Method for estimating time of arrival of received signals for ultra wide band impulse radio systems

The present invention relates generally to radio communication systems, and more particularly to determining a time of arrival of a received signal in a wireless communications network.

To estimate a distance between a transmitter and a receiver in a wireless communications network, the transmitter can send a signal to the receiver at a time t 1 . The receiver, as soon as possible, returns a reply signal to the transmitter. The transmitter measures the time of arrival (TOA) of the reply signal at time t 2 . An estimate of the distance between the transmitter and the receiver is the time for the signal to make the round trip divided by two and multiplying by the speed of light, i.e.:

D

=

Accurate time resolution of ultra wideband (UWB) signals facilitates very precise positioning capabilities based on signal TOA measurements. Although a theoretical lower bound for TOA estimation can be achieved by using maximum likelihood methods, those prior art methods are impractical due to a need for extremely high sampling rates and a large number of multipath components of the signal.

Another method is correlation-based. That method serially searches possible signal delays of a signal received via a first signal path and takes a very long time to estimate the TOA of the received signal.

Moreover, the signal received from the first path does not always have a strongest correlation output, which can result in an inaccurate TOA estimate by the prior art correlation-based methods.

Therefore, there is a need for a time of arrival estimation method that overcomes the problems of the prior art.

The invention provides a method for estimating a time of arrival of a signal received in a wireless communication system. An energy in a frame of a received signal is measured to determine a block in the frame, the block representing a coarse time of arrival of the received signal. Time-delayed versions of a template signal are combined with the block of the received signal to select a particular template signal. The particular template signal identifies a particular chip representing a fine time of arrival of the received signal.

FIG. 1 is a block diagram of frame-based time intervals;

FIG. 2 is a flow diagram of the method for estimating a time of arrival according to the invention;

FIG. 3 is a block diagram of determining a coarse time of arrival of a received signal according to the invention; and

FIG. 4 is a flow diagram for determining a fine time of arrival of a received signal according to the invention.

System Structure and Method Operation

Our invention provides a method for estimating a time of arrival (TOA) of a signal at a radio transceiver in a wireless communications network. For the purpose of this description, the transceiver estimates the TOA for a received signal. However, it should be understood that the transceiver can transmit and receive.

As shown in FIG. 1 , wireless impulse radio transceivers mark time in terms of frames 101 , blocks 102 , and chips 103 . Frames are longer than blocks, which are longer than chips. Each frame includes multiple blocks. Each block includes multiple chips.

A binary phase shift keying (BPSK) time hopping (TH) impulse radio (IR) transmitted signal can be represented by:

where w tx (t) is a transmitted ultra wideband (UWB) pulse having a duration T c , E is a transmitted pulse energy, T f is a frame time, N f is a number of pulses representing one information symbol, T s =N f T f is a symbol interval, and b └j/N f ┘ ε{+1,−1} is a binary information symbol. In order to smooth a power spectrum of the transmitted signal and to allow the channel to be shared by multiple users without causing collisions, a time-hopping (TH) sequence c j ε{0, 1, . . . , N c −1} is assigned to each transmitter, where N c is the number of chips per frame interval, that is, N c =T f =T c .

Additionally, random polarity codes, a j 's, can be included. The polarity codes are binary random variables taking values ±1 with equal probability. The values are known at the receiver.

Consider the following channel model

where a l is a channel coefficient for an l th path, L is a number of multipath components, and τ TOA is a TOA of the received signal. From equations (1) and (2), and considering effects of antennas, the received signal can be represented by:

where s rx (t) is given by

with w rx (t) denoting the received UWB pulse with unit energy. Assuming a data aided TOA estimation method using a training sequence, we consider a training sequence of b j =1∀j.

In this case, equation (4) can be expressed as

For simplicity, we assume that the signal always arrives in one frame duration, i.e., τ TOA <T f , and there is no inter-frame interference (IFI); that is, T f ≧(L+c max )T c or, equivalently, N c ≧L+c max , where c max is a maximum value of the TH sequence. Note that the assumption of τ TOA <T f does not restrict the invention. In fact, it is enough to have τ TOA <T s for the invention to work when the frame is sufficiently large and predetermined TH codes are used.

Moreover, even if τ TOA <T s , an initial energy detection can be used to determine the arrival time within a symbol uncertainty.

Two Step TOA Estimation Method

One of the most challenging tasks in time of arrival estimation is to obtain a reliable TOA estimate in an acceptable time interval under the constraint of sampling rate. In order to have a low power and low complexity receiver, we use symbol-rate sampling in our preferred embodiment.

The invention provides a method for estimating a TOA that can perform TOA estimation from symbol-rate samples in less time than prior art methods, and at chip-level resolution.

As shown in FIG. 2 , the invention estimates 200 a TOA of a received signal 201 at a particular chip 202 of a particular block 301 in a frame.

A first step according to the invention determines 300 a block representing a coarse TOA 301 of the received signal 201 based on a signal measurement of energy of the received signal.

As shown in FIG. 3 , the signal 201 received during a frame 101 with N b blocks 102 has a highest measured energy during a third block 301 . Therefore, the coarse time of arrival is determined to be during the third block 301 .

In a second step, a fine time of arrival 202 of the received signal is estimated by applying a change detection method 400 , which combines multiple, time-delayed versions of a template signal to the received signal to identify a fine time of arrival 202 . The template signals are transmitted signals corresponding to the received signal. In the preferred embodiment, the time delays are one-chip intervals.

FIG. 4 illustrates the change detection method 400 for determining the fine TOA 202 according to the invention.

As shown in FIG. 4 , after the third block 301 is determined, the received signal is combined 410 with each of multiple, time-delayed versions 406 of the template signal 405 according to the block 301 . The combining 410 produces a set of combined values 411 , one for each combination of a time-delayed template with the received signal. The template signal associated with a combined value that matches a correlation value 415 is selected 420 to identify the chip that represents the fine TOA 202 .

We express the TOA (τ TOA ) in equation (3) as follows:

τ TOA =kT c =k b T b +k c T c   (6)

where kε[0,N c −1] is the TOA in terms of the chip interval T c , T b is the block interval including B chips (T b =BT c ), and k b ε[0,N c /B−1] and k c ε[0,B−1] are integers that determine, respectively, in which block and chip the signal arriving via the first signal path.

For simplicity, the TOA is assumed to be an integer multiple of the chip duration T c . In a practical application, sub-chip resolution can be obtained by employing a delay-lock-loop (DLL) after the TOA estimation with chip-level uncertainty.

Coarse TOA Estimation

As shown in FIG. 3 , the coarse arrival time of the signal in the frame interval is determined 300 , i.e., which block in the frame represents the arrival time of the received signal, e.g. a UWB pulse. Without loss of generality, we assume that the frame time T f is an integer multiple N b of block size T b , i.e., T f =N b T b . In order to have reliable decision variables in this step, energies from N 1 different frames of the incoming signal for each block can be combined. Hence, the decision variables are expressed as

Then, k b in (6) is estimated as

In other words, we select the block with the largest signal energy.

The parameters of this step that can be optimized are the block size T b (N b ) and the number of frames N 1 , from which energy is measured.

Fine TOA Estimation from Low Rate Correlation Outputs

After determining the coarse arrival time, the second step estimates the fine TOA k c according to equation (6). Ideally, chips k c ε[0,B−1] need to be searched for fine TOA estimation, which corresponds to searching kε[{circumflex over (k)} b B,({circumflex over (k)} b +1)B−1], with chips {circumflex over (k)} b determined from equation (9).

However, in some cases, the first signal path can be in one of the blocks preceding the block having the highest energy level due to multipath effects. Therefore, instead of searching a single block, multiple blocks kε[{circumflex over (k)} b B−M 1 ,({circumflex over (k)} b +1)B−1], with M 1 ≧0, can be searched for the fine TOA in order to increase the probability of detecting of a first path of the received signal. In other words, in addition to the block with the largest signal energy, we perform an additional search over M 1 chips by applying templates with relatively long time delays.

For notational simplicity, let U={n s , n s +1, . . . , n e } denote an uncertainty region, i.e., possible TOA of the first path of the received signal, where n s ={circumflex over (k)} b B−M 1 and n e =({circumflex over (k)} b +1)B−1 are the start and end points of the uncertainty region in the frame.

In order to estimate the fine TOA, i.e., a TOA at a chip-level resolution, we consider combinations of the received signal with time-shifted versions 406 of the template signal 405 . For delay iT c , we obtain the following output:

where N 2 is the number of frames over which the combination output is obtained, and s temp (t) is the template signal given by

From the combination outputs for different delays, the object is to determine the chip in which the first signal path has arrived. According to the block interval T b and considering the multipath components in the received signal, which is typical for indoor UWB systems, we can assume that the block starts with a number of chips with noise-only components and the remaining chips start with signal plus noise components. Assuming that the statistics of the signal paths do not change in the uncertainty region, we can express different hypotheses approximately as follows:

H 0 :z i =η i ,i=n s , . . . , n f ,

H k :z i =η i ,i=n s , . . . , k− 1,

z i =N 2 √{square root over (E)}α i−k+1 +η i ,i=k, . . . ,n f ,  (12)

for kεU, where η n 's denote the i.i.d. output noise distributed as N(0, α n 2 ), and σ n 2 =N 2 N 0 /2, α 1 , . . . , α n f −k+1 are independent channel coefficients, assuming n f −ns+1≦L, and n f =n e +M 2 , with M 2 being the number of combination outputs that are considered out of the uncertainty region in order to have reliable estimates of the unknown parameters of α.

Due to very high resolution of UWB signals, it is appropriate to model the channel coefficients approximately as

where p is a probability that a channel tap arrives in a given chip, d l is the phase of α l , which is ±1 with equal probability, and |α l | is the amplitude of α l , which is modeled as a Nakagami-m distributed random variable with parameter Ω; that is,

for α≧0, m≧0.5 and Ω≧0, where Γ is the Gamma function.

According to equation (12), the TOA estimation problem can be considered as a change detection problem. Let θ denote the unknown parameters of the distribution of α; that is, θ=[p m α]. Then, the log-likelihood ratio (LLR) is determined as:

where p θ (z i |H k ) denotes the probability distribution function (p.d.f) of the correlation output under hypothesis H k , with unknown parameters given by θ, and p(z i |H 0 ) denotes the p.d.f. of the correlation output under hypothesis H 0 . Because θ is unknown, the maximum likelihood (ML) estimate can be obtained first for a given hypothesis H k and then that estimate can be used in the LLR expression. In other words, the generalized LLR approach can be taken, where the TOA estimate is expressed as

However, the ML estimate is usually complicated. Therefore, simpler estimators, such as a method of moments (MM) estimator can be used to obtain the parameters. The nth moment of a random variable X having the Nakagami-m distribution with parameter Ω is given by

Then, from the correlator outputs {z i } i n f =k+1, the MM estimates for the unknown parameters can be determined by:

with μ j denoting the jth sample moment given by

Then, the chip having the first signal path can be obtained as

where θ MM (k)=[p MM m MM ΩMM] is the MM estimate for the unknown parameters.

Let p 1 (z) and p 2 (z), respectively, denote the distributions of η and N 2 √{square root over (E)}d|α|+η. Then, the generalized LLR for the k th hypothesis is given by

and Φ denoting a confluent hypergeometric function given by [7]

Note that the p.d.f. of N 2 √{square root over (E)}d|α|+η,p 2 (z), is obtained from equations (14), (24) and the fact that d is ±1 with equal probability.

The TOA estimation rule can be expressed as

Note that equation (12) assumes that the block always starts with noise-only components, followed by the arriving signal. However, in practice, there can be cases where the first step determines a block of all noise components. By combining large number of frames; that is, by choosing a large N 1 in equation (7), the probability of selecting a block 301 having only noise can be reduced. However, a large N 1 also increases the estimation time. Hence, there is a trade-off between the estimation error and the estimation time.

In order to prevent erroneous TOA estimation when a noise-only block is selected, a one-sided test can be applied using the known distribution of the noise outputs. Because the noise outputs have a Gaussian distribution, the test compares the average energy of the outputs after the estimated change instant to a threshold.

In other words, if

1 n f - k ^ + 1 ⁢ ∑ i = k ^ n f ⁢ z i 2 < δ 1 ,

then the block is considered a noise-only block, and the two-step process is repeated.

Another improvement of the invention can be achieved by checking whether the block 301 includes signal from all paths; that is, determining if the fine TOA is actually prior to block 301 . Again, by following a one-sided test approach, we can check the average energy of the correlation outputs before the estimated TOA against a threshold and detect an all-signal block if the threshold is exceeded.

However, for very small values of the TOA estimate {circumflex over (k)}, there can be a significant probability that the signal from the first path arrives before the current observation region because the distribution of the correlation output after the first path includes both the noise distribution and the signal plus noise distribution with some probabilities as expressed by equation (13).

Hence, the test can fail even though the block is an all-signal block. Therefore, some additional correlation outputs before {circumflex over (k)} can be employed as well when calculating the average power before the TOA estimate. In other words, if

1 ⁢ ⁢ 1 k ^ - n s + M 3 ⁢ ∑ i = n s - M 3 k ^ - 1 ⁢ z i 2 > δ 2 ,

the block is considered as an all-signal block, where M 3 ≧0 additional outputs are used depending on {circumflex over (k)}.

When it is determined that the block includes all signal outputs, the TOA is expected to be in one of the previous blocks. Therefore, the uncertainty region, i.e., the observation block, is shifted backwards, and the change detection method is repeated.

Although the invention has been described by way of examples of preferred embodiments, it is to be understood that various other adaptations and modifications may be made within the spirit and scope of the invention. Therefore, it is the object of the appended claims to cover all such variations and modifications as come within the true spirit and scope of the invention.