Patent Application
20230170975
The present disclosure relates generally to network nodes in wireless communication networks and more specifically to receiver subarray cooperation in downstream receiver processing circuitry.
Technology advances, capacity demands, and higher frequency operation have led to a rise in the number of base station receive antennas and radio-near processing. Receiver antenna arrays are partitioned into subarrays, each with their own independent and dedicated processing circuitry.
According to some embodiments, a method of operating a network node a communication network is provided. The method includes receiving, by processing circuitry of the network node, a first upstream-processed signal associated with an original signal. The method further includes receiving, by the processing circuitry in the network node, a second upstream-processed signal, associated with the original signal. The method further includes determining, by the processing circuitry in the network node, a covariance associated with the first upstream-processed signal and the second upstream-processed signal. The method further includes obtaining, by the processing circuitry in the network node, a processed received signal based on the first upstream-processed signal, the second upstream-processed signal, and the covariance.
According to other embodiments, a network node operating in a communication network is provided. The network node includes processing circuitry and memory coupled to the processing circuitry. The memory has instructions stored therein that are executable by the processing circuitry to cause the network node to perform operations. The operations include receiving, by processing circuitry of the network node, a first upstream-processed signal associated with an original signal. The operations further include receiving, by the processing circuitry in the network node, a second upstream-processed signal, associated with the original signal. The operations further include determining, by the processing circuitry in the network node, a covariance associated with the first upstream-processed signal and the second upstream-processed signal. The operations further include obtaining, by the processing circuitry in the network node, a processed received signal based on the first upstream-processed signal, the second upstream-processed signal, and the covariance.
According to other embodiments, a network node operating in a communication network and adapted to perform operations is provided. The operations include receiving, by processing circuitry of the network node, a first upstream-processed signal associated with an original signal. The operations further include receiving, by the processing circuitry in the network node, a second upstream-processed signal, associated with the original signal. The operations further include determining, by the processing circuitry in the network node, a covariance associated with the first upstream-processed signal and the second upstream-processed. The operations further include obtaining, by the processing circuitry in the network node, a processed received signal based on the first upstream-processed signal, the second upstream-processed signal, and covariance.
According to other embodiments, a computer program is provided. The computer program includes program code to be executed by processing circuitry of a network node operating in a communication network, whereby execution of the program code causes the network node to perform operations. The operations include receiving, by processing circuitry of the network node, a first upstream-processed signal associated with an original signal. The operations further include receiving, by the processing circuitry in the network node, a second upstream-processed signal, associated with the original signal. The operations further include determining, by the processing circuitry in the network node, a covariance associated with the first upstream-processed signal and the second upstream-processed signal. The operations further include obtaining, by the processing circuitry in the network node, a processed received signal based on the first upstream-processed signal, the second upstream-processed signal, and the covariance.
According to other embodiments, a computer program product is provided. The computer program product includes a non-transitory storage medium including program code to be executed by processing circuitry of a network node operating in a communication network, whereby execution of the program code causes the network node to perform operations. The operations include receiving, by processing circuitry of the network node, a first upstream-processed signal associated with an original signal. The operations further include receiving, by the processing circuitry in the network node, a second upstream-processed signal, associated with the original signal. The operations further include determining, by the processing circuitry in the network node, a covariance associated with the first upstream-processed signal and the second upstream-processed signal. The operations further include obtaining, by the processing circuitry in the network node, a processed received signal based on the first upstream-processed signal, the second upstream-processed signal, and the covariance.
In various embodiments described herein, processing architecture and algorithms for a receiver with multiple subarrays is described with parts of the processing occurring at each subarray, and another part occurring downstream fed by the subarrays. Combining the up-stream processed signals by downstream processing circuitry results in overall receiver performance that approaches that of a full array with unrestricted processing.
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this application, illustrate certain non-limiting embodiments of inventive concepts. In the drawings:
Inventive concepts will now be described more fully hereinafter with reference to the accompanying drawings, in which examples of embodiments of inventive concepts are shown. Inventive concepts may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of present inventive concepts to those skilled in the art. It should also be noted that these embodiments are not mutually exclusive. Components from one embodiment may be tacitly assumed to be present/used in another embodiment.
The following description presents various embodiments of the disclosed subject matter. These embodiments are presented as teaching examples and are not to be construed as limiting the scope of the disclosed subject matter. For example, certain details of the described embodiments may be modified, omitted, or expanded upon without departing from the scope of the described subject matter.
Constrained processing per subarray can result in performance loss compared to unconstrained processing with the full array, which may prevent the array investment being fully exploited.
Various embodiments described herein propose a process for subarray cooperation, so that overall receiver performance approaches that of a full array with unrestricted processing. The subarrays produce outcomes including appropriate information, and those outcomes are then further processed together downstream.
Some embodiments do not violate the constrained architecture, in the sense that some embodiments do not require any direct communication among subarrays or any feedback to the subarrays. The combination of subarray output and downstream processing results in performance close to that of the unrestricted full array receiver.
As discussed herein, operations of communication device UE may be performed by processing circuitry 703 and/or transceiver circuitry 701. For example, processing circuitry 703 may control transceiver circuitry 701 to transmit communications through transceiver circuitry 701 over a radio interface to a radio access network node (also referred to as a base station) and/or to receive communications through transceiver circuitry 701 from a RAN node over a radio interface. Moreover, modules may be stored in memory circuitry 705, and these modules may provide instructions so that when instructions of a module are executed by processing circuitry 703, processing circuitry 703 performs respective operations (e.g., operations discussed below with respect to Example Embodiments relating to wireless communication devices).
As discussed herein, operations of the RAN node may be performed by processing circuitry 803, network interface 807, and/or transceiver 801. For example, processing circuitry 803 may control transceiver 801 to transmit downlink communications through transceiver 801 over a radio interface to one or more mobile terminals UEs and/or to receive uplink communications through transceiver 801 from one or more mobile terminals UEs over a radio interface. Similarly, processing circuitry 803 may control network interface 807 to transmit communications through network interface 807 to one or more other network nodes and/or to receive communications through network interface from one or more other network nodes. Moreover, modules may be stored in memory 805, and these modules may provide instructions so that when instructions of a module are executed by processing circuitry 803, processing circuitry 803 performs respective operations (e.g., operations discussed below with respect to Example Embodiments relating to RAN nodes).
According to some other embodiments, a network node may be implemented as a core network CN node without a wireless transceiver. In such embodiments, transmission to a wireless communication device UE may be initiated by the network node so that transmission to the wireless communication device UE is provided through a network node including a transceiver (e.g., through a base station or RAN node). According to embodiments where the network node is a RAN node including a transceiver, initiating transmission may include transmitting through the transceiver.
As discussed herein, operations of the CN node may be performed by processing circuitry 903 and/or network interface circuitry 907. For example, processing circuitry 903 may control network interface circuitry 907 to transmit communications through network interface circuitry 907 to one or more other network nodes and/or to receive communications through network interface circuitry from one or more other network nodes. Moreover, modules may be stored in memory 905, and these modules may provide instructions so that when instructions of a module are executed by processing circuitry 903, processing circuitry 903 performs respective operations (e.g., operations discussed below with respect to Example Embodiments relating to core network nodes).
The RAN node may include network interface circuitry 1007 (also referred to as a network interface) configured to provide communications with other nodes (e.g., with other base stations) of the RAN and/or core network CN. The network node may also include downstream processing circuitry 1003 (also referred to as a processor, e.g., corresponding to processing circuitry 4170) coupled to the transceiver circuitry, and memory circuitry 1005 (also referred to as memory) coupled to the processing circuitry. The memory circuitry 1005 may include computer readable program code that when executed by the downstream processing circuitry 1003 causes the downstream processing circuitry 1003 to perform operations according to embodiments disclosed herein. For example, downstream processing circuitry may combine outputs from receivers 1010, 1020. According to other embodiments, downstream processing circuitry 1003 may be defined to include memory so that a separate memory circuitry is not required. Modules may be stored in memory 1005, and these modules may provide instructions so that when instructions of a module are executed by processing circuitry 1003, 1013, 1023, processing circuitry 1003, 1013, 1023, performs respective operations. In additional or alternative embodiments, another network node (e.g., a UE or a CN node) may have multiple receivers similar to receiver 1010, 1020 that are each associated with an antenna subarray and a downstream processor for performing similar operations as RAN node 1000.
In some embodiments, a MIMO system is provided in which the transmitter sends M streams and the receiver has N antennas. In some examples, linear receiver techniques are used which assume that N M. Furthermore, since the antennas are split into subsets, the size of each subset M. The following disclosure describes embodiments in which the array is split into two subsets of equal size (N/2) and that N/2≥M. However, the subarrays may be of inequal size and the array may be divided into any number of subarrays.
In some embodiments, a system equation for describing a vector, r, of received symbols is given by
r=Hs+w
where s is a vector of M symbols from a modulation constellation (e.g., QAM) and H is a channel matrix of size N×M. In some examples, the channels have an average energy of one. The total noise, w, can include an interference signal plus a white noise signal,
w=H
i
s
i
+v
The total noise covariance of size N×N can be written as
R
w
=E
i
H
i
H
i
H
+E
v
I
where Ei and Ev are the interference and white noise energy, respectively. Channel parameters can be estimated from the received signal, aided by pilot symbols embedded in transmitted signals.
A maximum likelihood (ML) demodulator can search for a candidate vectors for s that minimizes the metric:
m(ŝ)=(r−Hŝ)HRw−1(r−Hŝ)
This may require a full search, which can become infeasible for a large constellation or a large number of streams.
In some embodiments, in order to facilitate downstream processing, a receive signal domain representation of the receive signal can be transformed to a transmit signal domain representation, along the lines of a sphere decoder (SD). A SD can search among the candidate transmit signals {tilde over (s)} within a progressively shrinking sphere centered at an initial guess {tilde over (s)}. Overall, the SD can be computationally efficient while remaining equivalent to the ML demodulator. Various shortcuts can further speed up the SD, with corresponding performance penalties.
A receive signal domain representation can be transformed to a transmit domain representation using a left inverse F of H, of size M×N, given by:
F=(HHRw−1H)−1HHRw−1
The received signal, r, can be transformed into
{tilde over (s)}=Fr
where {tilde over (s)} is the ML estimate of s, if s were a Gaussian vector. Since s belongs to a discrete set, {tilde over (s)} can be an initial guess for s. The candidates of {tilde over (s)} can be searched in the neighborhood of ŝ for the final answer. {tilde over (s)} can be written as
{tilde over (s)}=s+x
where x=Fw with covariance:
R
x
=FR
W
F
H
of size M×M, which can be written as
R
x=(HHRw−1H)−1
In some embodiments, given {tilde over (s)}, the SD finds the best candidate using a modified metric given by:
n(ŝ)=({tilde over (s)}−ŝ)HRx−1({tilde over (s)}−ŝ)
The SD can exploit the Hermitian (complex symmetric) property of the covariance matrix. It can perform a Cholesky decomposition, which computes a triangular “root” matrix K such that:
K
H
K=R
x
−1
The triangular property can enable the solution of linear systems by back substitution. It is also an ingredient in facilitating the restriction of the search to a sphere. We refer to finding the best candidate using n(ŝ) under the sphere restriction as the sphere restricted search.
In some embodiments, in addition to the modulation symbols, a demapper can use the metrics of the SD to produce hard and soft values for the modem bits that map into the modulation symbols. The hard or soft bit values can be fed to the channel decoder.
In some embodiments, channel estimates can be computed by using pilot symbols. Note that matrix inverses appear on several occasions. In practice, except for very small dimensions, they can be approximated using well known numerical methods. In some embodiments, a covariance matrix can be estimated from received data. In additional or alternative embodiments, a square root of the covariance can be computed directly.
Various embodiments herein are illustrated by the structure in
In additional or alternative embodiments, all the N antennas see independent channels. The full array receiver will then capture all the channel diversity, whereas the individual subarray receivers will only capture their limited share. However, it is possible to recover the full diversity by appropriately combining subarray outputs using downstream processing circuitry.
In additional or alternative embodiments, interference suppression is a more difficult problem with subarrays. The interference can be treated as colored noise with covariance Rx and can be suppressed with a linear operation. The full array receiver has knowledge of the complete Rx (of its estimate in practice). Each subarray, however, only has incomplete knowledge of the noise color, reflected in a submatrix of covariance Rx as described below, so some of the color information is lost on it. Downstream processing captures the partial color information from the subarrays and makes up some of the loss of color information with a less granular version. An averaging effect can smooth over some of the finer color structure and can explain the residual loss.
In some embodiments, the receive antennas can be split into two subsets, A and B, of equal size N/2, which represent two subarrays. Without loss of generality, the subsets can be assumed to be contiguous in the vector notation. The received signal and the channel can be split accordingly
The individual subarray receivers observe their own partial received signals
r
A
=H
A
s+w
A
r
B
=H
B
s+w
B
The total covariance can be rewritten as
where RwA and RwB are the separate covariances of wA and wB respectively, and QwAB is the cross covariance of wA and wB.
Extrapolating from the original full array receiver, the subarray receiver A can be expected to estimate its own covariance RwA, and similarly for B. So, when A and B join forces downstream, the knowledge of RwA and RwB will be available and can be exploited. The real problem is that neither A nor B can estimate QwAB, so there is a loss of information compared to the full array receiver.
Subarray receivers for A and B following the same approach as before, with a transform to the signal domain. That is, A observes rA and produces
{tilde over (s)}
A=(HAHRwA−1HA)−1HAHRwA−1rA=FArA=S+FAwA=s+xA
where the previous notation has been modified to refer to subarray A. The covariance of xA can be written as
R
xA
=F
A
R
wA
F
A
H
which can also be written as
R
xA=(HAH,RwA−1AHA)−1
The structure of the subarray processing is shown in
Similarly, B observes rB and goes through the same steps to produce
{tilde over (s)}
B
=s+F
B
w
B
=s+x
B
and also the matrices FB and RxB (just replacing subscript A with B). The structure of the subarray processing is shown in
A (or B) can be treated as a complete standalone receiver if the last step of the SD, namely the sphere restricted search, is applied to {tilde over (s)}A to find the best candidate symbol, as well as applying the demapper to find hard or soft values. However, the performance may be limited by the reduced size of the subarray of A (or B).
A and B output {tilde over (s)}A and {tilde over (s)}B for downstream processing (along with any other quantity needed such as FA, FB, etc.).
In some embodiments, {tilde over (s)}A and {tilde over (s)}B are processed together to improve the overall performance and approach the performance of the full array. In some examples, {tilde over (s)}A and {tilde over (s)}B are combined into a new improved estimate {tilde over (s)}comb which can be input to the sphere restricted search of the SD.
The noise terms xA and xB are correlated, with covariance RxAB that can be estimated based on the information available at the downstream processing phase. The estimated covariance can be used in a similar way as the unconstrained processing or the subarray processing.
In order to combine {tilde over (s)}A and {tilde over (s)}B into a single vector of size M, they can be treated as a single received signal with colored noise. To do so, {tilde over (s)}A and {tilde over (s)}B can be stacked to get a new channel model,
where the noise
has covariance RxAB. Assuming that
then RxAB can be related to Rw, as
R
xAB
=GR
w
G
H
The special double identity channel
In order to combine the elements of {tilde over (s)}AB into the estimate {tilde over (s)}comb another transform can be performed using J as the channel model and RxAB as the noise covariance. That is, a transformation matrix is formed
F
AB=(JHRxAB−1J)−1JHRxAB−1
and {tilde over (s)}comb is obtained as
{tilde over (s)}
comb
=F
AB
{tilde over (s)}
AB
=s+y
AB
of dimension M. The noise term
has covariance
R
yAB
=F
AB
R
xAB
F
AB
H
The signal {tilde over (s)}comb and the covariance RyAB are fed to the sphere restricted search of the SD. The process is shown in
Recall that RxAB=GRwGH, and that QwAB, the cross covariance term of Rw, cannot be estimated by either subarray processing phases, so it is not available in the downstream processing phase.
With some further manipulations, RxAB can be written as
The matrices RxA and RxB were described earlier and can be estimated in their respective subarray phases. The cross covariance QxAB can be written as
Q
xAB
=F
A
Q
wAB
F
B
H
again highlighting the missing submatrix QwAB.
In order to estimate RxAB, the previously obtained estimates RxA and RxB can be used. Then QxAB can be estimated directly from the signals {tilde over (s)}A and {tilde over (s)}B.
Alternatively, the whole of RxAB can be estimated directly from the stacked signal {tilde over (s)}AB.
Using the estimate of RxAB, RyAB can be determined, which is needed in the SD.
In some embodiments, estimating RxAB, which is needed in the downstream processing, does not violate the limitation of the subarray constraint. RxAB can be thought of as an averaged version of Rw However, QxAB has a size of M×M, which is smaller than QwAB of size N/2×N/2, thus some of the fine grained information about the noise color may be irreversibly lost in the averaging due to FA and FB.
In some embodiments, J is assumed to be a double identity matrix. This may only be true if channel parameters are estimated perfectly. In some examples, receivers operate in a range of channel quality where channel parameters are estimated quite well. In some embodiments, J is modeled as a double identity matrix. In which case it is not estimated in the parameter estimation block in
Operations of a network node will now be discussed with reference to the flow charts of
At block 1110, first upstream processing circuitry 1013 receives, via antenna subarray 1030a, a first version of an original signal. In some embodiments, the first version of the original signal is a first receive signal domain representation of the original signal.
At block 1120, first upstream processing circuitry 1013 obtains a first upstream-processed signal based on the first version of the original signal. In some embodiments, the first upstream-processed signal is a transmit signal domain version/representation of a first receive signal domain version of the original signal received at a first receiver 1010 in the RAN node 1000. In additional or alternative embodiments, obtaining the first upstream-processed signal includes transforming the first receive signal domain representation of the original signal to a first transmit signal domain representation of the original signal.
At block 1130, second upstream processing circuitry 1023 receives, via antenna subarray 1030b, a second version of the original signal. In some embodiments, the second version of the original signal is a second receive signal domain representation of the original signal.
At block 1140, second upstream processing circuitry 1023 obtains a second upstream-processed signal based on the second version of the original signal. In some embodiments, the second upstream-processed signal is a transmit signal domain version/representation of a second receive signal domain version of the original signal received at a second receiver 1020 in the RAN node 1000. In additional or alternative embodiments, obtaining the second upstream-processed signal includes transforming the second receive signal domain representation of the original signal to a second transmit signal domain representation of the original signal.
At block 1150, downstream processing circuitry 1003 receives the first upstream-processed signal. In some embodiments, receiving the first upstream-processed signal includes receiving the first upstream-processed signal from the first upstream processing circuitry 1013 in the RAN node 1000. In additional or alternative embodiments, the first upstream-processed signal can be received from a remote receiver independent from the network node.
At block 1160, downstream processing circuitry 1003 receives the second upstream-processed signal. In some embodiments, receiving the second upstream-processed signal includes receiving the second upstream-processed signal from the second upstream processing circuitry 1013 in the RAN node 1000. In additional or alternative embodiments, the second upstream-processed signal can be received from a remote receiver independent from the network node.
At block 1170, downstream processing circuitry 1003 determines a common characteristic (in this example, a covariance) associated with the first upstream-processed signal and the second upstream-processed signal. The common characteristic is determined based on both the first upstream-processed signal and the second upstream-processed signal being associated with the original signal. In some embodiments, the common characteristic includes a correlated covariance.
At block 1180, downstream processing circuitry 1003 obtains a processed received signal based on the first upstream-processed signal, the second upstream-processed signal, and the common characteristic. In some embodiments, obtaining the processed received signal includes performing a maximum likelihood process on the first upstream-processed signal and the second upstream-processed signal using the common characteristic. In additional or alternative embodiments, performing the maximum likelihood process includes sphere decoding the first upstream-processed signal and the second upstream-processed signal using the common characteristic.
In some embodiments, the communication network is a new radio, NR, network.
Various operations of
Many variations and modifications can be made to the embodiments without substantially departing from the principles of the present inventive concepts. All such variations and modifications are intended to be included herein within the scope of present inventive concepts. Accordingly, the above disclosed subject matter is to be considered illustrative, and not restrictive, and the examples of embodiments are intended to cover all such modifications, enhancements, and other embodiments, which fall within the spirit and scope of present inventive concepts. Thus, to the maximum extent allowed by law, the scope of present inventive concepts are to be determined by the broadest permissible interpretation of the present disclosure including the examples of embodiments and their equivalents, and shall not be restricted or limited by the foregoing detailed description.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/IB2020/054220 | 5/4/2020 | WO |
