EP3254420A1

EP3254420A1 - Communication device and communication method for joint estimation of channel parameters

Info

Publication number: EP3254420A1
Application number: EP15787524.6A
Authority: EP
Inventors: Baicheng Xu; Peter Almers; Junshi Chen; Jianjun Chen
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2015-10-23
Filing date: 2015-10-23
Publication date: 2017-12-13
Also published as: WO2017067602A1

Abstract

A communication device (10) for determining channel parameters of a communication channel is provided. The communication device (10) comprises a gross power estimator (12), configured to estimate a gross power of a received signal, based upon channel estimates of the communication channel. Moreover, it comprises a cross correlator (13), configured to perform a cross correlation of the channel estimates and thereby determine at least three cross correlation metrics. Furthermore, it comprises a joint estimator (14), configured to jointly estimate the channel parameters of the communication channel based upon the determined gross power of the received signal and the determined at least three cross correlation metrics.

Description

Communication device and communication method for joint estimation channel parameters

TECHNICAL FIELD

The invention relates to a communication device and a communication method, especially using an OFDM communication standard.

BACKGROUND

Beyond 3G mobile communication systems, e.g., the 3GPP LTE standard specifies multiple input multiple output - MIMO, hybrid automatic repeat request - HARQ, and carrier aggregation - CA, as well as other features to provide increasing data rates. At the same time, advanced detectors, not specified by the 3GPP standard, e.g., interference rej ection combining - IRC, sphere decoder - SD, are introduced to receivers to reach an optimal performance .

Further, besides advanced detectors, proper fine tuning according to different wireless channels is also very important to reach optimal reception. However, the fine tuning relies on accurate parameter estimates of the wireless channel.

Channel parameters mentioned here refer to Doppler spread, delay spread and signal-to-noise-ratio - SNR. In pilot based orthogonal frequency division multiplex - OFDM systems, the channel estimates of a certain grid can be used for channel parameter estimation . E.g., the grid mentioned here can be the pilot grid. In 3GPP LTE, the transmitter transmits pilots known to the receiver, which are referred as dedicated reference symbols - RS . Different RSs, e.g., CRS, DMRS and MBSFN RS, are introduced in LTE for data demodulation and other purposes. Beam-forming is applied for DMRS with granularity of 1~3 resource blocks - RBs. CRS and MBSFN RS are transmitted transparently without beam-forming in the whole bandwidth. The differences among RSs make it difficult to have one generic solution of channel parameters estimation for all the RSs.

So far, the channel parameters have been estimated separately. For example a Doppler spread estimation can be based on a time direction correlation. Also a delay spread estimation can be based on a frequency direction correlation. Moreover an SNR estimation can be based on RSs. These approaches though lead to a low estimation accuracy and to a high computational complexity for estimating all channel parameters. Also the available information is not used to its full extent in these approaches. Moreover it is not possible to cover dedicated demodulation reference symbols - DMRS by these approaches. If the channel estimation is not accurate enough, a low reception quality follows.

SUMMARY

Accordingly, an object of the present invention is to provide an apparatus and method, which allow for an accurate reception of a transmitted signal while at the same time only requiring a low computational complexity.

The object is solved by the features of claim 1 for the apparatus and claim 17 for the method. Further it is solved by the features of claim 18 for the associated computer program. The dependent claims contain further developments. According to a first aspect of the invention, a communication device for determining channel parameters of a communication channel is provided. The communication device comprises a gross power estimator, configured to estimate a gross power of a received signal, based upon channel estimates of the communication channel. Moreover, it comprises a cross correlator, configured to perform a cross correlation of the channel estimates and thereby determine at least three cross correlation metrics. Furthermore, it comprises a joint estimator, configured to jointly estimate the channel parameters of the communication channel based upon the determined gross power of the received signal and the determined at least three cross correlation metrics. It is thereby possible to determine the channel parameters accurately and with a low computational complexity . According to a first implementation form of the first aspect, the channel parameters are a delay spread, a Doppler spread and a signal-to-noise-ratio of the communication channel. The joint estimator is then configured to jointly estimate the delay spread, the Doppler spread and the signal-to-noise-ratio of the communication channel based upon the determined gross power of the received signal and the determined at least three cross correlation metrics. An accurate and low computational complexity determining of the delay spread, Doppler spread and signal-to-noise-ratio is thereby possible. According to a further implementation form of the first aspect or the previous implementation form, the communication device further comprises a channel estimator, configured to determine channel estimates of the communication channel from the received signal. By this measure, it is not necessary to provide channel estimates to the communication device, but they can be determined within the communication device. According to a first implementation form of the previous implementation form, the channel estimator is configured to determine the channel estimates of the communication channel based upon known pilot symbols within the received signal. An especially accurate and low computational complexity determining of the channel estimates is thereby possible.

According to a further implementation form of the first aspect or the previous implementation forms, the gross power is comprised of a pilot symbol power and a noise power. An accurate determining of the gross power is thereby possible.

According to a further implementation form of the first aspect or the previous implementation forms, the gross power estimator is configured to estimate the gross power of the received signal according to:

wherein

P is the gross power, is an available channel estimate of a sub-carrier k in an OFDM symbol /, is an index set of OFDM symbols which have available channel estimates, is an index set of sub-carriers which have available channel estimates in OFDM symbol I, and

is a number of available channel estimates used for gross power computation. thereby possible to especially accurately determine the gross

According to a further implementation form of the first aspect or the previous implementation forms, each cross correlation metric comprises a correlation value, a time distance value and a frequency distance value. The time distance value is a distance between two OFDM symbols used for the correlation. The frequency distance value is a distance between OFDM sub-carriers used for the correlation. The cross correlator is then configured to determine the cross correlation value based on the time distance, the frequency distance and the channel estimates. The cross correlator is then furthermore configured to determine the at least three cross correlation metrics by determining the correlation value, the time distance value and the frequency distance value from the performed cross correlation. An especially accurate and low computational complexity determining of the correlation metrics is thereby possible.

According to a further implementation form of the first aspect or the previous implementation forms, the cross correlator is configured to determine the correlation value for a cross correlation metric according to

c{ _g,bg) ₌

_N(corrX,_ag,bg) ^{i|W+b_flesW}^ _{fe|feife+¾es}(if)j Hi_ikh.i + _bgik + a_g wherein

is the correlation value,

pj(corrX,a_g,bg) is the number of correlations,

is a frequency distance,

bg is a time distance,

1 is an OFDM symbol index,

k is a sub-carrier index, H,k is a conjugated complex of the available channel estimate of a sub-carrier k in an OFDM symbol Z, and

H+b_a,k+a_n the available channel estimate of a sub-carrier 1+b, in an OFDM symbol k+a,

An especially accurate and low computational complexity cross correlation can thereby be achieved.

According to a further implementation form of the first aspect or the previous implementation forms, the cross correlator is configured to determine the correlation value of a cross correlation metric according to

wherein

is the correlation value,

fl(corrX,a_g,b_g) is the number of correlations,

is a frequency distance,

b_e is a time distance,

1 is an OFDM symbol index,

k is a sub-carrier index,

Re() is an operation to get the real part of a complex value, hl,k is a conjugated complex of the available channel estimate of a sub-carrier k in an OFDM symbol I, and hi+bg,k+a_g the available channel estimate of a sub-carrier l+b_g in an OFDM symbol k+a_g.

In this alternative, a further increase in correlation accuracy can be achieved.

According to a further implementation form of the previously described two implementation forms, the cross correlator is configured to, if two determined cross correlation metrics correspond to

c( 'l^fcpl)

pj (corrX,\ a_p \,\b_p\) _ pj (corrX,a_p,b_p) ₊ ^(corrX ,a_q,b_q) ^ _{a n d} wherein the cross correlator is configured to, if two determined correlation metrics corresponding to and do not fulfill either

set the correlation value and correlation numbers as follows:

A very low computational complexity and a high correlation accuracy can thereby be achieved in this case.

According to a further implementation form of the first aspect or the previously described implementation forms, the joint estimator is configured to determine the channel parameters based upon ₅(^d°P))P

using M correlation values c^^a°'^bo ... , (^aM-i'^bM-i) _f

wherein

is a pre-defined correlation function of delay spread, dependent from the frequency normalized by delay spread, s^^deV> is the delay spread to be estimated, which is normalized in terms of sub-carrier spacing, is a pre-defined correlation function of Doppler spread, dependent from the time normalized by Doppler spread, s(^doP) is the Doppler spread to be estimated, which is normalized in terms of pilot symbol spacing, a₀ is a frequency distance of a correlation metric 0, b₀ is a time distance of a correlation metric 0, a_M_i is a frequency distance of a correlation metric M-l, b_M_i is a time distance of a correlation metric M-l,

P is the gross power,

P is a pilot signal power, and σ² is a power of noise and/or interference added to the channel estimates hl,k^■

An especially accurate joint estimation can thereby be achieved.

According to a first implementation form of the previously described implementation form, in case of the cross correlator having a determined three cross correlation metrics fulfilling ai=ao and bi=0, the joint estimator is configured to determine the channel parameters based upon:

wherein the joint estimator is configured to

- compute the Doppler spread s^^dop-⁾ based on a ratio r^⁰¹ wherein r(01) _ _c( ₀,b₀)^_c( ₁,b₁) _ _c(a₀,b₀)j_c(a₀,0) _→ p(dop) ■ s^^del^

- compute the delay spread s^^del^ based on a ratio r*-⁰²-¹ and s^^dop wherein r(02) _ _c( ₀,b₀) /_c(a₂,b₂) _f

compute the signal power P by

c(ai,i>i)

P = , and compute the noise power by

o² = P-P

compute the signal-to-noise ratio by SNR = 4 .

In this special case, a high accuracy and low complexity can be achieved.

According to a second implementation form of the second to last described implementation form, in case of the cross correlator having determined three cross correlation metrics fulfilling a2=0, the joint estimator is configured to determine the channel parameters based upon:

wherein the joint estimator is configured to - compute the delay spread s^-^deV> based on a ratio r^⁰² wherein

r(02) _ _c(a₀,b₀) /_c(a₂,b₂) _ _c(a₀,b₀) /_c(0,b₀) _→ pidel) ^ . _s{del) ^

r(02) ₌ p(del (_a . _s(del) _r

- compute the Doppler spread based on a ratio r*-⁰¹-¹ and wherein r(01) _{= c}(a₀.b₀) /_c(aifii) _r

s(dop) _{= arR min} ( ^ ^)^ ) _ ₍₀₁₎

- compute the signal power by

- compute the noise power by

σ²=Ρ-Ρ, _and

- compute the signal-to-noise-ratio by

SNR = ·

Also for this case, a very accurate and low complexity joint estimation can thereby be achieved. According to a third implementation form of the third to last described implementation form, in case the cross correlator has determined three cross correlation metrics fulfilling a₀≠ai≠a₂≠0 and b₀≠bi≠b₂≠0, the joint estimator is configured to - compute two ratios r^⁰¹ r^^{0 >} each from a pair of arbitrary correlation metrics of c^^ai,bi i = 0,1,2 , preferably as follows:

₍₀₁₎ _ cCo-fro) F^(a (^de'))F(^d"P)(b (^doP))

and c(a₀,i>o) F^^del a₀-s^^del^)F^^d°P b₀-s^^do^)

c(a₂,h₂) ^→ F(^dei)( ₂-s(^dei))F(^iioP)(b₂-s(^iioP))

- perform a maximum likelihood search to derive the delay spread ' and Doppler spread s^^dov) from the two ratios r^⁰¹ r^⁰²⁾ as follows

(del) άορ)1 ₌

] =

- derive the power P based upon the delay spread ' and the Doppler spread s^^dov) and any correlation metric of c^^am,bm m = 0,1,2 as follows:

^{P =} F^ l)(a_m-s(^del))F(^d°P)(b_m-s(^d°P))

- compute the noise power by σ² = P -P , and

- compute the signal-to-noise-ratio SNR by SNR = -^.

Also for this case, a low computational complexity high accuracy joint estimation can thereby be reached.

According to a fourth implementation form of the fourth to last implementation form, in case of the cross correlator having determined more than three cross correlation metrics, the joint estimator is configured to select three cross correlation metrics, compute an initial delay spread, an initial Doppler spread, an initial signal power, and an initial noise power, define a cost function including all correlation metrics determined by the cross correlator and determine the delay spread, the Doppler spread, the signal power and the noise power by minimizing the cost function. An especially high accuracy of the estimation can thereby be achieved.

According to an implementation form of the previous implementation form, the joint estimator is configured to minimize the cost function by two-dimensional sweeping or by two repetitions of one-dimensional sweeping. It is thereby possible to decide for focusing either on accuracy or on computational complexity. The two different options also influence the required hardware resources, e.g. memory.

According to a second aspect of the invention, a method for determining channel parameters of a communication channel is provided. The method comprises estimating in gross power of a received signal, based upon channel estimates of the communication channel. Moreover, the method comprises performing a cross correlation of the channel estimates and thereby determining at least three cross correlation metrics. Finally, the method comprises jointly estimating the channel parameters of the communication channel based upon the determined gross power of the received signal and the determined at least three cross correlation metrics. It is thereby possible to determine the channel parameters with a high accuracy and low computational complexity.

According to a third aspect of the invention, a computer program with a program code for performing the previously described method according to the second aspect when the computer program runs on a computer, is provided . It is important to note that in addition to the above described features, the communication device can furthermore comprise a receiver for receiving the signal, which then hands on the received signal to the channel estimator.

Moreover, it is important to note that the communication device furthermore can comprise further processing units, further processing the determined channel parameters. The same is true for the described communication method.

Generally, it has to be noted that all arrangements, devices, elements, units and means and so forth described in the present application could be implemented by software or hardware elements or any kind of combination thereof. Furthermore, the devices may be processors or may comprise processors, wherein the functions of the elements, units and means described in the present applications may be implemented in one or more processors. All steps which are performed by the various entities described in the present application as well as the functionality described to be performed by the various entities are intended to mean that the respective entity is adapted to or configured to perform the respective steps and functionalities. Even if in the following description or specific embodiments, a specific functionality or step to be performed by a general entity is not reflected in the description of a specific detailed element of that entity which performs that specific step or functionality, it should be clear for a skilled person that these methods and functionalities can be implemented in respect of software or hardware elements, or any kind of combination thereof. BRIEF DESCRIPTION OF DRAWINGS

The present invention is in the following explained in detail in relation to embodiments of the invention in reference to the enclosed drawings, in which

Fig. 1 shows an embodiment of the communication device in a block diagram;

Fig. 2 shows a further embodiment of the communication device;

Fig. 3 shows an exemplary cross correlation;

Fig. 4 shows first exemplary results;

Fig. 5 shows second exemplary results;

Fig. 6 shows third exemplary results;

Fig. 7 shows fourth exemplary results, and

Fig. 8 shows an embodiment of the method in a flow diagram. DESCRIPTION OF EMBODIMENTS

First we demonstrate the construction and function of different embodiments of the communication device according to the first aspect of the invention along Fig. 1 - Fig. 3. Then, along Fig. 4 - Fig. 7, benefits of the embodiments of the invention are described. Finally, along Fig. 8, details of the function of an embodiment of the communication method of the second aspect of the invention are described. Similar entities and reference numbers in different figures have been partially omitted.

Gross power is the power of channel estimates including the pure signal power and the power of noise/interference added to the channel estimates. P = Equation 2-1

Where, hik is the available channel estimate of sub-carrier k in OFDM symbol Z; is the index set of OFDM symbols which have available channel estimates;

S_i is the index set of sub-carriers which have available channel estimates in OFDM symbol I, and

]]^(GP) j__s ^he number of available channel estimates used for gross power computation .

Note that, P can be averaged over different Tx antennas and Rx antennas to get a more accurate value. P can be averaged or filtered over different SFs (subframes) to get a more accurate value.

In Fig. 1, an embodiment of the communication device 10 according to the first aspect of the invention is shown. The communication device 10 comprises a gross power estimator 12, a cross correlator 13 and a joint estimator 14. The joint estimator 14 is connected to the gross power estimator 12 and to the cross correlator 13.

Channel estimates of the communication channel, through which a signal is transmitted, are provided to the gross power estimator 12 and to the cross correlator 13. The gross power estimator 12 estimates a gross power of the received signal based upon the channel estimates of the communication channel. At the same time, the cross correlator 13 performs a cross correlation of the channel estimates and thereby determines at least three cross correlation metrics.

The estimated gross power is provided to the joint estimator 14 by the gross power estimator 12. The at least three correlation metrics are provided to the joint estimator 14 by the cross correlator 13. The joint estimator then determines the channel parameters, especially the delay spread, the Doppler spread and the signal-to-noise-ratio of the communication channel based upon the gross power of the received signal and the determined at least three cross correlation metrics. The joint estimator 14 then outputs these values. They can be used for further processing .

In Fig. 2, a more detailed embodiment of the communication device 10 according to the first aspect of the invention is shown. Here, the communication device 10 additionally comprises a channel estimator 11, which is provided with the received signal and determines the channel estimates of the communication channel through which the received signal is received. The channel estimator 11 especially uses known pilot symbols within the received signal to determine the channel estimates. The channel estimates are then provided to the gross power estimator 12 and to the cross correlator 13. Apart from this difference, Fig. 1 is identical to Fig. 2.

Not shown in Fig. 1 and Fig. 2 are further elements of the communication device 10, which are not relevant for understanding the present invention. Especially it is pointed out that the channel parameters, especially the delay spread, the Doppler spread and the signal-to-noise-ratio are further used within the communication device 10 for performing a reception of the signal.

In the following, the function of the gross power estimator 12 is explained in detail:

The channel estimates received by the gross power estimator 12 change in amplitude and phase between different time- and frequency-instances (e.g. between different OFDM signals and frequency subbands) due to the Doppler spread and delay spread.

A cross correlation determines the correlation of two channel estimates with a certain distance in both time and frequency direction. It is possible to determine a correlation value of any two channel estimates. At the same time as determining the gross power by the gross power estimator 12 a cross correlation is performed by the cross correlator 13 based on the channel estimates. The function of the cross correlator 13 is described in the following:

One kind of cross correlation can be identified by lagX = |^J , which is a pair of lags, lag a is for frequency direction (e.g. frequency subbands) and lag b is for time direction (e.g. OFDM signals) . The Lags are signed, and thereby give the direction of the cross correlation. Moreover the absolute distance of two channel estimates is determined. When the lags in both time and frequency direction are not 0, the amplitude of correlation metric includes mixed information of fading caused by delay spread and Doppler spread. Note that, it should be avoided that lagX = and lagX = [_^] are configured for cross correlation at the same time, which will not result in any additional information. Assume there are G cross correlations identified by matrix L with size

[a ^Qc-il

^ ... j_j J? each column defines one lagX , i.e., one kind of cross correlation. Then cross correlation can be done by two steps:

1. Compute the cross correlation corresponding to lagX , g = 0,1,— G - 1 , by

_N{corrX,ag,b_g) L{l \ l,l+b_gES(V} ^{_fe|_fe l,k 'H+b_a,k+a_a Equation 2 -2

Where, is the number of correlations corresponding to lagX

^■ that, c^^ae^,ba) can be averaged over different Tx antennas and different Rx antennas to get a more accurate value. c^^as^,bs^ can be averaged or filtered over different SFs to get a more accurate value.

Note that, if beam forming is applied for RS (e.g., DMRS) , the two channel estimates for cross correlation must undergo the same pre-coder for beam forming. That is, it is not allowed that the correlation window spans over different precoding matrices.

If there are only very small or no frequency/timing offset remaining in the channel estimates, Equation 2-3 could be also used instead of Equation

2-2 for the c computation .

= Re I _N(corrX,a_g,bg) L{l \ l,l + b_gES(L)} ^{fc|_fe,_{fe + £¾eSi} ⁽(^,K^f))}\ h_l* _kh_{l+b k+a} Equation 2 -3 Where, ffe(-) is the operation to get the real part of a complex value.

2. Combine the cross correlation and update L

If two cross correlation metrics corresponding to and fulfill

{dp ⁼ —⁽Xq (^Q- ⁼ ®Ί

I_J — I_J or _ the two correlation metrics are combined and the corresponding correlation numbers are accumulated.

_C(M,M) = cW₊cW Equation 2-4

2 _ pj(corrX,a_p,b_p) ₊ pj(corrX,a_q,b_q) Equation 2-5

Otherwise, = N<^,corrX-^ag-^bg⁾ .

That is, the sign of lags in L are removed and the element number in L maybe

h_h \ ,a_m,b_m≥ 0,m = 0,1,^■■■ M— 1,M≤

"0 ^" "Af lJ

G accordingly for following description of joint estimation.

In Fig. 3, an example for choosing estimates for the cross correlation is shown. In the time- frequency-diagram, different channel estimates and their respective time- and frequency-distance are shown.

M correlation metrics, c^^a°^,b° ... ,c^^aM-^1,bM-^ and one gross power P will provide M + l equations as shown in Equation.

(_r(a₀,b₀) F^(de°(a₀ ^■ s^{deV))F^{dop)(b₀ -_s ^(dop))P

c(a_M_₁,b_M_₁) ₌ p(de (_aM__l . . _s(dop))p Equation 2-6

= ρ + σ² Where,

is the pre-defined correlation function of delay spread, x is the frequency normalized by delay spread, this function could be also replaced by a look-up-table; _s(dei) ^_s ^j _e ^elay spread to be estimated, which is normalized in terms of sub-carrier spacing, i.e., the lag unit in frequency direction;

is the pre-defined correlation function of Doppler spread, x is the time normalized by Doppler spread, this function could be also replaced by a look-up-table; _s(dop) j__s i-^g D₀pp]__er spread to be estimated, which is normalized in terms of RS (OFDM) symbol spacing, i.e., the lag unit in time direction; P is the pure signal power, and

σ² is the noise/interference power added to channel estimates .

After the cross correlation metrics including the correlation value, the time distance and the frequency distance have been determined, these values as well as the estimated gross power are handed to the joint estimator 14. The detailed function of the jointed estimator 14 is explained in the following:

Theoretically, it is possible to use the above described four equations to derive the four unknown variables, delay spread, Doppler spread, signal power and noise power. Signal power and noise power will further derive p

the SNR at the very end by SNR =—. Then, three cross correlation metrics are needed which fulfill at least the following constraints: . c^^b^,b₂≠ 0

The following sections show the detailed operation for different to derive the unknown variables.

Case 1: Three cross correlation metrics available with a_l = a₀,b_i = 0:

In this case, Equation could be simplified as below,

Equation 2-7

4 steps are needed for this case.

1. Compute the Doppler spread s^^dop^ based on ratio r^⁰¹-¹

r(oi) _{= c}( ₀,b₀)/_c(a₁,b₁) _{= c}(a₀,b₀)/_c(a₀,o)_→ pidop)^ . _s(dei)) Equation 2-8

According to _r ⁽°ⁱ⁾ = f^(d°P)(&₀. _s ^(d°P)) , Doppler spread s^(dop) can be determined .

2. Compute the delay spread based on r*-⁰²-¹

r(02) _{= c}( ₀,b₀)/_c(a₂,b₂) Equation 2-9 s^(de _can be derived by Equation 2-10.

Equation 2-10

If a₂ = 0, Equation 2-10 can be simplified as, _s(del) _ _arg _mjn, F^<-^delXa_(i-x)F(^do (b₀-s(^d° ) _ (₀₂)

F(dop)(b₂-_s(dop)) Equation 2-11

If b₂ = b_0r Equation 2-10 can be simplified as,

s(del) _ _arg _mjn> fiff! _ _r(02) Equation 2-12

F ^delXa₂-x)

0 and b₂ = b_0r Equation 2-10 can be simplified as, arg min{I F^(de°( ₀ ^■ x) - r⁽⁰²⁾|} Equation 2-13 c(ai,i>i)

3. Compute the real signal power by P = Equation 2-14

F(^d^(a₀-s(^d^)) 4. Compute the noise power by o² = P— P and SNR by SNR =—

Case 2: Three cross correlation metrics available with a₂ = 0,b₂

In this case, Equation 2-6 could be simplified as below,

Equation 2-15

4 steps are needed for this case.

1. Compute the delay spread s^^deV> based on ratio r^^{0 >}

r(02) _{= c}(a₀,b₀)/_c(a₂,b₂) _{= c}(a₀,b₀)/_c(o,b₀)_→ p{dei)^_aQ . _s(de0) Equation 2-16 According to = can be determined. 2. Compute the Doppler spread based on r*-⁰¹-¹

{del) can be derived by Equation

(dop) _ _arg _mjn F(^fe')(a (^fe'))F(^d"P)(b₀-y) _(0l)

F(^dei)(a_j.s(dei))_F(dop) (b_j.y) Equation 2 -17 y

If b₁ = 0 , Equation 2-17 can be simplified as,

Equation 2 -18

If a = a₀ , Equation 2-17 can be simplified as, Equation 2 -19

If % = a₀ and ¾ = 0 , Equation 2-17 can be simplified as, s(do_P) ₌ -y) - _r ⁽⁰¹⁾ |} Equation 2 -20

3. Compute the real signal power by P = P

4. Compute the noise power by σ =P— P and SNR by SNR=— Case 3: Three Cross Correlation Metrics Available with an≠ a^≠ a₂≠ 0 and &o≠ ≠ b₂≠ 0:

This is the most difficult case. Usually it can be avoided by choosing a proper configuration for the lags.

4 steps are needed for this case.

1. Select two pairs of equations from the equations of = 0,1,2 to compute two ratios

As an example, the two ratios can be computed as below. Equation 2-21

c(a₀,i>o) f (dei) ^_ao ._s(dei)-)_F(dop) (¾₀._s(dop)-,

)( Equation 2-22

2. Derive the delay spread and Doppler spread from the two ratios above based on max-likelihood.

{del) ^(dop)

Equation 2-23

3. Derive P based on s^(dei), s^(dop) and any equation of ^(am,bm), m = 0,1,2

Equation 2-24

Compute the noise power by o² = P— P and SNR by SNR

Case 4: More Than Three Cross Correlation Metrics Available:

If more than three cross correlation metrics are available, three of the cross correlation metrics together with the estimated gross power are used to compute initial estimates of delay spread, Doppler spread, signal power and noise power.

Further on, one or more cost functions are defined to include all the determined cross correlation metrics. Fine estimates of delay spread, Doppler spread, signal power and noise power are determined by minimizing the cost function (s) .

Note that, the better the initial estimates are, the lower the resulting complexity in minimizing the cost functions are, since the searching range is much smaller.

There are several ways to define the cost function (s) and do the minimizing. Two solutions based on iteration are given below:

2D sweeping:

1. Define cost function as Equation eO,y) y)P) | ² Equation 2 -25

Where, w_m is the weight of c^^am'^bm which may depend on the quality of c^^am'^bm^> ;

P is the latest signal power estimation and P is from the initial estimate for the first iteration.

2. Update the delay spread and Doppler spread by minimizing the cost function.

[s^(de°,s^(dop)l = arg min{e(x,y)} Equation 2 -26

x,y

To reduce the complexity, x may be searched within a small range around latest delay spread and y may be searched within a small range around latest Doppler spread. 3. Update the signal power P.

Derive P based on _s(dei)_{f s}(do_P) _{and any} equation of ^(am _' ^bm), m = 0,1, ---M - 1 c(am.b_m)

^{P =} F^(a_m-s(^))F(dop)(b_m-s(dop)) Equatⁱon 2-27

Or derive P based on = 0,1,--- M - 1 _p__V -l w_mc⁽°m.½⁾ . . _{n Q} ^^_ '™=⁰F(«)(_¾,(*'))F(i»i')(i,_m,(*!')) Equatⁱon - a

4. Update the weights w_n optionally based on latest s(^d°ri and P This step though can be skipped if the weights are fixed or the weights are constant as 1, i.e., no weights.

5. Repeat the steps 2-4 if the iteration number is less than the pre-defined one. Otherwise, go to step 6

-_> ~ p

6. Compute the noise power by o =P— P and SNR by SNR=—. 2xlD sweeping:

1. Define cost functions as Equation 2-29 and Equation 2-30. e_fd{x) quation 2-29e_t Equation 2 -30

Where, w_m is the weight of c^^am,bm) , which may depend on the quality of c^^am,bm) ; P, s(^doP) and s^^del^ are the latest signal power estimate, latest Doppler spread estimate and latest Doppler spread estimate respectively. They are from the initial estimation for the first iteration. 2. Update the delay spread by minimizing the cost function of e_fd(x). r_s(^de ₌ arg minje_fd(x)} Equation 2-31 x is swept within a certain range around the latest delay spread estimates . 3. Update the Doppler spread by minimizing the cost function of e_td(y).

[s^(dop)l = arg min{e_td (y)} Equation 2-32 y

y is swept within a certain range around the latest Doppler spread estimates . 4. Update the signal power P.

Derive P based on _s(dei)_{f s}(do_P) _{and any} equation of ^(am'^bm), m = 0,1, -- - 1

c(am.b_m)

^{P =} F^(a_m-s(^))F(dop)(b_m-s(dop)) Equation 2-33

Or derive P based on _s(dei)^ _s(do_P) _{and all} equations of c^(am,bm), m = 0,1, · · · Μ - 1 _p _ _V -l w_mc(°m.½) . . _n /_i r - L_m= o f_e-Cc^dei)) ((a„_m^-s((^dd^ee^l))F_e((^dd°₀Pp))((b_m-.s,((^dd°₀Pp)))\ i-quation zi -^j$

5. Update the weights w_m optionally based on latest ' , ^v> and P This step can be skipped if the weights are fixed or the weights are constant as 1, i.e., no weights.

6. Repeat step 2~5 if the iteration number is less than the pre-defined one. Otherwise, go to step 7

7. Compute the noise power by o² = P— P and SNR by SNR = The previously shown approach improves the accuracy of determining the delay spread and Doppler spread estimation because of the introduction of multiple cross correlation metrics. Moreover it improves the SNR estimation because of the consideration of delay spread and Doppler spread.

For given channel estimates, this approach is able to exhaust the useful information. The delay spread, Doppler spread and SNR are estimated jointly, which makes good use of all the available information to improve the channel parameter estimation performance.

In the following, parameters of the tests resulting in the results Fig. 4 - Fig. 7 are given:

In the following, along Fig. 4 - 7, effects of the embodiments of the present invention are shown.

In Fig. 4, the performance of joint estimation in terms of delay spread and Doppler spread for an AWGN channel are shown. In Fig. 5, the performance of joint estimation in terms of signal-to-noise-ratio for an AWGN channel is shown. In Fig. 6, the performance of joint estimation in terms of delay spread and Doppler spread for an EVA70 channel are shown. In Fig. 7, the performance of joint estimation in terms of signal-to-noise-ratio for an EVA70 channel is shown.

It is clear from Fig. 4 - Fig. 7, that the performance of the above-described approach is significantly better, at least equal to the previously employed approach for all signal-to-noise-ratio ranges.

Finally, along Fig. 8, an embodiment of the communication method according to the second aspect of the invention is shown in a flow diagram. In a first step 100, a signal is received. In a second step 101, channel estimates are determined. The steps 100 and 101 are optional steps. In a third step 102, a gross power is estimated based upon the channel estimates. In a fourth step 104, a cross correlation is performed based upon the channel estimates. The steps 102 and 103 are performed simultaneously. Although, they are displayed here as successive steps, this is not to be understood as a chronological order. Finally, in a fifth step 104 a joint estimation of the delay spread, the Doppler spread and the signal-to-noise-ratio is performed based upon the determined gross power and the determined cross correlation.

The invention is not limited to the example specified above. Especially it is not limited to specific types of communication devices. The communication devices could be mobile telephones, base stations, routers, etc. Also, the communication device can be a measuring device such as a measuring receiver. The characteristics of the exemplary embodiments can be used in any advantageous combination.

The invention has been described in conjunction with various embodiments herein. However, other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure and the appended claims. In the claims, the word "comprising " does not exclude other elements or steps and the indefinite article "a" or "an" does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in usually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. A computer program may be stored/distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the internet or other wired or wireless communication systems.

In the following, some acronyms are explained:

Claims

1. A communication device (10) for determining channel parameters of a communication channel, comprising:

- a gross power estimator (12), configured to estimate a gross power of a received signal, based upon channel estimates of the communication channel ,

- a cross correlator (13), configured to perform a cross correlation of the channel estimates and thereby determine at least three cross correlation metrics,

- a joint estimator (14), configured to jointly estimate the channel parameters of the communication channel based upon the determined gross power of the received signal and the determined at least three cross correlation metrics.

2. The communication device (10) of claim 1,

wherein the channel parameters are a delay spread, a Doppler spread, and a signal-to-noise-ratio of the communication channel, and

wherein the joint estimator (14) is configured to jointly estimate the delay spread, the Doppler spread, and the signal-to-noise-ratio of the communication channel based upon the determined gross power of the received signal and the determined at least three cross correlation metrics .

3. The communication device (10) of claim 1 or 2, further comprising a channel estimator (11), configured to determine the channel estimates of the communication channel from the received signal.

4. The communication device (10) of claim 3, wherein the channel estimator (11) is configured to determine the channel estimates of the communication channel based upon known pilot symbols within the received signal.

5. The communication device (10) according to any of the claims 1 to 4, wherein the gross power is comprised of a pilot symbol power and a noise power .

6. The communication device (10) according to any of the claims 1 to 5, wherein the gross power estimator (12) is configured to estimate the gross power of the received signal according to:

wherein

P is the gross power, is an available channel estimate of a sub-carrier k in an OFDM symbol /, is an index set of OFDM symbols which have available channel estimates, is an index set of sub-carriers which have available channel estimates in OFDM symbol I, and is a number of available channel estimates used for

gross power computation.

7. The communication device (10) according to any of the claims 1 to 6, wherein each cross correlation metric comprises a correlation value, a time distance value and a frequency distance value, and

wherein the time distance value is a distance between two OFDM symbols used for the correlation,

wherein the frequency distance value is a distance between OFDM subcarriers used for the correlation,

wherein the cross correlator (13) is configured to determine the cross correlation value based on the time distance, the frequency distance, and the channel estimates, and

wherein the cross correlator (13) is configured to determine the at least three cross correlation metrics by determining the correlation value, the time distance value, and the frequency distance value from the performed cross correlation.

8. The communication device (10) according to any of the claims 1 to 7, wherein the cross correlator (13) is configured to determine the correlation value for a cross correlation metric according to:

_N(corrX,a_g,bg)∑{l \ l,l_+bge_S(^L)} ^^ k.k+ igES^) ^htk^hl + b_g,k + a_g wherein

c (a_g,b_g) j__{s corre}i_ation value,

_N (corrx, cig.bg) _{is the num}]3_{er Q}f correlations,

a_g is a frequency distance,

b_g is a time distance,

1 is an OFDM symbol index,

k is a sub-carrier index, is a conjugated complex of the available channel estimate of a sub-carrier k in an OFDM symbol I , and hi+bg,k+a_g the available channel estimate of a sub-carrier l+b_g in an OFDM symbol k+a_g.

9. The communication device (10) according to any of the claims 1 to 7, wherein the cross correlator (13) is configured to determine the correlation value of a cross correlation metric according to:

_C ⁽¾.¾⁾ = Re (_w(corri_afl¾)∑{i\_tll+bg€sW)∑{_{fe|feife+¾es}W} ¾.¾₊¾,/.₊¾) , wherein

c(^a _S'b_e) j__s ^he correlation value,

N(^corrX'^ag'^bg) i_s the number of correlations,

a_g is a frequency distance,

b_g is a time distance,

1 is an OFDM symbol index,

k is a sub-carrier index,

Re ( ) is an operation to get the real part of a complex

value, h_l*_k is a conjugated complex of the available channel estimate of a sub-carrier k in an OFDM symbol Z, and hi+b_g,k+a_g the available channel estimate of a sub-carrier l+b_g in an OFDM symbol k+a_g.

10. The communication device (10) according to claim 8 or 9, wherein the cross correlator (13) is configured to, if two determined cross correlation metrics corresponding to and fulfill either or

combine the two correlation metrics and accumulate the corresponding correlation numbers as follows: _{a n d} wherein the cross correlator (13) is configured to, if two determined cross correlation metrics corresponding to and do not fulfill either

(dp CLq

[ bp = b,

set the correlation valued and correlation numbers as follows: c(l¾l_'l*el = = N(^c»^rrXag ^bg) _

11. The communication device (10) according to any of the claims 1 to 10,

wherein the joint estimator (14) is configured to determine the channel parameters based upon:

using M correlation values c^^a°'^bo ... , (^aM-i'^bM-i) _f

wherein

is a pre-defined correlation function of delay spread, dependent from the frequency normalized by delay spread, s^^deV> is the delay spread to be estimated, which is normalized in terms of sub-carrier spacing, is a pre-defined correlation function of Doppler spread, dependent from the time normalized by Doppler spread, s(^d°P) is the Doppler spread to be estimated, which is normalized in terms of pilot symbol spacing, a₀ is a frequency distance of a correlation metric 0, b₀ is a time distance of a correlation metric 0, a_M-i is a frequency distance of a correlation metric M-l, b_M-i is a time distance of a correlation metric M-l,

P is the gross power,

P is a pilot signal power, and σ² is a power of noise and/or interference added to the channel estimates

12. The communication device (10) according to claim 11,

wherein, in case the cross correlator (13) has determined three cross correlation metrics fulfilling a₁ = a_o,b₁ = 0, the joint estimator (14) is configured to determine the channel parameters based upon:

wherein the joint estimator (14) is configured to

- compute the Doppler spread s^^dop^ based on a ratio r^⁰¹ wherein

r(01) _ _c( ₀,b₀)^_c( ₁,b₁) _ _c( ₀,b₀)j_c(a₀,0)_→ F^^dop) [b_Q■ s(^de')) compute the delay spread s^^del^ based on a ratio r*-⁰²-¹ and s^^dop wherein r(02) _ _c (a₀,b₀) /_c(a₂,b₂) ^

compute the signal power P by

c(ai,i>i)

F(^d^ (a₀ -s(^d^)) and compute the noise power by

o² = P-P

compute the signal-to-noise ratio by

SNR =^■½

13. The communication device (10) according to any of claims 11-12 wherein, in case the cross correlator (13) has determined three cross correlation metrics fulfilling a₂ = 0,b₂ = b_0r the joint estimator (14) is configured to determine the channel parameters based upon:

wherein the joint estimator (14) is configured to

- compute the delay spread s^-^deV> based on a ratio r^⁰² wherein

r(02) _ _c (a₀,b₀) /_c (a₂,b₂) _ _c(a₀,b₀) /_c (0,b₀) _→ p idel) ^ . _s{del) ^ r(02) ₌ p(del (_a . _s (del) _r

- compute the Doppler spread based on a ratio r*-⁰¹-¹ and wherein r(01) _{= c}(a₀.b₀)/_c(aifii) _r

- compute the signal power by

- compute the noise power by

σ² = Ρ - Ρ , _and

- compute the signal-to-noise-ratio by SNR = -^.

14. The communication device (10) according to any of claim 11-13 wherein, in case the cross correlator (13) has determined three cross correlation metrics fulfilling a₀≠ <¾≠ a₂≠ 0 and b₀≠ b_x≠ b₂≠ 0, the joint estimator (14) is configured to

- compute two ratios r^⁰¹ r^^{0 >} each from a pair of arbitrary correlation metrics of c^^ai,bi i = 0,1,2 , preferably as follows:

₍₀₁₎ _ cCo-fro) F^ (a₀-5(^de'))F(^d"P) (b₀-5(^doP))

c(^a _l<⁶ _l) F(dei)

- perform a maximum likelihood search to derive the delay spread

Doppler spread s^^dov) from the two ratios r^⁰¹ r^⁰²⁾ as follows

- compute the noise power by σ² = P -P , and - compute the signal-to-noise-ratio SNR by

15. The communication device (10) according to any of claims 11-14 wherein, in case the cross correlator (13) has determined more than three cross correlation metrics, the joint estimator (14) is configured to

- select three cross correlation metrics,

- compute an initial delay spread, an initial Doppler spread, an initial signal power, and an initial noise power,

- define a cost function including all correlation metrics determined by the cross correlator (13) , and

- determine the delay spread, the Doppler spread, the signal power, and the noise power by minimizing the cost function.

16. The communication device (10) according to claim 15,

wherein the joint estimator (14) is configured to minimize the cost function by two-dimensional sweeping or by two repetitions of one-dimensional sweeping.

17. A Method for determining channel parameters of a communication channel, comprising:

- estimating (102) a gross power of a received signal, based upon channel estimates of the communication channel,

- performing (103) a cross correlation of the channel estimates and thereby determining at least three cross correlation metrics,

- jointly estimating (104) the channel parameters of the communication channel based upon the determined gross power of the received signal and the determined at least three cross correlation metrics.

18. A computer program with a program code for performing the method according to claim 17 when the computer program runs on a computer.