Measurement-assisted dynamic frequency-reuse in cellular telecommuncations networks

TECHNICAL FIELD OF THE INVENTION

0001The invention is related, in general, to the field of wireless telecommunications and, in particular, to improvements in frequency-reuse in cellular telecommunications.

BACKGROUND

0002Frequency reuse patterns are cell-based schemes for assigning the frequency channels available within a particular cellular telecommunications system. The most basic unit of any frequency reuse pattern is a cell. Each cell within a frequency reuse pattern is assigned a number of frequency channels. A plurality of cells are then associated together and referred to as a cluster and utilizes all of the frequency channels available to a particular cellular telecommunications system. Groups of clusters are then used to provide a cellular coverage area within the cellular telecommunications system and the frequency channels allocated for one cluster are reused in other clusters. The scheme for recycling or reassigning the frequency channels throughout the serving coverage area is referred to as a reuse plan. The distance between a first cell using a particular frequency channel within a first cluster and a second cell using the same frequency channel within a second cluster is further known as a reuse distance.

0003The reuse of the same frequency channels by a number of different cells implies that cells may suffer from co-channel interferences. It is therefore desirable for the received strength of the serving carrier (C) within each cell to be higher than the total co-channel interference level (/). As a result, the higher the carrier to interference (C/I) value, the better the speech quality. A higher <i>C</i>/<i>I</i> value is obtained partly by controlling the channel reuse distance. The larger the reuse distance between adjacent cells utilizing the same frequency channels, the lesser the co-channel interferences created between those cells. The <i>C</i>/<i>I</i> ratio is further related to a frequency reuse plan (<i>N</i>/<i>F</i>) where <i>N</i> indicates the number of sites included within a single cluster and F indicates the number of frequency groups. For example, the <i>C</i>/<i>I</i> ratio is directly related to the following equation: <i>D_R</i>=(<i>3</i>*<i>F</i>)^{<i>1</i>/<i>2</i>}·<i>R</i>, where: <i>D_R</i> is the reuse distance; <i>F</i> is the number of frequency groups; and, <i>R</i> is the radius of a cell. Accordingly, the larger the <i>F</i> value, the greater the reuse distance. However, it is not always desirable to use a larger <i>F</i> value to increase the <i>C</i>/<i>I</i> ratio. Since the total number of available frequency channels (<i>T</i>) is fixed within a particular mobile network, if there are <i>F</i> groups, then each group will contain <i>T</i>/<i>F</i> channels. As a result, a higher number of frequency group (<i>F</i>) would result in a fewer channels per cell and lesser call capacity. Furthermore, in a packet data cellular system (such as in Evolved UTRA) packet transmission takes place over a shared channel where resources are shared by several users. This means a very large number of users may have to compete for the limited resources, reducing peak user bit rate and thereby increasing the packet delay transmission. Increased packet delay is undesirable as it adversely affects the service quality.

0004For most cellular systems, capacity is not a major issue when the system initially goes into operation. Therefore, in order to achieve a high <i>C</i>/<i>I</i> value and to improve the quality of speech connection, a high frequency reuse plan (<i>N</i>/<i>F</i>), such as 9/27, is initially used. However, as the capacity increases, the cellular telecommunications network has to resort to a lower frequency reuse plan, such as a 7/21 or 4/12, to allocate more frequency channels per cell. Moreover, the success of such systems requires that they are able to offer high peak bitrate and shorter packet transmission delay already during their initial deployment.

0005In cellular communication there are generally two main modes of operation for duplex transmission on uplink and downlink: Frequency Division Duplex (FDD) and Time Division Duplex (TDD), with their usage typically dependent on the frequency band used. FDD uses paired band where uplink and downlink transmission takes place at different carrier frequencies. Generally, there is also a fixed relation between a frequency band used for uplink and downlink transmission. TDD is used on unpaired bands where common carrier frequency is used for uplink and downlink transmission. One potential advantage with TDD is that frequency bands are more efficiently used. Secondly, the total available radio resources which are defined in terms of uplink and downlink time slots can be dynamically interchanged. This means asymmetric traffic between uplink and downlink can be better handled by adjusting uplink and downlink capacity (<i>i.e.,</i> time slots).

0006Other art in a related technical field is disclosed in <patcit id="pcit0001" dnum="WO03005752A1"><text>WO 03/005752 A1</text></patcit>, which discloses a method in an RNC of controlling radio resources assigned to a communication between a terminal and a radio network applying spread spectrum and micro diversity, comprising an RNC and fixed transceivers. However, the resources that are controlled are the active sets of transceivers or transmission power of the transceivers in the active sets. They are controlled to optimise the micro diversity gain. Furthermore, the need to add or remove transceivers from the active sets or to adapt transmission power levels is assessed based on measurements of propagation channel parameters which are used to determine the energy distribution over the propagation paths in the propagation profiles. The measurements comprise power measurements taken on the pilot channels or signals.

0007Accordingly, there is a need in the art for improved methods, and systems for employing such methods, to optimize frequency reuse in cellular communications systems.

SUMMARY OF THE INVENTION

0008To overcome the deficiencies of the prior art, the present invention discloses a method for dynamically distributing resources to a plurality of cell regions in a cellular communications network according to claim 1 and a corresponding first network node according to claim 16. Preferred embodiments are disclosed in the dependent claims. The novel method can be implemented in a conventional radio network resource controller, such as a Radio Network Controller, or other node, in a Global System for Mobile communications (GSM) and Evolved Universal Terrestrial Radio Access Network (E-UTRAN) telecommunications network. The E-UTRAN will use orthogonal frequency division multiple access (OFDMA) in the downlink and single carrier frequency division multiple access (SC-FDMA) in the uplink. The E-UTRAN will employ both time division duplex (TDD) and frequency division duplex (FDD) mode of operations. In both OFDMA and SC-FDMA based systems, the available bandwidth is sub-divided into several resource blocks or units as defined, for example, in 3GPP TR 25.814: "Physical Layer Aspects for Evolved UTRA". According to this document, a resource block is defined in both time and frequency. According to the current assumptions, a resource block size is 180 KHz and 0.5 ms in frequency and time domains, respectively. The overall uplink and downlink transmission bandwidth can be as large as 20 MHz. The principles of the invention, however, are not limited to a particular technology standard, but are adaptable to most conventional wireless network-topologies and technologies.

BRIEF DESCRIPTION OF THE DRAWINGS

0009A more complete understanding of the method and apparatus of the present invention may be had by reference to the following detailed description when taken in conjunction with the accompanying drawings, wherein: <ul id="ul0001" list-style="none" compact="compact"><li><figref idref="f0001">FIG. 1</figref> illustrates an exemplary cell having two frequency-reuse regions;</li><li><figref idref="f0001">FIG. 2</figref> illustrates a first exemplary dynamic frequency-reuse scheme;</li><li><figref idref="f0002">FIG. 3</figref> illustrates a second exemplary dynamic frequency-reuse scheme;</li><li><figref idref="f0002">FIGS. 4-A and 4-B</figref> illustrate exemplary scenarios for triggering frequency-reuse re-allocation;</li><li><figref idref="f0003">FIGS. 5-A and 5-B</figref> illustrate exemplary scenarios for triggering frequency-reuse re-allocation;</li><li><figref idref="f0003">FIG. 6</figref> illustrates a first network topology in which the principles of the invention can be implemented;</li><li><figref idref="f0004">FIG. 7</figref> illustrates a second network topology in which the principles of the invention can be implemented; and,</li><li><figref idref="f0004">FIG. 8</figref> illustrates an exemplary method for dynamic frequency-reuse re-allocation in accordance with the principles of the invention.</li></ul>

DETAILED DESCRIPTION OF THE DRAWINGS

0010In a simple scenario, sets of channels (i.e., carrier frequencies) are assigned to each cell with a certain frequency reuse pattern. In this case, there is no partition within the cell in terms of channel assignment and the assigned carrier frequencies can be used throughout the cell. In another scenario, as illustrated in <figref idref="f0001">Figure 1</figref>, a cell can be divided into two (or more) regions. In the example illustrated in <figref idref="f0001">Figure 1</figref>, the two regions are concentric. In the inner region 101 of the cell, the frequency reuse is 1, whereas in the outer region 102 (cell border region) frequency reuse is k (k>1). In the downlink for a given service, a user equipment (UE; <i>e</i>.<i>g</i>., a wireless terminal) requires lower base station transmitted power in the inner region compared to that in the outer region (i.e. cell border region) of the cell. In accordance with UE position and mobility profile, the base station transmitted power is generally controlled by dynamically compensating the loss due to distance and fading behavior. In the uplink, power control can also be used; <i>i.e.,</i> a UE transmits with lower power when close to the cell and with higher power when in the cell border region. The main advantage of this approach is that carriers are more efficiently utilized and interference in the cell border is minimized. In principle, a cell can be divided into multiple frequency reuse regions. The most usual and practical scenario, however, is that of two partitions as illustrated in <figref idref="f0001">Figure 1</figref>. One problem with conventional frequency reuse schemes is that fixed resource assignment in different reuse regions leads to inefficient resource utilization. This is due to the fact that loads in different regions (<i>e</i>.<i>g</i>., inner cell and cell border regions) can vary over time, but the resources are not reassigned between different regions on a dynamic basis.

0011Utilizing two different reuse partitions in a cell significantly reduces inter-cell interference in the cell border. There is, however, throughput degradation due to the fixed resource assignment in different partitions. The present invention recognizes that efficient realization of dynamic, or semi-dynamic, variable frequency reuse requires measurement reports from radio base stations, or access points, and possibly user terminal measurements. Conventional measurements, such as cell load, transmit power, received power, and Block Error Rate (BLER), however, are not sufficient for a dynamic frequency reuse scheme.

0012According to the principles of the invention, certain measurements performed by radio access points or user terminals are reported to a controller on a periodic basis or in response to a predefined triggering event. Based on the reported measurements (<i>e</i>.<i>g</i>., resource activity), the controller dynamically distributes resources between cell regions with different frequency reuse. The controller can further improve the resource assignment in different regions by utilizing other measurements, such as transmitted power statistics above a predefined threshold or channel quality statistics in neighboring cells above a threshold.

0013Referring now to <figref idref="f0001">Figure 2</figref>, illustrated is a first exemplary dynamic frequency-reuse scheme. In this example, each cell is partitioned into two regions 201, 202 for the purpose of resource assignment. The cell border region 202 of each cell can be defined by any state-of-the art technique, such as those based on the measurement reporting of received signal strength or received quality measured by a user terminal on a common pilot signal. In this example, let G∈{C_1, C₂,...,C_N} be the set of available channels (<i>e</i>.<i>g</i>., frequency carriers/frequency chunks/time-frequency resource block) to be assigned by a resource controller 203 to each cell with two partitions; in this example, the resource controller 203 is a Radio Network Controller (RNC) in a Global System for Mobile Communications (GSM) cellular network. Set G is divided into two sub-sets of resources: H and S, where set H∈{C₁, C₂,---,C_M} is initially assigned to the inner cell region 201 with reuse-1 and set S∈{C_M+1,....,C_N} is initially assigned to the cell border region 202 with reuse-<i>k</i>. The RNC 203 then directs each network node, such as a radio base station, associated with cell region, or a wireless terminal in communication through said each cell region, to measure and report radio resource-related data (211), as described hereinafter. The RNC 203 then receives measurement reports from the network nodes, or wireless terminals, of the radio resource-related data. In response, the RNC 203 then dynamically reallocates the distribution of resources between cell regions as a function of the radio resource-related data.

0014It should be noted that the channel assignments to different cell regions can be carried out independently for uplink and downlink channels. The interference relation can be different for uplink channels than for downlink channels. Furthermore, cell regions other than concentric cells may be an alternative, such as illustrated in <figref idref="f0002">Figure 3</figref>. In the example depicted in <figref idref="f0002">Figure 3</figref>, each cell includes a sub-region per neighboring cell, and S is divided into a 3-reuse with S1, S2 and S3. Only one of the S-sets, however, is used in the adjacent areas of surrounding cells in order to achieve lower uplink interference for the other two subsets.

0015In order to dynamically switch the radio resources between different cell regions, the radio access point (<i>e</i>.<i>g</i>., radio base station) associated with the cell, and/or a wireless terminal in communication through the cell, performs certain radio resource-related data measurements which are reported to a radio network resource controller. Radio resource-related data can be: (1) resource activity per channel, wherein the resource activity per channel is defined as the ratio of the time during which a channel is scheduled to the measurement period; (2) aggregate resource activity per channel group, wherein the aggregate resource activity per channel group is defined as the average or x^th percentile of the resource activities of all the channels in a group over a measurement period; (3) the number of transmitted power samples that exceed a threshold over a measurement period; and, (4) channel quality samples, per channel in a neighboring cell region, that exceed a quality threshold over a measurement period.

0016For resource activity per channel data, the radio network resource controller directs a radio access point to measure and report resource activity per channel (µ) in each cell region, where the resource activity per channel (µ) is defined as the ratio of the time during which a channel (<i>e</i>.<i>g</i>., time-frequency resource block, frequency chunk) is scheduled (T_s) to the measurement time period (T_m). The measurement period T_m can be set by the radio network resource controller or can be a default value. The time during which a channel is scheduled T_s is measured in the radio access point by a scheduler. The radio access point can measure the resource activity (µ) of all channels used in both downlink and uplink. The radio network resource controller can specify a number of parameters and events to the radio access point for the purpose of resource activity (µ) measurement reporting. For example the radio access point can be directed to: report resource activity (µ) if it's above a certain threshold (µ > x₁); report resource activity (µ) if it's below a certain threshold µ < x₂); or report resource activity (µ) on carriers/chunks whose quality is above a minimum signal quality level (γ_min) over time T₁. The parameters <i>x</i>₁, <i>x</i>₂, γ_min and T₁ can be set by the radio network resource controller or can be default values used by the radio access point. Using the same principles as stated above, the radio network resource controller can also request a radio access point to report aggregate resource activity per channel group. It indicates the overall activity of <i>K</i> (<i>K</i>>1) channels. A channel group is a set of at least two or more contiguous or non-contiguous channels in the frequency domain. A group may also comprise all channels used in one cell region; in this embodiment, the measurement would depict the global resource utilization status of several, or all, channels in a cell region. The main advantage of aggregate resource activity per channel group is that it requires less signaling overhead. The scheduling of channels (<i>e</i>.<i>g</i>., time-frequency resource block, chunk) to the users is carried out by a scheduler, which is located at the base station. Therefore, the base station can easily measure resource activity both on uplink and downlink channels and report the results to the network controller. The radio access point reports the resource activity per channel and a corresponding resource identification (ID) to the radio network resource controller. Similarly, the radio access point reports the aggregate resource activity per channel group and a corresponding resource group identification (G-ID) or region ID to the radio network resource controller. The measurement reporting can be either event triggered, where an event is specified by the system parameters, or it can be periodic. The event-triggered approach reduces the signaling overheads between the radio access point and the radio network resource controller. Based on resource activity reports, the radio network resource controller can then reallocate channels in different cell regions.

0017For transmitted power samples that exceed a threshold over a measurement period data, the radio network resource controller directs a radio access point or wireless terminal to measure and report the number of transmitted power samples which are above a certain threshold, measured per channel, over a measurement period (T_m). The measurement is performed by a wireless terminal for uplink transmit power statistics and by a radio access point for downlink transmit power statistics. In both cases, the radio network resource controller sets the power threshold and the measurement period. Transmitted power sample data can also be aggregated, in which case the power statistics are collected for all channels used in a cell region over the measurement period. A wireless terminal will report this measurement only for the channels allocated to it, while a radio access point can collect power statistics for all downlink channels.

0018For channel quality samples, per channel in a neighboring cell region, that exceed a quality threshold over a measurement period, the radio network resource controller directs a wireless terminal to measure and report the number of channel quality samples which are above a certain threshold, measured per channel in a neighbor cell, over a measurement period (T_m). The criteria to estimate the channel quality can be based on received total power on the channel, carrier-to-interference ratio (CIR); and, received signal strength indication (RSSI), radio link level block error rate (BLER), packet loss rate, <i>etc</i>. The radio network resource controller specifies the thresholds, measurement period and neighbor cells. Channel quality sample data can also be aggregated, in which case the channel quality statistics are collected for all channels used in a neighbor cell over the measurement period. The measurement is performed by a wireless terminal for downlink channels and by a radio access point for uplink channels.

0019Various algorithms are known in the prior art that can be used by a radio network resource controller to reallocate resources as a function of the radio resource-related data included in measurement reports according to the principles of the invention. The measurements can be used to assist the radio network resource controller in dynamic, or semi-dynamic, assignment of channels in different cell regions. The assigned resources can then be utilized by the scheduler for the corresponding cell region.

0020<figref idref="f0002">Figures 4-A and 4-B</figref> illustrate exemplary scenarios for triggering frequency-reuse re-allocation. As illustrated in those figures, the resource activity per channel (µ) reports indicate to the radio network resource controller that an overload situation exists in the outer cell region (4-A) or the inner cell region (4-B). The radio network resource controller can then dynamically reallocate radio resources between those cell regions to balance the load.

0021<figref idref="f0003">Figures 5-A and 5-B</figref> illustrate further exemplary scenarios for triggering frequency-reuse re-allocation. As illustrated in those figures, if resource activity is high (5-A) or low (5-B) in both inner cell and outer cell regions, the radio network resource controller can still reallocate radio resources between different cell regions. In such cases, carrier reassignment can, for example, be based on quality level. For instance, the carrier whose quality is above a certain threshold (γ₁) or below another threshold (γ₂) can be assigned to the cell border region, or <i>vice versa.</i> The parameters γ₁ and γ₂ can be set either by the radio network resource controller or then can be default parameters.

0022<figref idref="f0003">Figure 6</figref> illustrates a first network topology in which the principles of the invention can be implemented. In this embodiment, the radio access network architecture is characterized by a central radio network resource controller, such as a Radio Network Controller (RNC) 603 in a Global System for Mobile Communications (GSM) telecommunications network, which controls a plurality of radio access points 601, such as GSM Radio Base Stations (RBS). In principle, however, the principles of the invention can also be implemented in a distributed architecture, without a central controller, wherein one or more nodes function as the radio network resource controller; such a network is illustrated in <figref idref="f0004">Figure 7</figref>. In that embodiment, the measurements are exchanged directly between the radio access points. Measurements can also be exchanged between the radio access points via wireless terminals (not shown). In such embodiments, the measurement reports from a radio access point can be broadcast to all wireless terminals or directly transmitted to one or more specific terminals. The wireless terminals can then pass this information to other neighboring radio access points or to other wireless terminals in a neighboring cell, which can then pass it to their own wireless access points. The wireless access points can in this way mutually decide which resources are to be allocated in different cell regions.

0023Finally, reference is to <figref idref="f0004">Figure 8</figref>, which summarizes the exemplary method described herein for dynamic frequency-reuse reallocation. In step 801, a radio network resource controller directs a first network node associated with a first cell region, or a wireless terminal in communication through the first cell region, to measure and report radio resource-related data. The radio resource-related data is selected from the group consisting of: (1) resource activity per channel, wherein the resource activity per channel is defined as the ratio of the time during which a channel is scheduled to the measurement period; (2) aggregate resource activity per channel group, wherein the aggregate resource activity per channel group is defined as the average or x^th percentile of the resource activities of all the channels in a group over a measurement period; (3) the number of transmitted power samples that exceed a threshold over a measurement period; and (4) channel quality samples, per channel in a neighboring cell region, that exceed a quality threshold over a measurement period. Next, in step 802, the radio network resource controller receives at least one measurement report of the radio resource-related data. In step 803, the radio network resource controller, as a function of the radio resource-related data in the first cell region, dynamically reallocates the distribution of resources between the first cell region and at least a second cell region.

0024The cell regions in uplink and downlink are not necessarily equal in size. Also, depending upon the type of services in operation, the traffic load can be asymmetrical in uplink and downlink directions. In FDD mode, the measurements are independently performed on uplink and downlink radio resources. This means in FDD the interference mitigation based on the measurements shall be done independently on uplink and downlink. In TDD mode, separate measurements are also to be done on uplink and downlink radio resources (<i>i</i>.<i>e</i>., uplink and downlink slots). But since radio resources (time slots) are shared between uplink and downlink, the interference mitigation would require efficient and dynamic coordination between uplink and downlink channel allocation in the time domain in step 803 (<i>i.e.,</i> increase uplink and decreasing downlink time slots, or <i>vice versa).</i>