processing logic" should be understood as
selected steps/decision blocks and/or hardware/software/firmware
for implementing the selected steps/decision blocks. "Decision
block" means a step in which a decision is made based on some
condition, and process flow may be altered based on whether the
condition is met or not met.
[0037] The expression ad hoc in reference to nodes of an array of
nodes is used to signify that at least some (or all) of the ad hoc
nodes have their own physical clocks, and the nodes are
"untethered" in the sense that they may be (1) free to move, in
absolute terms (e.g., with respect to a point with fixed
coordinates, including scatterers and nodes of the other array),
and to move with respect to each other and (2) free to rotate
around one or more axes. Some constraints on the movements of some
or all of the nodes need not necessarily vitiate their untethered
or ad hoe character; for example, nodes that are free to move in
only one or two dimensions (and not all three dimensions) may still
be ad hoc nodes, whether or not they rotate around any axis.
Examples of such nodes may include radios carried by a squad of
soldiers; radios onboard different aircrafts, water vessels or
buoys, land vehicles, satellites and similar nodes. The nodes may
be ad hoc even if they are not used in TR-communications; for
example, nodes of a phased-array that is beamforming in a selected
direction (rather than beamforming on a selected spot/item or
spot-focusing, as is typically the case with TR communications) may
also be ad hoc nodes.
[0038] Some definitions have been explicitly provided above. Other
and further explicit and implicit definitions and clarifications of
definitions may be found throughout this document.
[0039] FIG. 1 illustrates in a high level, block-diagram manner,
selected components of arrays 105 and 110. We illustrate and
describe the array 105 in some detail, with the understanding that
the array 110 may be identical to the array 105, substantially the
same as the array 105, or similar to the array 105. The array 105
includes ad hoc nodes 105-N that may communicate with each other,
and synchronize their respective clocks aligning time, phase, and
frequency). As shown, the array 105 includes five distributed
cooperating nodes, 105-1 through 105-5. In similar examples, the
array 105 may include any number of a plurality of nodes 105-N, for
example, 2, 3, 4, 5, 6, 7, 8, 9,10, or more.
[0040] The nodes 105-N may be within LoS or NLoS of each other, and
may communicate directly with each other via side channel links
120. As shown, the links 120-1, 120-2, and 120-4 connect the node
105-3 to each of the nodes 105-1, 105-2, and 105-4, respectively
and the link 120-5 connects the node 105-4 to the node 105-5. The
node 105-3 may thus communicate directly with each of the nodes
105-1, 105-2, and 105-4. The node 105-3 may communicate with the
node 105-5 indirectly, through the node 105-4 and the links 120-4
and 120-5. This is just one example. More generally, any of the
nodes 105-N may be connected by such side channel link 120 to any
of the other nodes 105-N, and any of the nodes 105-N may lack a
direct link to any other node (or nodes) 105-N, and communicate
with such other nodes 105-N through intermediate nodes and multiple
(two or more) links. The side channel links 120 may be implemented,
for example, using short-range radio frequency (RF) link such as a
Bluetooth.RTM. link WiFi, or other short-, medium-, and
longer-range RF technologies. As discussed in some of the related
and commonly-owned patent documents listed below, the side channel
links 120 may also he implemented using non-RF technologies and
transmission media, including optical technologies, such as
free-space or guided optics, and sound/acoustic (ultrasound)
technologies. Moreover, different types of side channel links 120
may be present within the same array, for example, both RF and
non-RF links.
[0041] The array 110 includes ad hoc nodes 110-M that, may
communicate with each other, and synchronize their respective
clocks (i.e., aligning time, phase, and frequency). As shown, the
array 110 includes five distributed cooperating nodes, 110-1
through 110-5; and side channel links 130-1, 130-2, 130-3, 130-4,
and 130-5.
[0042] The inter-node distance of each of the arrays 105 and 110
may he much smaller (by a factor of at least 10, at least 100, at
least 1000, or even greater) than the inter-array distances between
any of the nodes 105 and any of the nodes 110. For example, each of
the distances between any two nodes 105 may be less than 1/10, less
than 1/100, or less than 1/1000 than any of the distances between
any selected node 105 and any selected node 110. In examples,
however, the inter-node distance of either or both arrays 105 and
110 are not much smaller (as "much smaller" is explained at the
beginning of this paragraph) than the inter-array distances.
[0043] FIG. 2 illustrates selected elements of an apparatus 200
configured in accordance with one or more features described in
this document. The apparatus 200 may be any of the cooperative
nodes of the arrays 105 and 110. The apparatus may include a
processor 205; storage device(s) 210 (which may store program code
for execution by the processor 205); an RF receiver 220 configured
to receive radio frequency signals, such as sounding signals and
their reflections/backscatter and information from other nodes of
the same array and from nodes of other arrays; an RF transmitter
215 configured to transmit radio frequency signals, such as
sounding signals, collaborative communications to nodes of other
arrays, and information for other nodes of the same array; one or
more RF transmit and receive antennas 225 coupled to the receiver
220 and the transmitter 215; and a non-RF processing module 227,
such as an optical or acoustic transceiver and associated signal
processing devices. A bus 230 couples the processor 205 to the
storage device 210, to the receiver 220, to the transmitter 215,
and to the non-RF processing module 227. The bus 230 allows the
processor 205 to read from and write to these devices, and
otherwise to control operation of these devices. In embodiments,
additional receivers and/or transmitters are present and coupled to
the processor 205.
[0044] In examples, the arrays 105 and 110 communicate in whole or
in part using Time-Reversal techniques Time Reversal techniques may
combine (1) sounding of a channel with (2) applying pre-filtering
to a transmission, e.g., time-reversing the channel impulse
response (the channel response from one object to another) and
convolving it with data to he sent or with some other signal such
as a pulse/burst or another waveform (which may be a
well-autocorrelated waveform). "Sounding" and its inflectional
morphemes refer to transmitting a signal for the purpose of
obtaining information about the channels, for example, for forming
TR signals. Sounding may also be opportunistic, that is, the
sounding signal may be transmitted for another purpose but also
used for obtaining the channel state information. The sounding
signal may be a sharp pulse approaching an impulse, a Gaussian
burst, or another appropriate burst with substantially flat
frequency response in the communication band, and having a good
autocorrelation function (i.e., approaching that of an impulse
function), as is known in communication theory and related fields
(e.g., CDMA, autocorrelation radar).
[0045] Time-reversal techniques (for communication and other
purposes) and sounding are described in several commonly-owned and
related patent documents, including the following:
[0046] 1. U.S. patent application Ser. No. 13/462,514, Publication
Number 2012-0328037, entitled ANTI-GEOLOCATION, filed on 2 May
2012;
[0047] 2. International Patent Publication WO/2012/151316
(PCT/US2012/36180), entitled DISTRIBUTED CO-OPERATING NODES USING
TIME REVERSAL filed 2 May 2012;
[0048] U.S. patent application Ser. No. 14/114,901, Publication
Number 2014-0126567, entitled DISTRIBUTED CO-OPERATING NODES USING
TIME REVERSAL filed on 30 Oct. 2013;
[0049] 4. U.S. Provisional Patent Application Ser. No. 61/481,720,
entitled DISTRIBUTED CO-OPERATING NODES USING TIME REVERSAL FOR
COMMUNICATIONS, SENSING & IMAGING, filed on 2 May 2011;
[0050] 5. U.S. Provisional Patent Application Ser. No. 61/540,307,
entitled DISTRIBUTED CO-OPERATING NODES USING TIME REVERSAL FOR
COMMUNICATIONS, SENSING & IMAGING, filed on 28 Sep. 2011;
[0051] 6. U.S. Provisional Patent Application Ser. No. 61/809,370,
entitled APPARATUS, METHODS AND ARTICLES OF MANUFACTURE FOR
COLLABORATIVE BEAMFOCUSING OF RADIO FREQUENCY EMISSIONS, filed on 7
Apr. 2013;
[0052] 7. U.S. Provisional Patent Application Ser. No. 61/829,208,
entitled APPARATUS, METHODS, AND ARTICLES OF MANUFACTURE FOR
COLLABORATIVE BEAMFOCUSING RADIO FREQUENCY EMISSIONS, filed on 30
May 2013;
[0053] 8. International Patent Application PCT/US2014/033234,
entitled DISTRIBUTED CO-OPERATING NODES USING TIME REVERSAL, filed
7 Apr. 2014;
[0054] 9. U.S. patent application Ser. No. 14/247,229, Publication
Number US 2014-0301494, entitled DISTRIBUTED CO-OPERATING NODES
USING TIME REVERSAL, filed on 7 Apr. 2014;
[0055] 10. U.S. Provisional Patent Application Ser. No. 61/881,393,
entitled APPARATUS, METHODS, AND ARTICLES OF MANUFACTURE FOR
COLLABORATIVE ARRAY COMMUNICATIONS INCLUDING BEAMFOCUSING OF
EMISSIONS, filed on 23 Sep. 2013;
[0056] 11. U.S. patent application Ser. No. 14/476,738, Publication
Number US 2015-0085853, entitled SYNCHRONIZATION OF DISTRIBUTED
NODES, filed on 4 Sep. 2014;
[0057] 12. U.S. patent application Ser. No. 14/494,580, Publication
Number 2015-0173034, entitled SYNCHRONIZATION OF DISTRIBUTED NODES,
filed 23 Sep. 2014;
[0058] 13. U.S. Provisional Patent Application Ser. No. 62/126,437,
entitled TIME REVERSAL IN WIRELESS COMMUNICATIONS, filed on 27 Feb.
2015;
[0059] 14. International Patent Publication WO/2016/137898
(PCT/US2016/018968) entitled TIME REVERSAL IN WIRELESS
COMMUNICATIONS (OFDM), filed 22 Feb. 2016;
[0060] 15. U.S. Provisional Patent Application Ser. No. 62/196,885,
entitled TIME REVERSAL IN WIRELESS COMMUNICATIONS, filed on 24 Jul.
2015; and
[0061] 16. U.S. patent application Ser. No. 15/217,944, entitled
WIRELESS SENSING WITH TIME REVERSAL, filed 22 Jul. 2016.
[0062] Each of the patent documents described above is hereby
incorporated by reference, including specification, claims,
figures, tables, and all other matter in the patent document. We
may refer to these documents and other commonly-owned patent
documents specifically identified throughout the present document
collectively as "incorporated applications" and "incorporated
patent documents."
[0063] Turning now again to FIG. 1, recall that each of the arrays
105/110 has a plurality of nodes, with at least one antenna per
node. (Each node can have a single antenna or a sub-array of
antennas.) In the following discussion, we will consider the array
105 as the initiating array with N nodes, and the array 110 as the
receive array with M nodes; generally these are the two ends of the
desired communication link, though cascaded arrangements will be
described as well. The array 105 is initiating in the sense that it
begins the beamforming process by sending the original sounding
signal; the roles of the two arrays may be reversed.
[0064] The nodes within each of the arrays can communicate locally
(with other nodes of the same array), for example, for data
distribution, synchronization, coordination, and other purposes.
The nodes of the array 105 may be ad hoc nodes with separate
individual clock references; the nodes may be tethered with a
common clock reference; or the array 105 may have a combination of
both types of nodes, ad hoc and tethered (not ad hoc) nodes.
Similarly, the nodes of the array 110 may be ad hoc nodes, tethered
nodes, or the array 110 may have a combination of both types of
nodes. At each of the arrays 105/110, local time, phase, and
frequency synchronization/alignment of nodes is performed using,
for example, the processes described in the incorporated patent
documents. Thus, in each of the arrays 105/110, a time base common
to the respective array is established with a common reference
mirror and emission time set for all local array nodes. The
reference mirror and emission times may be used to perform channel
sounding, beamforming, and retrodirection.
[0065] For the array-to-array communications, target array
discovery may be a challenge when the transmit power is too limited
to provide sufficient link budgets to enable an individual node of
the array 105 to communicate across the link to an individual node
of the array 110, and/or versa. Therefore, no individual radio
within the transmit array may be able to communicate with a node
within the other array. But the transmit power may be sufficient
for a communication link between the arrays to be established using
beamforming array gains. This is due to the increased link budget
resulting from multiple array nodes. An iterative TR (ITR)
algorithm may be used to determine the transmit and receive antenna
beamforming weights to increase the link performance. Note that in
this document the meaning of "ITR" is descriptive in the context of
array-to-array communications, and therefore may be somewhat
different from the meaning of "ITR" in some of the incorporated
patent documents; the meanings of "ITR" in this document and the
incorporated patent documents, however, are analogous, mutatis
mutandis, as a person of average skill in the art of RF
communications will understand after careful perusal of this
document and the incorporated patent documents. The term Reciprocal
Convergence Algorithm or process means ITR in the array-to-array
context, as will become clear.
[0066] An exemplary ITR process between two arrays may start with
an initial state where the array-to-array link has not yet been
established. In this case, the arrays may close the link between
them using a number of methods described herein. This may be
referred to as a "cold start" condition.)
[0067] Array-to-array beamforming uses a mechanism to derive the
transmit and receive beamforming weights at each end of the link
between the two arrays, e.g., the link between the arrays 105 and
110. With channel reciprocity in a time division duplex (TDD)
configuration and when the transmit array and receive array are
identical in size, number of nodes, and geometry, the transmit
beamforming and receive beamforming weights may be identical at
each node; that is, for a given node the transmit weights are the
same as receive weights, though the weights will typically vary
from node to node. Below we describe various approaches to
establish coherent array-to-array beamforming using ad hoc and/or
tethered nodes.
[0068] If the layout of the array (e.g., the array 105) is
unavailable at the beginning of the to beamforming process, the
array may use a time-coordinated emission to close the link to the
other array (e.g., to the array 110). Recall that the nodes within
the array 105 may be tethered and/or ad hoc. The ad hoc nodes may
first he synchronized/aligned to other ad hoc nodes of the array
105, and to the tethered nodes if such are present in the array 105
in addition to the ad hoc nodes. A common time, t.sub.0, may be
used as a reference for the array 105 nodes to emit a coordinated
sounding. The array 105 nodes may all emit the sounding at the time
t.sub.0 or time-stagger their soundings based on their relative
distance that is derived during the clock-frequency
synchronization/alignment step. Relative distance may be to the
center of mass of the array (with each node having the same weight,
for example), the most central node, or any other metric that
derives relative distances along the nodes within a distributed
array. If the nodes are tethered and their relative locations are
not known, the time of emission may also be varied from sounding to
sounding in some time-varying manner, for example, random or
following some predetermined order or variations, within a
predetermined or dynamically-determined time boundary .DELTA.t. The
bounds on .DELTA.t may be set to the maximum inter-node
time-of-flight of the array 105, or somewhat higher (e.g., 50%
higher).
[0069] Thus, after the nodes of the array 105 are
synchronized/aligned, the array may emit a sounding at a
coordinated time. The collective signal may exhibit, on average, an
incoherent gain of N, statistically varying between decoherent to
coherent at the other array 110. This is in part due to unknown
channel delays from each node of the array 105 to the nodes of the
array 110, causing the sounding signals not to align perfectly at
the nodes of the array 110. The fluctuation between coherent and
decoherent may be due to the randomized carrier phases of the
signals incident on the array 110. Statistically, the coordinated
sounding from the array 105 may achieve sufficient link budget
increases to be detected by the array 110, so that a beamforming
training session may begin. The array 105 may then use a
coordinated incoherent search signal to converge towards focused
communication to the array 110.
[0070] Once the nodes 105 agree on a signal emission time and time
window duration, the nodes 105 may emit a signal in a coordinated
manner, for example, simultaneously (e.g., if the location of the
nodes 110 of the other array are unknown), or with some coarse
weighting (if an approximate location of the nodes 110 or the
direction from the array 105 to the array 110 is known). Note that
the array 105 does not need to have beamforming weights for ideal
or near ideal coherent addition at the array 110. Thus, the
following exemplary procedures may be followed for the initial
coordinated incoherent sounding:
[0071] 1. A sounding signal may be emitted from the array 105 in a
blind manner, i.e., with every node 105 emitting at agreed upon
time. The array 110 capture of the sounding signal will then depend
on the incoherent gain statistics; thus, multiple soundings may be
needed for an incoherent sounding signal from the array 105 to be
identified at the array 110.
[0072] 2. Beamforming weights may be estimated for the sounding
signal at the array 105 if a priori information on location and/or
direction of the array 110 is known at the array 105. The
information on the layout of the array 105 (which may also be
needed to estimate the beamforming weights) may be derived, for
example, from GPS or inter-node time-of-flight data.
[0073] FIG. 3 illustrates an example of an emitted sounding signal
300, such as an initial sounding signal emitted by the nodes of the
array 105. In the Figure, the sounding signal 300 appears
substantially as a Gaussian waveform for illustrative purposes, but
can be another waveform, particularly a well-autocorrelated
function as is discussed above in the context of explaining time
reversal.
[0074] While a single node 105 emitting a sounding signal
independently may not have the ability to trigger a beamforming
weight training initialization because its signal may be buried in
noise after path loss and fading, conditions may exist such that a
coordinated sounding from the array 105, statistically, has
sufficient incoherent gain to overcome link budget deficit and the
sounding may be detected by one or more of the nodes of the array
110.
[0075] The array 110 nodes are thus triggered by the initial
incoherent beamforming sounding signal from the array 105 to begin
the TR-based back-and-forth training, sequence during which
beamforming weights are determined at both arrays. The array 110
nodes are synchronized/aligned, which may be done, for example, in
response to the initial sounding from the array 105, or earlier.
The nodes of the array 110 set a common Time Reversal
capture-retransmission window, and simultaneously transmit the
Time-Reversed sounding signal received from the array 105. The
Time-Reversed signal transmitted by the array 110 is encapsulated
(contained) in this window. This latter signal is the sounding
signal of the array 110 transmitted to the array 105 as part of the
first back-and-forth beamfocusing iteration of the two arrays.
[0076] The sounding signal transmitted by the array 110 as part of
the first iteration is received by the nodes of the array 105,
Time-Reversed, and re-transmitted hack to the array 110 in a common
capture-retransmission window of the array 105. This latter
sounding signal is the first sounding signal of the second
back-and-forth beamfocusing iteration; it will typically be better
focused than the initial sounding signal from the array 105 to the
array 110.
[0077] The latest sounding signal emitted by the array 105 is
received by the array 110. Because of the better focusing, there
should be some coherent gain, and the received signal may be
stronger than the sounding signal received by the array 110 in the
first iteration. The array 110 then Time-Reverses it and
re-transmits it back to the array 105 in a common window of the
array 110. This is the second part of the second iteration.
[0078] The back-and-forth sounding may be continued as needed, to
focus each of the arrays 105/110 on the other array 110/105. Once
sufficient beamforming is achieved, the two arrays may communicate
and exchange data across the array-to-array link with the
transmit/receive beamforming gains that have been determined as
part of the beamforming operation. Sufficiency of the beamforming
may be determined, for example, by reference to some performance
metric or change in the performance metric (e.g., SNR for one or
both sides, and/or change in the SNR from one iteration to a
following iteration). The sufficiency of beamforming may also be
determined by reference to the changes of the beamforming weights
in one or both arrays. For example, a metric aggregating
differences between each current weight and the same weight as
determined in the immediately previous (or another previous)
iteration may be computed, and sufficiency may be declared when
this "aggregated differences metric" is less than some
predetermined limit. The aggregated differences metric may be based
on the mean square computations.
[0079] The back-and-forth iterative beam focusing process may be
repeated as needed, for example, to maintain link performance in
either or both directions, as determined based on some link metric
such as a bit/symbol/block error rate, or SNR. The process may also
or instead be repeated periodically, at other predetermined or
random times, or in response to other factors. When repeated, the
process may but need not begin at the "cold start" condition; the
process can use the latest beamforming weights as the starting
point.
[0080] To recapitulate, the arrays 105 and 110 use time reversal to
obtain the (relatively) coherent signal based on the (relatively)
incoherent signal through multiple back-and-forth iterations,
converging towards the optimal or near-optimal array-to-array
beamforming weights. The array 110 nodes may initially be triggered
by the incoherent beamforming training request to begin the
Time-Reversal-based back-and-forth array weight training sequence.
The array 110 nodes set a common Time Reversal capture window, and
retransmit simultaneously the signal encapsulated in this window
for nodes 110-1 through 110-M. The array 105 nodes capture the Time
Reversed coordinated transmission from the array 110, and time
reverse and retrodirect the captured signal back to the array 110.
The array 110 repeats the Time Reversal mirroring back to the array
105. These steps may be repeated in a back-and-forth manner, which
we may call the "Reciprocal Convergence Algorithm" in this
document. Through a series of back-and-forth Time Reversal
retrodirections, the array emissions may converge to the optimal or
near-optimal beam-patterns for communication with beamforming
weights discovered automatically, without calculation, without
explicit channel estimation, and without a priori node layout
knowledge requirements.
[0081] FIG. 4 and FIG. 5 illustrate selected aspects of
simulations, of signals observed on one of receivers on each end of
the link for ad hoc node array layouts, and demonstrate the
Reciprocal Convergence processes for array weight discovery,
transforming the beamforming from uninformed to informed, for each
individual node when both arrays are located within LoS. The arrows
in FIG. 5 indicate the direction in which the sounding signals are
emitted. Aspects of the convergence of exemplary ITR or Reciprocal
Convergence processes are illustrated in FIG. 6A and FIG. 6B. FIG.
6A illustrates a simulated convergence of linear 10-node Array A
and 10-node Array B to a focused beam (one array to the other
array) through a series of Time Reversal iterations, showing how
the beam pattern converges from an omnidirectional search to a more
focused beamformed pattern. FIG. 6B illustrates a simulated
convergence of a 10-node ad hoc Array A and 10-node ad hoc array B
with non-linear geometries. The broader beam shape at convergence
in FIG. 6B may be due to the ad hoc array-to-array beamforming
energy maximization for coherent transmit to all the array
members.
[0082] FIG. 7 illustrates selected steps of a Reciprocal
Convergence Algorithm process 700 performed between two ad hoc
arrays. It should be noted, however, that tethered and other to
arrays may be used in an analogous manner. To facilitate
discussion, we will refer to the arrays 105/110 discussed
above.
[0083] The process 700 begins at a flow point 701, where the
devices performing the process 700 (e.g., the nodes of the arrays
105/110) are powered and initialized. Initialization of an ad hoc
array may include frequency, phase, and time
synchronization/alignment of the individual is nodes.
[0084] In step 705, an initial sounding signal is transmitted by
the nodes of one of the arrays, for example, the nodes of the array
105. The initial sounding signal may be, for example, an
omnidirectional or substantially omnidirectional signal with a good
autocorrelation function, such as a pulse, burst, Gaussian
function, etc. An omnidirectional emission may be similar to an
emission from a dipole radiator, and need not have beamforming
weights. Note, however, (that the use of beamforming weights even
for the initial signal is not excluded, particularly when some a
priori information is available. The nodes of the array 105 may
transmit the initial sounding signal simultaneously, at t.sub.0.
Here, the sounding signal is in effect a signal combining the
transmissions of the multiple nodes of the array 105.
[0085] In step 710, the sounding signal emitted most recently by
the first array is received by the nodes of the other array, for
example, the array 110. The reception window is finite; it may be
made sufficiently long to receive all significant reflections
(multipath components) of the most recent sounding signal from the
array 105.
[0086] The signals received in the step 710 are time-reversed in
step 715 and current beamforming weights at the array 110 are thus
computed.
[0087] In decision block 720, a test is performed and a decision is
made whether the array(s) 110 and/or 105 need to continue the
focusing process or sufficient focusing has been achieved. The
decision may be made, for example, by reference to some performance
metric or change in the performance metric, as has already been
mentioned. The test may consider metrics or changes on both sides,
i.e., at the array 110 and/or at the array 105. In a specific
variant, the relative order of the step 715 and the decision block
720 is reversed.
[0088] If the test indicates that the process should continue, the
latest beamforming weights are used to emit a sounding signal by
the array 110, in step 725, and the process flow continues to step
730. Otherwise, the process flow proceeds to a flow point 799, to
be repeated as needed or otherwise (possibly repeated
continually).
[0089] In the step 730, the sounding signal emitted in the most
recent step 725 is received by the nodes of the array 105. The
reception window here is also finite; it may be made sufficiently
long to receive all significant reflections (multipath components)
of the most recent sounding signal from the array 110.
[0090] The signals received in the step 730 are time-reversed in
step 735 and current beamforming weights at the array 105 are thus
computed.
[0091] In decision block 740, a test is performed and a decision is
made whether the array(s) 105 and/or 110 need to continue the
focusing process or sufficient focusing has been achieved. The
decision may be made, for example, by reference to some performance
metric, or change in the performance metric, as has already been
mentioned. The test may consider metrics or changes on both sides,
i.e., at the array 105 and/or at the array 110. The test in the
decision block 740 may be the same or analogous to the test in the
decision block 720, mutatis mutandis. In a specific variant, the
relative order of the step 735 and the decision block 740 is
reversed.
[0092] If the test indicates that the process should continue, a
sounding signal with the latest beamforming weights is emitted by
the array 105, in step 745, and the process flow returns to the
step 710. Otherwise, the process flow proceeds to the flow point
799, to be repeated as needed (possibly repeated continually).
[0093] Thus, the steps 710-745 may be repeated sequentially, i.e.,
each time each of the reflected sounding signals is time-reversed
and re-emitted. After several iterations, as determined in the
decision blocks 720/740, each array 105/110 may be focused on the
other array, and the process may then terminate at the flow point
799. The two arrays 105/110 may then send each other (in either
direction or in both directions) data-carrying signals with
beamforming transmit/receive gains. It should be noted that the
sounding signals may also carry or be accompanied by some data; in
particular, one of the arrays may signal the other array that the
condition of its decision block (720 or 740, as the case may be)
has been met.
[0094] Beamforming weights for the array-to-array communications
may converge from an incoherent initialization to focused
beamforming through a series of iterations. Beamforming weights
bootstrap incoherent coordinated sounding towards array-to-array
beamforming weights in in a relatively simple manner. The Tx and Rx
beamforming weights are the TR pre-filters derived at each end of
the link as the portion of the iteration in a given direction is
completed.
[0095] In the case where any individual node within a first array
(e.g., the array 105) is unable to close a link to any of the nodes
in the other array (e.g., the array 110) located at the other end
of the communication link, all of the nodes or a sub-set of
multiple nodes of the first array may be used to initiate a
beamforming process. If information on the relative antenna layout
of the beamforming array is available (e.g., via GPS,
time-of-flight-based computation, or another method), the arrays
can use an initial guess on the proper transmit and receive
beamforming weights for a location of the other array or in the
general direction of the other array. The array layout may enable
the first array to calculate the weights to beamform to the desired
location or in the desired direction. The first array (e.g., the
array 105) may not have ideal beamforming array gain to the desired
location, but can still use directionality of the coherent energy
to reach the other array (e.g., the array 110). The first array may
use a relatively wider beamforming shape to span more area in a
search mode and can converge towards a sharper beam for focused
scanning. Multiple angles and directions may be swept until a
beamforming initiation has been acknowledged by the other array.
Sidelobes may also be used to reach the other array if the main
beam is not directed to the other array. The coherent energy gained
from the coordinated emission from the first array incident on the
other array may initiate a beamforming training process if there is
sufficient signal to trigger a beamforming request received by any
single node, a subset of nodes, or all of the nodes of the other
array. In this way, knowledge of the layout of the nodes of the
first array may be used to improve the initiating sounding signal
in the first iteration, and to sweep the area until the other array
responds. FIG. 8 illustrates selected aspects of sweeping a main
beam 810 and sidelobes 820A and 820B, in the counterclockwise
direction as is indicated by the pointed arrow.
[0096] The iterative TR (ITR, Reciprocal Convergence) process that
beamforms the signal at two arrays may be repeated for
transmissions through intermediate arrays, in a cascaded manner,
FIGS. 9A and 9B illustrate selected aspects of such technique.
Here, Array B calculates the beamforming weights to communicate
with Array A and also calculates the beamforming weights to
communicate with Array C, as shown in FIG. 9A. Once the weights are
known, Array A may communicate with Array C through Array B, as is
shown in FIG. 9B. Array B may, for example, communicate with Array
A and Array C in a TDD and/or FDD fashion; or simultaneously at the
same frequency if the nodes of Array B are divided so that one set
communicates with Array A and the other communicates with Array
C.
[0097] In the presence of scatterers in the environment affecting
signals between two arrays, the beamforming weights may be adjusted
in such a way that the signal power transmitted by the nodes of one
array (e.g., array 105) arrives coherently at the other array
(e.g., array 110), because is of the properties of Time Reversal
algorithms in NLoS channels. The iterative process may follow the
steps described above (e.g., the process 700) to determine the
beamforming weights at each of the arrays. To summarize selected
aspects:
[0098] 1. The array 105 nodes agree on a signal emission time and
time window transmission and capture time duration.
[0099] 2. The array 105 nodes initially send the sounding signal to
the array 110 in an omnidirectional way, or using some a priori
available information (such as the general direction of the array
110 and the relative layout of the nodes of the array 105); or
after adjusting the beamforming weights in subsequent
iterations.
[0100] 3. The array 110 nodes agree on a signal capture and
emission times, a time window for transmission, and a capture time
duration, time-reverse the received signals, and send them back to
the array 105.
[0101] 4. The array 105 nodes agree on a signal emission time and
time window transmission and capture time duration, receive the
signals from the array 110, time-reverse the signals, and send the
signals back to the array 110.
[0102] The iterative process may continue until, for example, at
least two successive iterations indicate convergence towards the
directions between the array 105 and the array 110. Here and
elsewhere in this document, convergence may be indicated, for
example, by mean square of differences between corresponding
beamforming or TR pre-filter coefficients in successive iterations
falling below a predetermined limit.
[0103] Scatterer nulling may be added to the steps of the processes
described above. An iterative process may then proceed as
follows:
[0104] 1. The array 105 nodes send sounding signals either
collectively or separately to identify the scatterers (static or
semi-static objects) by checking the received echoes. Scatterer
nulling is described, for example, in U.S. Provisional Patent
Application Ser. No. 62/196,885, entitled TIME REVERSAL IN WIRELESS
COMMUNICATIONS, filed on 24 Jul. 2015; and in U.S. patent
application Ser. No. 15/217,944, entitled WIRELESS SENSING WITH
TIME REVERSAL, filed 22 Jul. 2016.
[0105] 2. The array 105 sets its initial transmit and receive
weights to the TR pre-filter IS derived from the scatterer echoes
in order to null the scatterers.
[0106] 3. The array 105 nodes send a sounding signal in an
omni-directional way to the array 110 after adjusting the
beamforming weights to account for scatterer nulling (so the signal
is not strictly omni-directional).
[0107] 4. The array 110 nodes agree on a signal emission time, time
window transmission, and capture time duration; time reverse the
received signal and send it back to the array 105.
[0108] 5. The array 105 nodes agree on a signal emission time, time
window transmission, and capture time duration; receive the
sounding signal from the array 110, time reverse it, and send the
time-reversed signal back to the array 110.
[0109] The iterative process may continue until, for example, at
least two successive iterations indicate convergence towards the
directions between the array 105 and the array 110.
[0110] FIG. 10 illustrates selected steps of an example of a
Reciprocal Convergence Algorithm process 1000 with scatterer
nulling performed between two ad hoc arrays. FIG. 11 illustrates
selected steps of another example of a Reciprocal Convergence
Algorithm process 1100 with scatterer nulling performed between two
ad hoc arrays; the steps/blocks of the process 1100 may be
analogous to the similarly-numbered step/blocks of the process 700
described above, with an addition of a new step 1103 (clutter
characterization and nulling) and "11" substituted for "7" in
numbering of the individual steps blocks. It should be noted that
tethered and other arrays may be used in an analogous manner.
[0111] In embodiments, one of the arrays (e.g., the array 105) can
be responsible for performing the entire alignment. This array not
only focuses its sounding signal at each receive node of the other
array (e.g. the array 110), but adjusts the phase and any time
delays of the signals so that the receivers do not have to perform
any alignment process.
[0112] We turn now to the TR beam nulling algorithm when trying to
null clutter or undesired objects. As has been indicated in this
document and the related patent documents, TR algorithms may
suppress the eigenmodes coherent energy associated with the clutter
objects whose milling is desired. Let us assume that the nodes of
an array identify two objects, i and j, in the array's field of
view. Furthermore, let us assume that the total number of antennas
of all nodes is M. (This similarly applies to a single node with M
antennas rather than M distributed is nodes.) The pre-TR-filter
weights assigned to each of the M antennas encompass the
beam-focusing TR pre-filter vector targeting objects i and j, while
nulling object i signal in the direction of object j and vice
versa, or nulling object j altogether. Such antenna weights can be
derived when the correlation between the CIR.sub.i and CIR.sub.j,
which are the CIRs (channel impulse or other responses) between
Tx/Rx nodes and objects i and j, respectively, is low. Therefore,
these two objects are uncorrelated (or orthogonal to each other) or
slightly correlated, and hence they can be distinguished from each
other using TR and SVD (Singular Value Decomposition) methods, such
as Zero Forcing (ZF), Modified Zero-Forcing (MZF), and Dirty Paper
Coding (DPC).
[0113] If we denote by H(.omega.) the CIR matrix where each row is
the CIR between the nodes' M antennas and each of the uncorrelated
objects, then the matrix H.sup.H(.omega.)H(.omega.) should be
invertible. The TR pre-filter TRF(.omega.) that encompasses the
beam-focusing and nulling weights at each of the Tx/Rx antennas is
defined such that TRF(.omega.) is the product of a linear
transformation matrix A and a diagonal power matrix P restricted by
the total maximum transmit power, i.e., TRF(.omega.)=A P.
Zero-Forcing is when A is defined as A=H.sup.H (H.sup.HH).sup.-1;
Modified Zero-Forcing is when A is defined as A=H.sup.H
(H.sup.HH+(N.sub.0/P.sub.avg)I).sup.-1; and DPC is when A is
defined as A=H.sup.HR.sup.-1 and R is an upper triangular
matrix.
[0114] Hence the signal focused on objects i and j can be written
as Y=G(.omega.)X+n, where G(.omega.)=H(.omega.) TRF(.omega.), n is
the noise, and X is the sounding signal vector. Then, the nulling
condition is written as y.sub.i.apprxeq.g.sub.iix.sub.i+n.sub.i and
y.sub.j.apprxeq.g.sub.jj x.sub.j+n.sub.j, where the cross-objects
interference terms corresponding to g.sub.ij are negligible due to
the application of one of the three beam nulling algorithms
mentioned above.
[0115] In some embodiments, the array nodes may be located at
multiple wavelengths apart causing high side lobes in the converged
radiation pattern after performing ITR. In this case, additional
complex weights may be applied to the Tx and Rx antennas (in
combination with the above-described nulling mechanism) to reduce
the side lobe and increase the energy transmitted toward the Rx
array. In this case, the array-to-array algorithm may be combined
with additional complex weights at the Tx and Rx using various
array synthesis algorithms such as steering vectors, windowing
functions, and genetic algorithms. Furthermore, using directional
antennas at the Tx and Rx sides can reduce these sidelobes when the
Tx and Rx broad orientation is known.
[0116] The methods described in this document may also be applied
to multi-user TR communication when, the Access Point (AP) is
communicating with multiple users simultaneously. In this case, the
TR pre-filters assigned the each of the AP antenna may be selected
such that each users' signal is beam-focused to that specific user
while nulling it in the direction of other users to reduce the
crosstalk between users. Such transmit beamforming and nulling,
also referred to as space division multiple access, enable signal
separation by directing one or multiple beams simultaneously
towards users at different spatial locations without creating
crosstalk between the signals to different users. These algorithms
can be performed in the time and frequency domain signal
processing.
[0117] The features described throughout this document may be
present individually, or in any combination or permutation, except
where the presence or absence of specific elements/limitations is
inherently required, explicitly indicated, or otherwise made clear
from the context.
[0118] Although the process steps and decisions (if decision blocks
are present) may be described serially in this document, certain
steps and/or decisions may be performed by same and/or separate
elements in conjunction or in parallel, asynchronously or
synchronously, in a pipelined manner, or otherwise. There is no
particular requirement that the steps and decisions be performed in
the same order in which this description lists them or the Figures
show them, except where a specific order is inherently required,
explicitly indicated, or is otherwise made clear from the context.
Furthermore, not every illustrated step and decision block may be
required in every embodiment in accordance with the concepts
described in this document, while some steps and decision blocks
that have not been specifically illustrated may he desirable or
necessary in some embodiments in accordance with the concepts. It
should be noted, however, that specific
embodiments/variants/examples use the particular order(s) in which
the steps and decisions (if applicable) are shown and/or
described.
[0119] The instructions (machine executable code) corresponding to
the method steps of the embodiments, variants, and examples
disclosed in this document may be embodied directly in hardware, in
software, in firmware, or in combinations thereof. A software
module may be stored in volatile memory, flash memory, Read Only
Memory (ROM), Electrically Programmable ROM (EPROM), Electrically
Erasable Programmable ROM (EEPROM), hard disk, a CD-ROM, a DVD-ROM,
or other form of non-transitory storage medium known in the art.
Exemplary storage medium or media may be coupled to one or more
processors so that the one or more processors can read information
from, and write information to, the storage medium or media. In an
alternative, the storage medium or media may be integral to one or
more processors.
[0120] This document describes in detail the inventive apparatus,
methods, and articles of manufacture for array-to-array beamforming
with time reversal techniques, including using ad hoc and tethered
arrays. This was done for illustration purposes and, therefore, the
foregoing description is not necessarily intended to limit the
spirit and scope of the invention(s) described. Neither the
specific embodiments of the invention(s) as a whole, nor those of
its (or their, as the case may be) features necessarily limit the
general principles underlying the invention(s). The specific
features described herein may be used in some embodiments, but not
in others, without departure from the spirit and scope of the
invention(s) as set forth herein. Various physical arrangements of
components and various step sequences also fall within the intended
scope of the invention(s). Many additional modifications are
intended in the foregoing disclosure, and it will be appreciated by
those of ordinary skill in the pertinent art that in some instances
some features will be employed in the absence of a corresponding
use of other features. The embodiments described above are
illustrative and not necessarily limiting, although they or their
selected features may be limiting for some claims. The illustrative
examples therefore do not necessarily define the metes and bounds
of the invention(s) and the legal protection afforded the
invention(s).
* * * * *