Asymmetric double-sided two-way ranging in an ultrawideband communication system

Abstract

In an ultra-wideband (“UWB”) communication system comprising a pair of UWB transceivers, an asynchronous two-way ranging method for closely estimating the time-of-flight between the transceivers after the exchange of only 3 messages between the transceivers. In an alternate asynchronous two-way ranging method, the time-of-flight between the transceivers may be closely estimated after the exchange of only 4 messages between the transceivers.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates generally to wireless communication systems, and, in particular, to a wireless communication system adapted to perform asymmetric double-sided two-way ranging.

2. Description of the Related Art

In general, in the descriptions that follow, we will italicize the first occurrence of each special term of art which should be familiar to those skilled in the art of ultra-wideband (“UWB”) communication systems. In addition, when we first introduce a term that we believe to be new or that we will use in a context that we believe to be new, we will bold the term and provide the definition that we intend to apply to that term. In addition, throughout this description, we will sometimes use the terms assert and negate when referring to the rendering of a signal, signal flag, status bit, or similar apparatus into its logically true or logically false state, respectively, and the term toggle to indicate the logical inversion of a signal from one logical state to the other. Alternatively, we may refer to the mutually exclusive Boolean states as logic_0 and logic_1. Of course, as is well known, consistent system operation can be obtained by reversing the logic sense of all such signals, such that signals described herein as logically true become logically false and vice versa. Furthermore, it is of no relevance in such systems which specific voltage levels are selected to represent each of the logic states.

By way of example, in an ultra-wideband (“UWB”) communication system, a series of special processing steps are performed by a UWB transmitter to prepare payload data for transmission via a packet-based UWB channel. Upon reception, a corresponding series of reversing steps are performed by a UWB receiver to recover the data payload. Details of both series of processing steps are fully described in IEEE Standards 802.15.4 (“802.15.4”) and 802.15.4a (“802.15.4a”), which are expressly incorporated herein in their entirety by reference. As is known, these Standards describe required functions of both the transmit and receive portions of the system, but specify implementation details only of the transmit portion of the system, leaving to implementers the choice of how to implement the receive portion.

One or more of us have developed certain improvements for use in UWB communication systems, which improvements are fully described in the following pending applications or issued patents, all of which are expressly incorporated herein in their entirety:

“A Method and Apparatus for Transmitting and Receiving Convolutionally Coded Data”, U.S. Pat. No. 7,636,397, issued 22 Dec. 2009;

“A Method and Apparatus for Generating Codewords”, U.S. Pat. No. 7,787,544, issued 31 Jul. 2010;

“A Method and Apparatus for Transmitting and Receiving Convolutionally Coded Data”, U.S. Pat. No. 8,358,709, issued 22 Jan. 2013; and

“Receiver for Use in an Ultra-Wideband Communication System”, U.S. Pat. No. 8,437,432, issued 7 May 2013;

“Convolution Code for Use in a Communication System”, U.S. Pat. No. 8,677,224, issued 18 Mar. 2014;

“Adaptive Ternary A/D Converter for Use in an Ultra-Wideband Communication System”, U.S. Pat. No. 8,436,758, issued 7 May 2013; and

“Receiver for Use in an Ultra-Wideband Communication System”, U.S. Pat. No. 8,760,334, issued 24 Jun. 2014.

In prior art wireless communication systems, one known approach for measuring the distance from device A to device B is to measure the round trip times for an electromagnetic signal such as a group of radio pulses to get the time of flight, T_f. We can then multiply this time, T_f, by the speed of radio waves (i.e., the speed of light) to find the distance from A to B.

For example, in FIG. 2, A transmits a packet, P1, to B. B receives this signal, and a short time later, D_b, it transmits a packet, P2, back to A. The packet, P2, arrives at A a time R_a after transmitting packet P1. So we have the relationship:R_a=2T_f+D_b  [Eq. 1]So:

$\begin{array}{cc}{T}_f=\frac{\left(R_a-D_b\right)}{2}& \left[\mathrm{Eq}.2\right]\end{array}$

In practice, in A and B, the times are measured by real clocks, Clock A and Clock B, respectively. Because it may not be possible to synchronize these clocks with each other, they will run either faster or slower than an ideal clock, but if they are synchronized to a reference frequency generator, they will run at what can be assumed to be a constant frequency over the duration of the signal exchange. Let us say that Clock A and Clock B run respectively at k_a and k_b times the frequency of an ideal, true, clock. Any time measurements will be multiplied by these constants, k_a or k_b. Let us denote the actual time estimates for R_a as {circumflex over (R)}_a, for D_a as {circumflex over (D)}_a and similarly {circumflex over (R)}_b and {circumflex over (D)}_b are the estimates of R_b and D_b. Since {circumflex over (R)}_a is measured at A by A's clock:{circumflex over (R)}_a=k_aR_a  [Eq. 3a]Similarly:{circumflex over (D)}_a=k_aD_a  [Eq. 3b]{circumflex over (R)}_b=k_aR_b  [Eq. 3c]{circumflex over (D)}_b=k_aD_b  [Eq. 3d]

We could use these estimates for R_a, etc., in Eq. 2 to develop an estimate, {tilde over (T)}_f, of T_f. However, if the values of k_a or k_b are unknown, then we cannot calculate R_a, etc. But, if we use {circumflex over (R)}_a, {circumflex over (D)}_a, etc., as estimates for R_a, D_a, etc.:{circumflex over (R)}_a=2{tilde over (T)}_f+{circumflex over (D)}_b  [Eq. 4a]{circumflex over (R)}_a=2{tilde over (T)}_f+{circumflex over (D)}_b  [Eq. 4b]then we introduce a measurement error. For example, if we consider {tilde over (T)}_f1 to be an estimate of T_f:

$\begin{array}{cc}{\tilde{T}}_{f1}=\frac{\left({\hat{R}}_a-{\hat{D}}_b\right)}{2}=\frac{\left(k_a{R}_a-k_b{D}_b\right)}{2}& \left[\mathrm{Eq}.5\right]\end{array}$ then the error in this estimation is:

$\begin{array}{cc}{\tilde{T}}_{f1}-T_f=\frac{\left(\left(k_a-1\right)R_a-\left(k_b-1\right)D_b\right)}{2}& \left[\mathrm{Eq}.6\right]\end{array}$

For typical UWB systems, e.g., those based on the IEEE802.15.4a UWB physical layer, (k_a−1) and (k_b−1) are up to 20 ppm, i.e., 20×10⁻⁶, and it is important to keep the error below 100 ps)(1×10⁻¹⁰, which means the delays, e.g., {circumflex over (D)}_b, must be kept below about 5 μs. The delay includes the time to send the packet, which is typically >100 μs, so it is not practical to implement such short delays.

An improvement on this has been suggested by Fleming, et al. (U.S. Pat. No. 6,400,754), and later by Menzer, et al. (U.S. Pat. No. 7,843,379). In these schemes, a second round trip messaging round is implemented, for a total of 4 transmitted messages. The packet P2, sent from B not only acknowledges the arrival of P1 but its arrival at A initiates a delay, {circumflex over (D)}_a, in A before transmission of a third packet P3 by A. This third packet is received at B and the time for this second round trip is measured at B. In both of these schemes, it is suggested that the delay times {circumflex over (D)}_a and {circumflex over (D)}_b be made symmetric, i.e., be made to be equal or nearly equal. It is pointed out by Menzer that if this is done and if two estimates are calculated for each of the two round trips and if the two estimates are averaged, the errors almost cancel each other out and the overall error can be made to be acceptably small. Using this prior art method, the error in the average of the two estimates is:

$\begin{array}{cc}\frac{\left({\tilde{T}}_{f1}-T_f\right)+\left({\tilde{T}}_{f2}-T_f\right)}{2}=\frac{\left(\left(k_a-1\right)R_a\left(k_b-1\right)D_b\right)}{4}+\frac{\left(\left(k_b-1\right)R_b\left(k_a-1\right)D_a\right)}{4}& \left[\mathrm{Eq}.7a\right]\\ \frac{\left({\tilde{T}}_{f1}-T_f\right)+\left({\tilde{T}}_{f2}-T_f\right)}{2}=\frac{\left(\left(k_a-1\right)R_a\left(k_b-1\right)D_b\right)}{4}+\frac{\left(\left(k_b-1\right)R_b\left(k_a-1\right)D_a\right)}{4}& \left[\mathrm{Eq}.7b\right]\end{array}$ While this is certainly true, we submit that it is an unreasonable burden on systems to require the response delays to always be symmetric, i.e., approximately equal.

We submit that what is needed is an improved method and apparatus for use in the receiver of a wireless communication system to determine ranging. In particular, we submit that such a method and apparatus should provide performance generally comparable to the best prior art techniques, but allow asymmetric delays to be used without significantly reducing accuracy.

BRIEF SUMMARY OF THE INVENTION

In accordance with a preferred embodiment of our invention, we provide an asymmetric double-sided two-way ranging method using three messages, P₁, P₂, and P₃, to complete a pair of round-trip delay measurements between a first device, A, having a first clock, C_a, and a second device, B, having a second clock, C_b. In accordance with our method, in A, we transmit P₁ at a selected point in time T₀ relative to C_a; then, in B, after an unknown time-of-flight, T_f, we receive P₁ at a time T₁ relative to C_b; and, after a first transmit delay, {circumflex over (D)}_b, we transmit P₂ at a time T₂ relative to C_b. Again in A, after a first response delay, {circumflex over (R)}_a, relative to T₀, we receive P₂ at a time T₃ relative to C_a; and after a second transmit delay, {circumflex over (D)}_a, we transmit P₃ at a time T₄ relative to C_a; and then, in B, after a second response delay, {circumflex over (R)}_b, relative to T₂, we receive P₃ at a time T₅ relative to C_b; and develop an estimate, {tilde over (T)}_f, of T_f in accordance with a selected one of the following:

${\tilde{T}}_{\mathrm{fa}}=\frac{1}{2}\frac{{\hat{R}}_a{\hat{R}}_b-{\hat{D}}_a{\hat{D}}_b}{{\hat{R}}_a+{\hat{D}}_a};$ ${\tilde{T}}_{\mathrm{fb}}=\frac{1}{2}\frac{{\hat{R}}_a{\hat{R}}_b-{\hat{D}}_a{\hat{D}}_b}{{\hat{R}}_b+{\hat{D}}_b};\mathrm{and}$ ${\tilde{T}}_{\mathrm{fab}}=\frac{{\hat{R}}_a{\hat{R}}_b-{\hat{D}}_a{\hat{D}}_b}{{\hat{R}}_a+{\hat{D}}_a+{\hat{R}}_b+{\hat{D}}_b}.$

In accordance with a further embodiment of our invention, we enhance our method by developing, in B, an estimate of a distance between A and B as a function of {tilde over (T)}_f.

The methods of our invention may be embodied in computer readable code on a suitable computer readable medium such that when a processor executes the computer readable code, the processor executes the respective method.

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

Our invention may be more fully understood by a description of certain preferred embodiments in conjunction with the attached drawings in which:

FIG. 1 illustrates, in block diagram form, one embodiment of a receiver adapted for use in a UWB communication system;

FIG. 2 illustrates, in time flow form, one embodiment of a method adapted to determine ranging in accordance with our invention; and

FIG. 3 illustrates, in time flow form, one other embodiment of a method adapted to determine ranging in accordance with our invention.

In the drawings, similar elements will be similarly numbered whenever possible. However, this practice is simply for convenience of reference and to avoid unnecessary proliferation of numbers, and is not intended to imply or suggest that our invention requires identity in either function or structure in the several embodiments.

DETAILED DESCRIPTION OF THE INVENTION

Recall that we do not know the real delay times, D_a and D_b, or the round trip times, R_a and R_b. All we have are estimates which we know are in error by a much larger amount than the time of flight we are trying to estimate. Indeed, the amount of error in these estimates is impossible to know without access to a perfect clock.

We know from Eq. 4:

$\begin{array}{cc}{D}_a=\frac{{\hat{D}}_a}{k_a}& \left[\mathrm{Eq}.8a\right]\end{array}$

And similarly:

$\begin{array}{cc}{D}_b=\frac{{\hat{D}}_b}{k_b}& \left[\mathrm{Eq}.8b\right]\end{array}$

Then from Eq. 8a and Eq. 1:

$\begin{array}{cc}{R}_a=2T_f+\frac{{\hat{D}}_b}{k_b}& \left[\mathrm{Eq}.9a\right]\end{array}$

And similarly:

$\begin{array}{cc}{D}_b=\frac{{\hat{D}}_b}{k_b}& \left[\mathrm{Eq}.9b\right]\end{array}$

Then from Eq. 9a and Eq. 3a:

$\begin{array}{cc}{\hat{R}}_a=2k_a{T}_f+\frac{k_a{\hat{D}}_b}{k_b}& \left[\mathrm{Eq}.10a\right]\end{array}$

Similarly, from Eq. 9b and Eq. 3b:

$\begin{array}{cc}{\hat{R}}_b=2k_b{T}_f+\frac{k_b{\hat{D}}_a}{k_a}& \left[\mathrm{Eq}.10b\right]\end{array}$

In Eq. 10a and Eq. 10b, above, we have elapsed time periods, {circumflex over (R)}_a, {circumflex over (R)}_b, {circumflex over (D)}_a and {circumflex over (D)}_b, that we can measure using the respective, local clocks of A and B. But we have no convenient way of measuring k_a or k_b; and these clock skew errors swamp the value of T_f. We have discovered, however, that if we multiply {circumflex over (R)}_a by {circumflex over (R)}_b, the bulk of the value of the product will be the product of {circumflex over (D)}_a and {circumflex over (D)}_b. Thus, for this term in the product, the k_a and k_b constants effectively cancel each other out. Let us see where this leads:

From Eq. 10a and Eq. 10b:

$\begin{array}{cc}{\hat{R}}_a{\hat{R}}_b=4k_a{k}_b{T}_f^2+\frac{k_a{\hat{D}}_b}{k_b}\frac{k_b{\hat{D}}_a}{k_a}+2k_b{T}_f\frac{k_a{\hat{D}}_b}{k_b}+2k_a{T}_f\frac{k_b{\hat{D}}_a}{k_a}& \left[\mathrm{Eq}.11\right]\\ {\hat{R}}_a{\hat{R}}_b=4k_a{k}_b{T}_f^2+{\hat{D}}_a{\hat{D}}_b+2T_f\left(k_a{\hat{D}}_b+k_b{\hat{D}}_a\right)& \left[\mathrm{Eq}.12\right]\\ {\hat{R}}_a{\hat{R}}_b-{\hat{D}}_a{\hat{D}}_b=4k_a{k}_b{T}_f^2+2T_f\left(k_a{\hat{D}}_b+k_b{\hat{D}}_a\right)& \left[\mathrm{Eq}.13\right]\end{array}$

Then from Eq. 10a and Eq. 13:

$\begin{array}{cc}\frac{{\hat{R}}_a{\hat{R}}_b-{\hat{D}}_a{\hat{D}}_b}{{\hat{R}}_a+{\hat{D}}_a}=\frac{4k_a{k}_b{T}_f^2+2T_f\left(k_a{\hat{D}}_b+k_b{\hat{D}}_a\right)}{2k_a{T}_f+\frac{k_a{\hat{D}}_b}{k_b}+{\hat{D}}_a}& \left[\mathrm{Eq}.14\right]\end{array}$

On the left hand side, taking out 2Tf and multiplying above and below by k_b:

$\begin{array}{cc}\frac{{\hat{R}}_a{\hat{R}}_b-{\hat{D}}_a{\hat{D}}_b}{{\hat{R}}_a+{\hat{D}}_a}=2T_f{k}_b\frac{2k_a{k}_b{T}_f+k_a{\hat{D}}_b+k_b{\hat{D}}_a}{2k_a{k}_b{T}_f+k_a{\hat{D}}_b+k_b{\hat{D}}_a}& \left[\mathrm{Eq}.15\right]\end{array}$

So finally:

$\begin{array}{cc}\frac{{\hat{R}}_a{\hat{R}}_b-{\hat{D}}_a{\hat{D}}_b}{{\hat{R}}_a+{\hat{D}}_a}=2T_f{k}_b\approx 2T_f& \left[\mathrm{Eq}.16a\right]\end{array}$

Similarly:

$\begin{array}{cc}\frac{{\hat{R}}_a{\hat{R}}_b-{\hat{D}}_a{\hat{D}}_b}{{\hat{R}}_a+{\hat{D}}_b}=2T_f{k}_a\approx 2T_f& \left[\mathrm{Eq}.16b\right]\end{array}$

So now we have two possible estimates for T_f. Since k_a and k_b are very close to 1, i.e., 0.99998<k_a, k_b<1.00002, we can estimate T_f as follows:

$\begin{array}{cc}{\tilde{T}}_{\mathrm{fa}}=\frac{1}{2}\frac{{\hat{R}}_a{\hat{R}}_b-{\hat{D}}_a{\hat{D}}_b}{{\hat{R}}_a+{\hat{D}}_a}& \left[\mathrm{Eq}.17a\right]\\ {\tilde{T}}_{\mathrm{fb}}=\frac{1}{2}\frac{{\hat{R}}_a{\hat{R}}_b-{\hat{D}}_a{\hat{D}}_b}{{\hat{R}}_b+{\hat{D}}_b}& \left[\mathrm{Eq}.17b\right]\end{array}$

Note that these estimates are very close to the actual T_f because k_a and k_b are very close to one, and, in particular, their accuracy is independent of the delays employed at A and at B.

Whether a system should best use Eq. 17a or Eq. 17b would depend on which clock it expects to be more accurate. For example, if the system consisted of devices with high accuracy clocks in Role B, above, and tags with low accuracy clocks in Role A, above, then it should use Eq. 17a. If it expects neither to be more accurate than the other and it cannot readily calculate the accuracy of either, it would be most accurate to use the average result from Eq. 17a and Eq. 17b since this will always be as good as, or better than, the worst of the two. We have found by experimentation with typical values of delay and clock offset that this average can be approximated by the following:

$\begin{array}{cc}{\tilde{T}}_{\mathrm{fab}}=\frac{{\hat{R}}_a{\hat{R}}_b-{\hat{D}}_a{\hat{D}}_b}{{\hat{R}}_a+{\hat{D}}_a+{\hat{R}}_b+{\hat{D}}_b}& \left[\mathrm{Eq}.18\right]\end{array}$

Note that the value of {circumflex over (R)}_a is close to the value of {circumflex over (D)}_b, since {circumflex over (D)}_b makes up most of the time for this particular round trip measurement. This means that in the denominator of formulas Eq. 17a, Eq. 17b and Eq. 18, {circumflex over (D)}_b can be used instead of {circumflex over (R)}_a or {circumflex over (R)}_a can be used instead of {circumflex over (D)}_b without greatly reducing the accuracy. Similarly, {circumflex over (D)}_a can be used instead of {circumflex over (R)}_b and {circumflex over (R)}_b can be used instead of {circumflex over (D)}_a, leading, e.g., to:

$\begin{array}{cc}{\tilde{T}}_{\mathrm{fD}}=\frac{1}{2}\frac{{\hat{R}}_a{\hat{R}}_b-{\hat{D}}_a{\hat{D}}_b}{{\hat{D}}_a+{\hat{D}}_b}& \left[\mathrm{Eq}.19\right]\end{array}$

Systems which Benefit from Flexible Response Delays

By way of example, let us assume that one device device sends a packet, P1, to many, say 5, tags. Each tag then responds with a packet to this device in successive responses: Tag 1 responds with packet 2a after time t; Tag 2 responds with packet 2b after time 2t; Tag 3 responds with packet 2c after time 3t; Tag 4 responds after time with packet 2d 4t; and Tag 5 responds with packet 2e after time St. Now, the device closes off the round with a final packet 3. Each tag can now calculate its distance from the device after a sequence of just 7 messages. If the device had used SDS-TWR, it would be forced to have the same delay for each tag interaction and a minimum of 3 messages per tag, or 15 messages would be required. In accordance with our invention, the number of packets required is N+2 instead of 3N. By thus allowing asymmetric delays, our invention results in a significant reduction in airtime and power consumption.

Consider now a system with a mobile tag (on an asset, say) that sends a packet P1 received by many fixed devices in an infrastructure, where three of which reply in turn with packets P2a, P2b, P2c, after which the tag sends P3 received by all three devices. Then, using our invention, each of the three devices independently calculates its distance to the tag. These three distances can then be combined in an infrastructure-based-solver to locate the tag by triangulation. This allows the tag/asset to be located after sending 2 messages and receiving 3. If symmetric timings were needed, as in the prior art, then this process would require a minimum of 6 transmissions and 3 receptions to complete.

In the case of a peer-to-peer network of N mobile nodes where each node wants to find its distance to every other peer node as part of solving their relative location, then this is ½N(N−1) distance measurements. For example, for a 5 node system, this consists of 10 distinct distance measurements. With the prior art symmetric double-sided ranging, this requires 3 messages per distance measurement. In some cases, there may also be a need to send an additional message to communicate the results, which could be 1 per distance measurement or just 1 per node containing all the results which that node calculated. This is then a total of 35 to 40 messages in the 5 node example case. Using our asymmetric ranging invention, as illustrated in FIG. 2, the ranging exchanges can be combined and completed with just two transmissions per node, i.e., 10 messages in the case of the 5 node example. This is achieved as follows:

Let us define the three messages of the ranging exchange as: the Poll, P, sent by the initiator; the Response, R; and the Final message, F, that completes the two round trips; and, further, define the result communicated as a time-of-flight report message, T. If we enumerate these messages as P, R, F and T, with subscripts indicating the source and destination node addresses, and number the nodes from 1 to 5, then the 10 ranging exchanges for the 5 nodes can be achieved with the 10 messages listed in table 1:

TABLE 1
Example 5 node peer-to-peer optimized ranging solution
Message	Sender	Message
#	Node	Content	Description
1	1	P12 P13 P14 P15	Initiating Poll from node 1 to the others.
2	2	R21 P23 P24 P25	This message is the Response to node 1, and, also serves as
			node 2's initiating Poll to nodes 3, 4 and 5.
3	3	R31 R32 P34 P35	Responses to nodes 1 and 2, and also serves an Initiating Poll
			to nodes 2 and up.
4	4	R41 R42 R43 P45	Node 4's Responses to nodes 0, 1 and 2, and Poll to node 5.
5	5	R51 R52 R53 R54	Gives Response to nodes 0, 1, 2 and 3.
6	1	F12 F13 F14 F15	Final Messages completing the ranging exchange between
			node 1 and each of the other nodes 2 to 5.
7	2	T21 F23 F24 F25	TOF message reporting the node-1 to node-2 distance. And
			the Final Messages completing the ranging exchanges
			between node 2 and each of the other nodes 3 to 5.
8	3	T31 T32 F34 F35	TOF messages reporting node 3's distances to nodes 1 and 2.
			And the Final Messages completing the ranging exchanges
			between node 3 and each of the other nodes 4 and 5.
9	4	T41 T42 T43 F45	TOF messages reporting node 4's distances to nodes 1, 2 and
			3. And the Final Message completing the ranging exchange
			between node 4 and node 5.
10	5	T51 T52 T53 T54	TOF messages reporting node 5's distances to nodes 1, 2, 3
			and 4.

As can be seen, this then is a substantial saving on message traffic (saving battery power and air-time). However, the ranging exchanges are highly asymmetric: in the Table 1 example, above, the ranging exchange between node 1 and node 5 starts with the poll from node 1 at message #1, then the response from node 5 is at message #5 and the final from node 1 is message #6. If these message times are in units approximately 200 μs, then these two round trips are asymmetric with timings of approximately 800 μs versus 200 μs. This scheme then only works well when the asymmetric nature of the communications does not lead to a large ranging error. Fortunately, however, there are many other examples where flexible response delays are an advantage.

Possible Ranging Schemes

As noted above, FIG. 2 illustrates a two-way ranging exchange using three messages to complete a pair of round-trip delay measurements. In this scheme, a first device, A, transmits a first message, P₁, at a selected point in time (step 10). After an unknown time-of-flight, T_f, a second device, B, receives P₁ (step 12). After a first delay, {circumflex over (D)}_b, that is characteristic of B, B transmits a second message, P₂ (step 14). After T_f, A receives P₂ (step 16). After a second delay, {circumflex over (D)}_a, that is characteristic of A, A transmits a third message, P₃ (step 18). Finally, after T_f, B receives P₃ (step 20). Using a selected one of either Eq. 17a and Eq. 17b, or, perhaps, using only Eq. 18, we can now calculate a reasonably close estimate of T_f.

As shown in FIG. 3, our asymmetric ranging method can also be performed using four messages to comprise two pairs of round-trip measurements, where each pair is separated in time by some relatively small but otherwise arbitrary interval. In accordance with this alternative scheme, the first round-trip measurement consists of a first message, P₁ (steps 22 and 24) and a first response, P₂ (steps 26 and 28), and the later, second round-trip measurement also consists of a second message, P₃ (steps 30 and 32) and a second response, P₄ (steps 34 and 36), but sent in the opposite direction to those of the first round-trip measurement. Again, using a selected one of either Eq. 17a and Eq. 17b, or, perhaps, using only Eq. 18, we can now calculate a reasonably close estimate of T_f.

Thus it is apparent that we have provided an improved method and apparatus for use in the receiver of a wireless communication system to determine ranging. Although we have so far disclosed our invention only in the context of a packet-based UWB communication system, we appreciate that our invention is broadly applicable to other types of wireless communication systems, whether packed-based or otherwise, that perform ranging using response time stamps. Further, we submit that our invention provides performance generally comparable to the best prior art techniques but more efficiently than known implementations of such prior art techniques.

Claims

1. An asymmetric double-sided two-way ranging method using three messages, P1, P2, and P3, to complete a pair of round-trip delay measurements between a first device, A, having a first clock, Ca, and a second device, B, having a second clock, Cb, the method comprising the steps of: [1.1] in A, transmitting P1 at a selected point in time T0 relative to Ca;[1.2] in B: [1.2.1] after an unknown time-of-flight, Tf, receiving P1 at a time T1 relative to Cb; and[1.2.2] after a first transmit delay, {circumflex over (D)}b, transmitting P2 at a time T2 relative to Cb;[1.3] in A: [1.3.1] after a first response delay, {circumflex over (R)}a, relative to T0, receiving P2 at a time T3 relative to Ca; and[1.3.2] after a second transmit delay, {circumflex over (D)}a, transmitting P3 at a time T4 relative to Ca; and[1.4] in B: [1.4.1] after a second response delay, {circumflex over (R)}b, relative to T2, receiving P3 at a time T5 relative to Cb; and[1.4.2] developing an estimate, {tilde over (T)}f, of Tf in accordance with a selected one of:

2. The method of claim 1 further including the step of: [1.4.3] developing an estimate of a distance between A and B as a function of {tilde over (T)}f.

3. An asymmetric double-sided two-way ranging method using four messages, P1, P2, P3 and P4, to complete a pair of round-trip delay measurements between a first device, A, having a first clock, Ca, and a second device, B, having a second clock, Cb, the method comprising the steps of: [3.1] in A, transmitting P1 at a selected point in time T0 relative to Ca;[3.2] in B: [3.2.1] after an unknown time-of-flight, Tf, receiving P1 at a time T1 relative to Cb; and[3.2.2] after a first transmit delay, {circumflex over (D)}b, transmitting P2 at a time T2 relative to Cb;[3.3] in A: [3.3.1] after a first response delay, {circumflex over (R)}a, relative to T0, receiving P2 at a time T3 relative to Ca;[3.4] in B, transmitting P3 at a selected point in time T4 relative to Cb;[3.5] in A: [3.5.1] after an unknown time-of-flight, Tf, receiving P3 at a time T5 relative to Ca; and[3.5.2] after a second transmit delay, {circumflex over (D)}a, transmitting P4 at a time T6 relative to Ca; and[3.6] in B: [3.6.1] after a second response delay, {circumflex over (R)}b, relative to T4, receiving P4 at a time T7 relative to Cb; and[3.6.2] developing an estimate, {tilde over (T)}f, of Tf in accordance with a selected one of:

4. The method of claim 3 further including the step of: [3.6.3] developing an estimate of a distance between A and B as a function of {tilde over (T)}f.

5. An asymmetric two-way ranging circuit configured to perform the method of any preceding claim.

6. A wireless receiver comprising an asymmetric two-way ranging circuit according to claim 5.

7. A wireless transceiver comprising a wireless receiver according to claim 6.

8. A wireless communication system comprising a wireless transceiver according to claim 7.

9. A non-transitory computer readable medium including executable instructions which, when executed in a processing system, causes the processing system to perform the steps of a method according to any one of claims 1 to 4.