DETECTION IN A WIRELESS COMMUNICATIONS NETWORK

TECHNICAL FIELD

Embodiments herein relate to a network node and methods therein. In some aspects, they relate to detecting bandlimited interference in a single-carrier Uplink (UL) from a User Equipment (UE) in a wireless communications network.

BACKGROUND

In a typical wireless communication network, wireless devices, also known as wireless communication devices, mobile stations, stations (STA) and/or User Equipments (UE)s, communicate via a Wide Area Network or a Local Area Network such as a Wi-Fi network or a cellular network comprising a Radio Access Network (RAN) part and a Core Network (CN) part. The RAN covers a geographical area which is divided into service areas or cell areas, which may also be referred to as a beam or a beam group, with each service area or cell area being served by a radio network node such as a radio access node e.g., a Wi-Fi access point or a radio base station (RBS), which in some networks may also be denoted, for example, a NodeB, eNodeB (eNB), or gNB as denoted in Fifth Generation (5G) telecommunications. A service area or cell area is a geographical area where radio coverage is provided by the radio network node. The radio network node communicates over an air interface operating on radio frequencies with the wireless device within range of the radio network node.

3GPP is the standardization body for specify the standards for the cellular system evolution, e.g., including 3G, 4G, 5G and the future evolutions. Specifications for the Evolved Packet System (EPS), also called a Fourth Generation (4G) network, have been completed within the 3rd Generation Partnership Project (3GPP). As a continued network evolution, the new releases of 3GPP specifies a 5G network also referred to as 5G New Radio (NR).

Frequency bands for 5G NR are being separated into two different frequency ranges, Frequency Range 1 (FR1) and Frequency Range 2 (FR2). FR1 comprises sub-6 GHz frequency bands. Some of these bands are bands traditionally used by legacy standards but have been extended to cover potential new spectrum offerings from 410 MHz to 7125 MHz. FR2 comprises frequency bands from 24.25 GHz to 52.6 GHz. Bands in this millimeter wave range, referred to as Millimeter wave (mmWave), have shorter range but higher available bandwidth than bands in the FR1.

Multi-antenna techniques may significantly increase the data rates and reliability of a wireless communication system. For a wireless connection between a single user, such as UE, and a base station, the performance is in particular improved if both the transmitter and the receiver are equipped with multiple antennas, which results in a Multiple-Input Multiple-Output (MIMO) communication channel. This may be referred to as Single-User (SU)-MIMO. In the scenario where MIMO techniques is used for the wireless connection between multiple users and the base station, MIMO enables the users to communicate with the base station simultaneously using the same time-frequency resources by spatially separating the users, which increases further the cell capacity. This may be referred to as Multi-User (MU)-MIMO. Note that MU-MIMO may benefit when each UE only has one antenna. Such systems and/or related techniques are commonly referred to as MIMO.

Communication between a sender and receiver takes place over an unsafe, unreliable and noisy channel thus distorting the transmitted signal.

Distortion in communications means the alteration of the waveform of an information-bearing signal, such as an audio signal representing sound or a video signal representing images, in an electronic device or communication channel.

UL interference due to non-linear channel distortion, e.g., any signal degradation (excluding noise) on a received UL signal, remains a difficult problem to detect and can only to a limited extent be handled by linear Minimum Mean Square Error (MMSE) methods. UL interference detection methods based on non-linear models, although capable of representing the distortion, may however in general be costly in terms of computational resources and processing power.

As hinted above, mmWave communication systems uses higher carrier frequencies, bandwidths such as band of spectrum between 30 GHz and 300 GHz, and lower power levels, and it is an important part of the 5G system. mmWave communication systems will increase the demand for highly non-linear Digital Signal Processing (DSP) methods to characterize and represent unwanted signals. These methods find application in filtering and detection algorithms which are needed to maintain high Signal-to-Noise (SNR) ratio and thus reliable communication. However, these methods are often associated with a high increase in computational resources (i.e. computational complexity) and power consumption. SUMMARY

An object of embodiments herein is to provide an improved way of detecting UL interference in a wireless communications network.

Implicit methods form an interesting subset of non-linear signal processing methods as they rely on relations between transform spaces, e.g. the time-domain and frequency- domain, to implement a strong non-linearity in a selected target space implicitly. Kernel- based signal processing is an attractive family of implicit methods due to its ability to model a vast space of signals. However, also Kernel-based signal processing methods for bandlimited interference detection may be improved. For example, noise and detection function imperfections, e.g. due to kernel design trade-off, add to the error of these methods. For interference detection methods it is important to eliminate these source of errors and stabilize the output from the methods. Due to this a variance of the detection output may in some cases be unnecessary high.

Embodiments herein disclose kernel-based methods which use a weighted average of outputs from a kernel similarity metric. In other words, embodiments herein may use temporal information to detect UL interference.

According to an aspect of embodiments herein, the object is achieved by a method performed by a network node for detecting bandlimited interference in a single-carrier Uplink, UL, channel from a User Equipment, UE, in a wireless communications network.

The network node receives a data sequence of the single-carrier UL.

The network node obtains a first sequence. The first sequence comprises a sequence of an integer of samples comprising ideal UL data of a given configuration associated to characteristics of the single-carrier UL.

The network node further obtains a second sequence, which second sequence comprises a sequence of the integer of samples from the received data sequence.

The network node calculates two or more squared Maximum Mean Discrepancy, MMD, distance-measure values by comparing the first sequence with the second sequence over two or more windows of samples according to a defined kernel-based statistical two-sample test, wherein a length of a respective window of the two or more windows of samples is smaller than the integer of samples.

The network node calculates a further distance-measure value by weighting the calculated two or more squared MMD distance-measure values. The network node decides whether or not bandlimited interference is present in the single-carrier UL, based on the calculated further distance-measure value.

According to another aspect of embodiments herein, the object is achieved by a network node configured to detect bandlimited interference in a single-carrier Uplink, UL, from a User Equipment, UE, in a wireless communications network. The network node is further configured to:

- receive a data sequence of the single-carrier UL,

- obtain a first sequence, which first sequence is adapted to comprise a sequence of an integer of samples comprising ideal UL data of a given configuration associated to characteristics of the single-carrier UL,

- obtain a second sequence, which second sequence is adapted to comprise a sequence of the integer of samples from the received data sequence,

- calculate two or more squared Maximum Mean Discrepancy, MMD, distance- measure values by comparing the first sequence with the second sequence over two or more windows of samples according to a defined kernel-based statistical two-sample test, wherein a length of a respective window of the two or more windows of samples is smaller than the integer of samples,

- calculate a further distance-measure value by weighting the calculated two or more squared MMD distance-measure values, and

- decide whether or not bandlimited interference is present in the single-carrier UL, based on the calculated further distance-measure value.

Since the network node calculates the further distance-measure value by weighting the calculated two or more squared MMD distance-measure values and then decides whether or not bandlimited interference is present in the single-carrier UL based on the further distance-measure value interference detection is more robust against noise and design trade-offs of the kernel. Thus, embodiments herein allow for reliably detecting bandlimited UL interference. For example, a variance of the further distance-measure value may be reduced compared to a variance of the individual MMD distance-measure values due to weighting of the calculated two or more squared MMD distance-measure values.

Some advantages provided by embodiments herein e.g. comprises a minimized computational complexity. There is no need to consider the stability of the detection output as this is automatically considered due to weighting of the input values to the method, e.g., the two or more squared MMD distance-measure values.

Another advantage is a simplified design process. The inputs to the method are easy to deduce from a given frequency band allocation and an input single-carrier UL data stream which makes configuration of embodiments herein simple. For example, parameters used by the kernel may be deduced easily from the band allocation and some sample data.

BRIEF DESCRIPTION OF THE DRAWINGS

Examples of embodiments herein are described in more detail with reference to attached drawings in which:

Figure 1a is a schematic block diagram illustrating embodiments of a wireless communications network.

Figure 1b is a flowchart depicting a reference method in a network node.

Figure 2a is a flowchart depicting a method in a network node according to embodiments herein.

Figure 2b is a flowchart depicting a method in a network node according to embodiments herein.

Figure 2c is a further flowchart depicting a method in a network node according to embodiments herein.

Figure 2d is a yet further flowchart depicting a method in a network node according to embodiments herein.

Figure 3a-b are schematic block diagrams illustrating embodiments of a network node.

Figure 4 schematically illustrates a telecommunication network connected via an intermediate network to a host computer.

Figure 5 is a generalized block diagram of a host computer communicating via a base station with a user equipment over a partially wireless connection,

Figures 6-9 are flowcharts illustrating methods implemented in a communication system including a host computer, a base station and a user equipment.

DETAILED DESCRIPTION Figure 1a is a schematic overview depicting a wireless communications network 100 wherein embodiments herein may be implemented. The wireless communications network 100 comprises one or more RANs and one or more CNs. The wireless communications network 100 may use a number of different technologies, such as mmWave communication networks, Wi-Fi, Long Term Evolution (LTE), LTE-Advanced, 5G, NR, Wideband Code Division Multiple Access (WCDMA), Global System for Mobile communications/enhanced Data rate for GSM Evolution (GSM/EDGE), Worldwide Interoperability for Microwave Access (WiMax), or Ultra Mobile Broadband (UMB), just to mention a few possible implementations. Embodiments herein relate to recent technology trends that are of particular interest in a 5G context, however, embodiments are also applicable in further development of the existing wireless communication systems such as e.g. WCDMA and LTE.

A number of network nodes operate in the wireless communications network 100 such as e.g., a network node 110. The network node 110 provides radio coverage in one or more cells which may also be referred to as a service area, a beam or a beam group of beams, such as e.g. a cell 11.

The network node 110 may be any of a NG-RAN node, a transmission and reception point e.g. a base station, a radio access network node such as a Wireless Local Area Network (WLAN) access point or an Access Point Station (AP STA), an access controller, a base station, e.g. a radio base station such as a NodeB, an evolved Node B (eNB, eNode B), a gNB, an NG-RAN node, a base transceiver station, a radio remote unit, an Access Point Base Station, a base station router, a transmission arrangement of a radio base station, a stand-alone access point or any other network unit capable of communicating with UEs, such as e.g. a UE 120, within the service area served by the network node 110 depending e.g. on the first radio access technology and terminology used. The network node 110 may communicate with UEs such as the UE 120, in DL transmissions to the UEs and UL transmissions from the UEs.

A number of wireless devices operate in the wireless communication network 100, such as e.g. the UE 120. The UE 120 may also referred to as a device, an loT device, a mobile station, a non-access point (non-AP) STA, a STA, a user equipment and/or a wireless terminal. The UE 120 may communicate via one or more Access Networks (AN), e.g. RAN, to one or more core networks (CN). It should be understood by the skilled in the art that “wireless device” is a non-limiting term which means any terminal, wireless communication terminal, user equipment, Machine Type Communication (MTC) device, Device to Device (D2D) terminal, or node e.g., smart phone, laptop, mobile phone, sensor, relay, mobile tablets or even a small base station communicating within a cell.

The UE 120 may be served by the network node 110, e.g., when being located in cell 11.

Methods herein may be performed by the network node 110. As an alternative, a Distributed Node (DN) and functionality, e.g. comprised in a cloud 135 as shown in Figure 1 , may be used for performing or partly performing the methods herein.

According to examples herein, embodiments herein are related to Kernelized interference detection in a single band-limited uplink.

For example, a method is provided that is linear in appropriately chosen non-linear features (i.e. kernel evaluations K(x, a) where a is a sample point) and show that non- linear interference may be separated from a single-carrier UL channel from a UE.

According to embodiments herein, compact and translation-invariant kernels are advantageously used herein as they have a useful frequency domain interpretation.

For use in communication, embodiments herein may use kernels with compactly supported and bandlimited spectrum.

Example embodiments herein allow for detecting bandlimited UL interference without explicitly transforming the input sample sequence of a single-carrier UL to frequency domain, filtering out the target signal characteristics and correlating it to a target template. Instead, embodiments herein rely on an implicit representation of above operations thus minimizing computational complexity.

According to some reference embodiments herein, a kernel is used in a detection method based on a two-sample test for determining if a sample sequence of a single- carrier UL conforms to a specifically designed empirical probability density describing an ideal UL or to noise, referred to as ideal UL data of a given configuration (e.g., bandwidth, center frequency) associated to characteristics of the single-carrier UL.

The method thus decides D(X) if a received data sequence X comprises bandlimited interference. Which may be described as:

D(X) = True if interference is present in sequence “X” else D(X) = False Figure 1b illustrates a reference method which may be performed by for example the network node 110 for detecting bandlimited interference in a single-carrier UL from the UE 120 in the wireless communications network 100.

The reference method may be described as a statistical method using a Saturated sine kernel. This kernel corresponds to a spectrum with two rectangular functions, which may be normalized, representing (1) bandlimited noise and (2) UL channel data. The reference method uses this kernel in a kernelized two-sample test. The test value that is outputted from the test is then compared to a threshold to produce a True/False value D(X) denoting whether the incoming signal contains interference. In other words, a threshold function is applied to the test values outputted from the test.

Schematically the information flow through the reference method according to the arrows, later also referred to as a non-robust method.

However, if the threshold function only considers a single output from the test it may be sensitive to fluctuations and outliers. For instance, in environments with high levels of noise or strong interference relative the UL the output variability from the test may lead to a high rate of false alarms or misdetections in the detector.

Embodiments herein disclose a temporal robust method that may improve the reference method by applying a threshold to a weighted linear combination of test values. As will be disclosed in more detail below a robust interference detector may operate on a window of test values.

A number of embodiments will now be described, some of which may be seen as alternatives, while some may be used in combination.

Figure 2a shows example embodiments of a method performed by the network node 110 for detecting bandlimited interference in a single-carrier UL from the UE 120 in the wireless communications network 100 in an improved way, e.g. such that it is more robust against noise and design trade-offs of the kernel.

Embodiments herein are used to construct a low-variance kernelized interference detecting method for a single-carrier UL. The above-mentioned reference method, i.e. , the kernelized two-sample tests, for detecting if a sample sequence conform to a specifically designed empirical probability density of an ideal UL channel or noise, may be improved in the following way. As with the above-mentioned reference method embodiments herein thus takes a decision D(X) if a received data sequence X contains bandlimited interference,

D(X) = True if interference is present in sequence “X” else D(X) = False

Figure 2b shows example embodiments of a method performed by the network node 110 for robustly detecting bandlimited interference in a single-carrier UL from the UE 120 in the wireless communications network 100. The method will also be described with further reference to Figure 2a.

Bandlimited interference when used herein means that the interference is limited to a bounded set of frequencies that overlap with the band allocation of the single-carrier UL.

The method comprises the following actions, which actions may be taken in any suitable order. Optional actions are referred to as dashed boxes in Figure 2b.

Figure 2c and Figure 2d show further details of the example embodiments.

Action 201

According to an example scenario, the network node 110 is about to communicate with the UE 120. It may therefore receive a data sequence in UL from the UE 120. The network node 110 wants to know if there is any bandlimited UL interference. The network node 110 receives a data sequence of the single-carrier UL. The data sequence may mean a sequence of data packets. The data sequence of the single-carrier UL is received from the UE 120.

Action 202

The network node 110 may define a kernel-based statistical two-sample test. The kernel-based statistical two-sample test may form a squared Maximum Mean Discrepancy (MMD) distance-measure also referred to as the test function herein. The kernel-based statistical two-sample test may be referred to as the two-sample test herein. This will be explained more in detail below.

The kernel-based statistical two-sample test that forms a squared MMD distance- measure may be a suitable test to use for detecting any bandlimited interference. This is since it is efficient and use a specific signal metric, i.e. the kernel, to detect the interference. The two-sample test may use any number of pairs of samples for each respective sequence. The two-sample test may define a squared MMD distance-measure, wherein an MMD represents a distance between probability distributions. When defining the kernel-based statistical two-sample test, a suitable kernel is advantageously used. In some embodiments, the defining of the kernel-based statistical two-sample test is based on a kernel with compactly supported spectrum, such as the Saturated Sine kernel, which will be defined below in equation (2). The defining of the kernel-based statistical two-sample test may be based on a kernel constructed for detecting noise and interference with or without UL.

Action 203

The network node 110 obtains a first sequence Y. The first sequence is referred to as Y herein. The first sequence Y comprises a sequence of an integer N of samples. The N samples comprises ideal UL data of a given configuration. The given configuration is associated to characteristics of the single-carrier UL.

The first sequence Y may be obtained by being generated at the network node 110 as the detection method initializes. The first sequence Y is needed to be an input to the kernel-based statistical two-sample test.

Action 204

The network node 110 obtains a second sequence X. The second sequence is referred to as X herein. The second sequence X comprises a sequence of the integer N of samples from the received data sequence.

The second sequence X may be obtained directly from the network node’s 110 receiver. The second sequence X is needed to be a further input to the defined kernel- based statistical two-sample test.

Action 205

The network node 110 calculates two or more squared Maximum Mean Discrepancy, MMD, distance-measure values T[i], T[i-1 ] by comparing the first sequence Y with the second sequence X over two or more windows of samples according to the defined kernel-based statistical two-sample test. A length W of a respective window of the two or more windows of samples is smaller than the integer of N samples. For example, the number of samples within the window may be smaller than the integer of samples.

For example, the network node 110 may calculate a first squared MMD distance- measure value T[i] by comparing the first sequence Y with the second sequence X over a first window of samples. Then the network node 110 may calculate a second squared MMD distance-measure value T[i-1 ] by comparing the first sequence Y with the second sequence X over a second window of samples.

Thus, the network node 110 may compare the first sequence Y with the second sequence X according to the defined kernel-based statistical two-sample test. This may be performed by entering the first sequence Y and the second sequence X as inputs to the defined kernel-based statistical two-sample test.

The result of comparing of the first and second sequences will be needed in next actions as a basis to decide whether or not bandlimited interference is present in the single-carrier UL.

In some embodiments, comparing of the first sequence Y with the second sequence X according to the defined kernel-based statistical two-sample test corresponds to an implicit UL signal isolation filtering and comparing in a frequency domain of the received data sequence. This may mean that the received signal is compared to a known template in the frequency domain where characteristics of an ideal UL channel is well-defined.

This will be described more in detail below.

Action 206

As illustrated in Figure 2c the network node 110 calculates a further distance- measure value O by weighting the calculated two or more squared MMD distance- measure values T[i], T[i-1].

In some embodiments calculating the further distance-measure value O comprises calculating a weighted linear combination of the first squared MMD distance-measure value T[i] of the two or more squared MMD distance-measure values T[i], T[i-1] and one or more delayed squared MMD distance-measure values T[i-1 ] of the two or more squared MMD distance-measure values T[i], T[i-1 ].

As illustrated in Figure 2c and in Figure 2d embodiments herein may implement the weighting, such as temporal weighting, as a logical neural network (LNN), which may be shallow.

As illustrated in Figure 2d the network node 110 may calculate the further distance- measure value O by calculating a combined output value O from two LN Ns comprising a first LNN associated with a first class of output values from the first LNN, and a second LNN associated with a second class of output values from the second LNN. The respective LNN may be responsible for improving robustness of a single class of output values, e.g., either a positive or negative class. The robustness may be defined as a robustness against fluctuations of the test values. As further illustrated in Figure 2d in some embodiments herein calculating the combined output value O is based on calculating a disjunction of a first output value 0+ of the first class of output values from the first LNN and a second output value O- of the second class of output values from the second LNN. The first output value 0+ and the second output value O- are each calculated based on the first squared MMD distance- measure value T[i] and the one or more N-1 delayed squared MMD distance-measure values T[i-1], ... T[i-N],

Each LNN may comprise a number of rules each defined by a decision multiplied by its importance.

Embodiments herein may further implement the weighting either non-recursively, e.g. the LNN only depends on delayed input test values, or recursively, e.g., the LNN also depends on previous outputted weighted test values, such as previous outputted temporal weighted test values. The input to the weighting at time “i” is N-1 delayed test values (i.e. N inputs in total), e.g., T[i], T[i-1] ... , T[i-N],

Thus, in some further embodiments the network node 110 calculates the further distance-measure value O further based on previously calculated further distance- measure values O. For example, the network node 110 may calculate the further distance-measure value O recursively, e.g., each LNN may further be based on previously calculated further distance-measure values of two or more previously calculated squared MMD distance-measure values. In other words, the LNN may also depend on previous outputted weighted test values.

Action 207

The network node 110 decides whether or not bandlimited interference is present in the single-carrier UL, based on the calculated further distance-measure value O. The bandlimited interference may be non-linear UL interference.

In some embodiments deciding whether or not bandlimited interference is present in the single-carrier UL based on the calculated further distance-measure value O comprises:

-when the further distance-measure value O fulfils a criterion, deciding, that bandlimited interference is not present, and

-when the further distance-measure value O does not fulfil the criterion deciding, that bandlimited interference is present.

Thus, the network node 110 decides whether or not bandlimited interference is present in the single-carrier UL. Deciding whether or not bandlimited interference is present in the single-carrier UL may comprise applying the threshold to the weighted linear combination of test values.

For example, when the further distance-measure value O is above a first threshold then D(X) is false and bandlimited interference is not present.

In another example, when the further distance-measure value O is below the first threshold then D(X) is true and bandlimited interference is present.

It may also be the other way around:

In other embodiments when the further distance-measure value O is above a second threshold then D(X) is true and bandlimited interference is present. Then when the further distance-measure value O is below the second threshold then D(X) is false and bandlimited interference is not present.

It should be noted that the first threshold and the second threshold may have the same value or a different value.

In some embodiments, the network node 110 tunes the capability of detecting bandlimited interference in the single-carrier UL of the defined kernel-based statistical two-sample test. This may be performed by adjusting any one or more out of:

- The size of the window of the two-sample test. E.g., meaning the length of the sample sequences X and Y.

- Data template parameters of sequence Y. E.g., meaning the parameters that controls the characteristic of the generated ideal UL data.

- Kernel parameters. Meaning the attributes that control the behaviour of the kernel.

By using embodiments of the method described above, e.g. the following advantages are provided.

- Minimized computational complexity as mentioned above. The developed detection method may implicitly correspond to a UL signal isolation filter and comparison in the frequency domain without requiring costly explicit construction of such signal manipulation operations.

- Controllable and/or verifiable detection capability. This is since the interference detection capability of the provided method may be tuned by adjusting, for instance, the window size of the two-sample test. As statistical power is related to the window size the capability of the method may be verified and validated by testing the method with known inputs and tuning it as needed. - A simplified installation process as the parameters used in the method are easy to deduce from a given band allocation and an input single-carrier LIL data stream, thus making a configuration of the provided embodiments simple.

The above embodiments will now be further explained and exemplified below. The embodiments below may be combined with any suitable embodiment above.

Defining a kernel-based statistical two-sample test

As mentioned above in Action 202, the network node 110 may define a kernel- based statistical two-sample test. The kernel-based statistical two-sample test forms a squared MMD distance-measure.

A kernel is used for defining the kernel-based statistical two-sample test as mentioned above. The kernel is a positive-semidefinite function K defined by a non- standard inner product through a unique feature map (...). Equation (1) defines such as kernel:

Embodiments disclosed herein may use a Saturated Sine kernel, defined by equation (2):

Comparing the first and second sequence in action 205.

As mentioned above, the network node 110 compares the first sequence Y with the second sequence X according to the defined kernel-based statistical two-sample test. This may be performed by entering the first sequence Y and the second sequence X as inputs to the defined kernel-based statistical two-sample test. The result of comparing of the first and second sequences will be a basis to decide whether or not bandlimited interference is present in the single-carrier LIL.

Let Y be a sequence of N samples from an ideal UL data generator for a given configuration, such as the first sequence Y as mentioned above, and referred to as GeneratedULSequence below. Further, let X be a length N sequence from a received (RX) data stream, such as the second sequence X as mentioned above. In that case, the kernel will be used to compare the two sequences X and Y. For this the kernel-based statistical two-sample test using e.g., a squared, MMD distance-measure, is defined, which may be described as equation (3):

An evaluation of K(x,y) is zero if x and y are orthogonal and hence not linearly- dependent. Furthermore, as the kernel is constructed for detecting Noise with or without LIL and therefor lower absolute value of MMD[X,Y] may correspond to detection of interference.

Definitions used in embodiments herein (the temporal-robust method)

Let D = [0,1] be the unit interval and the set of possible values, such as probability values that interference is present, from a test function, i.e. , the real numbers, then the embodiments herein (temporal robust method) may depend on these operators:

• Conjunction operator which is commutative, associative, order- preserving and satisfies for all values

• Disjunction operator which is commutative, associative, order- preserving and satisfies x for all values

• Complement operator D -> D which is order-reversing and satisfies and

• Membership operator have a unique maximum at some “center” value c E V and decreases to 0 as values increasingly differ from the center.

• Weight operator is an arbitrary n-ary function assigning weights to “delayed” values.

Embodiments herein may further define: (i.e. multiplication) (i.e. sum-minus-product) (i.e. negation) (i.e. Gaussian membership with center “c” and scaling “y" ) (i.e. Weighted linear combination of values x _i with coefficients

However alternative operators are possible (as long as they conform to the specific operator definition given above) and sometimes beneficial. For instance, polynomials used as weight operators may enable non-linear temporal relations, triangular functions as membership operator may simplify digital hardware implementation

LNNs

As mentioned above in action 205 the LNN comprises two sub-networks. Each sub- network may be responsible for improving robustness of a single class of output values, e.g., either the positive or negative class.

Each subnetwork may in turn be composed of a number of rules rj each defined by a decision di, e.g., weighting function evaluation, multiplied by its importance importance α _i , e.g., conjunction of membership values. A number of rules in the positive LNN may be denoted #r+. A number of rules in the negative LNN may be denoted #r-. The output values of the LNNs, i.e. the first output value 0+ and the second output value O-, may be given by: where is called the “antecedent” and “consequent” of rule respectively and given by,

In conclusion the temporal weighting is defined by,

Training of the temporal-robust method

Assuming a fixed number of rules and fixed number of delayed test values (N- 1 ), or variable number that is decided at initialization by ex. cross-validation, the positive and negative LNN in the temporal-robust method may be trained in three stages:

• Stage #0 obtain training dataset, such as a pre-process training dataset.

• Stage #1 decide membership operator parameters, e.g., center and scaling.

• Stage #2 decide the weight operator parameters, e.g., coefficients of delayed test values. The three-stage training algorithm may be given a matrix of M test values T = ) and corresponding classification which is either positive

(+1 ) or negative (-1 ) and then, the method may continue with

1 . Creating an extended test-value matrix where is a vector of delayed test values, i.e otherwise. We let denote the matrix with M ₊ (M_) rows containing the delayed test samples that have a corresponding positive (negative) classification, i.e. there is a for each where y _t = +1 (y _i = —1).

2. Computing the antecedent of all rules by deciding the membership operators using the Expectation-Maximization (EM) method that compute Gaussian Mixture Models (GMM), a. One #r ₊ —component (N+1 )-variate GMM of T ⁺ . b. One #r_ —component (N+1 )-variate GMM of T~.

The EM algorithms yields a GMM with mean μ _i and covariance Σ _i by maximizing the log-likelihood given data with M ^± vectors as per,

E-Step: For i = 1 , ... ,M ^± set and update parameters,

The mean “ μ _i ” and variance “Σ _i,i " of each component and GMM marginal variable is used to construct one membership function with center “ μ _i ” and scaling “Σ _i,i ".

Other clustering algorithms may substitute the EM method, in particular for non- gaussian membership functions.

3. Computing the consequent of all rules by deciding Weighted Least Square (WLS) regressors, a. For each rule r ₊ ^ in the positive LNN: Let V be a M x M diagonal matrix column-vector of length M ₊ with all entries equal to +1 then is the WLS coefficients and is the weight operator (and in extension the consequent) of rule b. For each rule in the negative LNN apply similar procedure as step (a) above to produce weight operator

An evaluation of K(x,y) is zero if x and y are orthogonal and hence not linearly- dependent. Furthermore, as the kernel is constructed for detecting Noise with or without UL and therefor lower absolute value of MMD[X,Y] may correspond to detection of interference.

Definitions used in embodiments herein (the temporal-robust method)

Let D = [0,1] be the unit interval and the set of possible values, such as probability values that interference is present, from a test function, i.e. , the real numbers, then the embodiments herein (temporal robust method) may depend on these operators:

• Conjunction operator which is commutative, associative, order- preserving and satisfies for all values

• Disjunction operator which is commutative, associative, order- preserving and satisfies for all values

• Complement operator which is order-reversing and satisfies and -il = 0.

• Membership operator have a unique maximum at some “center” value c e V and decreases to 0 as values increasingly differ from the center.

• Weight operator is an arbitrary n-ary function assigning weights to “delayed” values.

Embodiments herein may further define:

(i.e. multiplication)

(i.e. sum-minus-product)

(i.e. negation) (i.e. Gaussian membership with center “c” and scaling “y" )

(i.e. Weighted linear combination of values x _t with coefficients

However alternative operators are possible (as long as they conform to the specific operator definition given above) and sometimes beneficial. For instance, polynomials used as weight operators may enable non-linear temporal relations, triangular functions as membership operator may simplify digital hardware implementation LNNs

As mentioned above in action 205 the LNN comprises two sub-networks. Each sub- network may be responsible for improving robustness of a single class of output values, e.g., either the positive or negative class.

Each subnetwork may in turn be composed of a number of rules rj each defined by a decision di, e.g., weighting function evaluation, multiplied by its importance importance α _i, , e.g., conjunction of membership values. A number of rules in the positive LNN may be denoted #r+. A number of rules in the negative LNN may be denoted #r-. The output values of the LNNs, i.e. the first output value O+ and the second output value O-, may be given by: where is called the “antecedent” and “consequent” of rule respectively and given by,

In conclusion the temporal weighting is defined by,

Training of the temporal-robust method

Assuming a fixed number of rules (#r _± ) and fixed number of delayed test values (N- 1), or variable number that is decided at initialization by ex. cross-validation, the positive and negative LNN in the temporal-robust method may be trained in three stages:

• Stage #0 obtain training dataset, such as a pre-process training dataset.

• Stage #1 decide membership operator parameters, e.g., center and scaling.

• Stage #2 decide the weight operator parameters, e.g., coefficients of delayed test values. The three-stage training algorithm may be given a matrix of M test values T = and corresponding classification which is either positive (+1) or negative (-1) and then, the method may continue with

1. Creating an extended test-value matrix where is a vector of delayed test values, i.e. otherwise. We let denote the matrix with M ₊ (M_) rows containing the delayed test samples that have a corresponding positive (negative) classification, i.e. there is a for each where y _t = +1 (y _i = -1).

2. Computing the antecedent of all rules by deciding the membership operators using the Expectation-Maximization (EM) method that compute Gaussian Mixture Models (GMM), a. One #r ₊ -component (N+1)-variate GMM of b. One #r_ -component (N+1)-variate GMM of

The EM algorithms yields a GMM with mean and covariance ii by maximizing the log-likelihood given data with

M ^± vectors as per,

E-Step: For i = 1 , ... ,M ^± set for k = 1 #

M-Step: For k = 1 , ... , and update parameters,

The mean “μ _i ” and variance of each component and GMM marginal variable is used to construct one membership function with center “μ _i ” and scaling

Other clustering algorithms may substitute the EM method, in particular for non- gaussian membership functions.

3. Computing the consequent of all rules by deciding Weighted Least Square (WLS) regressors, a. For each rule r _+& in the positive LNN: Let V be a M x M diagonal matrix where and Y ⁺ be a column-vector of length M ₊ with all entries equal to +1 then is the WLS coefficients and is the weight operator (and in extension the consequent) of rule b. For each rule in the negative LNN apply similar procedure as step (a) above to produce weight operator Above three stages decides the Gaussian membership operator parameters, such as center and scaling, and the weight operator parameters, e.g. the coefficients, and therefor the LNN(s) in the Temporal weighting block is fully defined and may be used by applying new (unseen) test values.

However, for the special case of Gaussian membership operators it is beneficial to use the EM algorithm (which is a soft-clustering method) since it is very tractable, compared to most other clustering methods, and in most cases achieve good results.

Addition of the weighting function may improve the prediction accuracy from 39.7%, i.e. the result of the non-robust method, to 74.9%, i.e. the result of the temporal robust method, according to experiments.

Some embodiments allow for “non-continuous” groups of probability values as the positive and negative distributions may overlap such that one distribution may be covered by two non-overlapping regions and separate weighting functions. For example, in a scenario there may be two groups, i.e., two weight components, that cover the positive distribution and one group cover the negative distribution. In this case the prediction accuracy may be raised from 42.2% to 86.2% by applying weighting.

Deciding whether or not bandlimited interference is present 206.

As mentioned above, the network node deciding whether or not bandlimited interference is present in the single-carrier UL. The deciding is based on the comparing, in Action 205, of the first sequence Y with the second sequence X according to the defined kernel-based statistical two-sample test.

The deciding of whether or not bandlimited interference is present in the single- carrier UL, D(X) may be determined using a thresholding function:

The threshold Threshold may in some embodiments be decided at design time during factory calibration or in the field where a real UL may be obtained.

To perform the method actions above, the network node 110 is configured to detect bandlimited interference in a single-carrier UL from the UE 120 in the wireless communications network 100. The network node 110 may comprise an arrangement depicted in Figures 3a and 3b.

The network node 110 may comprise an input and output interface 300 configured to communicate with UEs such as e.g. the UE 120. The input and output interface 300 may comprise a wireless receiver (not shown) and a wireless transmitter (not shown).

The network node 110 may further be configured to, e.g. by means of a receiving unit 310, receive the data sequence of the single-carrier UL.

The network node 110 may further be configured to, e.g. by means of a defining unit 320, define the kernel-based statistical two-sample test, adapted to form the squared MMD distance-measure.

The network node 110 may further be configured to, e.g. by means of an obtaining unit 330, obtain the first sequence Y, which first sequence Y is adapted to comprise the sequence of an integer N of samples comprising ideal UL data of the given configuration associated to characteristics of the single-carrier UL.

The network node 110 may further be configured to, e.g. by means of the obtaining unit 330, obtain the second sequence X, which second sequence X is adapted to comprise the sequence of the integer N of samples from the received data sequence.

The network node 110 is further configured to, e.g. by means of a calculating unit 340, calculate the two or more squared MMD distance-measure values T[i], T[i-1 ] by comparing the first sequence Y with the second sequence X over two or more windows of samples according to the defined kernel-based statistical two-sample test, wherein the length W of the two or more window of samples is smaller than the integer of N samples,

The network node 110 is configured to, e.g. by means of the calculating unit 340, calculate the further distance-measure value O by weighting the two or more squared MMD distance-measure values T[i], T[i-1].

The network node 110 is further configured to, e.g. by means of a deciding unit 350, decide whether or not bandlimited interference is present in the single-carrier UL based on the calculated further distance-measure value O. The network node 110 may further be configured to, e.g. by means of the deciding unit 350, decide whether or not bandlimited interference is present in the single-carrier UL based on the calculated further distance-measure value O by being configured to:

- decide that bandlimited interference is not present when the further distance-measure value O fulfils the criterion, and

- decide that bandlimited interference is present when the further distance- measure value O does not fulfil the criterion.

In some embodiments the network node 110 is configured to, e.g. by means of the calculating unit 340, calculate the further distance-measure value O based on calculating the weighted linear combination of the first squared MMD distance-measure value T[i] of the two or more squared MMD distance-measure values T[i], T[i-1 ] and one or more delayed squared MMD distance-measure values Ti-1 , ... Ti-N of the two or more squared MMD distance-measure values T[i], T[i-1 ].

In some further embodiments the network node 110 is configured to, e.g. by means of the calculating unit 340, calculate the further distance-measure value O further based on previously calculated further distance-measure values O.

The network node 110 may be further configured to, e.g. by means of the calculating unit 340, calculate the further distance-measure value O by calculating the combined output value O from the two LNNs comprising the first Logical Neural Network associated with the first class of output values from the first LNN, and the second LNN associated with the second class of output values from the second LNN. Then the network node 110 may be further configured to, e.g. by means of the calculating unit 340, calculate the combined output value O based on calculating the disjunction of the first output value O+ of the first class of output values from the first LNN and the second output value O- of the second class of output values from the second LNN. The first output value O+ and the second output value O- are each calculated based on the first squared MMD distance-measure value Ti and the one or more N-1 delayed squared MMD distance-measure values Ti-1 , ... Ti-N.

In some further embodiments the network node 110 is configured to, e.g. by means of the calculating unit 340, base each LNN on previously outputted further distance- measure values O of two or more previously calculated squared MMD distance-measure values. The network node 110 may be further configured to, e.g. by means of the calculating unit 340, to define each LNN such that it comprises the number of rules each defined by the decision multiplied by its importance.

The embodiments herein may be implemented through a respective processor or one or more processors, such as the processor 370 of a processing circuitry in the network node 110 depicted in Figure 3a, together with respective computer program code for performing the functions and actions of the embodiments herein. The program code mentioned above may also be provided as a computer program product, for instance in the form of a data carrier carrying computer program code for performing the embodiments herein when being loaded into the network node 110. One such carrier may be in the form of a CD ROM disc. It is however feasible with other data carriers such as a memory stick. The computer program code may furthermore be provided as pure program code on a server and downloaded to the network node 110.

The network node 110 may further comprise a memory 380 comprising one or more memory units. The memory 380 comprises instructions executable by the processor in the network node 110. The memory 380 is arranged to be used to store e.g. information, indications, symbols, data, configurations, and applications to perform the methods herein when being executed in the network node 110.

In some embodiments, a computer program 390 comprises instructions, which when executed by the respective at least one processor 370, cause the at least one processor of the network node 110 to perform the actions above.

In some embodiments, a respective carrier 395 comprises the respective computer program 390, wherein the carrier 395 is one of an electronic signal, an optical signal, an electromagnetic signal, a magnetic signal, an electric signal, a radio signal, a microwave signal, or a computer-readable storage medium.

With reference to Figure 4, in accordance with an embodiment, a communication system includes a telecommunication network 3210, such as a 3GPP-type cellular network, e.g. the wireless communications network 100, which comprises an access network 3211, such as a radio access network, and a core network 3214. The access network 3211 comprises a plurality of base stations 3212a, 3212b, 3212c, e.g. the network node 110, such as AP STAs NBs, eNBs, gNBs or other types of wireless access points, each defining a corresponding coverage area 3213a, 3213b, 3213c. Each base station 3212a, 3212b, 3212c is connectable to the core network 3214 over a wired or wireless connection 3215. A first user equipment (UE) such as a Non-AP STA 3291, e.g. the UE 120, located in coverage area 3213c is configured to wirelessly connect to, or be paged by, the corresponding base station 3212c. A second UE 3292 e.g. the UE 122, such as a Non-AP STA in coverage area 3213a is wirelessly connectable to the corresponding base station 3212a. While a plurality of UEs 3291 , 3292 are illustrated in this example, the disclosed embodiments are equally applicable to a situation where a sole UE is in the coverage area or where a sole UE is connecting to the corresponding base station 3212.

The telecommunication network 3210 is itself connected to a host computer 3230, which may be embodied in the hardware and/or software of a standalone server, a cloud- implemented server, a distributed server or as processing resources in a server farm. The host computer 3230 may be under the ownership or control of a service provider, or may be operated by the service provider or on behalf of the service provider. The connections 3221, 3222 between the telecommunication network 3210 and the host computer 3230 may extend directly from the core network 3214 to the host computer 3230 or may go via an optional intermediate network 3220. The intermediate network 3220 may be one of, or a combination of more than one of, a public, private or hosted network; the intermediate network 3220, if any, may be a backbone network or the Internet; in particular, the intermediate network 3220 may comprise two or more sub-networks (not shown).

The communication system of Figure 4 as a whole enables connectivity between one of the connected UEs 3291, 3292 and the host computer 3230. The connectivity may be described as an over-the-top (OTT) connection 3250. The host computer 3230 and the connected UEs 3291 , 3292 are configured to communicate data and/or signaling via the OTT connection 3250, using the access network 3211 , the core network 3214, any intermediate network 3220 and possible further infrastructure (not shown) as intermediaries. The OTT connection 3250 may be transparent in the sense that the participating communication devices through which the OTT connection 3250 passes are unaware of routing of uplink and downlink communications. For example, a base station 3212 may not or need not be informed about the past routing of an incoming downlink communication with data originating from a host computer 3230 to be forwarded (e.g., handed over) to a connected UE 3291. Similarly, the base station 3212 need not be aware of the future routing of an outgoing uplink communication originating from the UE 3291 towards the host computer 3230. Example implementations, in accordance with an embodiment, of the UE, base station and host computer discussed in the preceding paragraphs will now be described with reference to Figure 5. In a communication system 3300, a host computer 3310 comprises hardware 3315 including a communication interface 3316 configured to set up and maintain a wired or wireless connection with an interface of a different communication device of the communication system 3300. The host computer 3310 further comprises processing circuitry 3318, which may have storage and/or processing capabilities. In particular, the processing circuitry 3318 may comprise one or more programmable processors, application-specific integrated circuits, field programmable gate arrays or combinations of these (not shown) adapted to execute instructions. The host computer 3310 further comprises software 3311 , which is stored in or accessible by the host computer 3310 and executable by the processing circuitry 3318. The software 3311 includes a host application 3312. The host application 3312 may be operable to provide a service to a remote user, such as a UE 3330 connecting via an OTT connection 3350 terminating at the UE 3330 and the host computer 3310. In providing the service to the remote user, the host application 3312 may provide user data which is transmitted using the OTT connection 3350.

The communication system 3300 further includes a base station 3320 provided in a telecommunication system and comprising hardware 3325 enabling it to communicate with the host computer 3310 and with the UE 3330. The hardware 3325 may include a communication interface 3326 for setting up and maintaining a wired or wireless connection with an interface of a different communication device of the communication system 3300, as well as a radio interface 3327 for setting up and maintaining at least a wireless connection 3370 with a UE 3330 located in a coverage area (not shown) served by the base station 3320. The communication interface 3326 may be configured to facilitate a connection 3360 to the host computer 3310. The connection 3360 may be direct or it may pass through a core network (not shown) of the telecommunication system and/or through one or more intermediate networks outside the telecommunication system. In the embodiment shown, the hardware 3325 of the base station 3320 further includes processing circuitry 3328, which may comprise one or more programmable processors, application-specific integrated circuits, field programmable gate arrays or combinations of these (not shown) adapted to execute instructions. The base station 3320 further has software 3321 stored internally or accessible via an external connection.

The communication system 3300 further includes the UE 3330 already referred to. Its hardware 3335 may include a radio interface 3337 configured to set up and maintain a wireless connection 3370 with a base station serving a coverage area in which the UE 3330 is currently located. The hardware 3335 of the UE 3330 further includes processing circuitry 3338, which may comprise one or more programmable processors, application- specific integrated circuits, field programmable gate arrays or combinations of these (not shown) adapted to execute instructions. The UE 3330 further comprises software 3331, which is stored in or accessible by the UE 3330 and executable by the processing circuitry 3338. The software 3331 includes a client application 3332. The client application 3332 may be operable to provide a service to a human or non-human user via the UE 3330, with the support of the host computer 3310. In the host computer 3310, an executing host application 3312 may communicate with the executing client application 3332 via the OTT connection 3350 terminating at the UE 3330 and the host computer 3310. In providing the service to the user, the client application 3332 may receive request data from the host application 3312 and provide user data in response to the request data. The OTT connection 3350 may transfer both the request data and the user data. The client application 3332 may interact with the user to generate the user data that it provides. It is noted that the host computer 3310, base station 3320 and UE 3330 illustrated in Figure 5 may be identical to the host computer 3230, one of the base stations 3212a, 3212b, 3212c and one of the UEs 3291 , 3292 of Figure 4, respectively. This is to say, the inner workings of these entities may be as shown in Figure 5 and independently, the surrounding network topology may be that of Figure 4.

In Figure 5, the OTT connection 3350 has been drawn abstractly to illustrate the communication between the host computer 3310 and the use equipment 3330 via the base station 3320, without explicit reference to any intermediary devices and the precise routing of messages via these devices. Network infrastructure may determine the routing, which it may be configured to hide from the UE 3330 or from the service provider operating the host computer 3310, or both. While the OTT connection 3350 is active, the network infrastructure may further take decisions by which it dynamically changes the routing (e.g., on the basis of load balancing consideration or reconfiguration of the network).

The wireless connection 3370 between the UE 3330 and the base station 3320 is in accordance with the teachings of the embodiments described throughout this disclosure. One or more of the various embodiments improve the performance of OTT services provided to the UE 3330 using the OTT connection 3350, in which the wireless connection 3370 forms the last segment. More precisely, the teachings of these embodiments may improve the RAN effect: data rate, latency, power consumption and thereby provide benefits such as corresponding effect on the OTT service: reduced user waiting time, relaxed restriction on file size, better responsiveness, extended battery lifetime.

A measurement procedure may be provided for the purpose of monitoring data rate, latency and other factors on which the one or more embodiments improve. There may further be an optional network functionality for reconfiguring the OTT connection 3350 between the host computer 3310 and UE 3330, in response to variations in the measurement results. The measurement procedure and/or the network functionality for reconfiguring the OTT connection 3350 may be implemented in the software 3311 of the host computer 3310 or in the software 3331 of the UE 3330, or both. In embodiments, sensors (not shown) may be deployed in or in association with communication devices through which the OTT connection 3350 passes; the sensors may participate in the measurement procedure by supplying values of the monitored quantities exemplified above, or supplying values of other physical quantities from which software 3311, 3331 may compute or estimate the monitored quantities. The reconfiguring of the OTT connection 3350 may include message format, retransmission settings, preferred routing etc.; the reconfiguring need not affect the base station 3320, and it may be unknown or imperceptible to the base station 3320. Such procedures and functionalities may be known and practiced in the art. In certain embodiments, measurements may involve proprietary UE signaling facilitating the host computer’s 3310 measurements of throughput, propagation times, latency and the like. The measurements may be implemented in that the software 3311, 3331 causes messages to be transmitted, in particular empty or ‘dummy’ messages, using the OTT connection 3350 while it monitors propagation times, errors etc.

Figure 6 is a flowchart illustrating a method implemented in a communication system, in accordance with one embodiment. The communication system includes a host computer, a base station such as a AP STA, and a UE such as a Non-AP STA which may be those described with reference to Figure 4 and Figure 5. For simplicity of the present disclosure, only drawing references to Figure 6 will be included in this section. In a first step 3410 of the method, the host computer provides user data. In an optional substep 3411 of the first step 3410, the host computer provides the user data by executing a host application. In a second step 3420, the host computer initiates a transmission carrying the user data to the UE. In an optional third step 3430, the base station transmits to the UE the user data which was carried in the transmission that the host computer initiated, in accordance with the teachings of the embodiments described throughout this disclosure. In an optional fourth step 3440, the UE executes a client application associated with the host application executed by the host computer.

Figure 7 is a flowchart illustrating a method implemented in a communication system, in accordance with one embodiment. The communication system includes a host computer, a base station such as a AP STA, and a UE such as a Non-AP STA which may be those described with reference to Figure 4 and Figure 5. For simplicity of the present disclosure, only drawing references to Figure 7 will be included in this section. In a first step 3510 of the method, the host computer provides user data. In an optional substep (not shown) the host computer provides the user data by executing a host application. In a second step 3520, the host computer initiates a transmission carrying the user data to the UE. The transmission may pass via the base station, in accordance with the teachings of the embodiments described throughout this disclosure. In an optional third step 3530, the UE receives the user data carried in the transmission.

Figure 8 is a flowchart illustrating a method implemented in a communication system, in accordance with one embodiment. The communication system includes a host computer, a base station such as an AP STA, and a UE such as a Non-AP STA which may be those described with reference to Figure 4 and Figure 5. For simplicity of the present disclosure, only drawing references to Figure 8 will be included in this section. In an optional first step 3610 of the method, the UE receives input data provided by the host computer. Additionally or alternatively, in an optional second step 3620, the UE provides user data. In an optional substep 3621 of the second step 3620, the UE provides the user data by executing a client application. In a further optional substep 3611 of the first step 3610, the UE executes a client application which provides the user data in reaction to the received input data provided by the host computer. In providing the user data, the executed client application may further consider user input received from the user. Regardless of the specific manner in which the user data was provided, the UE initiates, in an optional third substep 3630, transmission of the user data to the host computer. In a fourth step 3640 of the method, the host computer receives the user data transmitted from the UE, in accordance with the teachings of the embodiments described throughout this disclosure.

Figure 9 is a flowchart illustrating a method implemented in a communication system, in accordance with one embodiment. The communication system includes a host computer, a base station such as a AP STA, and a UE such as a Non-AP STA which may be those described with reference to Figure 4 and Figure 5. For simplicity of the present disclosure, only drawing references to Figure 9 will be included in this section. In an optional first step 3710 of the method, in accordance with the teachings of the embodiments described throughout this disclosure, the base station receives user data from the UE. In an optional second step 3720, the base station initiates transmission of the received user data to the host computer. In a third step 3730, the host computer receives the user data carried in the transmission initiated by the base station. When using the word "comprise" or “comprising” it shall be interpreted as non- limiting, i.e. meaning "consist at least of".

The embodiments herein are not limited to the above described preferred embodiments. Various alternatives, modifications and equivalents may be used.