HK1092315B - Improvements in or relating to distributed radio units - Google Patents

Description

Improvements in or relating to distributed radio units

Technical Field

The present invention relates to a method in a communication system comprising an access network with a Radio Network Controller (RNC) and a radio base station (RBS or node B) and one or several user equipments (UE or phone), wherein the radio base station (RBS or node B) consists of a Main Unit (MU) performing baseband signal processing and one or more Radio Remote Units (RRUs) converting between baseband and radio frequencies and transmitting and receiving signals over one or more antennas covering a cell; the user equipment (UE or phone) moves closer and closer to another cell, the network notices this and then it will initiate a handover procedure in which a call will be transferred from one cell to another cell within the radio base station (RBS or node B) or to a cell in another radio base station (RBS or node B) in the communication network.

Background

Conventional radio base stations in cellular communication systems are typically located at a single location and the distance between the baseband circuitry and the radio circuitry is short, e.g. about one meter. Due to the high attenuation of the RF feed, the antenna is mounted near the radio circuit, e.g. about 20-100 meters. The antenna systems of conventional radio base stations are therefore installed in a limited geographical area-usually within the range of the same antenna mast or roof ridge. The distributed base station design, referred to as the master-remote design, separates the baseband and radio sections of the base station. The Main Unit (MU) performs baseband signal processing, while one or more Radio Remote Units (RRUs) convert between baseband and radio frequencies and transmit and receive signals through one or more antennas. Each RRU serves a certain geographic area or cell. A respective optical link connects the main unit to each of the plurality of radio remote units.

The main remote base (base) is an alternative to placing the RBS components in a single cabinet, e.g. size and cost can be reduced by placing only those components (here referred to as RF components) that are actually needed for radio transmission in each cell.

In a WCDMA network, the base station system comprises a main unit for digital processing and a plurality of remote radio units for radio processing. Each remote unit allows one carrier and can be installed up to 15km from the main unit. Connected by fiber optic or coaxial cables. Accordingly, the antenna system of the main remote RBS is thus spread over a wide geographical area.

An example of such a system is described in US-5,761,619.

Handover is naturally done in such networks and is the process of transferring a call from one cell to another. This is necessary to continue the call while the phone is traveling. Traditionally we have three general handover types: hard, soft, idle. The type of handover depends on the situation in which the handover is located.

Soft handover establishes a connection with a new RBS (also called node B) before disconnecting from the old RBS. This is possible because the WCDMA cells use the same frequency and because the mobile station uses a rake receiver. In handover, the mobile station assists the network. When the mobile station travels to the next coverage area, it detects a new pilot signal. The new base station then establishes a connection with the mobile station. The new communication link is established while the mobile station maintains the link with the old RBS.

Softer handover occurs between two sectors of the same RBS. The RBS decodes and combines the voice signals from each sector and forwards the combined voice frames to the RNC.

An interesting plan for the main remote architecture is to install 2 Radio Remote Units (RRUs) face to face along the highway, say 30km apart. This will result in cheap road coverage.

As mentioned, the prior art is that softer handover is always used between cells/TRXs within the RBS. The MS reports that its C-PICH is strong, while the RNC notices that it is from the same RBS and orders the RBS for a softer handover.

The RAKE receiver of the RBS has a limited window (in time) for where it accepts the UE uplink signal. Any signal falling outside the window is discarded and interference is generated.

For RBSs with co-located antennas this is not a problem. The propagation delay difference of the UE to these two antennas is very small-only equivalent to 10-50 meters. There may also be a delay difference in the receiver side, but the difference is only about 100 m. There is also a mechanism within the RBS to compensate for the latter.

For an RBS with distributed receivers, the propagation delay to the two antennas may be very different. If the antennas are 30km apart and the MS requests softer handoff at 10km from the first antenna and 20km from the second antenna, the propagation delay difference will amount to 10 km. Also, communication from the receiver side to the common RAKE receiver will be subject to different delays due to different cable lengths. This will undoubtedly constitute many km of corresponding air propagation delay, subject to wiring.

It should be noted that the worst case of cable delay differences and air propagation delay differences may not occur in the same configuration.

It is clear that in the current implementation, the two different UE signals (to antenna 1 and antenna 2) are likely to eventually delay in the RAKE beyond the window- +/-3 km.

Simulations show that for a road coverage solution as described above (already including equal delay calibration as described below), the risk of dropped calls is therefore about 7%.

Disclosure of Invention

It is an object of the present invention to provide an improved handover method in a system with distributed radio units.

A second object of the present invention is to provide a method such that a UE moving from one RRU coverage area to another will be handed over to the RRU it arrives to in the same secure way as we would if we used the traditional cell structure.

A third object of the present invention is to provide a method such that the receiver performs softer handover in the RBS.

A further object is to achieve delay calibration of digital transmissions arriving at a Rake receiver, depending on the distance to the area where handover is possible.

According to a first aspect of the present invention, there is provided a method in a communication system comprising an access network with a Radio Network Controller (RNC) and a radio base station (RBS or node B) and one or several user equipments (UE or phone), wherein the radio base station (RBS or node B) consists of a Main Unit (MU) performing baseband signal processing and one or more Radio Remote Units (RRU) converting between baseband and radio frequencies and transmitting and receiving signals over one or more antennas covering a cell; the user equipment (UE or phone) moves closer and closer to another cell, the network notices this and then it will initiate a handover procedure in which a call will be transferred from one cell to another cell within the radio base station (RBS or node B) or to a cell in another radio base station (RBS or node B) in the communication network.

-said handover procedure interacts with a memory containing a list (softer handover group) of said Radio Remote Units (RRUs) that can make softer handovers with each other using the same Rake receiver.

-the handover procedure is performed according to a selection from the list and the completion of the handover is in accordance with the following:

if the new cell is in the list (softer handover group) with another cell used by the user equipment (UE or phone), a softer HO is initiated to the RBS as usual.

If the new cell is not in the list (softer handover group) with another cell used by the user equipment (UE or phone), a soft handover is initiated in the RNC or RBS.

-said soft handover in a Radio Base Station (RBS) is a second order maximal ratio combining or selection combining with respect to a respective Rake receiver.

The selection of one of these two cases can be done either with the support of the radio network controller or locally in the Radio Base Station (RBS).

-the list (softer handover group) is formed from User Equipment (UE) measured delays.

-the reception time difference is used by the Radio Network Controller (RNC) or the radio base station to calculate the relative propagation delay between the new antenna and the user equipment compared to other active cells.

-the Radio Network Controller (RNC) can either include the new cell in the list (softer handover group) according to the measurement result or, if the Radio Network Controller (RNC) is not affected, the measurement result is forwarded to the Radio Base Station (RBS) and the RBS makes the decision.

-artificial delays are stored in the Radio Base Station (RBS) to enable the two signals from the two antennas to be received within the RAKE window so that softer handover can be made.

The delay equalization function makes the digital delay between the receiver/antenna and the RAKE receiver the same for all receivers/antennas.

-optimizing the delay to maximize the number of successful softer handoffs.

The main advantage of the invention is that it is possible to achieve a secure handover between cells of a distributed system according to the main remote principle.

The handover procedure is effective even when the antennas are located in geographical areas where delays would make handover according to the prior art difficult.

The problem addressed by the present invention can be solved by separating the data from the two RRUs all the way to the RNC. Obviously, this is not very efficient in terms of baseband resources or transmission resources. For this reason, the present invention is also advantageous.

Drawings

Embodiments of the invention will now be described, by way of example, with reference to the accompanying drawings, in which:

examples of air propagation delay differences and rake receiver windows are illustrated in fig. 1 and 2.

The static solution is shown in fig. 3 and 4.

The dynamic solution is shown in fig. 5 and 6.

The delay calibration is shown in fig. 7 and 8.

In fig. 9 and 10 delay compensation with equal delay solutions is shown.

Delay compensation with an optimal delay solution is shown in fig. 11 and 12.

Fig. 13 and 14 show self-learning optimal delay solutions.

Detailed Description

The following text sets forth a glossary of abbreviations used in the present patent specification to facilitate an understanding of the present invention.

RNC radio network controller

RBS radio base station

Radio base station in Node B WCDMA

WCDMA wideband code division multiplexing

UE user equipment

MU main unit

RRU radio remote unit

In fig. 1, the MU and two antennas covering the road connected to respective RRUs are shown. This may be the preferred solution when employing the main remote concept. The antennas are opposite to each other and the antennas cover two cells. When a mobile UE moves, it will traverse different cells. If the mobile phone is not engaged in a call, it will tell the network from time to time that it has moved to another cell. If the mobile UE is talking, it is of course necessary to keep talking while the UE is moving. The process of replacing communication with one cellular radio base station with another is referred to as handover. Even while a call is in progress, the mobile telephone scans for other cells and reports to the cellular network the strength of signals received from those cells. When the cellular network realizes that the mobile phone is moving closer to another cell, it will initiate a handover procedure in which the call will be transferred from one cell to another cell within the RBS or to another RBS within its cell.

The classical way to perform a soft handover consists in comparing the power levels received from different base stations adjacent to the mobile station. It is then decided that the mobile station is in soft handoff with all of these base stations so that their power levels received by the mobile station are within a range between the maximum power level received and the power reduction by one of the predetermined factors. This factor may be expressed in decibels and represents the switching window. The group of base stations with which a mobile station communicates represents the so-called "active set" of the mobile station. With respect to other aspects, the size of the active set is limited because the mobile station cannot support an unlimited number of links. In this way, the impact of soft handover is adjusted by acting with two parameters: the size of the active set and the switching window.

Examples of air propagation delay differences and rake receiver windows are illustrated in fig. 1 and 2. Labels 1 and 2 are different user equipments and a and are different RRUs.

A1 is the signal from UE1 received by RRU a, and B1 is the signal from UE2 received by RRU B.

As can be seen from fig. 2 and UE2, B1RRU falls outside the RAKE receiver window and will not be used for signal combining. Instead, it will generate interference.

For the UE2, which was successful, both a2 and B2 were within the RAKE receiver window and would be used for signal combining and improving signal quality.

The static solution is shown in fig. 3 and 4: a and B are far apart so they are placed in different softer HO groups. The solution then allocates one RAKE for each RRU and combines the results using a second combining stage. The disadvantage is that two RAKE's are used for UE2 as well.

A dynamic solution is shown in fig. 5 and 6, and we let the UE delay measurements decide the number of RAKE receivers to be used. We allocate the number of needed RAKEs based on the delay measurements reported by the UE.

The delay calibration is shown in fig. 7 and 8. Labels 1 and 2 are different user equipments, while a and B are different RRUs.

A1 is a signal from the UE received by RRU a.

The problem is that even if the air propagation delay is the same for RRU a and RRU B, the two signals will arrive at the RAKE receiver at very different times. This is due to the large difference in digital delay (from the RF receiver to the RAKE) caused by the very different fiber lengths.

In fig. 9 and 10 delay compensation with equal delay solutions is shown. Labels 1 and 2 are different user equipments, while a and B are different RRUs. A1 is a signal from UE1 received by RRU a. The "delay circuit x" is configured to make the digital delay the same for all x, i.e. all RRUs. In this example, delay circuit A will delay the signal from RRU A by 0us, while delay circuit B will delay the signal from RRU B by 50us (9980m faster than the speed of fiber).

In fig. 11 and 12 delay compensation with an optimal delay solution is shown. In this example, RRU B is placed in urban areas with the antenna facing the street below from the ridge. The switching region from a to B is very close to B.

When the UE is in the middle of the handover region, the "delay circuit x" is configured such that the digital delay + analog delay is the same for all x, i.e. all RRUs. In this example, "delay circuit a" would delay the signal from RRU a by 0us, while "delay circuit B" would delay the signal from RRU B by 50us (digital difference) + (dA-dB) x the speed of light in air.

Fig. 13 and 14 show self-learning optimized delay solutions.

Handover areas are typically derived from cell planning, but it is often difficult to determine an accurate area.

Also, the optimal delay adjustment (dA-dB) may be different throughout the handover region. The optimal adjustment needs to be weighted against the number of users for each of these relationships.

The solution stores the measured delays of the UE each time the UE requests a soft (softer) handover and derives the optimal delay from these measurements. This will indeed optimize the probability that a and B share the same softer handover group, since the history will depend on the user distribution in the handover area. In addition, the delay is determined by evaluating UE measured delays for a history of successful handovers between the relevant radio remote units.

Softer HO groups

The solution forms softer HO groups within the RBS, where each group comprises receivers that can do softer HO with each other, i.e. using the same RAKE. If the UE wants to extend its active set and the RNC agrees to use the new cell, the following selection is made:

if the new cell is within the same softer HO group as the other cell used by the UE, a softer handover is initiated to the RBS as usual.

If the new cell is not within the same softer HO group as another cell used by the UE, a soft HO is initiated in the RNC or RBS.

Soft HO in RBS is the second stage maximum ratio combining or selective combining but with respect to the respective RAKE receiver.

The choice of one of these two bold circles can be made locally in the RBS, not only with the support of the RNC: if the RBS is assigned to perform a softer HO, the RBS checks if the new cell is located within the softer HO group of any of the other used cells. If so, it uses the same RAKE that has been assigned to that softer HO group. If not, a new RAKE is established and the two outputs are combined. This can become completely transparent to the RNC.

Dynamically softer HO groups

A more advanced approach is to evaluate the CPICH reception time difference measured by the UE and use it to classify the cell as a softer HO candidate or a soft HO candidate.

The UE reports not only which new cell it wants to establish a connection with, but also the reception delay between the new cell CPICH and its own UL transmission time that it measures. Since the UL transmission time depends on the DL transmission time of the dedicated channel, the RNC or RBS can calculate the relative propagation delay between the new cell antenna and the UE compared to other active cells.

The RNC may or may not include the new cell in the softer HO group according to the measurement result. Alternatively, the RNC is not affected, but instead the measurement results are forwarded to the RBS (as is now) and the RBS makes the decision.

To maximize the probability that softer HO can be achieved, artificial delays should be added within the RBS to achieve that both signals (two antennas) are received within the RAKE window.

Normally, a delay equalization function should be added so that the digital delay between the receiver/antenna and the RAKE is the same for all receivers/antennas. Then only the difference in the air propagation delay will be of importance.

However, it is a goal that having equal digital delays will not always be optimal that MSs in the HO area have equal delays. If the antennas, output power and terrain are similar for both antennas, the HO area will be in the middle and equal delays are optimal. However, if the antenna tilt is different, the CPICH power is different or for any other reason the HO area is not in the middle of these sites, adjustments should be made to the delay for this purpose. This can be done manually at network planning time or dynamically according to actual UE reports (optimizing delay to maximize the number of successful softer HOs).

While the invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not to be limited to the disclosed embodiment, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims (11)

1. A method in a communication system comprising an access network with a radio network controller and a radio base station and one or several user equipments, wherein the radio base station consists of a main unit performing baseband signal processing and one or more radio remote units converting between baseband frequencies and radio frequencies and transmitting and receiving signals via one or more antennas covering a cell; the user equipment moving closer to another cell, which makes the network aware of and then it will initiate a handover procedure in which a call will be transferred from one cell to another cell within the radio base station or to a cell in another radio base station in the access network, characterized in that the handover procedure interacts with a memory containing a softer handover group of the radio remote units that are performing softer handovers with each other using the same Rake receiver; wherein the softer handoff groups are formed from delays measured by the user equipment including at least the over-the-air propagation delay for the user equipment to reach the antenna and the digital delay for the receiver to reach the RAKE receiver.

2. A method according to claim 1, characterized in that said handover procedure is performed according to a selection from said softer handover group and said handover is done in accordance with the following:

if the new cell is within the same softer handover group as another cell used by the user equipment, a softer HO is initiated to the radio base station as usual;

if the new cell is not in the same softer handover group as another cell used by the user equipment, a soft handover is initiated in the radio network controller or the radio base station.

3. A method according to claim 2, characterized in that said soft handover in the radio base station is a second order maximal ratio combining or selection combining with respect to the respective Rake receivers.

4. A method according to claim 2, characterized in that the selection of one of the two cases can be done locally in the radio base station or with the support of the radio network controller.

5. A method according to claim 1, characterized in that said softer handover group is formed on the basis of delays measured by the user equipment.

6. A method according to claim 5, characterized in that the CPICH reception time difference is used by the radio network controller or the radio base station to calculate the relative propagation delay between the new cell antenna and the user equipment compared to other active cells.

7. A method according to claim 5, characterised in that the radio network controller can include a new cell in the softer handover group depending on the measurement results or if the radio network controller is not affected the measurement results are forwarded to the radio base station and the radio base station makes this decision.

8. A method according to claim 1, characterized in that artificial delays are stored in the radio base station to enable the two signals from the two antennas to be received within the RAKE window, so that softer handover is possible.

9. A method according to claim 8, characterized in that the delay equalization function is such that the digital delay between the receiver/antenna and the RAKE receiver is the same for all receivers/antennas.

10. A method according to claim 1 or 8, characterized in that the delay is optimized to maximize the number of successful softer handovers.

11. A method according to claim 10, characterized in that the delay is determined by evaluating the delay measured by the user equipment of the history of successful handovers between the concerned radio remote units.