LATENCY

This application claims priority to European Patent Application entitled, SYSTEM FOR IDENTIFYING A LOCATION OF A MOBILE TAG READER, NO. EP14306895.5, filed on November 26, 2014 and forms a continuation in part (CIP) of a corresponding to-be-filed United States Patent Application (PF140319- US-PCT.) Field of the Invention

The present invention is directed to a radio frequency identification (RFID) tag system for reducing handoff delay in a wireless network and, in particular, for example, local area wireless computer networking (WiFi). Background

In the IEEE 802.11 standard, Extended Service Set (ESS) refers to two or more Basic Service Sets (BSS) whose respective Access Points (AP) communicate through a wired network, named Distribution System (DS). The BSS includes an AP antenna (ant). Its associated mobile stations communicate in the unlicensed Industrial Scientific and Medical (ISM) radio bands. When a mobile station (MS) moves beyond the radio range of an AP, and enters into a radio range of another BSS, the MS triggers a Handhoff process which can take from 200 ms and up to 1000 ms. Such a large delay range may be undesirable for delay sensitive applications, such as Voice over Internet Protocol (VoIP) . VoIP is a methodology and group of technologies for the delivery of voice communications and

multimedia sessions over Internet Protocol (IP). For the purpose of VoIP the recommended maximum end-to-end latency is 150 ms. In the IEEE 802.11 standard, the MS initiates the handoff process when the received signal strength and the signal-to-noise-ratio have decreased significantly. The MS can either passively or actively scan for new AP' s to associate next. In the case of a fast active scan which is faster than a passive scan, the MS broadcasts probe frames and waits for responses for a minimum channel time (Tmin), and continues scanning until a maximum channel time (Tmax) has elapsed, if at least one response has been received within Tmin. Thus, the time to probe (Tprobe) n channels is given by: n*Tmin < Tprobe < n*Tmax. This information is processed by the MS to decide which BSS to join next. In general, Tprobe constitutes 90% of the handoff delay. It may be desirable to reduce the handoff latency. An article, Published in Wireless Communications and Networking Conference (WCNC), 2010 IEEE, entitled, Handoff Management relying on RFID Technology, in the names of Apostolia Papapostolou and Hakima Chaouchi proposes to predict the next point of attachment (PoA) of an RFID-enabled MS by using topology information provided by the network with the collaboration of an RFID system.

The use of an RFID reader is also described in, for example, a published patent application No. WO 2005/071597. There, an RFID tag array-based "smart floor" system for navigation and location determination for guiding individuals includes a plurality of spaced apart RFID tags. Each RFID tag has memory having information stored therein including positional information and attributes of objects or structures disposed in proximity to the tags. The tags convey radio frequency (RF) signals including the positional information and the attributes in response to received electromagnetic excitation fields.

Long range RFID tag systems operating in the ultra high frequency (UHF) band have a range that is typically 12m in line of sight (LOS) conditions. This range could be drastically reduced by any blockage of the RFID tag or RFID reader caused by various kinds of obstacles such as people or furniture that results in a shadowing effect.

Because of multipath frequency selective fading, encountered in indoor

environments, significant level variations of the received RF signal are experienced even within a distance of a few centimeters. Figure 5 shows a graph presenting an example of variations of magnitudes of a received RF signal at UHF frequency within an indoor area having coordinates X and Y of 2mx2m, respectively. As shown in Figure 5, variation of the RF signal of up to 40 dBm could be noticed over distances of a few tens of centimeters. Such fast fading signal could, disadvantageously, prevent the activation of one RFID tag even for a transmission from a close RFID reader, situated in the same area; whereas, another RFID tag, located in a nearby area could, disadvantageously, be activated and thus read by the RFID reader located in the area in which the one RFID tag is located. As illustrated in the graph of Figure 6, a situation may happen in which an RFID tag located in a so-called fade at a region 66 with respect to a signal transmitted by an RFID reader situated in the same area. On the other hand, an RFID tag situated a different area might be located on a so-called crest 68.

Summary

In accordance with an aspect of the disclosure, a mobile station is configured for performing a handoff operation in a wireless communication network . It includes a radio frequency (RF) identification (RFID) transmitter for transmitting an interrogating RF signal. An RFID receiver stage is capable of receiving, in response to the interrogating RF signal, a plurality of responding RF signals generated in a plurality of RFID tags, respectively, located in a plurality of locations. A processor is configured to select, in accordance with a selection criterion, at least one of the responding RF signals containing information identifying a corresponding access point (AP) and to select, in accordance with the identifying information, the identified AP from a plurality of access points (AP's) for the mobile station to associate with in the handoff operation.

In accordance with another aspect of the disclosure, a mobile station performs a handoff operation in a wireless communication. It includes a radio frequency (RF) identification (RFID) transmitter for transmitting an interrogating RF signal that is applied to RFID tags arranged in RFID tag pairs. A given RFID tag pair includes a first RFID tag and a second RFID tag separated from each other by a distance for reducing fading effect. The given RFID tag pair of the RFID tag pairs is capable of generating, in response to the interrogating RF signal, a first responding RF signal, if at all, and a second responding RF signal, if at all, respectively. A processor is configured to select, in accordance with a difference between a magnitude of the first responding RF signal that is received by the processor and a magnitude of the second responding RF signal that is received by the processor, one of the magnitudes, when both the first and second responding RF signals are generated, and to select the magnitude of one

responding RF signal, when the given RFID tag pair generates merely the one responding RF signal. The processor is additionally configured to select a location from a plurality of locations. A given location of the plurality of locations contains a corresponding plurality of the RFID tag pairs, in a manner to reduce shadowing effect. The plurality of the RFID tag pairs contained in the given location generate at least a corresponding responding RF signal of a plurality of responding RF signals associated with the RFID tag pairs located in the given location. The selected location is selected in accordance with a sum of at least one selected magnitude of a plurality of selected magnitudes associated with the plurality of RFID tag pairs located in the selected location. The processor is further configured to select an access point (AP) from a plurality of access points (ΑΡ' s) for the mobile station to associate with in the handoff operation in accordance with the selected location.

Brief Description of Drawings Figure 1 schematically illustrates an example of a building having, in each location, RFID attached tag pairs, embodying a preferred embodiment;

Figure 2 illustrates a partial block diagram of an RFID tag pair of Figure 1 ; Figure 3 illustrates a partial block diagram of a mobile device containing an RFID tag reader, embodying a preferred embodiment;

Figure 4 illustrates a flow chart for explaining the operation of a processor of

Figure 3 ;

Figure 5 illustrates a graph presenting an example of variations of

magnitudes of a received RF signal; and

Figure 6 illustrates a graph demonstrating a fading effect of a received RF signal.

Detailed Description

Figure 1 schematically illustrates a residential structure such as a building 100. Building 100 includes, for example, four locations, a Room 1, a Room 2, a Room 3 and a Room 4, respectively. In each of Rooms 1 and 3 in building 100 of Figure 1, a corresponding wireless access point (AP) is installed, API and AP3, respectively. Each of API and AP3 allows wireless devices to connect to a network 200 using Wi-Fi, or related protocols. Each AP usually connects to a router (via a wired network) as a stand-alone device, but it can also be an integral component of the router itself. In each room of residential building 100, for example, in Room 1, four identical radio frequency identification (RFID) tag sets, an RFID tag pair 11 , an RFID tag pair 12, an RFID tag pair 13 and an RFID tag pair 14 are installed. RFID tag pairs 11, 12, 13 and 14 are spaced apart of each other so as to reduce the effect of shadowing. It should be understood that a set may include more RFID tags than a pair. However, in the preferred embodiment each set includes just a pair of RFID tags.

RFID tag pair 11 is embedded within or rigidly attached to a region 115-11 of a surface of a wall 50. Similarly, RFID tag pairs 12, 13 and 14 are embedded within or rigidly attached to a region 115-12 region, a region 115-13 and a region 115-14 of a surface of a wall 51, a ceiling 52 and a flooring 53, respectively. RFID tag pairs 11, 12, 13 and 14 are associated merely with Room 1 and with AP 1 and none is associated with any other Room or with another AP. An RFID tag pair 21, an RFID tag pair 22, an RFID tag pair 23 and an RFID tag pair 24 are similarly installed in room 2 and are associated with Room 2 and with AP3 and none is associated with any other Room or with another AP. An RFID tag pair 31, an RFID tag pair 32, an RFID tag pair 33 and an RFID tag pair 34 are similarly installed in Room 3 and are associated with Room 3 and with AP 3 and neither is associated with any other Room or with another AP. Lastly, an RFID tag pair 41, an RFID tag pair 42, an RFID tag pair 43 and an RFID tag pair 44 are similarly installed in room 4 and are associated with room 4 and with API and none is associated neither with any other Room or with another AP. Being associated with a given room indicates that the corresponding RFID tag pair is contained in the given room. Being associated with a given AP indicates that the corresponding AP would be the preferred AP to associate or re-associate with in a handoff process. Each RFID tag pair, for example, RFID tag pair 11 of Figure 2, forms a set that includes an RFID tag 1 la and an RFID tag 1 lb, each being of the passive type that does not require wired connection to a power supply for energization. Similar symbols and numerals in Figures 1 and 2 indicate similar items or functions.

RFID tag 11a, for example, of Figure 2 is paper thin constructed on a flexible plastic substrate 115 with adhesive for its fixation. Substrate 115 may be optically transparent or may have the wall/ceiling color to be visually unobtrusive. RFID tag 11a includes an etched directional antenna 110 having a hemispherical or half space radiation coverage, a tiny chip 114 which includes a memory 112 for storing data and a transceiver 113 that is coupled to antenna 110 in a conventional manner. The data in memory 112 may contain an identifier, not shown, identifying the room and region in which RFID tag 11a is installed which, in this example, is room 1, region 115-11 and wall 50 of Figure 1. It may also contain the RFID tag pair identifier which, in this example, identifies it as being included in RFID tag pair 11 and having an identification portion that differentiates it from the other RFID tag 1 lb, of the same RFID tag pair 11. In addition, it may contain necessary information required by the protocol of Wi-Fi wireless network 200 for addressing and associating or re-associating the preferred AP in a handoff process, which, in this example, is API. RFID tag 1 lb may be similarly constructed on flexible plastic substrate 115 as RFID tag 1 la but is attached to wall 50 of Figure 1 at a distance, d, from RFID tag 11a. RFID tag 1 la and RFID tag 1 lb may be

substantially identical except for the identifier, not shown, that differentiates RFID tag 1 lb from RFID tag 11a.

Figure 3 illustrates a partial block diagram of a mobile device, for example, a smart phone 60 capable of communicating in, for example, a Wi-Fi wireless network 200 via API and AP3 of Figure 1. Smart phone 60 also includes an RFID tag reader 61 forming both an RF receiver and an RF transmitter, embodying an advantageous feature. Similar symbols and numerals in Figures 1 , 2 and 3 indicate similar items or function. When smart phone 60 that contains Reader 61 of Figure 3 is carried by a user, not shown, it can be used for identifying, for example, a room from among rooms 1-4 of Figure 1 where mobile smart phone 60 is located and an AP that is most appropriate for association or re-association with smart phone 60 in a handoff process, when mobile smart phone 60 is located in such room .

RFID reader 61 of Figure 3 includes a processor 64 which may be realized as a digital signal processor (DSP). Processor 64 is coupled to a bus 65 for coupling processor 64 to a memory 66 and to a motion detector 67 such as an accelerometer. Processor 64 is coupled via an RFID transceiver 62 to an antenna 63.

The operation of processor 64 is explained in connection with a flow chart of Figure 4. Similar symbols and numerals in FIGURES 1, 2, 3 and 4 indicate similar items or functions. In an operation block 400 of Figure 4, processor 64 of Figure 3 waits for a movement or motion indication, not shown, from motion detector 67. The movement indication is tested in a decision block 401 of Figure 4 to determine whether smart phone 60 including RFID tag reader 61 has been moved. When motion detector 67 is indicative of a movement of portable RFID reader 61, the result is "yes" in decision block 401 of Figure 4. Consequently, in a block 402, processor 64 of Figure 3 is triggered to generate a transmission of an interrogation or selection RF signal 63a that is modulated to contain an address for sequentially and selectively addressing each RFID tag, for example, RFID tag 11a of Figure 2 of tag pair 11 of Figure 1. By not initiating the communication unless and until motion detector 67 becomes indicative of a movement of portable reader 61, current consumption of RFID reader 61 is, advantageously, reduced in a manner to conserve a charge in a battery, not shown, of smart phone 60.

It is known to use a variety of techniques to transmit and receive data to and from the corresponding RFID tag including amplitude modulation (AM), phase modulation (PM), and frequency modulation (FM). Furthermore, the data transmitted can be encoded using any of a variety of techniques, including frequency shift keying (FSK), pulse position modulation (PPM), pulse duration modulation (PDM), and amplitude shift keying (ASK).

In the example of Figure 1, RFID tag 11a of Figure 2 is exposed to an RF field produced by antenna 63 of Figure 3. Consequently, it absorbs energy from the RF transmissions of antenna 63 of reader 61 acting as transmitter and uses the absorbed energy for energizing chip 114 of Figure 2 to perform data retrieval and data transmission. Upon decoding transmission of interrogation or selection RF signal 63a of Figure 3 that targets, for example, RFID tag 11a, the aforementioned information stored in memory 112 of RFID tag 11a of Figure 2 is transmitted by antenna 110 and, in this example, a corresponding responding RF signal 11 la is received via antenna 63 of reader 61 of Figure 3 acting as a receiver.

In an operational block 403 of Figure 4, Processor 64 of Figure 3 detects responding RF signal 111a and stores in memory 66 a corresponding Receiver Signal Strength Indicator ( RSSI) value that is indicative of a magnitude of responding RF signals 111a received from RFID tag 11a of Figure 2 of tag pair 11. In addition, processor 64 of Figure 3 stores in memory 66 the corresponding identifier information such as the room number that, in this example, is Room land information capable of identifying the associated AP that, in this example, API is associated with Room 1 including addressing information. Such addressing information may be required by smart phone 60 for communicating with the associated APlusing the protocol of Wi-Fi network 200. Processor 64 of Figure 3 also stores in memory 66 a number identifying RFID tag 11a and RFID tag pair 11 from the information contained in responding RF signal 111a. Similarly, RFID reader 61 of Figure 3 communicates with RFID tag 1 lb of Figure 2 and with each of the remaining RFID tags of RFID tag pairs 12-14, 21-24, 31-34 and 41-44 of Figure 1 , during a sequence of time slots, respectively, not shown. The result is that the corresponding information for each responding RFID signal, if any, generated in RFID tag pairs 11-14, 21-24, 31-34 and 41-44 of Figure lis stored, similarly to that stored for responding RF signal 11 la, is stored in memory 66 of Figure 3.

If, in block 403, no responding RF signal is received by processor 64 of

Figure 3 from any of the targeted RFID tags of RFID tag pairs 11-14, 21-24, 31- 34 and 41-44 of Figure 1, represented by, T= 0, in a decision block 404 of Figure 4, processor 64 of Figure 3 will keep sending transmission of interrogation, as suggested by the "yes" path of decision block 404. On the other hand, if at least one responding RF signal has been received, in the aforementioned sequence of time slots, the "no" answer of decision block 404 indicates that an operation block 405 will follow.

In operation block 405 of Figure 4, processor 64 of Figure 3 selects the largest (or larger, in the case of a pair of responding RF signals) stored RSSI value of each set of responding RF signal pairs generated in each responding RFID tag pair. For example, assume that a responding RF signal 11 la of Figures 1 and 2 happens to have a larger RSSI value than that of a responding RF signal 111b, originated in RFID tags 11a and 1 lb, respectively, of Figure 2 of RFID tag pair 11. In this case, the RSSI value of responding RF signal 111a will be selected and stored. This would also be applicable in the special situation in which RF signal 111a is received but no responding RF signal 11 lb is detected. As indicated before, tag pair 11 is associated with region 115-11 of a portion of wall 50 of room 1 of Figure 1 where substrate 115 of Figure 2 is attached. RFID tag pair 11 is also associated with API . Similar communication process is applied with respect to each of the other responding RF signals originated in the corresponding RFID tag pairs 12-14, 21-24, 31-34 and 41-44 of Figure 1. For example, a stored RSSI value associated with a larger responding RF signal 121a of a set of responding RF signal pair that includes RF signal 121a and an RF signal 121b might be selected for representing tag pair 12 in further processing steps; whereas, in this example, a stored RSSI value of responding RF signal 121b will not be selected or excluded from having any further effect. Similarly, a stored RSSI value associated with a responding RF signal 21 la might be selected for representing tag pair 21 in following processing steps; whereas, in this example, a stored RSSI value of a responding RF signal 21 lb will not be selected. Likewise, a stored RSSI value associated with a responding RF signal 241a might be selected for tag pair 24; whereas, in this example, a stored RSSI value of a responding RF signal 241b will not be selected.

Thereafter, in an operation block 406 and a decision block 407, processor 64 of Figure 3 determines, in accordance with the information stored in memory 66, whether each of the responding RF signals has originated in a single room or associated with a single AP. In this example, it could be just Room 1 or API of Figure 1 , respectively.

In the majority of situations, it is more likely that all responding RF signals originate in a single room. Thus, if the answer in decision block 407 of Figure 4 is "yes", processor 64 of Figure 3 determines, in an operation block 410 of Figure 4, the room in which mobile station 60 of Figure 3 is located. In the example of Figure 1 , processor 64 of Figure 3 determines that mobile station 60 is located in Room 1. The same operation also determines the AP with which mobile smart phone 60 should preferably associate in a handoff process. In the example of Figure 1 , processor 64 determines that mobile smart phone 60 should preferably associate with API.

On the other hand, in rarely occurring situations, not all the responding RF signals originate in a single room or not all the responding RF signals are associated with a single AP , resulting in the answer, "no", in the aforementioned decision block 407 of Figure 4. Therefore, an operation block 408 will follow. Thus, operation block 408 follows when RFID reader 61 receives, in addition to, for example, responding RF signal 111a that originated in Room 1 and associated with API of Figure 1, also, for example, a responding RF signal 211a , a responding RF signal 21 lb or both. Responding RF signals 211a and 21 lb have been generated in RFID tag 21 of room 2 that is outside Room 1 in which Mobile smart phone 60 is presently located. Similarly, responding RF signals 211a and 21 lb, in this example, are associated with AP3 located in room 3. RFID reader 61 might receive, in addition, for example, a responding RF signal 241a or a responding RF signal 241b originated from RFID tag pair 24 or both that are located in Room 2 and are also associated with AP3. As explained later on, in operation block 408 and in the following operation blocks 409 and 410, using the information contained in or derived from the responding RF signals, processor 64 of Figure 3 determines the room number in which smart phone 60 is located and, similarly, the AP which Mobile smart phone 60 should preferably select to associate with in a handoff process.

A responding RF signal originated in, for example, tag pair 21 might be subject to the aforementioned multipath frequency selective fading problem encountered in indoor environments. Consequenltly and counter-intuitively, received responding RF signal 211a that are generated in Room 2 and associated with AP3 might happen to be even larger than received responding RF signal 111a generated in Room 1 of Figure 1, in which Mobile smart phone 60 is presently located, and associated with API. This could have led to an identification error resulting in a false determination. Such false determination indicates, for example, that the room in which Mobile smart phone 60 is located is room 2 and the AP with which Mobile smart phone 60 should preferably select to associate would be AP3 instead of a correct determination of API of Room 1, where it is actually located. To avoid such an error, the RFID tags of each RFID tag pair, for example, RFID tags 11a and 1 lb of the set of RFID tag pair 11 of Figure 2 are separated from each other by the aforementioned distance, d.

Distance, d, is selected to be greater than a coherence distance, λ/4, associated with the frequency of the radiated RF signal which, at 900 MHz, is approximately 8 cm. However, distance, d, is also selected to be smaller than λ/2 which at 900 MHz is approximately 16 cm. Because distance, d, is greater than the coherence distance, λ/4, it is unlikely that, for example, both responding RFID signals 111a and 111b generated in RFID tags 11a and 1 lb, respectively, will simultaneously encounter the multipath frequency selective fading problem. Thus, the multipath frequency selective fading problem that may be encountered in indoor environment such as in Room 1-Room 4 of Figure 1 is, advantageously, mitigated.

In the example referred to before, each selected responding RF signal 11 la,121a,21 la and 241a that was selected on the basis of having the larger RSSI value of the pair of responding RF signals generated in the corresponding RFID tag pair. In operation block 408 of Figure 4, the stored RSSI values of all the selected larger responding RF signals from each tag pairs located in the corresponding room, for example, the magnitude or RSSI of each of RF signals 111a and 121a originated in Room 1 of Figure 1 , are combined to form an accumulative

magnitude. The cumulative magnitude is formed by algebraically summed up RSSI of responding RF signals 111a and RSSI of responding 121a to produce a first sum associated with Room 1. Similarly, the stored RSSI values of each the selected larger responding RF signals, for example, of responding RF signals 211a and 241a, originated in Room 2 are also are combined to form an accumulative magnitude such as by being algebraically summed up to produce a second sum associated with Room 2. Similar operation is performed with respect to the stored RSSI values of all selected responding RF signals, if any, associated with each of Rooms 3 and 4 that are combined to form an accumulative magnitude of a third sum, if any, and an accumulative magnitude of a fourth sum, if any, respectively.

As shown in the example of Figure 1, the number of RFID tag pairs in each of Rooms 1-4 is equal to that in each of the other rooms. Accordingly, processor 64 of Figure 3 compares the first, second, third and fourth sums to one another for selecting the largest of the first, second, third and fourth sum , that is implemented in block 410 of Figure 4. In block 410 of Figure 4, processor 64 of Figure 3 determines, among Rooms 1-4, the particular Room in which smart phone 60 is located by determining the room associated with the largest of the first, second, third and fourth sums. In the same way, as explained later on in more details, processor 64 of Figure 3 determines the AP, API or AP3, associated with the largest of the first, second, third and fourth sums, as being the most appropriate AP to be selected for associating with in a handoff process.

As represented in preceding block 409 of Figure 4, processor 64 of Figure 3 may, in addition, calculate the probability of such room as being the correct room to contain smart phone 60 and of such AP as being the most preferable one to associate with in a handoff process. This probability is equal to a fraction having the largest of the first, second, third and fourth sums, as a numerator, and a sum total of the first, second, third and fourth sums, as a denominator.

Each responding RF signal associated with the largest of the first, second, third and fourth sums may contain sufficient information to be included in the protocol for initiating the communication between smart phone 60 and the selected AP, API or AP 3 of Figure 1, to associate with next in a handoff operation. As an alternative, memory 66 of RFID reader 61 of Figure 3 may contain in a table, not shown, information of the AP to associate with next when mobile station 60 is present in a given room. In the above example, the table will indicate that API is the appropriate AP to associate with when RFID reader 61 of Figure 3 is located in either Room 1 or Room 2. In another alternative, each responding RF signal that originate in Room 1 or Room 2 will contain sufficient information to enable the selection of API as the most appropriate AP to associate with in the handoff process.

In one embodiment, not shown, for each room in which Mobile smart phone 60 is located, the user can cause smart phone 60 to store in memory 66 information sufficient for enabling smart phone 60 to initiate Wi-Fi communication with the corresponding one AP, API or AP3, of Figure 1 which is the preferred AP to associate with next. For example, in a set-up or learning mode of operation, that can be initiated in response to a user command or automatically, smart phone 60 can determine by trial and error for a given room, in which Mobile smart phone 60 is located, the information required by smart phone 60 to initiate Wi-Fi

communication with the applicable AP, API or AP3, of Figure 1 to associate with next. This information is then stored in a local table, not shown, that is contained in, for example, memory 66 for future operations .

Following operation box 410 of Figure 4, when the determination of the next AP to associate with in a handoff operation has already been accomplished, smart phone 60 of Figure 1 continues the handoff process in network 200 of an operation block 411 of Figure 4.

Assume, for example, that smart phone 60 moves from Room 1 , of Figure 1 in which AP 1 is the preferable AP to associate with, to Room 2, in which AP 3 located in Room 3 is the preferable AP to associate with. For completing the handoff process, in order to verify that the aforementioned RFID-probe-response corresponds to AP 3 of Figure 1 , mobile smart phone 60 of Figure 3-can broadcast in Wi-Fi network 200 a New-AP-probe-request addressed to the next AP, AP 3, of Figure 1 , in a manner similar to that defined in the IEEE 802.11 (WiFi) handoff process. Thus, the aforementioned RFID-probe-response will terminate when mobile smart phone 60 of Figure 3 receives the New- AP-prove-response from AP 3 of Figure 1 , in this example . Thereafter, the Authentication Phase defined in, for example, the IEEE 802.11 (WiFi) handoff process, will follow in a conventional manner. This phase involves the request and transfer of credentials and other state information from AP3 of Figure 1 to mobile smart phone 60 of Figure 3. After successfully receiving from AP3 of Figure 1 the credentials and state information, mobile smart phone 60 of Figure 3 sends a Re-association Request message to AP 3 of Figure 1, which, in turn, sends a Move Request message to the previous AP which, in this example, is API. Then, in a conventional manner, AP3 completes the handoff process by sending a Reassociation Response message to mobile smart phone 60 of Figure 3.

Assume that the number of RFID tag pairs in different rooms of Rooms 1-4 of Figure 1 is unequal in a manner not shown in Figure 1. Accordingly, processor 64 of Figure 3 divides each of the first, second, third and fourth sums or

accumulative magnitudes by the number of RFID tag pairs in Rooms 1 , 2, 3 and 4, respectively, to produce a first average value, a second average value, a third average value and a fourth average value associated with the responding RF signals originated in Rooms 1, 2, 3 and 4, respectively. Processor 64 of Figure 3 compares the first, second, third and fourth average values to one another for selecting the largest of the first, second, third and fourth average values. The AP which is the most appropriate one to associate with in a handoff process would be the AP associated with the largest of the first, second, third and fourth aveage values. Similarly, the room in which mobile station 60 is located would be associated with the largest of the first, second, third and fourth average values.

In addition, processor 64 calculates in an operation block 409 of Figure 4 the probability, Pk, that such AP is the most appropriate one to associate with in a handoff process and that such room is the room in which mobile station 60 is located. This is performed by calculating a fraction having the largest of the first, second, third and fourth average values, as a numerator, and a sum total of the first, second, third and fourth average values, as a denominator.

Let Si _j be defined as the Maximum RSSI from a tag pair number j in room number i. Room number i assumes the value 1, 2, ...or R, such that "R" is also the total number of rooms. Tag pair number j assumes the values 1,2, 3, ...or Ti such that "Ti " is also the total number of tagged pairs in room i.

1. In case where all the rooms have the same number of tag pairs, for all i's Ti is equal to T.

For each room k, RFID reader 61 of Figure 3 calculates, a sum S _k of all Maximum RSSI obtained from the tag pairs situated in a room k. Let S _kj be defined as the Maximum RSSI of tag pair number in room k. It follows that S _k = ∑, ^■ ·¾ ^■ ; j = 1, ... , T. Thus, the room where RFID reader 61 of Figure 3 is most likely to be located is the room number for which the highest value S _k is obtained. Also, the probability P _k that the user is located in room k could be estimated as P _k = S _k /S where S is the sum total of all the S _k s or S = ∑ _k S _k . Similarly, the AP which Mobile smart phone 60 should preferably select to associate with is the AP associated with the RFID tags of room k for which the highest value S _k is obtained. 2. In case where the number of tag pairs is not the same in all the rooms, T _k represent the number of tag pairs in a room k.

For each room k, RFID reader 61 of Figure 3 calculates the average value of the Maximum of RSSI from tag pairs j situated in room k, denoted as Sav _k . Thus, Sav _k = S _k IT _k ; with S _k = ∑ S _kj ; j = l,2,..., T _k .

The room where RFID reader 61 of Figure 3 is located is the room number for which the highest value of Sav _k is obtained. The probability, P _k , that the user is located in room k could be estimated as P _k = Sav _k /Sav, where Sav is the sum of all Sav _k or Sav =∑ _k Sav _k ; k = 1,2, Similarly, the AP which Mobile smart phone 60 should preferably select to associate with is the AP associated with the RFID tags of room k for which the highest value Sav _k is obtained.