Apparatus, method, and computer-readable medium for PDP estimation in the frequency domain

Abstract

The present invention relates to, for example, power delay profile (PDP) estimation in a 5G system, which is necessary for channel estimation and demodulation. Based on channel statistical knowledge, the power delay profile is estimated directly in the frequency domain using the LMMSE property. The present invention adopts an approximation by estimating the second-order statistics of the channel autocorrelation function in the time domain, which is represented by the variance of the derivative of the original channel estimate in the frequency domain.

Description

Technical Field

[0001] The present invention generally relates to cellular radio networks and their implementation. With respect to different generations of technologies, the present invention may relate to any of the following: 2G, 3G, 4G, 5G or 6G access networks. Background Art

[0002] Current and future wireless communication systems, such as Long Term Evolution (LTE) or Fifth Generation (5G), also known as New Radio (NR), have envisioned using Multiple-Input Multiple-Output (MIMO) multi-antenna transmission technology. The increasing requirements for high throughput motivate wireless communication systems, such as 5G, to use millimeter wave (mmWave) frequencies due to the available high bandwidth.

[0003] However, the use of mmWave frequencies poses new challenges in MIMO performance.

[0004] The signal received by an antenna is the sum of several copies of the transmitted signal, each copy being characterized by its own attenuation and delay. The transmitted signal is typically reflected and scattered and reaches the receiver via multiple paths. When the relative path delays are on the order of one symbol period or more, the images of different symbols arrive simultaneously, causing Inter-Symbol Interference (ISI). The Power Delay Profile (PDP) gives the strength of the received signal through a multipath channel as a function of the delay spread.

[0005] The estimation of the PDP is necessary to perform channel estimation in the best possible way. Appropriate channel estimation is important for improving the demodulation performance on both the base station side and the mobile terminal side.

[0006] PDP estimation mechanisms are typically based on Inverse Fast Fourier Transform (IFFT) calculations and correlations, which require high CPU (Central Processing Unit) costs due to the complexity. In this case, the Channel Impulse Response (CIR) in the time domain is derived from the CIR in the frequency domain using IFFT, and then the instantaneous PDP parameters are estimated through successive correlations. The main additional drawback of this technique is the poor performance for low Signal-to-Noise Ratio (SNR) as well as for small bandwidth allocations. Summary of the Invention

[0007] The present invention introduces a new and efficient mechanism for estimating the Power Delay Profile (PDP) using the knowledge of channel statistics, based on the Linear Minimum Mean Square Error (LMMSE) or Minimum Mean Square Error (MMSE) properties. This only requires the calculation of the variance of the derivative of the original channel estimate in the frequency domain, thus minimizing the CPU cost. This and other detailed advantages are also listed at the end of the detailed description.

[0008] Power delay profile estimation is necessary for channel estimation and subsequent demodulation. The PDP is estimated directly in the frequency domain using the knowledge of the channel statistical characteristics. An approximation is used, which is achieved by estimating the second-order statistics of the channel autocorrelation function in the time domain, which is represented by the variance of the derivative of the original channel estimate in the frequency domain. Thereby, the use of the IFFT is avoided when performing this estimation in the time domain, or the use of a large number of consecutive correlations is avoided when performing this estimation in the frequency domain.

[0009] The present invention is a low-cost method, and it proposes a non-parametric estimation, in which the above-mentioned second-order statistics can be estimated without any assumptions on the PDP model. The method introduced according to the present invention is also effective in terms of performance.

[0010] The proposed method can be introduced both on the gNB (i.e., base station) side and on the mobile terminal side, because both of these entities must estimate the channel to demodulate the information.

[0011] Now, an improved method and a technical device for implementing the method have been invented, which alleviates the above problems through this method. Each aspect includes a method, a device, and a non-transitory computer-readable medium, which includes a computer program or a signal stored therein, and the features are as stated in the independent claims. Various details of the embodiments are disclosed in the dependent claims, the corresponding drawings, and the detailed description.

[0012] The scope of protection sought by the various embodiments of the present invention is set by the independent claims. The embodiments and features described in this specification that do not fall within the scope of the independent claims (if any) should be construed as examples to assist in understanding the various embodiments of the present invention.

[0013] According to a first aspect of the present invention, a device is introduced, including:

[0014] - Multiple-input multiple-output (MIMO) antennas;

[0015] - Components for receiving a demodulation reference signal (DMRS) vector in the frequency domain (21);

[0016] - Components for estimating an original channel estimate vector (23) from the DMRS vector received in the frequency domain (21);

[0017] - Components for calculating a derivative vector (25) of the original channel estimate vector (23);

[0018] - A component for determining a second-order statistical estimate (28), the determination being made by calculating the variance of the derivative vector (25) of the original channel estimate vector (23) and calculating the noise estimate variance (26), and by subtracting the calculated noise estimate variance (26) from the variance of the derivative vector (25) of the calculated original channel estimate vector (23);

[0019] - A component for the root mean square (RMS) delay spread (29) of each radio path from the determined second-order statistical estimate (28); and

[0020] - A component for estimating the power delay profile (PDP) by comparing the estimated RMS delay spread with a predetermined RMS delay spread threshold.

[0021] According to an embodiment of the first aspect of the present invention, the apparatus includes

[0022] - A component for calculating the variance of the noise estimate (26) by subtracting the reconstructed signal from the received signal.

[0023] According to an embodiment of the first aspect of the present invention, the component for estimating the RMS delay spread (29) of each radio path is configured to estimate the RMS delay spread (29) as follows

[0024] - where V is the determined second-order statistical estimate (28) and ΔF is the single carrier spacing in Hz.

[0025] According to an embodiment of the first aspect of the present invention, the apparatus includes

[0026] - A component for calculating the original channel estimate vector (23) for each TX / RX radio path k and at each DMRS position of the allocated physical resource block (PRB) according to the following formula:

[0027] For i = 0:nDMRS–1 and k = 0:nTX*nRX-1

[0028] H_RAW_DMRS k,i = r k,i .s k,i *

[0029] where nDMRS is the number of DMRS in all allocated PRBs, where r k,i is the sampled DMRS received at the (k,i) position, and s k,i * is the conjugate of the known DMRS symbol at the ((k,i) position.

[0030] According to an embodiment of the first aspect of the present invention,

[0031] - When k>1 is generated with more than one transmit antenna and one receive antenna, H_RAW_DMRS[i] of each radio channel path k is cascaded to generate

[0032] nDMRS = nDMRS * k, thereby obtaining the original channel estimation vector (23).

[0033] According to an embodiment of the first aspect of the present invention, the apparatus includes

[0034] - A component for calculating the derivative vector (25) of the original channel estimation vector (23), setting the difference between two DMRSs on the resulting distance, where the DMRSs are after the DMRS.

[0035] According to an embodiment of the first aspect of the present invention, the apparatus includes

[0036] - A component for calculating the derivative vector (25) of the original channel estimation vector (23) for each radio channel path k in the following manner:

[0037] For i = 0:nMeas - 1, α[i] = (H_RAW_DMRS[M + i] – H_RAW_DMRS[i]) / N where N is the number of resource elements between two DMRSs dedicated to measurement, where M = N / 2 = the resulting distance between two DMRSs dedicated to measurement of frequency drift, where nMeas is the number of measurements = nDMRS - M, which is also the length of the derivative vector (25) of the original channel estimation vector (23) on each radio channel path k.

[0038] According to an embodiment of the first aspect of the present invention, the apparatus includes

[0039] - A component for determining the second-order statistical estimate (28) as follows, where the second-order statistical estimate (28) corresponds to the variance of the derivative vector (25) of the original channel estimation vector (23):

[0040]

[0041] where V is the determined second-order statistical estimate (28), and [α] is the derivative vector (25) generated by the derivative operator (24), and is the average value, and is the noise estimation variance (26).

[0042] According to an embodiment of the first aspect of the present invention, the apparatus includes

[0043] - Components for performing minimum mean square error (MMSE) or linear minimum mean square error (LMMSE) based channel estimation based on an estimated power delay profile (PDP).

[0044] According to an embodiment of the first aspect of the present invention, the apparatus is a base station, and the base station includes components for estimating a channel to demodulate information.

[0045] According to an embodiment of the first aspect of the present invention, the apparatus is a mobile terminal, and the mobile terminal includes components for estimating a channel to demodulate information.

[0046] The apparatus mentioned herein and in related embodiments may include at least one processor; and at least one memory including computer program code, the at least one memory and the computer program code being configured to cause the performance of the apparatus in conjunction with the at least one processor.

[0047] According to a second aspect of the present invention, there is provided an apparatus, comprising:

[0048] - At least one processor; and

[0049] - At least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to be capable of at least:

[0050] - Transmit and / or receive wireless signals via multiple input multiple output (MIMO) antennas;

[0051] - Receive a demodulation reference signal (DMRS) vector in the frequency domain (21);

[0052] - Estimate an original channel estimation vector (23) from the received DMRS vector in the frequency domain (21);

[0053] - Calculate a derivative vector (25) of the original channel estimation vector (23);

[0054] - Determine a second-order statistical estimate (28) as follows, calculate the variance of the derivative vector (25) of the original channel estimation vector (23) and calculate the noise estimation variance (26), and subtract the calculated noise estimation variance (26) from the calculated variance of the derivative vector (25) of the original channel estimation vector (23);

[0055] - Estimate the root mean square (RMS) delay spread (29) for each radio channel path from the determined second-order statistical estimate (28); and

[0056] - Estimate the power delay profile (PDP) by comparing the estimated RMS delay spread with a predetermined RMS delay spread threshold.

[0057] According to an embodiment of the second aspect of the present invention, the apparatus is further caused to:

[0058] - Calculate the variance (26) of the noise estimate by subtracting the reconstructed signal from the received signal.

[0059] According to an embodiment of the second aspect of the present invention, wherein the component for estimating the RMS delay spread (29) for each radio circuit path is configured to estimate the RMS delay spread (29) as follows:

[0060] - where V is the determined second-order statistical estimate (28), and ΔF is the single-carrier spacing in Hz.

[0061] According to an embodiment of the second aspect of the present invention, the apparatus is further caused to:

[0062] - Calculate the raw channel estimate vector (23) for each TX / RX radio circuit path k and at each DMRS position of the allocated physical resource block (PRB) according to the following formula:

[0063] For i = 0:nDMRS–1 and k = 0:nTX*nRX-1

[0064] H_RAW_DMRS k,i = r k,i .s k,i *

[0065] where nDMRS is the number of DMRS in all allocated PRBs, and where r k,i is the sampled DMRS received at the (k,i) position, and s k,i * is the conjugate of the known DMRS symbol at the (k,i) position.

[0066] According to an embodiment of the second aspect of the present invention,

[0067] - In the case where more than one transmit antenna and one receive antenna result in k>1, the H_RAW_DMRS[i] of each radio circuit path k is concatenated to produce nDMRS = nDMRS*k, thereby obtaining the raw channel estimate vector (23).

[0068] According to an embodiment of the second aspect of the present invention, the apparatus is further caused to:

[0069] - Calculate the derivative vector (25) of the raw channel estimate vector (23) by differentiating two DMRSs over the resulting distance from one DMRS to the next, where the DMRS is after the DMRS.

[0070] According to an embodiment of the second aspect of the present invention, the apparatus is further caused to:

[0071] - For each radio path k, calculate the derivative vector (25) of the original channel estimation vector (23) in the following manner:

[0072] For i = 0:nMeas - 1, α[i] = (H_RAW_DMRS[M + i] –

[0073] H_RAW_DMRS[i]) / N

[0074] where N is the number of resource elements between two DMRSs dedicated to measurement, and where M = N / 2 = the distance between two DMRSs dedicated to frequency drift measurement, and where nMeas is the number of measurements = nDMRS - M, which is also the length of the derivative vector (25) of the original channel estimation vector (23) on each radio path k.

[0075] According to an embodiment of the second aspect of the present invention, the apparatus is further caused to:

[0076] - Determine the second - order statistical estimate (28) as follows, where the second - order statistical estimate (28) corresponds to the variance of the derivative vector (25) of the original channel estimation vector (23):

[0077]

[0078] where V is the determined second - order statistical estimate (28), [α] is the derivative vector (25) generated by the derivative operator (24) and is the mean value, and is the variance of the noise estimate (26).

[0079] According to an embodiment of the second aspect of the present invention, the apparatus is further caused to:

[0080] - Perform channel estimation of minimum mean - square error (MMSE) or linear minimum mean - square error (LMMSE) based on the estimated power delay distribution (PDP).

[0081] According to an embodiment of the second aspect of the present invention, the apparatus is a base station, and the base station can further estimate the channel to demodulate information.

[0082] According to an embodiment of the second aspect of the present invention, the apparatus is a mobile terminal, and the mobile terminal is further used to estimate the channel to demodulate information.

[0083] According to a third aspect of the present invention, a method is provided, including the steps of:

[0084] - Transmit and / or receive wireless signals through multiple - input multiple - output (MIMO) antennas;

[0085] - Receive a demodulation reference signal (DMRS) vector in the frequency domain (21);

[0086] - Estimate an original channel estimation vector (23) from the DMRS vector received in the frequency domain (21);

[0087] - Calculate a derivative vector (25) of the original channel estimation vector (23);

[0088] - Determine a second-order statistical estimate (28) as follows: calculate the variance of the derivative vector (25) of the original channel estimation vector (23) and calculate a noise estimation variance (26), and subtract the calculated noise estimation variance (26) from the calculated variance of the derivative vector (25) of the original channel estimation vector (23);

[0089] - Estimate the root mean square (RMS) delay spread (29) for each radio path based on the determined second-order statistical estimate (28); and

[0090] - Estimate the power delay profile (PDP) by comparing the estimated RMS delay spread with a predetermined RMS delay spread threshold.

[0091] In various embodiments according to the third aspect of the present invention, the method may include the various steps already disclosed in the various embodiments of the first and second aspects of the present invention above.

[0092] Furthermore, according to a further aspect, there is provided an apparatus comprising at least one processor and at least one memory, the at least one memory having computer program code stored thereon, the at least one memory and the computer program code being configured to, together with the at least one processor, cause the apparatus to perform at least the above method steps related to the third aspect of the present invention.

[0093] According to a fourth aspect of the present invention, there is provided a computer program comprising instructions stored thereon for performing at least the following operations:

[0094] - Transmit and / or receive wireless signals via multiple-input multiple-output (MIMO) antennas;

[0095] - Receive a demodulation reference signal (DMRS) vector in the frequency domain (21);

[0096] - Estimate an original channel estimation vector (23) from the DMRS vector received in the frequency domain (21);

[0097] - Calculate a derivative vector (25) of the original channel estimation vector (23);

[0098] - The second-order statistical estimate (28) is determined as follows: Calculate the variance of the derivative vector (25) of the original channel estimate vector (23) and calculate the noise estimate variance (26), and subtract the calculated noise estimate variance (26) from the calculated variance of the derivative vector (25) of the original channel estimate vector (23);

[0099] - Estimate the root mean square (RMS) delay spread (29) for each radio path based on the determined second-order statistical estimate (28); and

[0100] - Estimate the power delay profile (PDP) by comparing the estimated RMS delay spread with a predetermined RMS delay spread threshold.

[0101] In various embodiments according to the fourth aspect of the present invention, a computer program may include the respective steps (including instructions stored thereon for execution) disclosed in the various embodiments of the first, second, and third aspects of the present invention above.

[0102] According to a fifth aspect of the present invention, there is provided a non-transitory computer-readable medium comprising program instructions stored thereon for performing at least the following operations:

[0103] - Transmit and / or receive wireless signals via a multiple-input multiple-output (MIMO) antenna;

[0104] - Receive a demodulation reference signal (DMRS) vector in the frequency domain (21);

[0105] - Estimate an original channel estimate vector (23) from the DMRS vector received in the frequency domain (21);

[0106] - Calculate a derivative vector (25) of the original channel estimate vector (23);

[0107] - Determine a second-order statistical estimate (28) by calculating the variance of the derivative vector (25) of the original channel estimate vector (23) and calculating the variance of the noise estimate (26), and by subtracting the calculated variance of the noise estimate (26) from the calculated variance of the derivative vector (25) of the original channel estimate vector (23);

[0108] - Estimate the root mean square (RMS) delay spread (29) for each radio path based on the determined second-order statistical estimate (28); and

[0109] - Estimate the power delay profile (PDP) by comparing the estimated RMS delay spread with a predetermined RMS delay spread threshold.

[0110] In various embodiments according to the fifth aspect of the present invention, the computer-readable medium may include the respective steps (i.e., including program instructions stored thereon for execution) disclosed in the various embodiments of the first, second, third, and fourth aspects of the present invention above.

[0111] In a further option, a computer-readable storage medium (i.e., a computer-readable medium) according to a further aspect of the present invention includes code for use by a device, which when executed by a processor, causes the device to perform the various embodiments of the above method. BRIEF DESCRIPTION OF THE DRAWINGS

[0112] To more fully understand the example embodiments, reference is now made to the following description taken in conjunction with the accompanying drawings, in which:

[0113] Figure 1 shows an embodiment of the present invention, showing six demodulation reference signals (DMRS) in a single physical resource block (PRB),

[0114] Figure 2 shows a sorting that characterizes the implementation of the root mean square (RMS) delay spread estimation method according to an embodiment of the present invention,

[0115] Figure 3a shows a first example of processing in an allocated PRB in an embodiment of the present invention,

[0116] Figure 3b shows a second example of processing in an allocated PRB in an embodiment of the present invention,

[0117] Figure 4a shows a comparison of the demodulation performance between the current PDP estimation method according to the prior art and the PDP estimation method proposed according to an embodiment of the present invention, with the performance represented by the block error rate (BLER) for the coding scheme for MCS 0 and a delay spread of 30 ns,

[0118] Figure 4b shows a comparison of the demodulation performance between the current PDP estimation method according to the prior art and the PDP estimation method proposed according to an embodiment of the present invention, with the performance represented by the block error rate (BLER) for the coding scheme for MCS24 and a delay spread of 300 ns,

[0119] Figure 5 shows an exemplary network scenario according to at least some embodiments of the present invention, and

[0120] Figure 6 shows an example device capable of supporting at least some embodiments of the present invention. DETAILED DESCRIPTION

[0121] Suitable devices and possible mechanisms for performing power delay distribution (PDP) estimation are further described in detail below. In various embodiments of the present invention, power delay distribution (PDP) estimation can further be used for base stations or mobile stations, and even for channel estimation of these two elements. Although the following focuses on 5G networks, the embodiments further described below are by no means limited to implementation only in the said networks, but are applicable to any network that implements MU-MIMO transmission.

[0122] First, the theoretical aspects of the present invention are discussed.

[0123] The signal received by the antenna is the sum of several copies of the transmitted signal, and each copy has its own attenuation and delay. Due to the law of large numbers (central limit theorem), the received signal can be considered to be centrally Gaussian.

[0124] The power delay distribution gives the strength of the signal received through the multipath channel as a function of the delay spread.

[0125] The delay spread introduces a change in the frequency domain, which can be viewed as a linear frequency drift that is superimposed on the reference signal and is used to decompose the spectral operator, related to the development of "Karhunen-Loève", a technique that enables the fading to be represented as a superposition of uncorrelated Gaussian random variables and is thus independent.

[0126] The effect of non-stationarity in the frequency domain can be represented by additive composite noise, which drifts from one end of the selected sequence length to the other end, and its moving length is close to the coherence bandwidth.

[0127] The fading process is Gaussian, and its derivative, which is a linear transformation, is also Gaussian. Therefore, it can be uniquely characterized by its variance.

[0128] Studies have shown that the process defining the fading is independent, and the variance of the frequency domain derivative in the interval near the coherence bandwidth is proportional to the delay spread.

[0129] The optimal channel estimation technique that minimizes the MSE (mean square error) is the well-known linear minimum mean square error (LMMSE). LMMSE is based on the knowledge of channel statistics, which are represented by the power delay distribution (PDP).

[0130] It should be emphasized that the proposed process is independent of the type of channel sounding; therefore, the applied channel estimation technique can be LMMSE or MMSE; both work in a similar way in this case.

[0131] Due to the complexity of estimating the complete PDP, approximate methods that evaluate statistical estimates of the RMS delay spread are typically used. Mathematical demonstrations show that the variance of the derivative of the original channel estimate in the frequency domain corresponds to the second derivative at the origin of the autocorrelation function of the original channel estimate in the time domain, and this result helps in calculating the RMS delay spread in the time domain. With the help of the assumed PDP model, these statistics can perform accurate channel estimation.

[0132] Currently, the models used for PDP are the uniform model and the exponential decay model.

[0133] Note that when assuming an exponential PDP distribution, the average delay spread and the RMS delay spread are equal to the decay factor.

[0134] The main drawback of current PDP estimation and a general drawback for all techniques done in the time domain is the high computational complexity due to the use of the Fourier transform or the inverse Fourier transform (IFFT).

[0135] In the following embodiments describing the present invention, a new PDP estimation method based on directly utilizing the channel statistics of the received samples in the frequency domain is introduced.

[0136] Let X(f) be the channel impulse response in the frequency domain and X'(f) be the derivative of X(f) with respect to the frequency f.

[0137] The power delay profile R XX (u)i is defined as the autocorrelation of the channel response X(f):

[0138]

[0139] Deriving the above equation with respect to u gives:

[0140]

[0141] Under the assumption of autocorrelation stationarity:

[0142]

[0143] Taking the autocorrelation function at 0 in expression (2) results in:

[0144]

[0145] Deriving equation (2) again gives the second derivative of the autocorrelation function:

[0146]

[0147] From mathematical expressions (3) and (4), the most important observation and the basic property of the estimation suggestion are that, from the sampled frequency channels, the first and second derivatives can be obtained at the origin without any explicit calculation of the autocorrelation function.

[0148] As will be shown below, the knowledge of the two derivative values at the origin is linked in the time domain to the first and second order statistics of the PDP.

[0149] According to the Fourier transform properties, the function moments can be related to the derivatives of the Fourier transform.

[0150] Let h(t) be the power delay profile in the time domain and H(f) be its Fourier transform.

[0151] Using the general Fourier transform identity:

[0152] t n The Fourier transform of h(t) is the function

[0153] H (n) (f) is the n-th order derivative of H.

[0154] Therefore, for f = 0 and n = 1:

[0155]

[0156] Equation (5) expresses the average delay as a function of the first derivative at the origin of the Fourier transform of the time domain autocorrelation.

[0157] Similarly, for f = 0 and n = 2, the following result is obtained:

[0158]

[0159] Now, equations (3) and (5) yield the PDP for the first order statistical estimation, and equations (4) and (6) yield the PDP for the second order statistical estimation.

[0160]

[0161] The RMS delay spread follows equations (7) and (8):

[0162]

[0163] At this stage, it can be concluded that the calculation of the RMS delay spread applies the second order statistical estimation, and the second order statistical estimation will also be discussed in the practical aspects of the present invention and in the description related to Figure 2 related descriptions.

[0164] It is noted that a fundamental property is that the first - order and second - order statistics can be estimated without any assumptions about the PDP model. Thus, the proposed estimation method is a non - parametric estimation method, giving robustness against changes in the PDP model, which depends on the field location: indoor / outdoor, etc.

[0165] The first - order statistical estimation of the channel autocorrelation function corresponds to the average delay.

[0166] The second - order statistical estimation of the channel autocorrelation function corresponds to the delay variance.

[0167] For good robustness, an additional estimation of the average delay is required, as shown in Equation (9). However, D 2 should be lower than V for calculation, and this value (D 2 ) is always very low. Thus, the RMS delay spread is directly adopted in the following simulations without using D 2 .

[0168] Next, the steps for RMS delay spread estimation implementation are discussed in the embodiments.

[0169] Estimation of the original channel coefficients based on the OFDM reference symbols Z k for k = 1, …, N on one antenna.

[0170] Assume maximum delay spread filtering for the original channel estimate to reduce the noise on Z→X k for k = 1, …, N.

[0171] The derivative calculation is carried out as follows.

[0172] The calculations of D, V, and RMS delay spread are based on Equations (7), (8), and (9) using the samples X k and X′ k .

[0173] The effective time for different derivatives can be calculated:

[0174]

[0175] The validity of this expression assumes that the channel coherence bandwidth is much higher than p times the pilot separation bandwidth.

[0176] Using an increased step size p for derivative calculation can reduce the estimated noise variance.

[0177] The proof is as follows. X is a noisy estimate of the true channel R frequency response. X k = R k + W k , where R is the exact channel coefficient and W is the filtered noise sample, with assumed variance E(|W| 2) = σ 2 。

[0178] Using the statistical independence between R and W, we can obtain:

[0179]

[0180] Equation (11) shows that the p-th derivative V (p) the noise power on, for the value of is Therefore, the benefits of using an increased step size can be seen.

[0181] The delay spread can be estimated based on multiple p-derivatives, and then the best linear estimate under uncorrelated measurements is a weighted linear combination of p-th order estimates. The weights are the measured values of the reciprocals of the noise variances, normalized by their sum.

[0182]

[0183] Next, the practical aspects of the present invention are discussed.

[0184] This is discussed according to an embodiment, and exemplary structural selections and some exemplary parameter values within the presented concepts are combined below.

[0185] In the case of an OFDM 5G radio system with a single carrier spacing of ΔF = 30 kHz, each allocated PRB has six DMRSs of type 1, and the parameter N_DMRS_PER_PRB equal to 6 is generated through a dedicated antenna port. The raw channel estimate is calculated for each antenna port on the DMRS. Figure 1 Shows its implementation in one PRB in the described embodiment.

[0186] The raw channel estimate vector is calculated for each TX / RX channel propagation path k (i.e., TX / RX radio path k) and on each DMRS of the allocated PRB according to the following formula:

[0187] For i = 0:nDMRS–1 and k = 0:nTX*nRX-1

[0188] H_RAW_DMRS k,i = r k,i .s k,i *

[0189] Let nDMRS be the number of DMRSs in all allocated PRBs. The result is the H_RAW_DMRS vector of length nDMRS, where r k,i is the received sample DMRS at the (k,i) position, s k,i *is the conjugate of the known DMRS symbol at the (k,i) position.

[0190] Thus, the channel is roughly estimated in this way. Then, it can be assumed that the maximum delay spread filters the original channel estimate to reduce noise.

[0191] The goal is to estimate the selectivity of the channel to guide the estimation as faithfully as possible. The selectivity of the channel is a function of the RMS delay spread.

[0192] Here, the RMS delay spread can be evaluated by calculating the variance of the derivative of the original channel estimate in the frequency domain (see Equation (9) above), instead of the second derivative at the origin of the autocorrelation function of the original channel estimate in the time domain (see the theoretical aspects above). See also the detailed formulas for calculating "V" (i.e., the second-order statistical estimate 28) and "rms_ds" (i.e., the RMS delay spread 29) later. In other words, to clarify the terminology between the above theoretical aspects and the subsequent practical aspects, the variance of the derivative of the original channel estimate in the frequency domain (resulting in the second-order statistical estimate 28, i.e., "V", as Figure 2 shown in its respective description) corresponds to the second derivative at the origin of the autocorrelation function of the original channel estimate in the time domain.

[0193] In one embodiment, three PDPs characterized by short, medium, and long delay spreads are defined. The long delay spread of 300 ns normalized by 3GPP means that some components of the transmitted signal can reach the cyclic prefix, which corresponds to 2.38 μs for an SCS of 30 kHz. The short delay spread is applicable to delay spreads below 50 ns, and the medium delay spread is applicable to delay spreads below 150 ns. These three delay spread values will be used as detection thresholds.

[0194] The second-order statistical estimate of the PDP for each case is characteristic of the delay spread. It is possible to switch to one of the three possible PDPs based on the 3 thresholds of the delay spread and the estimate of the delay spread.

[0195] Figure 2 Shows the characteristic ordering of implementing the RMS delay spread estimation method in an embodiment according to the present invention.

[0196] First, based on the DMRS vector received in the frequency domain 21, the original channel estimate vector 23 is delivered according to the method described in the previous paragraph (by the original channel estimation block 22).

[0197] In the case where more than one transmit antenna and one receive antenna result in k>1, the H_RAW_DMRS[i] for each radio path k is concatenated, resulting in nDMRS = nDMRS*k.

[0198] Secondly, the derivative vector 25 of the original channel estimation vector 23 is then delivered by the derivative operator 24. The derivative of the original channel estimation is calculated by differentiating between two DMRSs over the resulting distance of DMRS followed by DMRS as shown in Figure 3a and Figure 3b , where each arrow represents a measurement. Figure 3a and Figure 3b show two examples processed in an allocated PRB. It can be seen that there are two resource elements (delta_re = 2) between two type 1 DMRSs. Therefore, the derivative vector 25 of the original channel estimation vector is constructed according to the number of allocated PRBs. In an embodiment of the present invention, this process is generally constructed as follows:

[0199] a. Let nDMRS be the number of received DMRS original channel estimations.

[0200] b. Let nPRB be the number of allocated PRBs, where PRB = nDMRS / N_DMRS_PER_PRB.

[0201] c. Let D_REF be the maximum distance of resource elements between two DMRSs, which is agreed upon for measurement.

[0202] d. Let nPrbMetric be the number of PRBs in which the measurement is performed.

[0203] e. Let N be the number of resource elements between two DMRSs dedicated to measurement. It is a multiple of D_REF. N = D_REF × nPrbMetric.

[0204] f. Let M be the distance between two DMRSs for which N / delta_re is dedicated to measuring the frequency drift, where delta_re is the distance of resource elements between two DMRSs (2 resource elements for type 1).

[0205] g. The number of measurements is nMeas = nDMRS - M, which is also the length of the derivative vector 25 of the original channel estimation vector 23 on each radio path k. The derivative vector 25 of the original channel estimation vector for each radio path k is calculated in a general manner:

[0206] For i = 0:nMeas - 1, α[i] = (H_RAW_DMRS[M + i] – H_RAW_DMRS[i]) / ND_REF is set to be equal to 2 resource elements in order to track the channel fluctuations as faithfully as possible in the case of long delay spread. The parameter nPrbMetric is also set to 1, resulting in N = 2 and M = 1. Also see Figure 3b .

[0207] Third, the second-order statistical estimate 28 is then delivered by the corresponding operator 27. It incorporates the variance of the derivative vector 25 of the original channel estimate vector 23 according to the following formula:

[0208]

[0209] [α] is the derivative vector 25 generated by the derivative operator 24, and is the average value, is the noise estimation variance 26, which can be estimated by any available method. V is the second-order statistical estimate 28.

[0210] The noise variance was previously calculated by any method and then the variance of the derivative vector of the original channel estimate was subtracted to obtain the second-order statistical estimate (V) 28 according to the previous formula. In an embodiment of the present invention, the noise variance can be estimated by subtracting the reconstructed signal from the received signal 26.

[0211] Fourth, in one embodiment, the RMS delay spread 29 for each radio circuit path k can be evaluated by where the scaling factor is 2π, and V is the second-order statistical estimate 28 calculated as described above, rms_ds is the RMS delay spread 29, and ΔF is the single carrier spacing (SCS).

[0212] Fifth, in an embodiment of the present invention, PDP detection is performed by comparing the estimated RMS delay spread with the RMS delay spread threshold 29.

[0213] In an embodiment of the present invention, the apparatus includes components for determining the RMS delay spread threshold as short, medium, and long delay spreads, where the short delay spread is below 50 ns, the medium delay spread is between 50 ns and 150 ns, and the long delay spread is between 150 ns and 300 ns.

[0214] Next, an example of implementation is presented, a specific set of parameters applied in an embodiment of the present invention.

[0215] ΔF = 30000 (Hz)

[0216] 1t1r transmission mode: 1 transmission antenna and 1 receiving antenna

[0217] nDMRS = 300

[0218] D_REF = 6

[0219] nPrbMetric = 2

[0220] N = D_REF × nPrbMetric = 6x2 = 12

[0221] M = N / 2 = 6

[0222] nPRB = nDMRS / N_DMRS_PER_PRB = 300 / 6 = 50

[0223] nMeas = nDMRS – M = 300 – 6 = 294

[0224] is the previous estimate

[0225] for i = 0; i < nMeas; i++

[0226] α i = (H_RAW_DMRS i+M – H_RAW_DMRS i ) / N

[0227] End

[0228]

[0229] Next, in some embodiments of the present invention, some results can be given based on the calculation principles presented above.

[0230] Hereinafter, the demodulation performance in some embodiments of the present invention is discussed.

[0231] Figure 4a Shows a comparison of the demodulation performance between the current PDP estimation method according to the prior art and the PDP estimation method proposed according to the embodiments of the present invention, where the performance is represented by the block error rate (BLER) of the coding scheme of MCS 0. In this example, a delay spread of 30 ns is used. An automatic PDP setting is selected here. The continuous line is related to the use of the current PDP estimator, and the dashed line is related to the use of the PDP estimation proposed according to the present invention.

[0232] It can be observed that in this example, the performance improvement is not so significant.

[0233] Figure 4b Shows a comparison of the demodulation performance between the current PDP estimation method according to the prior art and the PDP estimation method proposed according to the embodiments of the present invention. In this case, the performance is represented by the block error rate (BLER) of the coding scheme of MCS24. In addition, a delay spread of 300 ns is used. An automatic PDP setting is selected here. The continuous line is related to the use of the current PDP estimator, and the dashed line is related to the use of the PDP estimation proposed according to the present invention.

[0234] It can be observed that the proposed PDP estimator performs significantly better than the current estimator implemented in production (i.e., the prior art PDP estimator) at 300ns. When the signal-to-noise ratio (SNR) is above 23-25dB, the data segment error rate begins to drop significantly below the prior art. The maximum performance improvement is seen across the entire SNR range between 36dB and 42dB.

[0235] Another advantage of the proposed method is the reduced complexity, which will be explained in detail below.

[0236] A complexity comparison between the present invention and earlier prior art approaches is discussed next.

[0237] By means of the present invention, the IFFT process can be completely avoided, thereby achieving simpler (ie, fewer) calculations, which is a significant advantage of the present invention.

[0238] Next, let's take an example where a 25-PRB user bandwidth allocation results in 150 DMRSs. The method proposed in this invention requires 300 multiplications to calculate the derivative and variance, while the current implementation in the prior art requires 2048 multiplications to process the 256-point IFFT process and 4032 multiplications to process the time-based autocorrelation calculation. The ratio 6080 / 300 = 20.26, meaning that the new method according to this invention reduces operations by a factor of approximately 20. This result represents a significant improvement in method complexity (i.e., simplification).

[0239] It is also observed that the new method according to the present invention increases linearly with the number of PRBs, while the current implementation of the prior art increases the IFFT process part by the order of n×log2(n). Therefore, the complexity ratio will increase with the increase of the bandwidth allocation by a factor of approximately log2(n).

[0240] The proposed technique avoids the direct calculation of correlations and Fourier transforms that are usually used in known methods.The method proposed according to the present invention has the following main advantages over classic prior art techniques.

[0241] First, the method according to the present invention reduces complexity; for a 25 PRB allocation, the arithmetic operations are reduced by a factor of approximately 20.

[0242] Second, it provides similar performance for low latency spread cases, but the performance gap increases with increasing latency spread.

[0243] Third, the present invention provides better performance for low bandwidth allocations.

[0244] In summary, the present invention can be implemented at a very low CPU cost.

[0245] As a conclusive overview of the technical effects and advantages of the present invention, the following is pointed out. The present invention describes suitable apparatuses and possible mechanisms for performing power delay profile (PDP) estimation. In various embodiments of the present invention, the power delay profile (PDP) estimation can be further used in a base station or a mobile station, or even in both of these elements. In other words, the present invention includes (among other things) the following technical features:

[0246] - estimating the power delay profile (PDP) by comparing the estimated RMS delay spread and a predetermined RMS delay spread threshold; and respective components for performing this step.

[0247] A process for implementing the above steps is proposed herein. When performing PDP estimation, there are several technical effects and advantages, which have been listed above; foremost and most importantly, simpler processing can be achieved, and thus, there are a very low number of CPU processing operations and lower corresponding costs. In an embodiment, when the channel information is accurately estimated, this has a further impact on the demodulation performance. Therefore, the overall performance is improved compared to prior art solutions. Another advantage is that the proposed process (and respective devices) can be implemented both on the UE side and the base station side. This is because both of these entities need to estimate the channel in order to demodulate information. Therefore, the proposed procedures and corresponding devices are clearly required.

[0248] In an embodiment, the present invention can be applied to a 5G gNodeB, i.e., the corresponding base station in a 5G system. In an embodiment, the present invention can be applied to a 5G UE, i.e., the corresponding mobile terminal. In other embodiments, the described process (i.e., method) and the described device (i.e., apparatus) can be applied to other generations of telecommunication technologies, such as 2G, 3G, 4G, or 6G telecommunication systems.

[0249] The present invention can be implemented in various network elements and various network configurations. For example, the presented process can be applied to a base station of the applied network architecture. In an embodiment, the base station can be, for example, a BTS (base station in 2G), NodeB (base station in 3G), eNB (evolved NodeB; base station in 4G), gNB (next generation NodeB; base station of 5G system), CU (central unit part of 5G gNB) or DU (distributed unit part of 5G gNB). In an embodiment, the base station can be a 6G base station. In an embodiment, the presented process can be applied to a UE of the applied network architecture. Hereinafter, specific network scenarios and the device elements applied in that network will be discussed. It means to represent the environment regarding a specific embodiment, in which the PDP estimation method and the channel estimation method according to a specific embodiment of the present invention can be implemented and realized.

[0250] Figure 5illustrates an exemplary radio network scenario according to at least some embodiments of the present invention. According to Figure 5 the example scenario, there may be a wireless communication network, which includes user equipment, UE 110, access nodes, such as base station BS120 and core network element 130.

[0251] UE 110 may include, for example, a smart phone, a cellular phone, a machine-to-machine (M2M) node, a machine type communication (MTC) node, an Internet of Things (IoT) node, an automobile, an automotive telemetry unit, a notebook computer, a tablet computer, or other types of suitable UEs or mobile stations. Generally, a UE refers to any terminal device that may be capable of wireless communication. It may be a mobile device or a fixed device. By way of example and not limitation, a UE may also be referred to as a communication device, a terminal device, a user station (SS), a portable user station, a mobile station (MS), or an access terminal (AT). In Figure 5 the example system, UE 110 may communicate wirelessly with BS120 via an air interface 115, or communicate with a cell of BS120 wirelessly. In some example embodiments, BS120 may be considered as the serving BS of UE 110. UE 110 may also communicate with multiple BS120s and / or multiple cells of BS120 simultaneously.

[0252] BS120 may be connected to the core network 130 directly or via at least one intermediate node through an interface 125. The core network 130, in turn, may be connected to another network ( Figure 5 not shown in the figure) through an interface 135, through which a connection with the other network may be obtained, for example, through a global interconnected network. BS120 may also be connected to one or more other BSs through an inter-base station interface ( Figure 5 not shown in the figure).

[0253] UE 110 can be connected to BS120 via the air interface 115. The air interface 115 between UE 110 and BS 120 can be configured according to the radio access technology RAT, which UE 110 and BS120 are configured to support. Examples of cellular RATs include Long Term Evolution (LTE), New Radio (NR), which may also be referred to as 5th Generation (5G) radio access technology. For example, in the context of LTE, BS120 may be referred to as an eNB, while in the context of NR, BS120 may be referred to as a gNB. In any case, the example embodiments are not limited to any particular radio technology. Instead, the example embodiments may be utilized in any wireless communication network (which may be cellular or non-cellular technology) operating according to 3GPP standards, IEEE standards (such as, for example, IEEE local area networks based on 802.11), or may employ some other radio technology where improved transmit and / or receive performance, and in particular improved MIMO performance, is desired between an access node such as a BS and a UE.

[0254] The requirements for data throughput in wireless communication networks are constantly increasing, necessitating the use of a wide spectrum. Thus, for example, the frequency bands of 5G (also known as NR) are currently divided into different frequency ranges. Frequency Range 1 (FR1) includes frequency bands below 6 GHz, some of which are the bands traditionally used by previous standards, but this range has been extended to cover potential new spectrum offerings up to 7125 MHz. Another range is Frequency Range 2 (FR2), which currently includes frequency bands from 24.25 gigahertz (GHz) to 52.6 GHz. Frequencies in this range and above are referred to as millimeter-wave frequencies. The millimeter-wave frequency range is attractive because the available bandwidth is higher than that of the bands in FR1, which helps to provide data rates that meet the 5G requirements.

[0255] The described embodiments are particularly beneficial for millimeter-wave frequencies, including the FR2 range, but may equally apply to FR1 or any other frequency. As previously mentioned, although applicable to any wireless network, for simplicity, 5G is mainly focused on in the examples discussed. 5G is envisioned to use more base stations or nodes (the so-called small cell concept) than the current network deployments of LTE, including macro sites operating in cooperation with smaller local area network access nodes, and may also employ various radio technologies to achieve better coverage and enhanced data rates. 5G may consist of more than one radio access technology / radio access network (RAT / RAN), each optimized for certain use cases and / or spectrums. 5G mobile communications may have a wider range of use cases and related applications, including video streaming, augmented reality, different ways of data sharing, and various forms of machine-type applications, including vehicle safety, different sensors, and real-time control.

[0256] MIMO is one of the key enabling technologies for 5G wireless technology. The underlying principle of MIMO is to use multi-path transmit and receive antennas to increase the throughput and / or reliability of data transmission. The throughput can be increased by transmitting / receiving different data streams on multiple antennas, while the reliability can be increased by using multiple antennas to transmit / receive multiple versions of the same data.

[0257] Beamforming antenna arrays play an important role in 5G implementation. Although providing high bandwidth, millimeter-wave frequencies have higher propagation losses, which vary greatly depending on the environment. The smaller wavelength at higher carrier frequencies allows for smaller antenna element sizes, which gives the opportunity to place one or more (e.g., two, three, or more) relatively large antenna arrays at the UE. This in turn leads to various challenges in maintaining the expected performance.

[0258] Considering as a non-limiting example of 2x2 MIMO, the downlink (DL) MIMO performance (e.g., in the millimeter-wave frequency range, e.g., FR2) can be achieved by using polarization splitting (co-polar and cross-polar) of dual-fed antenna arrays at the base station (also known as gNB) and / or at the UE, where each polarization corresponds to one MIMO branch. The principle of this method is to achieve high and similar antenna gain performance in the MIMO channel while maintaining a compact spatial antenna design.

[0259] When designing a dual-polarized antenna array, it is very important to achieve high cross-polar discrimination (XPD). XPD can be defined as the ratio of the co-polarized component of a specified polarization to the orthogonal cross-polarized component over the sector or beamwidth angle.

[0260] The decorrelation at the antenna array can be obtained by ensuring that each fed antenna corresponds to a single polarization, and the resulting dual-fed polarization design is orthogonal. In this way, an antenna array with high XPD at the feed point can be designed. This method will ensure the full utilization of the two MIMO channels for line-of-sight (LoS) and / or non-line-of-sight (NLoS) operations, provided that the directions of the maximum gain and the orthogonal polarization are consistent at the antenna arrays at the UE and gNB.

[0261] In addition, the high requirements for antenna gain at millimeter-wave (e.g., FR2) frequencies will reduce its radiation beamwidth, thus requiring beam steering at the antenna array (or arrays) to cover the required angular space. The beam steering ability can be achieved by using adjustable phase shifters at each element of the antenna array, whereby the beam direction can be controlled electronically (phased array) rather than mechanically.

[0262] The XPD of any antenna (or antenna array) depends on its radiation pattern and can change dynamically as a function of the angle of departure (AoD) and / or the angle of arrival (AoA). This dependence increases with changes in the radiation pattern and electrical variations in the radiation pattern. Higher antenna gain patterns result in larger XPD variations in angular space. Phase-controlled arrays also exhibit increased XPD variations in angular space.

[0263] Therefore, the physical orientation of antennas at millimeter-wave frequencies will have a much greater impact on MIMO throughput than seen at Sub-6 GHz frequencies, where decorrelation of the UE is achieved through physical isolation between two receive antennas (each with random and different radiation patterns). In contrast, millimeter-wave architectures can utilize dual-orthogonal polarization antennas (or antenna arrays) designed for equal high-gain radiation patterns.

[0264] Figure 6 An example device capable of supporting at least some embodiments of the present invention is shown. Device 400 is shown in the figure and can be or be included in, for example, UE 110. Included in device 400 is a processor 410, which can include, for example, a single-core or multi-core processor, where a single-core processor includes one processing core and a multi-core processor includes more than one processing core. Processor 410 generally can include a control device. Processor 410 can include more than one processor. Processor 410 can be a control device. The processing core can include, for example, a Cortex-A8 processing core manufactured by ARM Holdings or a Steamroller processing core manufactured by Advanced Micro Devices, Inc. Processor 410 can include at least one Qualcomm Snapdragon and / or Intel Atom processor. Processor 410 can include at least one application-specific integrated circuit, ASIC. Processor 410 can include at least one field-programmable gate array FPGA. Processor 410 can be a means for performing the method steps in device 400. Processor 410 can be configured to perform actions at least in part through computer instructions.

[0265] The processor may include circuitry, or consist of circuitry or circuit components, which are configured to perform the various stages of the method according to the embodiments described herein. As used in this application, the term "circuitry" may refer to one or more or all of the following: (a) a pure hardware circuit implementation, such as an implementation solely in analog and / or digital circuitry, and (b) a combination of hardware circuitry and software, such as applicable: (i) a combination of analog and / or digital hardware circuitry and software / firmware, and (ii) any part of a hardware processor with software (including a digital signal processor), software, and memory working together to enable a device (such as UE 110 or BS 120) to perform various functions, and (c) hardware circuitry and / or a processor, such as a microprocessor or a part of a microprocessor, which requires software (such as firmware) to operate, but may not have the software when it does not require software to operate.

[0266] This definition of circuitry applies to all uses of the term in this application, including in any claims. As a further example, as used in this application, the term circuitry also encompasses an implementation of only hardware circuitry or a processor (or processors) or a part of a hardware circuitry or a processor and its (or their) accompanying software and / or firmware. The term circuitry also encompasses, for example and if applicable to a particular claim, an element such as a baseband integrated circuit or a processor integrated circuit for a mobile device or a similar integrated circuit in a server, a cellular network device, or other computing or network devices.

[0267] Device 400 may include a memory 420. The memory 420 may include a random access memory and / or a fixed memory. The memory 420 may include at least one RAM (running memory) chip, and / or at least one ROM (read-only memory). The memory 420 may include, for example, solid-state, magnetic, optical, and / or holographic memory. The memory 420 is at least partially accessible to the processor 410. The memory 420 may be at least partially included in the processor 410. The memory 420 may be a device for storing information. The memory 420 may include computer instructions that the processor 410 is configured to execute. When computer instructions configured to cause the processor 410 to perform a specific action are stored in the memory 420, and the device 400 as a whole is configured to operate under the direction of the processor 410 using the computer instructions from the memory 420, the processor 410, and / or its at least one processing core may be considered to be configured to perform the specific action. The memory 420 may be at least partially included in the processor 410. The memory 420 may be at least partially external to the device 400 but accessible to the device 400.

[0268] Device 400 may include a transmitter 430. Device 400 may include a receiver 440. The transmitter 430 and the receiver 440 may be configured to transmit and receive information respectively according to at least one cellular or non-cellular standard. The transmitter 430 may include more than one transmitter. The receiver 440 may include more than one receiver. The transmitter 430 and / or the receiver 440 may be configured to operate according to standards such as Global System for Mobile Communications, GSM, Wideband Code Division Multiple Access, WCDMA, 5G / NR, Long Term Evolution, LTE, IS-95, Wireless Local Area Network, WLAN, Ethernet, and / or Worldwide Interoperability for Microwave Access, WiMAX, etc.

[0269] Device 400 may include a Near Field Communication, NFC, transceiver 450. The NFC transceiver 450 may support at least one NFC technology, such as NFC, Bluetooth, Wibree, or similar technologies.

[0270] Device 400 may include a User Interface, UI, 460. The UI 460 may include at least one of a display, a keyboard, a touch screen, a vibrator, a speaker, and a microphone, the vibrator being arranged to signal the user by vibrating the device 400. The user may be able to operate the device 400 through the UI 460, such as answering incoming calls, making calls or video calls, browsing the Internet, managing digital files stored in the memory 420 or an accessible cloud, via the transmitter 430 and the receiver 440, or via the NFC transceiver 450, and / or playing games.

[0271] Device 400 may include or be arranged to accept a User Identity Module 470. The User Identity Module 470 may include, for example, a Subscriber Identity Module that can be installed in the device 400, a SIM card. The User Identity Module 470 may include information identifying the subscription of the user of the device 400. The User Identity Module 470 may include information that can be used to encrypt information, for authenticating the user identity of the device 400 and / or for facilitating communication information encryption and for billing the user of the device 400 for communications made via the device 400.

[0272] The processor 410 may be equipped with a transmitter for outputting information from the processor 410 to other devices in the device 400 via wires inside the device 400. Such a transmitter may include a serial bus transmitter, for example, outputting information to the memory 420 via at least one wire for storage therein. In addition to the serial bus, the transmitter may include a parallel bus transmitter. Similarly, the processor 410 may include a receiver for receiving information in the processor 410 from other devices in the device 400 via wires inside the device 400. Such a receiver may include a serial bus receiver, which, for example, receives information from the receiver 440 via at least one wire for processing in the processor 410. In addition to the serial bus, the receiver may include a parallel bus receiver.

[0273] The device 400 may include Figure 6 further devices not shown. For example, in the case where the device 400 includes a smartphone, it may include at least one digital camera. Some devices 400 may include a rear camera and a front camera, where the rear camera is for digital photography and the front camera is for video calls. The device 400 may include a fingerprint sensor, which is arranged to at least partially authenticate the user of the device 400. In some embodiments, the device 400 lacks at least one of the above devices. For example, some devices 400 may lack the NFC transceiver 450 and / or the user identity module 470.

[0274] The processor 410, the memory 420, the transmitter 430, the receiver 440, the NFC transceiver 450, the user interface 460, and / or the user identity module 470 may be interconnected in a variety of different ways via wires inside the device 400. For example, each of the aforementioned devices may be individually connected to the main bus inside the device 400 for the devices to exchange information. However, as will be understood by those skilled in the art, this is just an example, and various ways may be selected to interconnect at least two of the above devices without departing from the scope of the present invention according to the specific implementation.

[0275] According to an embodiment of the present invention, the UE 110 may monitor and / or evaluate the current MIMO quality. For example, the UE 110 may monitor a specific parameter or a set of parameters related to the MIMO quality (in a non-limiting embodiment, this may be the MIMO rank, a parameter related to the MIMO rank, or the UE that obtains the MIMO rank value and then reports it to the gNB, or any other suitable quality metric or metric). For example, the UE 110 may apply various algorithms to estimate the current MIMO rank based on measurements of reference signals (such as DMRS). In this case, the invertibility of the MIMO channel matrix may be used, and different metrics may be considered, such as singular values, condition numbers, and any other metric that can define the invertibility of the MIMO channel matrix. The average value of various metrics or any filtered value may also be used, and sliding windows, finite impulse response filters, infinite impulse response filters, or any order (such as first, second,..., nth order) should also be considered. The frequency average of the channel estimation of the MIMO channel may also be considered, such as the narrowband on the single-sided carrier DMRS or the full frequency bandwidth allocated to the UE, with different averaging techniques.

[0276] According to an embodiment of the present invention, the UE 110 may associate the MIMO transmit and / or receive quality with the power delay profile (PDP) of the input signal. As previously described, the PDP characterizes the intensity of the signal received through the multipath channel as a function of time delay. The PDP can thus indicate multipath components based on power peaks above a threshold level over a delay interval.

[0277] In an embodiment, the UE 110 may determine whether a possible better antenna configuration at the UE 110 can improve the transmit and / or receive quality (such as the MIMO rank or any other suitable quality metric) and / or throughput based on parameters associated with the current MIMO quality and PDP.

[0278] In an embodiment, the UE may use a reference signal to scan for new component carriers and / or angular power groups (i.e., APG). This may be a channel state information reference signal (i.e., CSI-RS) (e.g., the repetition is set to "on"), a synchronization signal block (SSB), a position reference signal, or some other existing or future suitable reference signal. Additionally, a dedicated reference signal may be defined in the standard for these purposes. In an embodiment, the UE may then configure a single or multiple (two or more) antenna arrays for MIMO reception of multiple component carriers with different angles of arrival (AoA), where the selected polarization configuration may be optimized for the best MIMO rank and / or combined power.

[0279] Thus, according to at least some embodiments, the UE can determine whether it is receiving different signal components (e.g., a second component) by evaluating the radio channel of the received signal for PDP. The PDP estimation can be performed for each polarization, where each polarization can correspond to a specific MIMO branch. The PDP estimation can be done with a wide beam. The UE can thereby avoid blindly performing a time-consuming full beam scan of the entire antenna array when not needed.

[0280] According to an embodiment, the findings of the present invention are utilized to configure antennas on the UE for better MIMO transmission and / or reception, e.g., in cases where the envisioned direct dual-polarization MIMO configuration fails to achieve a sufficiently high MIMO rank in either a LoS or NLoS environment.

[0281] In at least some embodiments, the UE 110 can receive one or more reference signals from a base station (BS), such as a demodulation reference signal (DMRS), a CSI-RS, or some other reference signal. The UE 110 can further use the one or more received reference signals to evaluate the MIMO quality (in non-limiting embodiments, this can be the MIMO rank, a parameter related to or indicative of the MIMO rank, or any other suitable quality metric or metric). Alternatively, or in addition to using reference signals, the UE 110 can use some other method or combination of methods to evaluate the MIMO quality.

[0282] In an embodiment, the UE 110 can calculate its current MIMO quality (e.g., rank, a parameter indicative of the rank, or any other suitable quality metric as described above) using the current antenna configuration. Next, the UE 110 can determine whether the current MIMO quality is sufficient. For example, in an embodiment, the UE 110 can compare its current MIMO quality with a predetermined threshold. In an embodiment, the threshold can be UE-driven, for example. In some other embodiments, the threshold can be network-controlled. Embodiments of the present invention can be applied to various MIMO configurations, such as 2x2 MIMO, 4x4 MIMO, or any other MIMO configuration.

[0283] If the MIMO quality is higher than the threshold, this can indicate in an embodiment that the current MIMO quality (e.g., rank) is sufficient. The UE 110 can keep the current antenna configuration unchanged in this case. The UE 110 can further wait for the next scheduled reference signal (e.g., DMRS, CSI-RS, or some other reference signal). The comparison with the threshold can also be implemented in the form of evaluating a true / false condition.

[0284] In an exemplary embodiment, a device, such as the UE 110 or the BS 120, can include means for performing the above embodiments and any combination thereof.

[0285] In an exemplary embodiment, a computer program may be configured to execute methods according to the methods of the above embodiments and any combination thereof. In an exemplary embodiment, a computer program product embodied on a non-transitory computer-readable medium may be configured to control a processor to execute processes (i.e., methods) including the above embodiments and any combination thereof.

[0286] In an exemplary embodiment, a device, such as UE 110 or BS120, may include at least one processor and at least one memory, the at least one memory including computer program code, wherein the at least one memory and the computer program code are configured to, together with the at least one processor, cause the device to execute at least the above embodiments and any combination thereof.

[0287] Hereinafter, different exemplary embodiments will be used to describe, as an example of an access architecture to which the embodiments can be applied, a radio access architecture based on Long Term Evolution Advanced (LTE-A) or New Radio (NR, 5G), but the embodiments are not limited to such architectures. Those skilled in the art can understand that, by appropriately adjusting parameters and procedures, the embodiments can also be applied to other types of communication networks with suitable devices. Some examples of other options suitable for the system are Universal Mobile Telecommunications System (UMTS) Radio Access Network (UTRAN or E-UTRAN), Long Term Evolution (LTE, the same as E-UTRA), Wireless Local Area Network (WLAN or WiFi), Worldwide Interoperability for Microwave Access (WiMAX), Personal Communication Service (PCS), Wideband Code Division Multiple Access (WCDMA), systems using Ultra-Wideband (UWB) technology, sensor networks, Mobile Ad-hoc Networks (MANETs), and Internet Protocol Multimedia Subsystem (IMS) or any combination thereof.

[0288] In an embodiment of the present invention, a base station is introduced, which includes:

[0289] - Multiple-Input Multiple-Output (MIMO) antennas;

[0290] - Components for receiving a Demodulation Reference Signal (DMRS) vector in the frequency domain (21);

[0291] - Components for estimating an original channel estimation vector (23) from the DMRS vector received in the frequency domain (21);

[0292] - Components for calculating a derivative vector (25) of the original channel estimation vector (23);

[0293] - A component for determining the second-order statistical estimate (28) as follows, calculating the variance of the derivative vector (25) of the original channel estimate vector (23) and calculating the noise estimate variance (26), and subtracting the calculated noise estimate variance (26) from the calculated variance of the derivative vector (25) of the original channel estimate vector (23);

[0294] - A component for estimating the root mean square (RMS) delay spread (29) for each radio path based on the determined second-order statistical estimate (28); and

[0295] - A component for estimating the power delay profile (PDP) by comparing the estimated RMS delay spread with a predetermined RMS delay spread threshold.

[0296] In an embodiment of the base station, the base station includes:

[0297] - A component for calculating the variance of the noise estimate (26) by subtracting the reconstructed signal from the received signal.

[0298] In an embodiment of the base station, the component for estimating the RMS delay spread (29) for each radio path is configured to estimate the RMS delay spread (29) by the following method

[0299] - where V is the determined second-order statistical estimate (28), and ΔF is the single carrier spacing in Hz.

[0300] In an embodiment of the base station, the base station includes:

[0301] - A component for calculating the original channel estimate vector (23) for each TX / RX radio path k and each DMRS position of the allocated physical resource block (PRB) according to the following formula:

[0302] For i = 0:nDMRS–1 and k = 0:nTX*nRX–1,

[0303] H_RAW_DMRS k,i = r k,i .s k,i *

[0304] where nDMRS is the number of DMRS in all allocated PRBs, where r k,i is the sampled DMRS received at the (k,i) position, and s k,i * is the conjugate of the known DMRS symbol at the (k,i) position.

[0305] In an embodiment of the base station,

[0306] - In the case where more than one transmit antenna and one receive antenna result in k>1, the H_RAW_DMRS[i] of each radio channel path k is cascaded to generate nDMRS = nDMRS*k, thereby obtaining the original channel estimation vector (23).

[0307] In an embodiment of the base station, the base station includes:

[0308] - A component for calculating the derivative vector (25) of the original channel estimation vector (23) as follows, differentiating two DMRSs at the obtained distance of DMRS followed by DMRS.

[0309] In an embodiment of the base station, the base station includes:

[0310] - A component for calculating the derivative vector (25) of the original channel estimation vector (23) for each radio channel path k:

[0311] For i = 0:nMeas-1, α[i] = (H_RAW_DMRS[M+i] – H_RAW_DMRS[i]) / N where N is the number of resource elements between two DMRSs dedicated to measurement, where M = N / 2 = the distance between two DMRSs dedicated to measuring frequency drift, where nMeas is the number of measurements = nDMRS-M, which is also the length of the derivative vector (25) of the original channel estimation vector (23) on each radio channel path k.

[0312] In an embodiment of the base station, the base station includes:

[0313] - A component for determining the second-order statistical estimate (28) as follows, where the second-order statistical estimate (28) corresponds to the variance of the derivative vector (25) of the original channel estimation vector (23):

[0314]

[0315] where V is the determined second-order statistical estimate (28), [α] is the derivative vector (25) generated by the derivative operator (24) and is the mean value, and is the variance of the noise estimate (26).

[0316] In an embodiment of the base station, the base station includes:

[0317] - A component for performing channel estimation based on minimum mean square error (MMSE) or linear minimum mean square error (LMMSE) based on the estimated power delay distribution (PDP).

[0318] In an embodiment of the present invention, a user equipment (UE) is introduced, including:

[0319] - Multiple-Input Multiple-Output (MIMO) antenna;

[0320] - Component for receiving a Demodulation Reference Signal (DMRS) vector in the frequency domain (21);

[0321] - Component for estimating an original channel estimation vector (23) from the DMRS vector received in the frequency domain (21);

[0322] - Component for calculating a derivative vector (25) of the original channel estimation vector (23);

[0323] - Component for determining a second-order statistical estimate (28) as follows: calculating the variance of the derivative vector (25) of the original channel estimation vector (23) and calculating the noise estimation variance (a26), and subtracting the calculated noise estimation variance (26) from the variance of the derivative vector (25) of the calculated original channel estimation vector (23);

[0324] - Component for estimating the Root Mean Square (RMS) delay spread (29) for each radio channel path based on the determined second-order statistical estimate (28); and

[0325] - Component for estimating the Power Delay Profile (PDP) by comparing the estimated RMS delay spread and a predetermined RMS delay spread threshold.

[0326] In an embodiment of the UE, the UE includes:

[0327] - Component for calculating the variance of the noise estimation (26) by subtracting the reconstructed signal from the received signal.

[0328] In an embodiment of the UE, the component for estimating the RMS delay spread (29) for each radio channel path is configured to estimate the RMS delay spread (29) as follows

[0329] - where V is the determined second-order statistical estimate (28), and ΔF is the single carrier spacing in Hz.

[0330] In an embodiment of the UE, the UE includes:

[0331] - Component for calculating the original channel estimation vector (23) at each DMRS position of each TX / RX radio channel path k and the allocated Physical Resource Block (PRB) according to the following formula:

[0332] For i = 0:nDMRS–1 and k = 0:nTX*nRX–1,

[0333] H_RAW_DMRS k,i = r k,i.s k,i *

[0334] where nDMRS is the number of DMRSs among all allocated PRBs, where r k,i is the sampled DMRS received at the (k,i) position, s k,i * is the conjugate of the known DMRS symbol at the (k,i) position.

[0335] In an embodiment of the UE,

[0336] - In the case where more than one transmit antenna and one receive antenna result in k>1, the H_RAW_DMRS[i] for each radio path k is concatenated to produce nDMRS = nDMRS*k, thereby obtaining the original channel estimation vector (23).

[0337] In an embodiment of the UE, the UE includes:

[0338] - A component for calculating the derivative vector (25) of the original channel estimation vector (23), differentiating between two DMRSs at the resulting distance of DMRS following DMRS.

[0339] In an embodiment of the UE, the UE includes:

[0340] - A component for calculating the derivative vector (25) of the original channel estimation vector (23) for each radio path k by:

[0341] For i = 0:nMeas-1, α[i] = (H_RAW_DMRS[M+i] – H_RAW_DMRS[i]) / N where N is the number of resource elements between two DMRSs dedicated to measurement, where M = N / 2 = the distance between two DMRSs dedicated to measuring frequency drift, where nMeas is the number of measurements = nDMRS-M, which is also the length of the derivative vector (25) of the original channel estimation vector (23) on each radio path k.

[0342] In an embodiment of the UE, the UE includes:

[0343] - A component for determining the second-order statistical estimate (28) as follows, where the second-order statistical estimate (28) corresponds to the variance of the derivative vector (25) of the original channel estimation vector (23):

[0344]

[0345] where V is the determined second-order statistical estimate (28), [α] is the derivative vector (25) generated by the derivative operator (24) and is the mean value, and is the variance of the noise estimate (26).

[0346] In an embodiment of the UE, the UE includes:

[0347] - components for performing channel estimation based on the minimum mean square error (MMSE) or linear minimum mean square error (LMMSE) based on an estimated power delay profile (PDP).

[0348] According to yet another aspect of the present invention, there is provided an apparatus including at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to perform at least the method of the third aspect of the present invention or an embodiment thereof.

[0349] According to some further aspects, there is provided a computer program, a computer program product, a computer-readable medium or a non-transitory computer-readable medium, which includes program instructions for causing a device to perform the method according to any one of the above aspects or an embodiment thereof.

[0350] Generally speaking, various embodiments of the present invention can be implemented in hardware or a dedicated circuit or any combination thereof. Although various aspects of the present invention can be illustrated and described as block diagrams or using some other diagrams, it can be well understood that these blocks, devices, systems, techniques or methods described herein can be implemented as non-limiting examples in hardware, software, firmware, dedicated circuits or logic, general-purpose hardware or controllers or other computing devices or some combination thereof.

[0351] The term "non-transitory" used herein is a limitation on the medium itself (i.e., tangible, rather than a signal), rather than a limitation on the persistence of data storage (e.g., RAM vs. ROM).

[0352] The above description has provided a complete and informative description of exemplary embodiments of the present invention through exemplary and non-limiting examples. However, various modifications and adaptations may be apparent to those skilled in the relevant art when reading in conjunction with the accompanying drawings and the appended examples. However, all such and similar modifications taught by the present invention will still fall within the scope of the present invention as represented by the appended claims.

[0353] In other words: Although the previous embodiments illustrate the principles of the embodiments in one or more specific applications, it is obvious to those of ordinary skill in the art that many modifications can be made in terms of form, usage, and implementation details without exercising creativity and without departing from the principles and concepts of the present invention. Therefore, the present invention is not intended to be limited unless limited by the claims set forth below.

[0354] However, in other words, the present invention is not limited to the embodiments presented above, and the present invention can vary within the scope of the claims.

[0355] List of Abbreviations

[0356] BLER: Block Error Rate

[0357] CDF: Cumulative Distribution Function

[0358] CIR: Channel Impulse Response

[0359] DMRS: Demodulation Reference Signal

[0360] DS: Delay Spread

[0361] gNB: gNodeB 5G New Radio

[0362] IFFT: Inverse Fast Fourier Transform

[0363] ISI: Inter-Symbol Interference

[0364] LCR: Level Crossing Rate

[0365] LMMSE: Linear Minimum Mean Square Error

[0366] PDP: Power Delay Profile

[0367] PRB: Physical Resource Block

[0368] RE: Resource Element

[0369] RMS: Root Mean Square

[0370] SCS: Subcarrier Spacing

[0371] SNR: Signal-to-Noise Ratio

[0372] TTI: Transmission Time Interval

Claims

1. A device for communication, comprising: - A multiple-input multiple-output (MIMO) antenna; - Components for receiving a demodulation reference signal (DMRS) vector in the frequency domain (21); - Components for estimating an original channel estimation vector (23) from the DMRS vector received in the frequency domain (21); - Components for calculating a derivative vector (25) of the original channel estimation vector (23); - Components for determining a second-order statistical estimate (28), the determination being made by calculating the variance of the derivative vector (25) of the original channel estimation vector (23) and by calculating a noise estimation variance (26), and by subtracting the calculated noise estimation variance (26) from the calculated variance of the derivative vector (25) of the original channel estimation vector (23); - Components for estimating a root mean square (RMS) delay spread (29) for each radio path from the determined second-order statistical estimate (28); And - Components for estimating a power delay profile (PDP) by comparing the estimated RMS delay spread with a predetermined RMS delay spread threshold.

2. The device according to claim 1, wherein the device comprises: - Components for calculating the noise estimation variance (26) by subtracting a reconstructed signal from a received signal.

3. The device according to claim 1, wherein the components for estimating the RMS delay spread (29) for each radio path are configured to estimate the RMS delay spread (29) as follows: - where V is the determined second-order statistical estimate (28), and ΔF is the single-carrier spacing in Hz.

4. The device according to claim 1, wherein the device comprises: - Components for calculating the original channel estimation vector (23) for each transmit / receive (TX / RX) radio path k and at each DMRS position of an allocated physical resource block (PRB) according to the following formula: For i = 0:nDMRS–1 and k = 0:nTX*nRX–1, H_RAW_DMRS k,i = r k,i .s k,i * where nDMRS is the number of DMRS in all allocated PRBs, and where r k,i is the sampled DMRS received at the (k,i) position, and s k,i * is the conjugate of the known DMRS symbol at the (k,i) position.

5. The device according to claim 4, wherein - In the case where more than one transmit antenna and one receive antenna result in k>1, the H_RAW_DMRS[i] for each radio path k is concatenated to produce nDMRS = nDMRS*k, thereby obtaining the original channel estimation vector (23).

6. The device according to claim 5, wherein the device comprises: - Components for calculating the derivative vector (25) of the original channel estimation vector (23) as follows: differentiating between two DMRSs at the resulting distance between consecutive DMRSs.

7. The device according to claim 6, wherein the device comprises: - Components for calculating the derivative vector (25) of the original channel estimation vector (23) for each radio path k as follows: For i = 0:nMeas-1, α[i] = (H_RAW_DMRS[M+i] – H_RAW_DMRS[i]) / N where N is the number of resource elements between two DMRSs dedicated for measurement, and where M = N / 2 = the distance between two DMRSs dedicated for frequency drift measurement, and where nMeas is the number of measurements = nDMRS-M, which is also the length of the derivative vector (25) of the original channel estimation vector (23) on each radio path k.

8. The apparatus according to claim 7, wherein the apparatus comprises: - means for determining the second-order statistical estimate (28) as follows, where the second-order statistical estimate (28) corresponds to the variance of the derivative vector (25) of the original channel estimation vector (23): where V is the determined second-order statistical estimate (28), and [α] is the derivative vector (25) generated from the derivative operator (24), and is the mean value, and is the noise estimate variance (26).

9. The apparatus according to claim 1, wherein the apparatus comprises: - means for performing channel estimation based on minimum mean square error MMSE or linear minimum mean square error LMMSE based on the estimated power delay profile PDP.

10. An apparatus for communication, comprising: - at least one processor; and - at least one memory storing instructions which, when executed by the at least one processor, cause the apparatus to at least: transmit and / or receive wireless signals via multiple-input multiple-output MIMO antennas; receive a demodulation reference signal DMRS vector in the frequency domain (21); estimate an original channel estimation vector (23) from the DMRS vector received in the frequency domain (21); calculate the derivative vector (25) of the original channel estimation vector (23); determine a second-order statistical estimate (28), the determination being by calculating the variance of the derivative vector (25) of the original channel estimation vector (23) and by calculating the noise estimation variance (26), and by subtracting the calculated noise estimation variance (26) from the calculated variance of the derivative vector (25) of the original channel estimation vector (23); estimate the root mean square RMS delay spread (29) for each radio path from the determined second-order statistical estimate (28); and estimate the power delay profile PDP by comparing the estimated RMS delay spread with a predetermined RMS delay spread threshold.

11. The apparatus according to claim 10, wherein the apparatus is further caused to: - calculate the noise estimation variance (26) by subtracting the reconstructed signal from the received signal.

12. The apparatus according to claim 10, wherein estimating the RMS delay spread (29) of each radio path is configured to estimate the RMS delay spread (29) as follows: - where V is the determined second-order statistical estimate (28), and ΔF is the single-carrier spacing in Hz.

13. The apparatus according to claim 10, wherein the apparatus is further caused to: - calculate the original channel estimation vector (23) for each TX / RX radio path k and at each DMRS position of the allocated physical resource block PRB according to the following formula: For i = 0:nDMRS – 1 and k = 0:nTX*nRX-1, H_RAW_DMRS k,i = r k,i .s k,i * where nDMRS is the number of DMRS in all allocated PRBs, and where r k，i is the sampled DMRS received at the (k,i) position, s k，i * is the conjugate of the known DMRS symbol at the (k,i) position.

14. The apparatus according to claim 13, wherein - in the case where k>1 is generated by more than one transmit antenna and one receive antenna, the H_RAW_DMRS[i] of each radio path k is cascaded to generate nDMRS = nDMRS*k, thereby obtaining the original channel estimation vector (23).

15. The apparatus according to claim 14, wherein the apparatus is further caused to: - calculate the derivative vector (25) of the original channel estimation vector (23) as follows: by differentiating two DMRSs at the resulting distance of DMRS followed by DMRS.

16. The apparatus according to claim 15, wherein the apparatus is further caused to: - calculate the derivative vector (25) of the original channel estimation vector (23) for each radio path k as follows: For i = 0:nMeas-1, α[i] = (H_RAW_DMRS[M+i] – H_RAW_DMRS[i]) / N where N is the number of resource elements between two DMRSs dedicated to measurement, and where M = N / 2 = the distance between two DMRSs dedicated to measurement of frequency drift, and where nMeas is the number of measurements = nDMRS-M, which is also the length of the derivative vector (25) of the original channel estimation vector (23) on each radio path k.

17. The apparatus according to claim 16, wherein the apparatus is further caused to: - determine the second-order statistical estimate (28), where the second-order statistical estimate (28) corresponds to the variance of the derivative vector (25) of the original channel estimation vector (23): where V is the determined second-order statistical estimate (28), and [α] is the derivative vector (25) generated from the derivative operator (24), and is the mean value, and is the noise estimate variance (26).

18. The apparatus according to claim 10, wherein the apparatus is further caused to: - perform channel estimation based on minimum mean square error MMSE or linear minimum mean square error LMMSE based on the estimated power delay distribution PDP.

19. A method for communication, comprising the steps of: - transmitting and / or receiving wireless signals through a multiple-input multiple-output MIMO antenna; - receiving a demodulation reference signal DMRS vector in the frequency domain (21); - estimating an original channel estimation vector (23) from the DMRS vector received in the frequency domain (21); - calculating a derivative vector (25) of the original channel estimation vector (23); - determining a second-order statistical estimate (28), the determination being by calculating the variance of the derivative vector (25) of the original channel estimation vector (23) and by calculating a noise estimation variance (26), and by subtracting the calculated noise estimation variance (26) from the calculated variance of the derivative vector (25) of the original channel estimation vector (23); - estimating the root mean square RMS delay spread (29) for each radio path from the determined second-order statistical estimate (28); and - estimating the power delay distribution PDP by comparing the estimated RMS delay spread with a predetermined RMS delay spread threshold.

20. A non-transitory computer-readable medium comprising program instructions stored thereon, the program instructions for at least performing the following: - Transmitting and / or receiving wireless signals via a multiple-input multiple-output MIMO antenna; - Receiving a demodulation reference signal DMRS vector in a frequency domain (21); - Estimating an original channel estimation vector (23) from the DMRS vector received in the frequency domain (21); - Calculating a derivative vector (25) of the original channel estimation vector (23); - Determining a second-order statistical estimate (28), the determination being by calculating a variance of the derivative vector (25) of the original channel estimation vector (23) and by calculating a noise estimation variance (26), and by subtracting the calculated noise estimation variance (26) from the calculated variance of the derivative vector (25) of the original channel estimation vector (23); - Estimating a root mean square RMS delay spread (29) for each wireless circuit path from the determined second-order statistical estimate (28); and - Estimating a power delay distribution PDP by comparing the estimated RMS delay spread and a predetermined RMS delay spread threshold.

Images