Field

The invention relates to apparatuses and methods for determining a direction of a wireless transmitter.

Background

Estimating the exact direction and proximity of wireless devices is a challenging problem. Directional antennas can be used to infer the direction of different wireless devices. Also, signal strength measurements can be used to infer the proximity of those devices.

However, multipath propagation and interference often cause significant signal strength fluctuations, thereby significantly impacting the accuracy of both direction and proximity detection. Summary

In a first aspect, this specification describes apparatus comprising an array of directional antennas each orientated in a different direction, at least one processor, and at least one memory, the at least one memory having computer-readable code stored thereon, the computer-readable code, when executed, causing the apparatus to measure a strength of a radio signal received at each of the directional antennas, to compare the strength of the radio signal received at each of the directional antennas with a first signal strength threshold, to identify a first set of at least one directional antennas from the array for which the strength of the received radio signal satisfies a first criterion with respect to the first signal strength threshold, and to determine a direction of a radio signal transmitter based on the direction in which at least one of the first set of directional antennas is orientated.

In a second aspect, this specification describes a method comprising measuring a strength of a radio signal received at each directional antenna of an array of directional antennas, each directional antenna in the array being orientated in a different direction, comparing the strength of the radio signal received at each of the directional antennas with a first signal strength threshold, identifying a first set of at least one directional antennas from the array for which the strength of the received radio signal satisfies a first criterion with respect to the first signal strength threshold, and determining a direction of a radio signal transmitter based on the direction in which at least one of the first set of directional antennas is orientated. In a third aspect, this specification describes computer readable code which, when executed by computing apparatus, causes the computing apparatus to perform a method according to the second aspect.

In a fourth aspect, this specification describes a non-transitory computer-readable storage medium having stored therein computer-readable code, which, when executed by computing apparatus, causes the computing apparatus to measure a strength of a radio signal received at each directional antenna of an array of directional antennas, each directional antenna in the array being orientated in a different direction, to compare the strength of the radio signal received at each of the directional antennas with a first signal strength threshold, to identify a first set of at least one directional antennas from the array for which the strength of the received radio signal satisfies a first criterion with respect to the first signal strength threshold, and to determine a direction of a radio signal transmitter based on the direction in which at least one of the first set of directional antennas is orientated.

In a fifth aspect, this specification describes apparatus configured to measure a strength of a radio signal received at each directional antenna of an array of directional antennas, each directional antenna in the array being orientated in a different direction, to compare the strength of the radio signal received at each of the directional antennas with a first signal strength threshold, to identify a first set of at least one directional antennas from the array for which the strength of the received radio signal satisfies a first criterion with respect to the first signal strength threshold, and to determine a direction of a radio signal transmitter based on the direction in which at least one of the first set of directional antennas is orientated.

In a sixth aspect, this specification describes apparatus comprising at least one processor, and at least one memory, the at least one memory having computer-readable code stored thereon, the computer-readable code, when executed, causing the apparatus to measure a strength of a radio signal received at each directional antenna of an array of directional antennas, each directional antenna in the array being orientated in a different direction, to compare the strength of the radio signal received at each of the directional antennas with a first signal strength threshold, to identify a first set of at least one directional antennas from the array for which the strength of the received radio signal satisfies a first criterion with respect to the first signal strength threshold, and to determine a direction of a radio signal transmitter based on the direction in which at least one of the first set of directional antennas is orientated.

In a seventh aspect, this specification describes apparatus comprising means for measuring a strength of a radio signal received at each directional antenna of an array of directional antennas, each directional antenna in the array being orientated in a different direction, means for comparing the strength of the radio signal received at each of the directional antennas with a first signal strength threshold, means for identifying a first set of at least one directional antennas from the array for which the strength of the received radio signal satisfies a first criterion with respect to the first signal strength threshold, and means for determining a direction of a radio signal transmitter based on the direction in which at least one of the first set of directional antennas is orientated.

In an eighth aspect, this specification describes a portable computing device comprising apparatus according to any one of the first, fifth, sixth and seventh aspects.

Brief Description of the Drawings

Example embodiments will now be described with reference to the accompanying drawings, in which:

Figure 1 is a schematic illustration of an example apparatus for determining the direction of a wireless transmitter;

Figure 2 is a schematic illustration of an array of directional antennas which forms part of the apparatus of Figure 1;

Figure 3A is a flow chart illustrating a method which may be performed by the apparatus of Figure 1;

Figure 3B is a schematic of the array of directional antennas which illustrates an aspect of the method of Figure 3A;

Figure 4A is a flow chart illustrating another method which may performed by the apparatus of Figure 1;

Figure 4B is a schematic of the array of directional antennas which illustrates an aspect of the method of Figure 4A;

Figure 5A is a flow chart illustrating another method which may be performed by the apparatus of Figure 1;

Figure 5B is a schematic of the array of directional antennas which illustrates an aspect of the method of Figure 5A; Figure 6 is a flowchart illustrating an aspect of the methods of Figures 3A, 4A and 5A; Figure 7 is a flowchart illustrating an alternative aspect of the methods of Figures 3A, 4A and 5A; and

Figures 8A and 8B are schematic illustrations of an experimental setup as used by the inventors.

Detailed Description

In the above-mentioned drawings and below-described embodiments, like reference numerals refer to like elements throughout.

Figure 1 is a schematic illustration of an apparatus 1 for determining the direction of a wireless radio transmitter (not shown). The apparatus 1 may be part of a larger device such as a mobile phone, a laptop, a PDA or a tablet computer. The apparatus 1 comprises a plurality of directional antennas 2, a controller 12 and at least one non-transitory computer-readable memory medium 14. The controller 12 is operable to control the operation of the plurality of directional antennas 2 under the control of computer-readable code 14A stored on the at least one memory medium 14. The controller 12 comprises at least one processor 12A which is operable to execute the computer-readable code 14A. The controller 12 may also comprise one or more application-specific integrated circuits (not shown). The at least one memory medium 14 may comprise one or more distinct memory mediums, such as but not limited to ROM, RAM and flash memory. Figure 2 is a schematic illustration of the plurality of directional antennas 2 of Figure 1. The plurality of directional antennas 2, which may be referred to as an array, comprises N directional antennas 20-1, 20-2, 20-3 ... 20-N each orientated in a different direction. The N directional antennas are distributed along a path. Preferably, the N directional antennas are distributed along a closed path. In the example of Figure 2, the closed path is circular. The N directional antennas 20-1, 20-2, 20-3 ... 20-N are arranged around a reference point 22. In this example, the array 2 comprises 18 antennas 20-1, 20-2... 20- 18. As such, in this example, N equals 18. Preferably, the directional antennas 20-1, 20- 2, 20-3 ... 20-N are equally spaced around the reference point 22. In other words, the angular separations between each antenna 20-1 and its two neighbouring antennas 20-N, 20-2 are equal to one another. Neighbouring antennas may also be referred to as adjacent antennas or successive antennas. In this example, the directional antennas 20- l, 20-2 ... 20-N are arranged in a circle with the reference point being at the centre of the circle 22. The plurality of directional antennas 2 may be provided in a common plane. Each directional antenna 20-1, 20-2, 20-3 ... 20-N is oriented in a radial direction relative to the reference point 22. In other words, the beam of each directional antenna 20-1, 20-2, 20-3 ... 20-N is oriented directly away from the reference point 22. Each of the directional antennas 20-1, 20-2, 20-3 ... 20-N is controllable by the controller 12. The controller 12 is operable, under the control of the computer-readable code 14A, to process radio signals received at each of the directional antennas 20-1, 20-2, 20-3 ... 20- N individually.

Figure 3A is a flowchart illustrating a first example of a method for determining the position of a wireless radio transmitter. The method of Figure 3A may be caused to be performed by the controller 12 under the control of the computer-readable code 14A. In step S3.1, the controller 12 switches on each of the directional antennas 20-1 ... 20-N. When switched on, the directional antennas 20-1 ... 20-N are operable to receive data packets in the form of radio signals transmitted by the radio transmitter 30 (see Figure 3B), the direction of which is being determined. The controller 12 may switch on each of the directional antennas sequentially or simultaneously. If they are switched on sequentially, it can be said that the antennas are rotated between.

In step S3.2, the controller 12 determines for one of the directional antennas 20-1 ... 20- N whether the number of data packets received by the directional antenna 20-1 ... 20-N is above a threshold number of data packets.

If a negative determination is reached in step S3.2, the operation proceeds to step S3.3 in which the antenna 20-1 ... 20-N which has received fewer than the threshold number of packets is disregarded. After this, at step 83.4a, the next antenna is selected and the method returns to step S3.2.

If a positive determination is reached in step S3.2, the operation proceeds to step S3.4. In step S3.4, an indication, or measure, of the radio signal strength (RSSI) of the data packets received by the antenna 20-1 ... 20-N is determined. The RSSI may be determined using the signal strength of the single data packet having the largest signal strength. Alternatively, it may be determined using the median signal strength for all data packets received at that antenna 20-1 ...20-N. Alternatively, the RSSI for a particular antenna may be determined using the mean signal strength for all data packets received by that directional antenna 20-1 ...20-N.

When determining the RSSI for each antenna 20-1 ...20-N, the controller 12 may be operable to disregard any data packets for which the signal strength is erratic. Data packets having an erratic signal strength may be a result of moving objects nearby or due to movement of the apparatus 1 or the wireless transmitter 30 during reception or transmission of the data packet. Erratic packets can be filtered and discarded using simple techniques. For example, the standard deviation of the signal strengths of all data packets received by a particular antenna 20-1 ...20-N may be calculated and any packets which are more than a certain number of standard deviations away from the mean signal strength may be disregarded for the purpose of calculating the RSSI of the antenna 20-1 ...20-N. In step S3.5, it is determined whether the last antenna has been processed. If not, at step 83.5a, the next antenna is selected and then steps S3.2 to S3.4 are repeated for this next directional antenna 20-1 ...20-N.

When step S3.5 reveals that the RSSI has been calculated for each antenna 20-1 ...20-N for which a positive determination was produced in step S3.2, the method proceeds to steps S3.6, S3.6a and S3.7. In these steps, the controller 12 determines which of the antennas 20-1 ...20-N have an RSSI satisfying a particular criterion. In this example, the criterion is that the value of the RSSI must be within a determined confidence interval of the RSSI of the antenna 20-1 ...20-N which has the highest RSSI. This value of the highest RSSI is hereafter referred to as MAX_RSSI.

More specifically, in step S3.6, the controller 12 determines the MAX_RSSI. Next, in step S3.6a, the controller 12 compares the RSSI for each antenna with a threshold defined by MAX_RSSI minus a particular confidence interval. Based on the comparison of step S3.6a, the controller 12 identifies in step S3.7 those antennas 20-1 ...20-N for which the RSSI is within the confidence interval of MAX_RSSI (i.e. is above the threshold). The confidence interval may be for example 3-4 decibel milliwatts. The exact value of the confidence interval may be determined by the controller 12 empirically. The confidence interval may be higher for situations in which the variation in RSSIs of the antennas is larger and may be smaller in situations wherein the variation in RSSIs is smaller. Those antennas 20-1 ...20-N for which the RSSI satisfies this specific criterion are hereafter referred as the HIGH_SET antennas.

Next in step S3.8, one or more clusters, or groups, of HIGH_SET antennas is identified based on the relative locations of the HIGH_SET antennas. Generally, a cluster of antennas consists of a succession of adjacent antennas for which the criterion is satisfied (in this example, antennas which are members of HIGH_SET). However, in some examples, a cluster may be interspersed with antennas for which the criterion is not satisfied. By allowing clusters of antennas for which the criterion is satisfied to be interspersed with antennas for which it is not satisfied, the error in the direction determination caused by data packets having an erratic RSSI may be reduced. In some examples, a cluster may comprise just a single antenna. Figures 6 and 7 illustrate in more detail examples of processes by which the controller 12 identifies clusters of antennas. These are described below.

In step S3.9, it is determined if the number of clusters returned in step S3.8 is equal to one. If the determination is negative, the operation returns to step S3.1. If the determination is positive, the operation proceeds to step S3.10. In step S3.10, the direction D of the wireless transmitter 30 is determined based on the directions in which each of the directional antennas in the identified cluster is orientated. The direction D is determined by identifying a direction between the directions of the two antennas in the cluster which are separated by the largest angle. This is described in more detail with reference to Figure 3B. Where a cluster comprising a single antenna has been identified, the direction D may be determined as the direction in which the single antenna is orientated.

Table 1, below, shows an example distribution of RSSI values detected by the antennas 20-1...20-N in the array 2.

Antenna number RSSI_antenna

20-1 -22dBm

20-2 -2odBm

20-3 -i6dBm

20-4 -lodBm

20-5 -lidBm 20-6 -i2dBm

20-7 -i8dBm

20-8 -24dBm

20-9 -26dBm

20-10 -28dBm

20-11 -35dBm

20-12 -38dBm

20-13 -39dBm

20-14 -40dBm

20-15 -4idBm

20-16 -37dBm

20-17 -27dBm

20-18 -25dBm

Table ι

Figure 3B is an illustration of the way in which the direction D of the wireless transmitter 30 is determined. For the purpose of this example, we will assume that the RSSI values are as shown in Table 1, and that the confidence interval C is 4dBm.

As can be seen from Table 1, the RSSI_MAX is -lodBm. Those antennas which are within the confidence interval (4dBm) of RSSI_MAX (i.e. the HIGH_SET antennas) are the fourth, fifth and sixth antennas 20-4, 20-5 and 20-6. In Figure 3B, the HIGH_SET antennas are denoted by vertical hatching. The non- HIGH_SET antennas are not hatched. In this example, the HIGH_SET antennas are located in a single cluster.

In methods such as that described with reference to Figure 3A (in which the criterion is that RSSI_MAX - RSSI_antenna < C), the direction D of the wireless transmitter 30 is determined by identifying a direction between the directions of the two antennas in the cluster that are separated by the largest angle. More specifically, the direction D is determined by identifying the bisector of the angle between the directions of the two antennas in the cluster that are separated by the largest angle. In this example, these antennas are the fourth and sixth antennas 20-4, 20-6. Consequently, the determined direction D of the wireless transmitter 30 happens to coincide with the orientation of the fifth directional antenna 20-4. It should be noted that, if the direction of the transmitter 30 had been determined using simply the direction of the antenna having the highest RSSI ("the single antenna approach"), a different result would have been produced. The improved accuracy of methods according to embodiments described herein compared with the single antenna approach is discussed below with reference to Figures 8A and 8B.

In some examples, the direction D of the wireless transmitter may determined by identifying the mean of the directions of all antennas in the cluster. In other words, the direction may be determined by identifying the direction of the sum of the vectors of the directions of all antennas in the cluster.

Figure 4A is a flowchart illustrating an alternative method by which the apparatus 1 can detect the direction of a wireless transmitter 30.

In Figure 4A, steps S4.1 to 84.5a are the same as steps S3.1 to 83.5a respectively as described in relation to Figure 3A. In steps S4.6, S4.6a and S4.7, the controller 12 determines which of the antennas 20-1 ...20-N have an RSSI satisfying a particular criterion. In this example, the criterion is that the value of the RSSI must be within a determined confidence interval of the RSSI of the antenna 20-1 ...20-N having the lowest RSSI. The value of the lowest RSSI is hereafter referred to as RSSI_MIN.

More particularly, in step S4.6, the controller 12 determines the RSSI_MIN. Next, in step S4.6a, the controller 12 compares the RSSI for each antenna with a threshold defined by MIN_RSSI plus a confidence interval. Based on the comparison of step S4.6a, in step S4.7, the controller 12 identifies all the directional antennas 20-1 ...20-N for which the measured RSSI is within the confidence interval of the RSSI_MIN (i.e. is below the threshold). As discussed with reference to step S3.6, the confidence interval may be determined based on the variation in all measured RSSIs. The plurality of antennas for which the measured RSSI is within the confidence interval from the

RSSI_MIN are hereafter referred to as LOW_SET antennas.

Next, in step S4.8, the controller 12 identifies one or more clusters of LOW_SET antennas. This may be carried out as described below with reference to Figures 6 and 7.

In step S4.9, it is determined if the number of clusters of LOW_SET antennas is equal to one. If a negative determination is reached, the method returns to step S4.1. If a positive determination is reached, the controller 12, in step S4.9, determines the direction D of the wireless transmitter 30 based on the orientations of the directional antennas within the single cluster. As in step S3.9, the direction D is determined by identifying a direction between the directions of the two antennas that are separated by the largest angle. However, in this example, the direction D is determined by taking the opposite of the identified direction.

Figure 4B is an illustration of the way in which the direction D of the wireless transmitter 30 is determined in step S4.9. For the purpose of this example, we will assume that the RSSI values are as shown in Table 1 above, and that the confidence interval C is 4dBm.

As can be seen from Table 1, the RSSI_MIN is -lodBm. Those antennas which are within the confidence interval (4dBm) of RSSI_MIN (i.e. the LOW_SET antennas) are the twelfth, thirteenth, fourteenth, fifteenth and sixteenth antennas 20-12, 20-13, 20-14, 20- 15 and 20-16. In Figure 4B, the LOW_SET antennas are denoted by cross-hatching. The non- LOW_SET antennas are not hatched. In this example, the LOW_SET antennas are located in a single cluster.

In methods such as that described with reference to Figure 4A (in which the criterion is RSSI_antenna - RSSI_MIN < C), the direction D of the wireless transmitter 30 is determined by identify a direction between the directions of the two antennas that are separated by the largest angle, and taking its opposite. More specifically, direction D is determined by calculating the opposite of the angular bisector of the separation between the directions of the antennas in the cluster that are separated by the largest angle. In this example, these antennas are the twelfth and sixteenth antennas 20-12, 20-16. The direction of the angular bisector is the direction of the fourteenth antenna 20-14. The direction D of the wireless transmitter 30 is determined to be the opposite of this which, in this example, is the direction of the fifth antenna 20-5.

In alternative examples, the direction D may be determined by identifying the opposite of the mean of the directions of all of the antennas in the cluster. In other words, the direction D may be determined by identifying the opposite of the direction of the sum of the vectors of the directions of all antennas in the cluster. In examples in which the cluster is constituted by a single antenna, the direction D is be determined by identifying the opposite of the direction in which the single antenna is orientated. Figure 5A is flow chart illustrating an example of another method by which the apparatus 1 of Figure 1 is operable to determine the direction D of a wireless transmitter 30. As will be understood from the below description, in this method the apparatus 1 is operable to determine the direction D of the wireless transmitter using the orientations of a first group of antennas for which the RSSI satisfies a first criterion and/or the orientations of a second group of antennas for which the RSSI satisfies a second criterion.

Steps S5.1 to 85.5a are the same as steps S3.1 to 83.5a respectively as described with reference to Figure 3A.

Steps S5.6, S5.6a, S5.7 and S5.8 are the same as steps S3.6, S3.6a, 3.7 and S3.8 respectively as described in relation to Figure 3A. In other words, the controller 12 determines the MAX_RSSI (step S5.6), compares the RSSI for each antenna with a threshold based on the MAX_RSSI (Ss.6a), identifies the HIGH_SET antennas based on the comparison (step S5.7), and identifies one or more groups or clusters of the

HIGH_SET antennas (step S5.8).

Next, the controller 12 performs steps S5.9, 85.9a, S5.10 and S5.11. These are the same as steps S4.6, S4.6a, S4.7 and S4.8 respectively as described with reference to Figure 4A. In other words, the controller 12 determines the MIN_RSSI (in step S5.9), compares the RSSI for each antenna with a threshold based on the MIN_RSSI (85.9a), identifies the LOW_SET antennas (in step S5.10) and based on the comparison identifies one or more groups or clusters of antennas (in step S5.11). In some examples, step 85.9a may comprise comparing with the threshold the RSSI's of only those antennas which were not identified as being members of the HIGH_SET.

Next, in step S5.12, the controller 12 determines if only one set of HIGH_SET antennas has been returned in step S5.8. If a positive determination is reached, the method proceeds to step S5.13 in which the direction D of the wireless transmitter is determined. This direction D is determined as described with reference to Figure 3B.

If a negative determination is reached in step S5.12, the method proceeds to step S5.14 in which the controller 12 determines if the number of clusters of LOW_SET antennas is equal to one. If a positive determination is reached, the controller 12 proceeds to step S5.13 in which the direction D of the wireless transmitter 30 is determined. The determination is carried out as described with reference to Figure 4B. If a negative determination is reached, the controller 12, in step S5.15, determines a direction based on each of the clusters of HIGH_SET antennas. This is determined by taking the direction of the angular bisector of the angle between the outermost antennas of each cluster.

Next, in step S5.16 the controller 12 determines a respective direction for each cluster of LOW_SET antennas. This is carried out by taking the opposite of the direction of the angular bisector of the angular separation between the two outermost antennas in each cluster.

Subsequently, in step S5.17 the controller 12 determines if any of the directions calculated using the clusters of HIGH_SET antennas is equal to a direction calculated using one of the clusters of LOW_SET antennas. If a positive determination is reached, the controller 12 proceeds to step S5.13 in which the direction D of the wireless transmitter is determined. Specifically, the controller 12 determines the direction that is common to a cluster of HIGH_SET antennas and a cluster of LOW_SET antennas to be the direction D of the wireless transmitter 30.

If a negative determination is reached in step S5.17, the controller 12 returns to step S5.1 in which the method is started again.

Figure 5B illustrates the way in which the controller 12 determines the direction D of the wireless transmitter using the orientations of HIGH_SET antennas and LOW_SET antennas.

Table 2, below, shows another example distribution of RSSI values detected by the antennas 20-1...20-N in the array 2. These values are used in the illustration of Figure

5B.

Antenna number RSSI_antenna

20-1 -28dBm

20-2 -25dBm

20-3 -24dBm

20-4 -2idBm

20-5 -i7dBm

20-6 -i6dBm 20-7 -i2dBm

20-8 -i9dBm

20-9 -i3dBm

20-10 -i4dBm

20-11 -23dBm

20-12 -4idBm

20-13 -38dBm

20-14 -35dBm

20-15 -40dBm

20-16 -37dBm

20-17 -27dBm

20-18 -25dBm

Table 2

As can be seen from Table 2, RSSI_MAX is -i2dBm and RSSI_MIN is -4idBm. The confidence interval for the example of Figure 5B is 4dBm. There are two clusters of HIGH_SET antennas (i.e. antennas for which RSSI_MAX - RSSI_antenna < C), which are denoted by vertical hatching. There are also two clusters of LOW_SET antennas (i.e. antennas for which RSSI_antenna - RSSI_MIN < C), which are denoted by cross- hatching. A first cluster of HIGH_SET antennas is formed by the sixth and seventh antennas 20-6, 20-7 and the second cluster is formed by the ninth and tenth antennas 20-9, 20-10. A first cluster of LOW_SET antennas is formed by the twelfth and thirteenth antennas 20-12, 20-13 and the second cluster is formed by the fifteenth and sixteenth antennas 20-15, 20-16. The directions calculated using the orientations of the antennas in the first and second clusters of the HIGH_SET antennas are denoted Dcm and DCH2. The directions calculated using orientations of the clusters from the

LOW_SET antennas are denoted DCLI and DCL ₂ .

As can be seen from Figure 5B, the direction DCL ₂ calculated using the second cluster of the LOW_SET antennas is equal to the direction Dcm calculated using the first cluster of the HIGH_SET antennas. Consequently, the controller 12 determines this direction to be the direction D of the wireless transmitter 30.

Figure 6 is a flowchart illustrating an example of a method by which the controller 12 identifies adjacent clusters or groups of antennas. This method may be used to perform steps S3.8, S4.8, S5.8 and S5.11. In step S6.1, an antenna counter i is set to zero. The value of the antenna counter indicates the number of the directional antenna that is being considered. For example, when 2 is equal to one, the first directional antenna 20-1 is being considered and when 2 is equal to twelve, the twelfth directional antenna 20-12 is being considered.

Next, in step S6.2, a current cluster field (CUR_CLUS) for temporarily storing indicators of antennas which are part of a current cluster is emptied.

Next, in step S6.3, the antenna counter 2 is incremented by one.

Subsequently, in step S6.4 the controller 12 determines if the antenna indicated by the antenna counter (A_i) is a member of the set of interest. In step S3.8 in which clusters of HIGH_SET antennas are identified, the HIGH_SET is the set of interest. As such, step S6.4 would comprise determining if the antenna indicated by the antenna counter 2 is a member of the HIGH_SET. Similarly, in step S4.8 in which clusters of LOW_SET antennas are identified, the LOW_SET is the set of interest

If it is determined in step S6.4 that the current antenna 2 is a member of the set of interest, the method proceeds to step S6.5 in which an indicator identifying the current antenna is added to the current cluster field.

Next, in step S6.6, the controller 12 determines if the antenna counter is equal to N (i.e. if it indicates the final antenna in the array). If a negative determination is reached, the method returns to step S6.3 in which the antenna counter 2 is incremented by one.

Subsequently, step S6.4 is performed in respect of the next antenna (i.e. the antenna indicated by the incremented antenna counter).

If a negative determination is reached in step S6.4, the method proceeds to step S6.7. In step S6.7, the controller 12 determines if the number of antennas indicated by the current cluster field is greater than zero. If it is determined that the number of antennas indicated by the current cluster field is greater than zero (i.e. that at least one antenna is indicated by current cluster field), the method proceeds to step S6.8.

In step S6.8, the controller 12 adds the cluster indicated by the current cluster field to a list of possible clusters (POSS_CLUS). Next, the method proceeds to step S6.9 in which it is determined if the current antenna i is the last antenna in the array 2. If a negative determination is reached, the method returns to step S6.2 in which the current cluster field is emptied. If a positive determination is reached, the method proceeds to step S6.10.

Returning now to step S6.6, if, subsequent to adding the current antenna to the current cluster field, it is determined that the current antenna is the final antenna in the array, the method proceeds to step S6.n. In step S6.ii, it is determined if the first antenna 20-1 is also member of the set of interest. If a positive determination is reached, the method proceeds to step S6.12, in which the current cluster (those antennas indicated by the current cluster field) is merged with the possible cluster which includes the first antenna (A_i_CLUS). The reason for this is that the first and last antennas 20-1, 20-N are located adjacent to one another and so, if both are members of the set of interest, they should be in the same cluster.

If a negative determination is reached in step S6.11, the method proceeds to step S6.13. In step S6.13, it is determined if the number of antennas indicated by the current cluster field is greater than zero. If a positive determination is reached, the method proceeds to step S6.14 in which the current cluster is added to a list of possible clusters. If a negative determination is reached, the method proceeds to S6.10.

After step S6.12, the method proceeds to step S6.10. Similarly, after step S6.14, the method also proceeds to step S6.10. In step S6.10, the controller 12 sorts the clusters identified in the list of possible clusters based on number of antennas per cluster (i.e. the size S of the cluster).

Next, in step S6.15, the controller 12 determines the size S of the largest group or cluster (MAX_S).

Subsequently, in step S6.16, all clusters having a size S equal to MAX_S are returned or output. It will thus be understood that only those clusters of antennas having a size equal to MAX_S are considered for the purpose of determining the direction D of the wireless transmitter 30. The method for identifying clusters of antennas as described with reference to Figure 6 does not allow antennas in a cluster to be interspersed with antennas which are not members of the set of interest. As such, erratic RSSI measurements may adversely affect the determination of the direction of wireless transmitters 30.

Figure 7 is an example of an alternative method for identifying clusters whilst reducing the impact of erratic RSSI measurements on the determination of the direction D of the wireless transmitter 30. In step S7.1 the antenna counter i is set to zero. Next, in step S7.2, a current gap counter (CUR_GAP) is set to zero. The current gap counter indicates the number of successive adjacent antennas which have not been identified in step S7.5 as being in the set of interest (i.e. HIGH_SET or LOW_SET). In step S7.3, the current cluster field (CUR_CLUS) is emptied. This is the same as step S6.2 of Figure 6.

Next, the antenna counter i is incremented by one. After this, the controller 12 proceeds to step S7.5 in which it is determined if the current antenna indicated by the antenna counter (A_z) is a member of the set of interest. If a positive determination is reached, the method proceeds to step S7.6 in which an indicator associated with the current antenna is added to the current cluster field. Subsequently, in step S7.7, it is determined if the current antenna is the last antenna (i.e. if z ^' =N). Following a negative

determination, the method returns to step S7.4 in which the antenna counter is incremented. Following a positive determination, the method proceeds to step S7.8.

Steps S7.8 to S7.13 are the same as steps S6.7 to S6.12 respectively, and so, for conciseness, a full description of these steps is not repeated here. Referring back to step S7.5, if a negative determination is produced, the method proceeds to step S7.14 in which the current gap counter is incremented by one. Subsequently, in step S7.15, it is determined if the current gap counter is greater than an allowed gap threshold (GAP_TH). The allowed gap threshold indicates the number of adjacent antennas which are not members of the set of interest that are allowed to be interspersed in a cluster. Following a determination that the current gap is greater than the current gap threshold, the method proceeds to step S7.16 in which it is determined if the number of antennas in the current cluster is greater than zero. If so, the current cluster of antennas is added to the list of possible clusters in step S7.17.

Subsequent to step S7.17, or in the event of a negative determination in step S7.16, the method proceeds to step S7.18 in which it is determined if the current antenna is the final antenna in the array 2 (i.e. if A_i = A_N . If the current antenna is the final antenna, the method proceeds to step S7.11. If the current antenna is not the final antenna, the method returns to step S7.2.

Returning now to step S7.15, if a negative determination is reached, the method proceeds to step S7.19 in which it is determined if the current antenna is the last antenna (i.e. if A_z = A_N . In the event of a negative determination, the method returns to step S7.4. In the event of a positive determination, the method proceeds to step S7.20.

In step S7.20, it is determined if the number of antennas in the current cluster is greater than zero. If a negative determination is returned (i.e. if the number of antennas in the current cluster field is zero), the method proceeds to step S7.11. If it is determined that the number of antennas in the current cluster is greater than zero, the method proceeds to step S7.21.

In step 7.21, a second antenna counter j is set to equal one. Subsequently, in step S7.22, it is determined if the antenna identified by the second antenna counter (A_j) is a member of the set of interest.

If in step S7.22 a positive determination is reached, the method proceeds to step S7.23 in which the current cluster is merged with the possible cluster which includes the antenna identified by the second antenna counter. If a negative determination is reached, the method proceeds to step S7.24.

In step 7.24, it is determined if the sum of the current gap counter and the second antenna counter is greater than the gap threshold. If a positive determination is reached, the method proceeds to step 87.24a in which the current cluster of antennas is added to the list of possible clusters. After this, the method proceeds to step S7.11. If a negative determination is reached in step S7.24, the method proceeds to step S7.25 in which the second antenna counter is incremented by one. After this, the method returns to step S7.22. Steps S7.22, S7.24 and S7.25 are repeated until it is determined in step S7.22 that A_j is a member of the set of interest or it is determined in step S7.24 that the sum of the current gap counter CUR_GAP and the second antenna counter j is greater than the allowed gap threshold.

The antenna grouping method of Figure 7 allows antennas in an identified cluster of antennas to be interspersed with antennas which are not members of the set of interest. Consequently, the impact of erratic RSSI measurements may be reduced.

It will be appreciated that the methods described with reference to Figures 3A, 4A and 5A are merely examples. As such, various steps of the methods may be performed in a different order or simultaneously with one another. In addition, some steps may be omitted or may differ slightly from those described above. For example, in Figures 3A, 4A and 5A, steps S3.2 to S3.4 (or their equivalents in Figures 4A and 5A) may be performed in respect of each antenna simultaneously. In some examples, step S3.1 (and its equivalents) may comprise activating just one of the antennas. The step may then be repeated in step S3.5 (or its equivalents) for each antenna. In Figure 5A, step S5.12 may be performed immediately following step S5.8, and steps S5.9 to S5.11 may be omitted if a positive determination is reached in step S5.12. In another variation, steps S5.9 to S5.11 may be performed before or concurrently with steps S5.6 to S5.8. In this variation, step S5.14 may be immediately following step S5.11 and steps S5.6 to S5.8 may be omitted if a positive determination is reached in step S5.11. Also in this variation, step S5.6a may comprise comparing with the threshold the RSSI's of only those antennas which were not identified as being members of the LOW_SET.

In some examples, the order of steps S5.12 and S5.14 may be reversed.

Regarding Figures 6 and 7, these are merely examples of two ways in which identification of clusters of antennas can be achieved. Different, for example non-iterative, approaches may be utilised instead. In other examples, steps S6.7 and S6.13 of Figure 6 may comprise determining if the number of antennas in the current cluster is greater than one. Similarly, step S7.16 of Figure 7 may comprise determining if the number of antennas in the current cluster is greater than one. In such examples, step 87.24a of Figure 7 is omitted. In these examples, all clusters are constituted by plural antennas. During development of the herein described embodiments, the inventors performed various experiments to assess the effect of using a cluster of antennas to determine the direction D of a wireless transmitter 30, as opposed to using just the orientation of the single antenna having the maximum RSSI.

Figures 8A and 8B illustrate aspects of an experimental setup that was used by the inventors. Specifically, Figure 8A shows an experimental array 7 of directional antennas 70-1, 70-2, 70-3, ... 70-6. It also shows the locations of the wireless transmitter 74-1, 74- 2, 74-2, 74-3, 74-4 relative to the array 7 at which RSSI measurements were made.

The array of directional antennas 7 comprises six directional antennas 70-1, 70-2, 70-3, ... 70-6 spaced equidistantly around a central reference point 72. Each of the directional antennas 70-1, 70-2, 70-3, ... 70-6 is oriented directly away from the reference point 72. The wireless transmitter used during the experiment was a Nokia N900 mobile telephone. The measurements were taken at four locations 74-1, 74-2, 74-2, 74-3, 74-4 around the antenna array 7. For each location, a measurement was taken with the mobile telephone in three different orientations relative to a radial axis of the antenna array upon which the mobile telephone was situated. The three orientations are illustrated in Figure 8B. The radial axis is shown by the dotted line labelled 78. These orientations are perpendicular 76-1 to the radial axis 78, horizontal 76-2 to the radial axis 78 and reverse horizontal 76-3 to the radial axis 78 (i.e. horizontal but rotated by 180 degrees).

Taking measurements at three mobile telephone orientations 76-1, 76-2, 76-3 for each of the four mobile telephone locations resulted in twelve different measurements. These twelve measurements were then repeated with the mobile telephone at three different distances from the array 7. The distance was calculated from the reference point 72 of the array 7 to a central point of the mobile telephone. The three distances used were 5cm, 10cm and 20cm.

Table 4 shows, for each distance from the array, the number of measurements out of twelve which yielded the correct direction of the mobile telephone. The second column shows the number of correct directions that were determined using only the directional antenna which received the maximum RSSI. The third column shows the number of correct determinations when a method according to Figures 3A and 6 was utilised to determine the direction D of the remote transmitter. As is clear from Table 4, the correct direction of the wireless transmitter was determined with far greater success when a method according to the examples described herein were utilised.

Distance MAX RSSI APRROACH CLUSTER APPROACH

5cm 7/12 (58.3%) 11/12 (91.6%)

10cm 6/12 (50.0%) 11/12 (91.6%)

20cm 4/12 (33-3%) 10/12 (83.3%)

Table 4

It should be realized that the foregoing embodiments should not be construed as limiting. Other variations and modifications will be apparent to persons skilled in the art upon reading the present application. Moreover, the disclosure of the present application should be understood to include any novel features or any novel combination of features either explicitly or implicitly disclosed herein or any generalization thereof and during the prosecution of the present application or of any application derived therefrom, new claims may be formulated to cover any such features and/or combination of such features.