US 20070270155 A1
A method for setting a power level in a wireless communication system given directional setting of an antenna including the steps of determining a gain adjustment factor that depends upon the directional setting of an antenna, and applying the gain adjustment factor to control a power level of a transmitted signal.
1. A method for setting a power level in a wireless communication system given a directional setting of an antenna, the method comprising the steps of:
determining a gain adjustment factor that depends upon the directional setting of an antenna; and
applying the gain adjustment factor to control a power level of a transmitted signal.
2. A method as in
3. A method as in
4. A method as in
5. A method as in
6. A method as in
7. A method for dynamically determining a steering direction of an antenna in a wireless system comprising:
providing a transmission quality metric applicable to describe a quality measure of a wireless signal;
applying the quality metric to a current steering position of the antenna;
repainting the antenna to a trial steering position;
reapplying the quality metric to the trial steering position;
comparing the quality metrics of the current and trial steering positions;
repeating the repainting by incrementally selecting a new trial steering position; and
selecting a best steering position by comparing each of the incremented trial steering positions.
8. The method of
repeating the repainting step through the selection step for a plurality of trial steering positions.
9. A method as in
identifying a frame offset in the sequence of the active time slots;
determining non-occupied time slots in the sequence based on a frame offset; and
transmitting information in directional non-active time slots.
10. A method as in
11. A method as in
computing an optimum setting from the directional information transmitted for a plurality of candidate directions.
12. A method for a user equipment device to acquire signals transmitted from a central base station in a wireless communication system, the method comprising the steps of:
selecting an omnidirectional mode for an antenna;
acquiring signals transmitted by one or more base stations with the antenna in the omnidirectional mode, and assigning such detected signals as a detected set;
for each available directional setting provided by the directional antenna, determining a received signal strength for each base station in the detected set;
for each such directional setting, determining a ratio of a detected cell having a higher receive signal strength and a sum of the signal strengths for the other detected cells; and
selecting the base station having the largest such ratio to be used as the active base station.
13. A method as in
14. A method as in
15. A method as in
This application is a continuation of U.S. patent application Ser. No. 10/434,437 filed on May 7, 2003, which claims the benefit of U.S. Provisional Application Ser. No. 60/378,698 filed on May 7, 2002, U.S. Provisional Application Ser. No.60/378,908 filed on May 7,2002 and U.S. Provisional Application Ser. No. 60/466,535 filed on Apr. 29, 2003 which are incorporated by reference as if fully set forth.
Wireless data communication networks of various types, including digital cellular systems, Wireless Local Area Networks (WLANs) and even personal area networks such as Bluetooth are increasingly viewed as an ideal connectivity solution for many different applications. These can be used to provide access to wireless equipped personal computers within home networks, mobile access to laptop computers and personal digital assistants (PDAs), as well as for robust and convenient access in business applications.
Indeed, it is estimated at the present time that approximately 10% of all laptop computers are shipped from the factory with wireless interface cards. One estimate is that this ratio will increase to 30% within the next two years. Certain microprocessor manufacturers, such as Intel, have incorporated wireless capability directly into processor chip platforms. These and other initiatives will continue to drive the integration of wireless equipment into computers of all types.
It is actually already possible in some cities to find “hot spots” where one can obtain connectivity to many different networks at the same time. Unfortunately, having tens, if not hundreds, of closely spaced networks means that interference becomes a problem. That is, although the most emerging wireless standards provide for robust signaling in the form of spread spectrum radio frequency modulation, or using Code Division Multiple Access (CDMA) over modulated subcarriers, crowding of the radio spectrum still increases noise and therefore decreases performance for all users.
The capacity of CDMA networks which use a frequency reuse factor of one is limited by both intercell and intracell interference. Techniques such as Multi User Detection (MUD) can be used to mitigate intracell interference. Intelligent management of channel power, code words, and time slots (that is, robust Radio Resource Management (RRM)) can also be used.
Of most interest to the present invention, it is also possible to use a directional, or adaptive, antenna to determine the optimal direction in which to transmit and receive signals. The directional antenna focuses the radiated power of such signals, so as to minimize interference with other transmissions.
One technique that can be used to mitigate intercell interference is a directional antenna on the mobile (remote) or so-called User Equipment (UE). To understand the advantages of doing this, consider a situation where adjacent base stations or sectors cause intercell interference on the forward link channels transmitted from Central Base Station Transceivers (BTSs) to the UEs. If these downlink (DL) channels have angular separation between the signals of the desired base station or sector and those of the interfering base stations or sectors, then the directional antenna on the UE can provide some amount of suppression of the interference. The exact amount depends on the angular separation, front-to-back ratio of the antenna, and the beam width of the antenna.
Users in adjacent cells or sectors also cause intercell interference on the uplink (UL) or reverse direction. If the directional antenna can be pointed such that most of the transmitted energy is directed to the desired base station and away from the adjacent cells or sectors, then the antenna can provide intercell interference suppression on the uplink as well. This interference suppression will manifest itself as a reduction of interference at the desired base station receiver.
The use of a directional antenna therefore contributes directly to improvements in link budget. It provides additional antenna gain over a standard omni directional antenna when operated in a directional mode. Depending on the algorithm used for steering, the additional gain can contribute directly to both the uplink and downlink link budgets. The directional antenna also reduces the effects of fading due to local scattering. The directivity of the antenna allows only a portion of the path structure created by the local environment to reach the receiver input, reducing the amount of fading. The required fast fade margin is therefore also reduced.
However, in order for the directional antenna to be most effective, it must be steered in the proper direction for both uplink and downlink. In packet switched Frequency Division Duplex (FDD) systems, the directions for both downlink and uplink typically must be the same since both the UL and DL carrier frequencies are active at the same time. Often, a compromise direction is thus picked to optimize reception in both directions.
However a Time Division Duplex (TDD) system has certain advantages over FDD when it comes to antenna steering:
Because the UE is half duplex, DL and UL pointing directions can be different, allowing an optimum direction to be selected in each case.
Because the DL and UL typically operate on the same frequency, under most conditions, the DL and UL path losses will be the same.
The frame structure of TDD is such that there are non-active time slots available for checking alternate antenna directions and computing the antenna steering metrics.
The present invention relates to a technique for setting an optimum direction for an antenna in a wireless communications system. The system has an automatically steerable antenna which can be selected to a number of different angular settings. The antenna is first set to a candidate setting. A metric is then measured with that setting; the metric is associated with utilization of the antenna at the candidate setting, such as by determining the relative quality of the signal received in the candidate setting. Such metrics are then taken for at least two different candidate settings and a best result, or optimum setting result is determined. The process is then repeated using different metrics for an optimum setting for different communication channels in the system. Specifically, optimum settings are determined for a communication channel operating in an uplink direction using a metric that is different for a communication channel operating in a downlink direction. The invention therefore may result in different optimum settings being determined for the uplink and downlink channels.
In the preferred embodiment the candidate settings provided by the steerable antenna include at least an omnidirectional mode, a right directional mode, and a left directional mode; although antennas having a lesser or greater number of directional modes may be employed.
Metrics are typically measured during the reception of appropriate signals such as pilot channel signals, but may also be taken on other types of signals such as data payload signals.
The invention is particularly advantageous, in a Time Division Duplex (TDD) system, since metrics can be taken during inactive time slots. In addition, coordination for uplink and downlink communication channels is more readily determined, since in typical TDD systems, either the uplink or the downlink is active in any given time slot.
In additional aspects the invention applies the determined optimum antenna setting to select different gain adjustment factors. The gain adjustment factors, which are indicative of the path loss associated with the system operating in the particular determined directional mode, are then applied as correct algorithms that control a power level of transmitted signals. Thus, for example, typical closed loop power control algorithms that are used to set uplink and downlink power level settings are adjusted for the antenna gain associated with a particular optimum setting.
In further aspects, the optimum directional settings may be stored in a register, memory, or other storage device and read out in synchronism with known assigned active time slots. Thus, a system which requires the antenna to be steered to a particular direction in a given time slot may be readily steered to another optimum direction in the next adjacent time slot.
Other aspects of the invention are directed to optimizing an initial acquisition mode. During an initial detection mode, system parameters are determined with the antenna in an omnidirectional setting. However, an additional set of procedures is advantageously performed after initial detection, in order to determine the best pointing angle for subsequent transmissions. In this situation, measurements taken during the omnidirectional mode are adjusted to achieve the same sensitivity, in terms of maximum expected path loss, as if the original search were conducted in a directional mode. The sensitivity improvement factor is determined as gain difference between an omnidirectional and directional mode, as measured during the subsequent processing. This aspect of the invention allows for extra gain in a directional mode to be used to provide still increased coverage for the system
The foregoing and other objects, features and advantages of the invention will be apparent from the following more particular description of preferred embodiments of the invention, as illustrated in the accompanying drawings in which like reference characters refer to the same parts throughout the different views. The drawings are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the invention.
A description of one preferred embodiment of the invention follows. This particular embodiment is for a cellular wireless communication-system that is known as Universal Mobile Telephone System (UMTS) Wideband Code Division Multiple Access (W-CDMA) as specified in the Telecommunications Industry Association (TIA) Ti.3GPP.25 series of specifications; it should be understood that the principles of the invention may be applied to other types of wireless systems, however.
In any cellular system there is generally a central Base Transceiver Station (BTS) or access point (AP), that communicates to remote units, or User Equipment (UE). Any such system is bi-directional in nature, that is, it must provide the ability to communicate from the BTS to the UE (downlink) and from the UE to the BTS (uplink). The duplexing of the uplink and downlink communication channels may be performed in two primary ways. These are Frequency Division Duplexing (FDD) and Time Division Duplexing (TDD). FDD separate carrier frequencies are utilized to facilitate the uplink and downlink communications. For Time Division Duplexing (TDD), the same frequency band is used for both the uplink and downlink. With TDD however, the uplink is active during certain time slots, while the downlink is active during others. The communication channels from user to user may be either time modulated in the separate time slots (TDMA) or may be Code Division Multiple Access (CDMA). Both TDD/CDMA and TDD/TDMA based access approaches are thus seen in use.
For TDD systems, the propagation channel is reciprocal meaning that the path loss and fading statistics of the channel will be the same in both the BTS to UE and UE to BTS transmission directions, since the same carrier code frequency is used in both directions.
Turning attention to
The antenna 10, in accordance with the preferred embodiment of the present invention, is a directional type antenna. It therefore consists of multiple radiating elements 12-1, . . . 12-n connected through a directional controller 14. The directional controller 14 accepts control inputs 16 to set a transmit or receive direction for the antenna elements 12. The directional controller 14 may consist of switches, phase shifters, or other components such as lumped impedances, to affect directivity of the combination of the antenna elements 12 in various ways that are known in the art.
A diplexer 18, in the case of a TDD system properly thought of as a transmit/receive switch 18, allows both the receiver 20 and transmitter 30 to be connected to the antenna subsystem 10.
The receiver 20 consists of a receive amplifier 21, radio frequency (RF) down converter 23, Voltage Controlled Oscillator (VCO) 22, and an in-phase and quadature demodulator 28 that consists of a phase shifter 25, a pair of mixers 26-1,26-2 and a pair of band pass filters 27-1, 27-2. In a manner which is well known in the art the receive amplifier accepts a received signal, amplifies it, and then provides it to the RF down converter 23.
Given an input frequency from the VCO 20 and possibly band select inputs 24, the RF down converter 23 shifts the incoming radio energy to intermediate carrier (IF) frequency signal. The I/Q demodulator 28 then provides receive data symbols as output in both an in phase (I) and quadrature (Q) channels.
For the transmitter 30, transmit data symbols are provided to an I/Q modulator 38 that itself consists of a pair of mixers 31-1, 31-2 quadrature phase shifter 32 and summer 33. Fed by a local reference signal, the I/Q modulator 38 provides a modulated signal to at an IF frequency to RF up converter 34. The transmit VCO 35 and any band select inputs 36 then up-convert the IF signal to a desired RF carrier frequency. The transmit amplifier 37 then provides the modulated transmit signal to the diplexer 18, which is in turn controlled by a transmit receive (T/R) input. A level detector circuit 38 may be used to detect the transmitted RF power level. The T/R input is controlled according to time slot allocations provided by the Base Station on a control channel. During some time slots, the diplexer 18 is set to receive mode (i.e., the downlink is active); at other times it is set to transmit mode (i.e., the uplink is active); and at still other times it is left open (inactive timeslots). A similar a level detect circuit 29 may also be used to determine the receive level detect signal level.
In accordance with this particular embodiment, the antenna subsystem 10 has three modes selectable by the directional control input 16. Turning attention to
Algorithms envisioned by the present invention utilize a measure of receive signal power and interfering signal power as measured at the UE. These can be provided by additional receive circuitry as shown in
An estimate of received interference power that is associated with adjacent base stations 51-2, 51-3, 51-4 can be provided by additional demodulators 63-1, 63-2, . . . 63P, summing circuit 64, and subtractor circuit 65. That is, by applying the additional demodulators having applied thereto inactive CDMA codes from BTS 51-1, that is the codes for channels that are not presently utilized by the UE, a relative signal power of such intracell interfering channels are individually provided by taking their sum in summing circuit 64. This is then removed from an estimate of the overall receive power (by subtractor circuit 65) to estimate the interfering code channel signal power (ICSP).
It can now be appreciated that the invention will need to select an optimum antenna mode setting for every particular active time slot in both the uplink (UL) and downlink (DL) directions.
An example situation may develop therefore as shown in
The overall result of applying an algorithm according to the present invention is that not only is the antenna position optimized, but also that a corrected path loss calculation may be made. Specifically, as shown in
In first step of the method, time slot assignments are obtained. These assignments are for both assigned active slots and assigned inactive slots on the downlink as was described in connection with
A next step for each monitored slot at state 92 at state 94 both the RCSP and ICSP are measured for each antenna mode. Thus, the antenna 10 is temporarily set in each of the three modes omni directional, right or left and both RCSP and ICSP are determined.
In the state 96, the RCSP and ICSP measurements for each mode of the antenna may then be filtered through appropriate low pass filtering and/or averaging calculations.
In state 98, the User Equipment (UE) then sends a report back to the Node B of these measurements. These would include at least the RCSP as measured for the active time slots as well as the ICSP as measured in both the active slots and inactive slots. This allows the Node B to determine the signal to interference ratio for active slots. With this information, it can then therefore assign active slots to User Equipment (UE) based upon observed signal to interference ratio (SIR) taken with various antenna settings.
From this state, in state 102 the best pointing direction can then be determined by computing SIR as a ratio of RCSP to ICSP for active and inactive slots. That is, the RCSP is taken for active slots and the ICSP measurements are taken from inactive slots (where it is known that only interference signal power was present). In this manner, an optimal antenna setting either omni directional, left or right can be determined from measurements taken.
The RCSP and ISCP active slot measurements in the downlink (DL) direction for other antenna pointing directions may be taken on appropriate signals and during times, for example, when no critical user data is being sent. This may be taken during transmission of pilot synchronization, or similar signals but it should be understood that the measurements can also be taken on data signals and in other instances.
Given a directional setting for the antenna, the forward (DL) path loss can then be estimated as shown in
In another manner of thinking, the estimate for path loss for both directions must be corrected. The forward link adjustment is used to correct path loss for antenna gain. Secondly, a reverse link adjustment is used to correct an actual reverse link transmit power level setting at the user equipment, the next time the user equipment attempts to transmit back to the base station
In alternate schemes, the base station loading of adjacent base stations may also be monitored in the state 154. That is, each base station may periodically broadcast a data indicative of how relatively busy it is, that is how many timeslots it is active in. If, in state 156 a neighboring base station is detected as being particularly loaded, then a direction is selected in state 158 that reduces interference to that particular neighbor. Thus, for example, the user equipment (UE) may determine that an adjacent base station 51-3 is particularly busy, having most of its time slots assigned to an active state. If this particular neighbor is relatively busier than other neighboring base stations, and especially if the presently assigned base station is not particularly busy, then a direction will be selected which reduces interference to the busy neighboring base station. This may therefore improve the overall system performance with minimum degradation to the desired link between LIE and the desired base station 51-1.
Again, even in the uplink direction, the path loss calculation should be adjusted for differences in gain of the uplink and downlink antenna direction settings, as in step 160.
In a TDD system, some sort of steering of the array must therefore be achieved in order to implement the optimized antenna setting. The steering of the array may be performed by software sending messages to the directional control 16 inputs if the software is fast enough. However, if the software cannot execute quickly enough, a hardware state machine that operates under control of the software may be needed. This Array Setting Coprocessor (ASCP) may be as simple as a register that contains the value for the next time slot. In such an embodiment, the values would be readout at the correct time.
A more sophisticated ACSP may require multiple registers that include the values for the next N array positions to be set. The software could write to this register prior to the actual arrival of the time slots. At the correct time, the ACSP would then apply the settings to control inputs 16 on the antenna array.
An example of a more sophisticated ACSP is shown in
Further aspects, as suggested by
Description for UMTS-TDD
The reader will now appreciate a more detailed description of how the invention can be used in a UMTD-TDD system environment. The following discussion assumes the reader is familiar with that specification, and the various channels it defines.
As defined in the above-referenced W-CDMA specification, the steady state case is either CELL FACH or CELL DCH. CELL DCH is the state in which the UE is actively engaged in voice and or data traffic. DL and UL slot/code/power allocations may be changing under RRM control and the UE may be mobile. CELL FACH is the state in which the UE does not have allocated resources, noise measurements are being reported for a future allocation by RRM, and the UE may be mobile.
The algorithm concept outlined in
1.11. RX Signal Power Measurements
Signal power measurements are also made by the UE in support of other RRM functions or a common control channel pilot power signal, P-CCPCH RSCP. The addition of a directional UE antenna requires that the P-CCPCH RSCP measurements be made at all possible pointing directions of the antenna. If a three-mode antenna is used with the UE, the P-CCPCH RSCP measurements must be made in the omni mode, left beam, and right beam mode. The P-CCPCH RSCP measurements are made only on slots on which the UE is receiving data (CELL DCH) on either traffic or broadcast channels. With a directional antenna, this measurement will-be for a selected one of the three beam positions (omni, left, right). The measurements for the other two directions can be made on the P-CCPCH when the UE is not required to receive data, since SIR in those directions may be degraded as compared to the current direction. Since the PCCPCH is always sent from the base station with the base station antenna in omni mode, measurement of the P-CCPCH receive power at other positions during non-active slots will be the same as the P-CCPCH receive power of the active slot at that same direction (assuming the slots are close in time, and assuming that some amount of averaging is done as in step 96 (
1.1.2 Intercell Interference Measurements
Intercell interference measurements are also made by the UE in support of RRM through time slot ISCP measurements. For the purposes of this discussion, ISCP is assumed to provide intercell interference power only. The directional UE antenna requires that the time slot ISCP measurements be made at all possible pointing mode directions of the antenna. If a three mode antenna 10 is used with the UE, the ISCP measurements thus are made in the omni, left beam, and right beam modes for each possible interfering code. For measurements of the active slot(s), the measurement can only be made at the current pointing position (although, steering the antenna would upset the receptor of an active channel). By definition, other positions for the active slot will have degraded SIR, so attempting to receive data (CELL DCH) at other positions may tend to degrade performance.
1.1.3 RX Signal Power Reporting
P-CCPCH RSCP measurements are also reported to the Radio Node Controller (RNC) periodically for power control, and for DL and UL time slot allocation purposes. With a directional antenna such as the three mode array 10, three different measurements are made of RSCP. All three measurements are used by the UE in determination of the future pointing directions for the antenna, but only the RSCP measurement of the current pointing direction is reported to the RNC. The raw RSCP measurements undergo various levels of filtering (step 92) before they are used by the UE for reporting (in step 98). The RSCP measurements for all directions are filtered in the same way as the current implementation, regardless if it is used for reporting or not. The reporting uses the average associated with the current pointing direction.
1.1.4 Intercell Interference Reporting
Time slot ISCP measurements are also reported to the RNC periodically for forward channel (DL) allocation purposes. The RNC specifies the time slots that are to be measured. With the three mode directional antenna, three different measurements are made of ISCP for each time slot. If the ISCP slot report is for an active slot (a slot being used by the UE), the ISCP value reported is the value taken from the current pointing direction of the antenna. If the ISCP slot report is for a non-active slot (a slot under consideration for future assignment), the ISCP value reported is the value taken from the direction of the antenna that provides the highest SIR. The raw ISCP measurements undergo various levels of filtering before they are used by the UE for reporting. The ISCP measurements for all directions and all slots are filtered in the same way as the current implementation, regardless if it is used for reporting or not. The reporting step 98 thus uses the average associated with current pointing direction for active slot ISCP reporting, and the average associated with the direction of maximum SIR for non-active ISCP slot reporting.
1.1.5 Pointing Direction for Active Slots
The current pointing direction of the UE antenna is determined by computing the SIR (ratio of RSCP to ISCP) for the active slot or slots for all pointing directions. The antenna position is then adjusted to point in the direction of highest SIR, just prior to the next slot time. The RSCP and ISCP values reported for the active time slot would be those values used to determine the SIR for the direction selected. It should be noted that if the Node B (the assigned Base Transceiver Station) is itself using directional beam former antennas, the SIR computed by the antenna steering algorithm will not be the same as the SIR maintained by thresholds in forward power control, since the RSCP for the P-CCPCH will be calculated with the base station transmitting in omni mode. However, since the directional antenna is reducing adjacent cell interference, the SIR trend over pointing directions with the Node B transmitting in omni should be the same as the SIR trend over pointing directions with the Node B transmitting with a beam former in relative terms.
1.1.6 Pointing Direction for Non-Active Slots
Every non-active DL slot measured by the UE will have an independent pointing direction. The pointing direction for non-active slots is the direction providing the highest SIR. If a non-active slot is allocated to the UE, the antenna position is then adjusted to point in the direction of the highest SIR just prior to the slot time. That slot is then considered an active slot.
1.1.7 Forward Path Loss
As we mentioned with
1.1.8 Closed Loop Forward Power Control
Integration of the directional antenna also affects forward (DL) power control when the antenna direction is changed for an active slot. The forward (DL) power control will potentially see a step change in forward power and more importantly a step change in SIR when the position is adjusted. If the antenna position is changed, the new position will have a higher SIR than the current position (or the position change would not have been made). The tendency will be for the forward (DL) power control loop to reduce power when the antenna position is changed. The reduction will take place over some period of time. During this period of time the link will operate above the target SIR (better link) until the forward (DL) power control recovers. Also during the forward (DL) power control loop adjustment period, the measurement averages of RCSP for the slots can continue, since those measurements are made on P-CCPCH, which is transmitted at a constant power.
In cases in which the UE is assigned multiple downlink slots, the direction for each slot could be different. Since the SIR for each slot is different even with an omni antenna, it is assumed that the forward power control maintains a set of variables for each downlink (DL) slot, such that the link performance is maintained independently for each slot. The directional antenna switching time is less than 100 ns, so the forward power control would not be affected on slot-to-slot direction changes if each slot were controlled independently. If a multi-slot allocation were controlled as one slot, the ISCP and RSCP values for the multi-slot would have to be averaged to determine a compromise direction for all the slots in the allocation. Otherwise the forward (DL) power control would be attempting to compensate for antenna direction changes.
1.1.9 AGC Attack Time
The Automated Gain Control (AGC) circuitry associated with receiver amplifier 21 (
1.2 Uplink (UL) Steering
The UL can be steered based on maximum received power at the Node B or on a set of metrics that minimizes the adjacent cell interference caused by the UE. Depending on the excess UE transmit power available, the UL steering algorithm may be a combination of the two. In cases where the UE is near maximum output power, the steering should be based on maximizing the receive power at the Node B. However, in cases where there is excess power available, the pointing direction should be such that it minimizes the adjacent cell interference caused by the UE.
Use of the directional antenna on the UL can mitigate intercell interference at an adjacent base station by minimizing the amount of energy a UE transmits to an ‘adjacent cell. Two algorithms could be used to point the UL to minimize intercell interference. The first approach is to measure the UE transmission at all adjacent Node B's and determine; the direction which maximizes the signal to adjacent cell interference. However, this requires coordination between not only adjacent Node B's but also between the Node B and the UE. A second preferred algorithm, described below, relies on the reciprocity between the uplink and downlink path losses to measure the forward path losses from adjacent base stations and infer the reverse link path losses. A pointing direction can be determined which maximizes the ratio of desired Node B power to adjacent Node B power(s). If the intra-frequency search algorithm is modified to accommodate the directional antenna then many of the necessary calculations are already available. The data for this algorithm is available in the UE and does not require coordination with the Node B.
There will be times when minimizing the adjacent cell interference on the UL will result in the loss of receive power at the desired Node B. If the UE is operating at near maximum path loss, any loss of received power at the Node B would be detrimental to performance of the UE. Under this condition, the UL should be steered solely on received power at the UE to maximize the UL power at the Node B.
1.2.1 RX Signal Power Measurements
The UE measures the DL power at all pointing directions, assumes reciprocity, and thus infers what the UL receive power will be at the Node B for a specific direction. RSCP for the P-CCPCH at all pointing directions is already being measured for the DL pointing algorithm. For each pointing direction, the UE will calculate an estimate of reverse link path loss to its Node B, based on a measurement of forward link path loss (through RSCP) at every pointing direction.
1.2.2 Intercell Interference Measurements
During assigned but idle time slots, the UE measures the DL power at all pointing directions for adjacent base stations, assumes reciprocity, and infers what the UL receive power will be at those Node B's, for a specific pointing direction. RSCP for the P-CCPCH must be measured for the adjacent Node B's during intra-frequency searching for potential handoff. Once the directional antenna is integrated, the intrafrequency search must also take into account the RSCP for adjacent Node B's at all pointing angles. Thus the data required to support intercell interference measurements on the UL can be obtained from the searcher. For each adjacent Node B, the UE will compute the reverse link path loss at all available pointing angles. This data will be used to determine the pointing direction for the UL.
1.2.3 Pointing Direction
As mentioned in connection with
Since all estimates are based on measurements of the forward link P-CCPCH, there is no slot-to-slot dependence for either the maximum received power direction or the minimum interference direction. Therefore, there is only one uplink direction for all uplink time slots.
An example calculation for the interference direction is shown in Table 1
In Table 1, even though left and omni provide the same receive power at the Node B, left beam mode would be selected because it provides 3.5 dB less interference power to the Adjacent Node Bs. Left and right beam modes are about the same in interference power, however right would require 5 dB more UL power (and thus more interference power) at the desired Node B, so left is the better choice.
A further level of sophistication would be to monitor the Node B loading during intra frequency search and then weight the UL interference calculations accordingly. For instance, in the above example adjacent Node B 51-2 would see a 5 dB increase in interference if left is selected over omni. If the UE could determine that Node B 51-2 is severely loaded by monitoring P-CCPCH of Node B 51-2, then the UE could potentially select omni instead of left to reduce the interference load at Node B 51-2.
1.2.4 Open Loop Power Control
As also mentioned in connection with
1.3 Coordination of UL and DL Pointing Changes
Because the Node B may itself be using a beam former, the UE should change the UL and DL pointing directions at a rate that is slower than that of the beam former at the Node B. It may also be advantageous to have the DL and UL directions change only at certain time intervals. If the Node B is using beam formers, the DL beam is being determined based on UL data. If the UL direction on the UE is changed, the Node B may adjust its UL beam based on that change and subsequently adjust its DL beam. The SIR numbers computed for the DL by the forward power control in the UE thus may no longer be valid for the new DL beam position. The SIR estimates for DL pointing are valid relative to each other because they were taken from the P-CCPCH in omni. At any time the UL UE pointing direction is adjusted, the forward power control may need some time to adjust if the Node B is using beam formers. The forward power control will also need some time to adjust if the DL UE direction is changed. In order to keep the number of forward power control transients down, it is recommended that the changes in UL and DL directions occur periodically, at the same time, and at a rate slower than both the beam former update and forward power control update.
1.4 Timing Adjustments
Any timing adjustments of slot times based on the path structure at either the UE or Node B receive may be affected by integration of a directional antenna. This is due to the fact that an omni-directional antenna, or the directional antenna in omni mode, will see all incident paths at the UE receiver, but the relatively high front to back ratio of the directional antenna combined with the directional mode (left or right) may result in a subset of the path structure received by the UE. Furthermore, the relative amplitude between the paths may change between the omni mode and a directional mode. If one assumes reciprocity in path structure, then any change in UE pointing direction on the UL will also affect the path structure seen at the Node B receiver.
During initial cell site selection, the UE computes RSCP for the P-CCPCH of each detected Node B. With the integration of the directional antenna, the RSCP must be measured at all pointing angles. The preferred procedure is to perform initial detection of cell sites in omni mode, and then further qualify each detected cell site with measurements at all pointing angles. Thus the initial detection is identical to detection with a single omni antenna. However, an additional set of procedures is performed after the initial detection to determine the best pointing angle for each detected cell. In particular, the multi-dwell (i.e., over multiple time slots) searcher used for the omni search is adjusted to achieve the same sensitivity in terms of maximum path loss, as it would have had if the search were conducted in directional mode. The sensitivity improvement is the absolute gain difference between omni and directional mode. This allows for extra gain in directional mode to be used for increased coverage. The final selection for initial acquisition can be made based on the highest RSCP over all angles for all detected cells, the maximum forward SIR over all angles for all detected cells, the maximum estimated reverse SIR, or a combination of the three. SIR can be used to select the forward link direction and received power was used to determine the initial reverse link pointing direction (reciprocity is assumed), for an FDD system. This is a compromise direction, which is selected based on the forward and reverse link directions.
For a TDD implementation, the forward SIR of interest is that which is maintained by the forward power control for an allocated slot time. Since this is not known prior to CELL_FACH, the SIR calculated by the searcher would be a value based on P-CCPCH SIR measurements of individual Node B's or a ratio of RSCP measurements between Node B's. Use of the cell with the highest RSCP would minimize the path loss between the UE and the Node B, this may be desirable under conditions when the downlink is marginal and the uplink would be near maximum output power (same DUUL direction). Under normal conditions choosing the cell with the highest ratio of RSCP to the other cells would most likely give the best performance when both uplink and downlink are considered together (same DUUL direction). Since the ratio of RSCP is the same criteria used to steer the UL during steady state, using this direction for initial connection with the Node B allows a potential UL beam former to maintain the same direction during call setup, and minimizes the impact to adjacent cells.
The TDD initial cell acquisition procedure can thus be as follows:
a. Select omni mode; acquire cells as in prior art implementations (detected set).
b. For each additional direction (left and right for a three mode antenna), compute the RSCP for each detected cell.
c. For each direction setting (omni, left, right) compute the ratio between the cell with the largest RSCP and sum of the RSCP for the other detected cells.
d. Select the cell/direction with the largest ratio (active set).
Ideally, cell reselection should be based on the same criteria as initial cell acquisitions. Measurement of each detected Node B's P-CCPCH RSCP should be used to calculate the pointing direction for the UL and DL together using the maximum interference ratio. However, since handover is based on measurements of RSCP reported to the Node B by the UE, the Node B has ultimate control over cell reselection. Without modification of the measurement reporting scheme and potentially the cell reselection algorithms at the base station, there are two options.
The first option is to collect the RSCP on the P-CCPCH of the monitored set at all possible pointing directions. The RSCP that gets reported or used for any configured measurement is the RSCP for the monitored set member at its best pointing direction. The UE maintains the history of the pointing directions associated with each set member, and if the base station commands a handover to a new cell, the UE uses the direction associated with the new active set member. This allows the base station algorithms to remain unaffected and oblivious to the UE antenna. However, from an interference mitigation standpoint, the cell selected based on the highest RSCP may not be the best from an interference standpoint.
The second option is to collect the RSCP on the P-CCPCH of the monitored set, and compute the best cell based on the maximum signal to interference ratio. When the best cell/direction based on interference differs from the best cell/direction based on highest RSCP, then the RSCP measurements reported to the Node B are biased by an amount to cause the Node B to allow handover to the best cell from an interference standpoint. This would only be done under conditions when the UE had excess transmit power to deal with any loss in receive power at the base station by making such a selection.
While this invention has been particularly shown and described with references to preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the scope of the invention encompassed by the appended claims