Wireless communication system with adaptive channel allocation

territorially invention

The present invention relates to wireless communication systems with Service areas, which are divided into a plurality of cells are Claim 1.

State of technology

Out EP-A-0731622 discloses a device and a method for adaptively dynamic channel allocation known in wireless communication networks. This known channel allocation system allocates channels different cells by the optimal partitioning of the available radio frequencies into non-overlapping Amounts, the optimum grouping of Mitbenutzerzellen and the best Assignment of the former to the latter. The aim is to optimize traffic coping capacity, which, given the large number of cells, as a maximization of a bottleneck capacity ratio expressed is. The money for a Cell is defined as the ratio of the number of radio frequencies, the cell are allocated, the number of radio frequencies needed are to meet blocking probability requirements. The solution to achieve an optimal non-regular channel assignment is on split two mathematical programs as master program be designated and as a subroutine. This will gradually with support of a channel amounts propagation technique between the solutions Master and the subroutine is implemented dissolved.

Out font "Channel Assignment Schemes for Cellular Mobile Telecommunications Systems: A Comprehensive Survey " I. Katzela - IEEE Personal Communications, June 1996 XP000593925, are channel allocation schemes for cellular mobile Telecommunications systems are known. According to this known Channel allocation schemes, a certain radio spectrum (or a Bandwidth) not in a lot of unaffiliated or another disturbing radio channels be divided. These channels can all be used simultaneously, while a satisfactory received radio signal is maintained. To a certain radio spectrum in such channels divide, can Many processes, such as frequency division, time division or code division be used.

Out <patcit><text>EP 0802695</text></patcit> are a self-configurable Channel allocation system and a method is known. This known auto-configurable channel assignment system can have its own channel allocations determine without prior planning or human intervention. Such a system is as an improvement in a wireless communication network, whose supply areas are divided into a plurality of cells, implemented, the improvement wherein cooperating by a Combination of: (1) an agent for autonomous collection of data the cells in terms of service signal / interference measurements and other relevant data; and (2) a means for automatically implementation a selected Channel allocation algorithm in response to data from the data collection means Provided, is implemented.

Out <patcit><text>US 5,491,837</text></patcit> are a method and a system for adaptive allocation of channels within a radio communication system, known in particular a cellular network. The assignment makes the Measurements, which are carried out by the mobile radio telephone, advantage and allocates channels due to the carrier-to-interference ratio to. Using an adaptive power control is applied to the Maintaining an acceptable carrier to interference ratio taken at the same time minimizing the transmission power.

the Coverage area of ​​a wireless communication system is in as cells known associated supply domains divided, where radio telephone users via radio links communicate with the serving base station the cell. The base station (BS) is connected to the mainland grid. Efficient utilization of the available Radio frequency spectrum is sequences by reusing the same Funkfre achieved in designated Kobe user cells by sufficient Distance are separated so that the combined, by Kokanal- and adjacent channel interference generated cells under tolerable Values ​​are.

Historical Seen based the allocation of radio frequencies (or channels) to Cells on regularity assumptions (Ie the same size, equally spaced Cells with uniformly distributed traffic loads) that the use of simple rules Kobe for identifying user cells and dividing the RF spectrum allow in channel sets. Since such regularity assumptions often do not apply, and therefore the rules of regular channel assignment is not necessarily füh to the efficient utilization of the RF spectrum&lt;?page 3?&gt;ren, however, has a channel allocation approach developed, as the non-regular is known channel assignment to treat this deficiency. Of the regular and the non-regular Channel allocation approach can both are classified as fixed channel allocation, fixed by a in relationship between cells and serving channels is.

in the Unlike allocation method fixed channel A as flexible channel assignment known new classification developed. Such flexible channel assignment methods exploit the capability a system for The remote, software-controlled retuning of the base station radios from, said capacity an adaptation of channel capacity allows to traffic fluctuations.

Wireless Systems are moving generally in the direction of digital technologies from traditional analog wireless systems, even though expected is that analog systems still a significant population of users for will provide some time. In the digital environment occur three Candidates out: TDMA ( "Time-Division Multiple Access "), Global System for Mobile (GSM) and Code Division Multiple Access (CDMA). The first two occur narrowband channels, the in separate timeslots drove three or eight calls can. The latter uses wider channels, the number of users at one time consider can and abutting each other in Cells can be reused.

From Standpoint of the service provider is seen from one of the fundamental differences between the two technologies, the need for channel assignment in the former, a requirement that is absent in the latter. As will emerge from the above discussion, the channel assignment for the currently used analog systems to a very high degree a requirement.

These executed in the prior art Channel assignment function is a both by the need Advance planning of such channel assignments and a requirement a considerable Data collection in. Moreover, such an achieved traditional channel assignment planning tends to only a sub-optimal Traffic load in the cells of a wireless system, and suboptimal traffic flow in such a system. In connection with the channel allocation is the setting or calibrating Cell base station RF transmit power.

On self-configurable wireless system is one in which autonomous Data collected and channels cells be allocated. The collected data will provide information that for the Channel allocation is required.

Summary the invention

A Object of the present invention is to provide a wireless communication system with the frequency / channel assignments be configured to cells themselves and RF transmission power level itself be calibrated using data, the wireless from the System are collected and than by system functionalities Part of the self-configuration process are determined.

correspondingly the present invention this object is achieved by the features solved by claim. 1

Improved embodiments the wireless invention Systems arise from the dependent claims.

The Data required for self configuration and self calibration are composed of signal strength measurements between handsets and base stations. The measurements can be either on the up or the downlink accomplished will. Software coordinates the data collection, the RF power setting and the channel assignment.

In one embodiment, of the invention, the required Data automatically from MAHO / MACA (ie Mobile Assisted Hand Off; Mobile Assisted Channel Assignment) functionalities of the IS 136 Air Interface Standard (AIS) provided, whereby an IS 136 system made capable is, without additional hardware to configure itself and to calibrate itself.

More accurate the system is started at the beginning with preliminary data. Thus can Signal strength measurements accomplished are to provide a collection of data that the for the efficient implementation &lt;?page 4?&gt;Selbstkonfigurierungs- and self-calibration is sufficient salgorithmen.

Short description tHE dRAWINGS

<figref idrefs="S40">1</figref> is a schematic of a regular Cell area layout of a wireless / cellular radiotelephone system;

<figref idrefs="S41">2</figref> is a cell locale in relation to the channel assignment admissibility;

<figref idrefs="S42">3</figref> is a block diagram of a wireless / cellular radiotelephone system;

<figref idrefs="S43">4</figref> is a block diagram of a mobile switching center for a wireless / cellular Radio telephone system;

<figref idrefs="S44">5</figref> shows a plurality of procedures according to the the invention configured mobile switching centers;

<figref idrefs="S45">6</figref> shows a functional illustration of a master and supply mobile switching center according to the invention;

<figref idrefs="S46">7</figref> is a flow process diagram of a traffic data collection process, so as implemented in the invention;

<figref idrefs="S47">8</figref> is a flow process diagram of a serving signal and interference data collection process, so as implemented in the invention;

<figref idrefs="S48">9</figref> is a Flussprozessidiagramm a channel (re) allocation procedure, as it is implemented in the invention;

<figref idrefs="S49">10</figref> is a flow process diagram of a self calibration process to RF transmission power level adjust.

Detailed description

1. Introduction

<figref idrefs="S40">1</figref> shows in schematic form a conventional regular hexagonal cell layout of a wireless cellular communication system. The mapping of the geographical coverage area via a hexagonal grid establishes a geometric pattern that a Allocation of frequencies permitted in a patterned disposition, whereby the reuse of these frequencies in a controlled repeatable regular Allocation model possible is. The cell areas are assigned to specific channel sets. Each channel set comprises a plurality of individual transmit and receive radio channels for use in the cell area. In which, in<figref idrefs="S40">1</figref> shown Model are "A" labeled cells Kobe user cells and all use the same channel set. The same thing applies to labeled with "B", "C" etc. Kobe user cells, all of which have their own assigned channel set.

Each Cell is an associate with a base station antenna system irradiated, said base stations with each other and / or with other may be connected to networks. When in <figref idrefs="S40">1</figref> exemplary shown Configuration by the antenna <figref>101</figref> an omnidirectional Directivity displayed, and the antenna <figref>102</figref> represents a directional antenna pattern, thereby sectorization of cells representing smaller coverage areas of the wedge type is.

generally known is a central aspect of cellular communications systems, the concept of frequency reuse. With frequency reuse, users can in different geographical locations (different cells) simultaneously use the same frequency channel, as indicated by gleichbenannte cells in <figref idrefs="S40">1</figref> for regular Channel assignment ready. Although the frequency reuse the can substantially increase spectral efficiency of a system, can it between on the common use of the same channel involved Cells to strong interference come, if orderly system design is missing. Frequency reuse or channel assignments are in Generally easier by using rules for identifying Kobe user cells and for partitioning the RF spectrum into channel sets implemented.

&lt;?page 5?&gt;

Channel assignment approaches can be broadly are classified into two categories: fixed and flexible. Fixed channel assignment fixes the relationship between cells and the channels serving them. Just the one cell allocated channels, compounds in this Cell supply, and each channel can simultaneously by all cells be used, which is assigned to the channel. An example of solid Channel assignment is regular channel assignment, by the regular repetition of a reuse pattern is characterized. Regular channel assignment is for a system with uniformly across cells Distributed traffic optimally.

If the traffic distribution is not uniform, can not be a solid regular Channel allocation are found, the channels cells depends on their traffic load allocates. [A process for achieving such a nonregular optimal Allocation is described in M. Benveniste, "Apparatus and Method for Non-Regular Channel Assignment in Wireless Communication Networks ", US Pat. No. 5,404,574, described.]

Flexible hand channel assignment methods exploit the capability of a system for remote, software-controlled retuning of the base station radios from, said capacity allows you to adjust channel capacity to traffic variations.

It should also it is well known that the quality of communication in wireless systems substantially from the ratio of the Received signal to interference (S / 1) or the bit error rate (BER) depends, the latter in turn of received signal and interference depends. The main disorders which are problematic, consist of two components: Kokanalstörungen and Adjacent channel interference. Kokanalstörungen are the interference from communication sources in the same frequency as the operating channel are matched. Adjacent channel interference originate from communication sources, the channels near the operating channel use in the frequency spectrum. To the desired voice or Datenübertragungsqualitat reach, has the ratio the received signal to the combined Kokanal- and adjacent channel interference on a specified threshold. RF transmission power levels are also a factor in determining the interference.

in the This section describes a methodology whereby a channelized wireless system the need for channel assignment planning can be avoided and instead its own channel assignment provisions performance, without advance planning or human intervention necessary are. The implementation of this methodology is here as a self-configurable wireless system called.

in the Below also describes a methodology by which a split in channels wireless system, the need for manual calibration RF power transmission level can avoid a base station. In fact, the calibrated System after an initial coarse power adjustment itself, without requiring a technician to intervene takes. This methodology is referred to by the term self-calibration, wherein the RF power setting to a software-based calculation reduced configuration the same system collected data as the Selbstkon used. The RF power level settings are determined, to restrictions consider the criteria for meet user-specific services and system performance. The self-calibration is continuous. It ends when the system has a stable state reached.

On according to self-configuring, self-calibrating system of the invention comprises an autonomous channel assignment and autonomous RF transmission power level setting for the Base stations. The self-configuration assigns channels automatically Cells in such a manner that a channel interference is avoided is. It includes the establishment of channel assignment admissibility criteria, which determine whether the simultaneous use of a channel by a Collection of base stations (or their handsets) is permitted. Such criteria can be used to create a neighbor list. The self-calibration has with the setting of RF transmission power levels do of base stations at a minimum level.

In a wireless system for self-configuration and self-calibration is capable of the RF planning is simply the problem of Selection of cell sites and setting of system configuration parameters reduced. Everything else that traditionally part of the RF planning, while was setting up a new system and its extension, is led by software, which is anchored in the system devices.

self-configuration and self-calibration using data collected by the system are while this works. This data consists of signal strength measurements between handsets and base stations, during this either on the uplink or the downlink accomplished who&lt;?page 6?&gt;the can. Generally it would additional radios at the base station for the collection of data required. A wireless communication system, the according to a working aspect of the invention under the IS 136 AIS, however, without any modification of the hardware structure of the wireless system be implemented. The MAHO / MACA- functionality provides the required signal strength measurements without the relevant need for additional radios to the cell sites. Indeed can necessary measurements on the downlink be performed by traveling handsets upon receipt of proper instructions. This process is coordinated by software to existing added software is.

2. self-configuration

I. Introduction

On self-configurable wireless system according to the invention offers advantages over prior art systems two main advantages. Most obvious is the elimination of Channel planning process itself and thus the elimination or substantial Reduction for this purpose necessary specialist staff - a significant cost advantage for the System operators. Secondly, the number of one to supply given traffic load necessary cell sites significantly reduced will. There are two mechanisms by which such a cell site reduction is achieved. One is the ability of the system channels, where appropriate, to relocate when traffic loads another time. If Traffic peaks over Move the system away, leads portability of channel capacity in fewer cell sites. The second is through increased traffic throughput, which is also accessible in static traffic conditions.

Selbstkonfigurierbarkeit includes two key features: Autonomous Data collection and a selected Channel allocation methodology.

the first feature, autonomous data collection, concerns for the administration necessary of voice channels Data. This includes Data interference environment characterize, and other data specific to the implemented are channel assignment algorithm relevant. In the preferred embodiment, using an ADCA algorithm, such data are in a Cell live loads. The interference data would at Startup of the system or when adding new cells or sectors collected. Both types of data can during normal operation collected on the performance of the system and without affecting will. Suitable methods for such autonomous data collection will be known to those skilled.

at a preferred embodiment, of the invention the selected Channel allocation method from the class of flexible channel assignment method selected, as already mentioned the ability a system for Contrasting that exploit software controlled retuning of the base station radios, the ability, allows an adjustment of channel capacity to traffic variations. However, it should be noted that the methodology of the invention is also applicable to fixed channel assignment methods.

Flexible Channel assignment methodologies comprise three categories: adaptive, dynamic and adaptive-dynamic channel allocation. This algorithm making resistant more than traditional channel assignment method and with current and planned standards cellular technology, infrastructure and radio interface compatible.

It Note that the Selbstkonfigurabilitätsmethodologie of the invention, including the self-calibration, and the channel allocation process underlying to analog, TDMA, GSM, TDMA-analog hybrid systems and GSM-analog hybrid systems are applicable. TDMA channels can as well as analog channels be treated, as long as you three connections per TDMA channel (a allow on each of the three time slots) and all three time slots are used by the same cell or the same sector. GSM channels are treated similarly, it being recognized that each channel has eight connecting slots.

II. Channel assignment

As already mentioned is in the preferred embodiment the invention, the channel allocation method from the class of flexible Channel assignment methodologies selected, this class the following specific categories includes: adaptive, dynamic and adaptive-dynamic. There is a brief discussion regarding this flexible channel assignment categories indicated.

&lt;?page 7?&gt;

A. Adaptive channel assignment

On Algorithm of adaptive channel assignment ( "ACA"), the channel assignments to the Traffic, by the optimal non-regular channel allocation for various periods is recalculated and observed data are used to to estimate expected traffic loads.

at ACA will calculate a new channel allocation newly when the traffic loads have changed in a statistically significant manner to a recalculation to justify. The time interval between successive Kanalneuzuteilungen could range from a half hour to eight hours. The reallocation will triggered by a test, the rejection the corresponding hypothesis tests equivalent. This results the mechanism for adjusting to traffic trend variations. The Adaptive channel assignment has no opportunity for adjustment to traffic variations due to randomness. The relationship between channels and cells located within the time interval between successive following similar Kanalneuzuteilungen as in non-regular Channel assignment determined.

B. Dynamic Channel Assignment

On Another method of setting the variable demand for channels dynamic channel allocation. Dynamic channel allocation is completely without the fixed relationship between channels and cells. It can more users access a channel, supplied as simultaneously could become. [See, eg, LG Anderson, "A Simulation Study of Some Dynamic Channel Assignment Algorithms in a High Capacity Mobile Telecommunications System ", IEE Trans. Commun., Vol. 21, No. 11, November 1973; R. Beck and H. Panzer, "Strategies for handover and Dynamic Channel Allocation in Micro-Cellular Mobile Radio Systems ", Proc. IEEE Vehicular Technol. Conference, 1989; LJ Cimini, Jr., GJ Foschini]. Due to its increased flexibility can a dynamic channel assignment algorithm caused both accidentally Activating traffic fluctuations as well as changes in trends.

If more users have access to a given channel, the idle time the channel frequently be reduced. Professionals is known, however, that not all dynamic channel allocation algorithms a capacity improvement relative to normal guarantee channel assignment. [See, eg, Beck and tanks, "Strategies for handover and Dynamic Channel Allocation in Micro-Cellular Mobile Radio Systems ", id.] Since the accumulation of a channel at a given time assigned Users of the dynamics of arrivals and statements depends on compounds , the average distance between them is greater than the minimum allowable from interference control reasons separation distance be. The through greater reuse distance Capacity decrease caused would the from greater flexibility in the Channel usage resulting potential gains cancel.

C. Adaptive-Dynamic channel assignment

Adaptive Channel allocation makes resistant more than regular Channel assignment, but the channel idle time reduction can not exploit, which is achieved when the restrictions on channel access loosened will. contrast Dynamic channel assignment allows flexibility in Use of channels through different cells can in storage However, capacity losses occur because a longer average reuse distance is realized. Adaptive dynamic Channel assignment ( "ADCA") combines the best the two approaches: The resistant Non-inferiority adaptive channel assignment with the ability of dynamic channel assignment, reduce channel idle time.

ADCA is adaptive channel assignment using a special form the dynamic channel assignment, channel borrowing. At the traditional Channel Borgen are channels Cells by the rules of regular channel assignment allocated. [See Anderson, "Simulation Study of Some Dynamic Channel Assignment Algorithms in a High Capacity Mobile Telecommunications System " id .; JS Engel and MM Peritsky, "Statistically optimum Dynamic Server Assignment in Systems with Interfering Servers ", IEEE Trans. Commun., Vol. 21, No. 11, November 1973.] cells then try out the allocated channels to use. In case of unavailability is on other channels accessed. A channel used by a different cell than its owner cell is referred to as a borrowed channel. A channel is only used by a cell when interference restrictions met are.

Of the Kanalborgeralgorithmus at ADCA is of the traditional Channel borrowing different than the channel allocation is not regular. you is not regular, such as in adaptive channel assignment. Due to the Kanalborgens can ADCA channel capacity to randomly caused fluctuations of transport and on &lt;?page 8?&gt;Activating traffic trend changes.

This ADCA algorithm can three basic features: a function of channel acquisition, a function of the channel release and a function of the channel (re) allocation. The function of the acquisition channel is called when a connection is started or passed. The function of the channel release is necessary only if the algorithm permits Kanalneuanordnungen. You will be called when one by one of the cell allocated channel supplied connection terminated or passed is to give the release a "borrowed" non-allocated channel is allowed, wherein the compound is then to the newly empty allocated Channel is transferred. takes the function of the channel (re) allocation out how channels to be allocated in the system so that the number of each Cell or each sector channels allocated satisfies a chosen criterion. To the Example, the number of allocated channels is proportional to the number necessary channels, and their relationship is maximum. The function is invoked when the channel allocation (New) needs to be calculated.

all Functions use the statistics of the supply signal and the disturbances. These statistics are from data collected by the system data estimated. Other required by the first two functions are the information Allocation status of channels for every Cell, supplied by the third function.

III. Criterion for admissibility a channel assignment

at most of the existing dynamic channel assignment algorithms is the permissibility a new connection on a channel determined by checking, if the channel in any of a specified list of neighbor cells is occupied. The list of disturbing Candidates is obtained either by different methods based on real-time or a priori measurements, or an analytical modeling constructed.

In certain cases is examining permissibility of Kokanalbenutzung for a pair of cells / sectors by the construction of a compatibility matrix, a square matrix with dimension equal to the number of cells / sectors in the wireless communications network is facilitated. The Elements of the matrix are either 1 or 0, which is indicated in each case, if a pair of the same channel cells / sectors use the same can or not. If the graph-Einfärbeansatz taken for the channel assignment is, is as an alternative a graph with the cells / -sectors in the system corresponding nodes and edges, the cells / sectors that are not the same channel can use at once, connecting respective nodes, constructed.

at two representations use the existing algorithms pairs approvals, a Collection of cells / sectors produce that same channel can use simultaneously. There these approvals without any information about other cells / sectors are derived, they are more stringent than necessary. The result is suboptimal, as a number of feasible combinations of Kokanalbenutzern can be excluded. Take, for instance, the example of Cells A and A 'in <figref idrefs="S41">2</figref>, If one pairs approvals the simultaneous use of a channel based on the assumptions and Criteria would deduce as used in the use of a re-use factor N = 7 (that is, It is assumed that there are six other in the same Kokanalbenutzer Distance as the pair in question are, as indicated by the dots in <figref idrefs="S41">2</figref> displayed) would this two cells authorization to use the same channel, denied. In a situation in which there are no other cells in the vicinity are sharing the same channel, which is by Kokanalbenutzung the cells A and A ', however, allowed.

consequently is herein a new test for admissibility of a new channel assignment disclosed, wherein the channel use in all cells / sectors in the interference environment the cell in question or in the sector concerned considered is.

Of the Admissibility test, we for Recommend a new channel assignment, taking into account the channel usage in all cells in the interference neighborhood of the cell in question. An example of citation [4] illustrates this approach. Alternative tests may similarly Example, using the bit error rate as a measure of the quality of a wireless connection can be derived.

Around algebraically expressing the compound, which requires the Signal-to-interference (S / I) ratio above a certain threshold value, we define the following terminology. Let j = 1, ..., J index of different cells i = 1, ..., J as j (the combination (i, j) designates a pair cells) &lt;?page 9?&gt;I_ij Kokanalinterferenzbeitrag of cell i to cell j (random variable) S_j Signal strength at Cell j (random variable) T threshold for signal-to-interference ratio I-α confidence level for S / I ratio requirement C_j Cell size, as the same channel Cell j uses

The S / I ratio requirement can be set as a probability statement as follows: <img img-content="mf" img-format="tif" he="20" wi="104" file="00170001.tif" />

Around the above as an equivalent writing deterministic inequality, one has the probability distribution of the S / I ratio know. Y is the value of this ratio, expressed in decibels. [Of the will have simplicity the index j omitted in other appendices]. This means: <img img-content="mf" img-format="tif" he="18" wi="91" file="00170002.tif" />Other treatments following it is assumed that Y is normally distributed. μ_γ and σ_γ² are the mean and the variance of Y, and R be the threshold value T of the S / I ratio, expressed in decibels. The equation (1) can be written as follows: <img img-content="mf" img-format="tif" he="13" wi="111" file="00170003.tif" />in which z is a normal random variable. The equivalent deterministic Constraint is as follows: <st32:df xmlns:st32="http://lighthouseip.com/">μ<st32:sub>γ</st32:sub> + z<st32:sub>α</st32:sub>σ<st32:sub>γ</st32:sub> ≥ R (4)</st32:df>where z_α the α-quantile Normal statistical variables.

The Equation (4) is the criterion for the admissibility of a channel assignment. The values ​​μ_γ and σ_γ hanging from the composition of the amount of C_j, The collection of cells for a mapping of the channel j are contemplated from. Your Values ​​are calculated using the assumption that the signals at all base stations, expressed in decibels, are independently distributed normal random variables and that the cumulative interference in the cell _j is also normally distributed, in Decibels expressed. It applies <st32:df xmlns:st32="http://lighthouseip.com/">Y = P - L (5)</st32:df>in which <st32:df xmlns:st32="http://lighthouseip.com/">L = 10 log<st32:sub>10</st32:sub>[.sigma..sub.i<st32:sub>k</st32:sub>j] (6)</st32:df><st32:df xmlns:st32="http://lighthouseip.com/">P = 10 log<st32:sub>10</st32:sub>S<st32:sub>j</st32:sub> (7)</st32:df>If μ_L the Average of the cumulative interference L in cell j is expressed in decibels σ_L² the variance of L is μ_ρ the Mean value of the power signal P in cell j, expressed in decibels σ_p² the variance of P is cov² (P, L) the covariance of P and L then, the mean values ​​and the variance of Y given by: <st32:df xmlns:st32="http://lighthouseip.com/">μ<st32:sub>γ</st32:sub> = E (Y) = (E (P) - e (L) = μ<st32:sub>ρ</st32:sub> - μ<st32:sub>L</st32:sub> (8th)</st32:df>&lt;?page 10?&gt;<st32:df xmlns:st32="http://lighthouseip.com/">σ<st32:sub>γ</st32:sub><st32:sup>2</st32:sup> = Var<st32:sub>(Y)</st32:sub> = Var<st32:sub>(P)</st32:sub> + Var<st32:sub>(L)</st32:sub> = σ<st32:sub>ρ</st32:sub><st32:sup>2</st32:sup> + σ<st32:sub>L</st32:sub><st32:sup>2</st32:sup> + 2 cov<st32:sup>2</st32:sup> (P, L) (9)</st32:df> μ_p and σ_p² are parameters based on empirical data estimated be that while operation of the system add up. μ_L, σ_L² and cov (P, L), with the composition the amount Cj vary, be a Leistungssummierungsverfahren calculated. The statistical parameters in the Leistungssummierungsberechnung be used, are also estimated from empirical data.

alternative tests can similarly Example, using the bit error rate as a measure of the quality of a wireless connection may be used.

IV. Embodiment a self-configurable wireless system

On self-configurable wireless system according to the present invention can without modification of the wireless system architecture implemented will. Instead, such an implementation is a Combination of software additions, Modifications to existing software, and adding radios at the cell sites to signal strength and interference measurements provide executed. By thus built on the existing system architecture, the existing functionality get this architecture and the Selbstkonfigurabilitätsmethodologie containing system is thus obtained the ability to switch between the existing and the self-configurable Mode of operation, if such a change is necessary or desirable should be. To a proper understanding of the invention, contemplated modifications to a wireless system to develop, first briefly illustrative example for a existing wireless system described.

A. Existing system configuration

On specific example of a Kokanal's <figref idrefs="S41">2</figref> shown. As in the parent application approvals for the Simultaneous use of a channel with the reuse factor N = 7 stated, would this two cells, the permission to use a common channel, denied. If no other cells in the vicinity of the same use channel, is a Kokanalnutzung by cells A and A 'is allowed.

On typical cellular System to which the invention itself configurable (including Self-calibration) may be applied, is in the block diagram from <figref idrefs="S42">3</figref> shown. There are several mobile switching centers (MSC), <figref>202</figref> and <figref>203</figref>Shown, the mobile telephone system, the with the public switched telephone network <figref>201</figref> (PSTN) connect. The switching of the MSCs connects multiple base stations (BS) <figref>210</figref>, Each feeding a cell coverage area. Each coverage area is as for an actual system typical irregular Limits shown as having. Each BS has radio transmit - / - receiving device and radiation antennas for the supply of mobile telephones <figref>250</figref> in your Cell coverage area.

In <figref idrefs="S43">4</figref> are represented the main functional elements of a mobile switching center. As is apparent from this figure, the control of the Verbindungsbearbeitungs- and channel assignment process in the MSC <figref>301</figref> anchored, wherein a plurality of processors are housed in a token ring architecture a two-process messaging (IMS) <figref>302</figref> are linked. The channels are assigned by the Exekutivzellularprozessor (ECP) <figref>303</figref> in the speech channel module Administration (VCA) module. In the present VCA module are typical enough, both fixed channel assignment and a simple dynamic channel assignment algorithm available.

After Initializing the system reads the VCA from a database the Channel assignments for all cells / sectors were specified. VCA keeps a record on the a cell or a sector available voice channels and on their Busy / idle status during system operation. When service is requested, is an idle selected channel following one of the available Algorithms for "interconnector Search". If the option the dynamic channel assignment (DCA) is used, marked VCA the for accessible the DCA algorithm channels and pursued the DCA released / -Sperrstatus a channel, ie whether a DCA channel in a disturbing Neighboring cell is occupied. When a DCA channel is selected, it marks in neighboring cells inhibited DCA as.

Of the ECP is also responsible for OA &amp; M functions, including collecting service measurements belongs. For example, the Number of offered connections, connection handoff requirements, blocked connections and dropped connections from the cell sites counted and are periodically charged to the ECP up. An ECP adjoint faster processor, the Opera&lt;?page 11?&gt;tion and management platform (OMP) <figref>305</figref>. helps the ECP with the OA &amp; M functions. The OMP is full for collecting service measurements responsible.

B. Self-Configurable system configuration

For the self-configurable wireless system of the invention remains the channel assignment at a preferred embodiment, centrally controlled. In a system with multiple MSC, such as in <figref idrefs="S44">5</figref> ready, the channel assignment function will be implemented in a hierarchical two-tier architecture. As shown in the figure, operates a as a master MSC <figref>401</figref> Marked MSC for calculating the channel allocation for the entire system. The other functions, eg traffic load, serving signal and interference statistics estimation, by the serving MSCs <figref>402</figref>. <figref>403</figref> &amp; <figref>404</figref> controlled (including also the Master MSC in relation to their own supply area).

The functional relationships between the master MSC, a serving MSC and a cell site are in <figref idrefs="S45">6</figref> shown. In an MSC to ECP and OMP share responsibility for according to the invention implemented new features, and new data collection functions be carried out by the cell sites. The functions of the inventive methodology and their relationship to each other are in the following subsections described and in flowchart form <figref idrefs="S46">7</figref>-<figref idrefs="S48">9</figref> ready.

(1) The ECP

In the ECP works VCA as usual, with modifications and additional Features which are described herein. A significant change is the elimination of a unique correspondence between a wireless device and Voice channels. Each cell or each sector has access to more channels than the number of radio devices at the cell site. If a radio channel request is received is selected VCA both a voice channel and a radio and instructed the cell site, the elected radio set to the selected Channel vote. The radios selection procedure stays unchanged, but the voice channel selection by the VCA is in the present system modified.

VCA results as date a list of a cell or a sector accessible channels and tracks the channel allocation status, ie whether a channel of is allocated or not allocated cell. This information supplied by the OMP and be updated every time when a new channel allocation is calculated. This process of channel (re) allocation is in the flow diagram of <figref idrefs="S48">9</figref> shown. accessible all of a cell channels be as DCA channels marked.

Around To determine whether an acceptable S / I ratio or an acceptable BER for a given channel assignment is obtained, VCA tracks the entire occurring on each channel in each cell disorders. This quantity is updated while channels be procured or enabled, and is using the relevant statistical parameters calculated. The estimate for the S / I ratio or the BER necessary statistical parameters of the OMP supplied and by the VCA for each cell or each sector recorded. These parameters change only when there are system configuration changes or new cells / -sectors be added. The process of determining serving signal and interference statistics is in the Flow chart of <figref idrefs="S47">8</figref> shown.

If Kanalumordnungen allowed are leading VCA also a counter the number of performed on each connection rearrangements. At a alternative embodiment contains the VCA a timer for guiding a record of the time interval since the last Kanalumordnung maintained at each connection, so that minimum distance is.

(2) The OMP

As in <figref idrefs="S45">6</figref> shown, collects the OMP <figref>506</figref> serving signal and interference measurements from the module <figref>518</figref> for Serving signal and interference measurements (SIMEAS) at a cell site <figref>507</figref> to estimate their statistical distribution parameters (ie mean, variance and Covariance) in the module Serving signal and interference statistics estimation (SISTAT). These parameters are required by both of the VCA and for the calculation of a channel allocation. The serving signal and interference statistics be first at the time of installation of a new system calculated. These statistics are then always recalculated when system configuration changes or additions new cells / sectors are.

&lt;?page 12?&gt;

The OMP collects also Traffic data from the module <figref>517</figref> for traffic measurements (tmeas) to a cell site <figref>507</figref> and estimates the offered loads in the module <figref>514</figref> for Traffic load statistics estimate (Tlstá). These estimates be of the module <figref>515</figref> Traffic load change test (TLCT) used to determine whether the traffic pattern enough changed has to require a reallocation of the channel. The process of Determining such traffic load statistics and determining whether a new channel allocation is to be calculated, is in the flow chart from <figref idrefs="S46">7</figref> shown.

As last determines the Master OMP <figref>504</figref> the allocation of channels to cells in the module <figref>513</figref> for the Calculating a new channel allocation (REALL).

In a multi-MSC system, it is desirable for all MSCs to compute channel allocations together to get the best results to obtain. Since it is possible efficient algorithms for large systems that a single OMP, the Master OMP, Traffic, serving signal and interference statistics from the other OMPs receives and the optimal channel allocation for the entire system calculated.

(3) The cell sites

the self-configurable wireless system will use an existing Function of the cell site, the measurement of power signals, in conjunction with an implemented on new cell site function, collecting interference measurements. By means of this invention, measurements of signals between Base stations and handsets using the MAHO / MACA functionalities collected. The SIMEAS module in the cell and collects supply noise measurements, the SISTAT from the module in the OMP needed in order to estimate the statistical distribution parameters used by the REALL module in the Master OMP and the VCA in the ECP to be. The home base stations of registered handsets instruct these handsets to match a specific control channel and the strength measure of received signals.

C. Estimation of statistical parameters

The admissibility a given channel assignment requires knowledge of the supply and interference in a mobile and base station receivers, regardless of whether the admissibility criterion is provided as a deterministic or probabilistic statement. In the inventive method the necessary parameters of measurements are derived, in which the base station during the normal operation of the system are carried out and those not to disturb. These are not real-time measurements of the supply and noise for the Compound (s), for which the feasibility checked by Kokanalbenutzung is. These measurements serve as a sample of a larger population for use for parameter estimation. Therefore, they are selectively ge gathers, and after a sufficient sample size collected was able Signal strength measurements stop, To only be resumed if new cells / sectors added be or when system configuration parameters are modified. Only a subset of parameters, near the immediate vicinity the system change must be re-estimated, because system changes have only a local impact.

Of the Section III above is an example of an examination of the admissibility a candidate channel assignment. In the preceding section , the configuration of a wireless system with the ability for collecting supply and noise measurements while the normal operation described. In the present section shows how the necessary statistical parameters collected Signal strength measurements estimated can be.

one Consider as an illustrative example, that the parameter, the estimated to be, are the following: μ_j Average the supply signal in the logical cell j σ_j variance the supply signal in the logical cell j μ_ij Average the interfering signal from the logical cell i to logical cell j σ_ij² Variance of the interference signal from the logical Cell i to the logical cell j cov_ikj² Covariance of the interfering signals from the logical Cells i and k to logical cell j

signals either on the uplink (From the mobile device to the base station) or the downlink be measured (from the base station to the mobile device), can for parameter estimation be used in both directions, as long as the proper power scaling was used to normalize the signal. In the following discussion is simplicity and &lt;?page 13?&gt;without loss of generality assumed that all signals in decibels on the uplink are measured and were normalized. It is further assumed that the following signals for the estimation of are parameters listed above available. Let us denote the nth individual measurement of the signal from a mobile device is received, which is supplied by cell j as DSj (n), n = 1, ..., N, and the signal from mobile devices by cell j is received, which are supplied by cell i, as DIij (n), n = 1, ..., N. Then, The following formulas are used to determine the parameters discussed above uplink appreciate.

<img img-content="mf" img-format="tif" he="70" wi="106" file="00260001.tif" />

the same Formulas apply for Downlink parameters, as long as one uses downlink measurements. When using uplink measurements it is enough, the relationship between the uplink and downlink signals to notice. If the uplink and downlink measurements a signal q as u (q) and d are referred to (q), then apply the following relationships: <st32:df xmlns:st32="http://lighthouseip.com/">u (DS<st32:sub>j</st32:sub>) = D (DS<st32:sub>j</st32:sub>) (15)</st32:df><st32:df xmlns:st32="http://lighthouseip.com/">u ((DI<st32:sub>ij</st32:sub>) = D (DI<st32:sub>ji</st32:sub>) (16)</st32:df>

(1) Berechnerische considerations

The following relationships abbreviations are used to recalculate a portion of the parameters to avoid: <st32:df xmlns:st32="http://lighthouseip.com/">u (μ<st32:sub>j</st32:sub>) + D (μ<st32:sub>j</st32:sub>) (17)</st32:df><st32:df xmlns:st32="http://lighthouseip.com/">u (σ<st32:sub>j</st32:sub><st32:sup>2</st32:sup>) = D (σ<st32:sub>j</st32:sub><st32:sup>2</st32:sup>) (18)</st32:df><st32:df xmlns:st32="http://lighthouseip.com/">u (μ<st32:sub>ij</st32:sub>) + D (μ<st32:sub>ij</st32:sub>) (19)</st32:df><st32:df xmlns:st32="http://lighthouseip.com/">u (σ<st32:sub>ij</st32:sub><st32:sup>2</st32:sup>) + D (σ<st32:sub>ji</st32:sub><st32:sup>2</st32:sup>) (20)</st32:df>

It there is no simple relationship between downlink and Aufwärtsstreckenkovarianz the signals of two logic cells which are third with a to disturb. It must both be calculated.

The estimate the necessary statistical parameters does not require any sampling data to keep in memory. It is sufficient that the following statistics to keep: &lt;?page 14?&gt;<img img-content="mf" img-format="tif" he="68" wi="85" file="00270001.tif" />

After Conclusion of a new measurement is therefore a simple multiplication and / or addition executed, to update the above relevant statistics. The measurement can then be discarded.

Finally must disturbance parameters not for all sorts estimated cell combinations will. These parameters are only for the cells in the interference environment each cell estimated.

D. Systeminitialisierungsbedingungen

On self-configurable wireless system according to the invention performs its own channel assignment by and collect the necessary data during the Normal operation. As long as the system is not running done, are However, no data available with which one can make a start. is for this purpose a simple Amount of input data are sufficient. This data consists of pairs of cells / sectors, not simultaneously use the same channel allowed.

Around to see that this amount of data is not enough, consider the discussed above embodiment the invention. The VCA needed the following data: the allocation status of channels in each cell and the serving signal and interference statistics, the necessary are to determine whether channel S / I or BER a specified meet requirement before channels selected will. The allocation status of channels, although for efficient but operation of the system is necessary, not necessary for the execution of VCA and can therefore be omitted from the system startup. Allen channels thus beginning a status "not allocated "given. While the system is running, accumulates it the necessary data for the calculation of a channel allocation.

The Determination of channel S / I or BER contrast requires data in respect this to the serving signal and interference statistics, which Data only becomes available are, when the system has been running for some time. Knowing the Pairs of cells / sec tors that are not the same channel at the same time may use, ranging However, in order compatibility ensure the S / I or BER requirements and is sufficient to go around the system. With the system operating accumulate improved data.

The necessary initialization data can be supplied by the operator will. Alternatively, this Data are collected autonomously by a system whose base station radios in the Are able to receive signals from other base stations. By employment can with the peer signal strength measurement one estimate, whether a pair of base stations use the same channel at the same time can.

As well can Initial RF power level over meeting be provided by the operator to the system up and running bring. These data allow the initial operation of the system. If the system during his Operation collects data, the self-calibration provides correspondingly the invention finally efficient RF power levels.

Among Use of this invention in the parent application, a system its operation at maximum efficiency begin as follows. Once initialization data has been specified, the system can begin, the necessary data for the self-configuration and Selbstkalibrierug by recording signal strength measurements between base stations and handsets, but not from customer handsets, &lt;?page 15?&gt;to collect. can at this point Handsets that are under the control of a technician, in the system coverage area for the express purpose of the collection of needed Data is moved. The system does not need to know that the handsets do not represent customer calls. Once enough data has been collected (And the optimum parameters were estimated) are customer calls admitted into the system.

In a general system (like an analog or IS-54 system) have to the handsets that for Initialization can be used to be engaged in a call. In an IS-136 system, however, it is sufficient that the mobile parts are registered.

correspondingly the claimed invention in this application, the MACA functionality of the IS-136 standards are used to the necessary provide data, wherein the mobile parts are idle.

3. Self-calibration

the The aim of the self-calibration is to adjust the power level for every Base station to the lowest level which is the base stations allows together an adequate signal strength for the service area with a specified probability provide. power level necessary to determine the coverage area of ​​a base station, which is determined by the model. The coverage area is the collection of the locations of handsets whose signals during Self-configuration were measured, determined. The model assigns these locations in an optimal manner to the base stations, whereby the Coverage area of ​​each base station and in turn its required RF power is determined.

the Self-calibration problem can therefore be considered as a special case of the stochastic "amount, superimposed" problem will. That is, Problem Description: There are given different mobile sites, which representing the coverage area. Now One wishes, mobile sites assign base stations and determine the least power level, the needed by the base stations is to their associated mobile device locations with a specific Likely to provide a signal of a desired thickness.

Out the collected data passes to an independent set M mobile sites AB, which is represented by the index m. Then determined based the signal strength measurements DI_ij^(N) and DS_j^(N) the relative Propagation loss α_mj the signal from the mobile device location m to base station j. The following mathematical programming model can be formulated to the optimum power level Basistations to find. Suppose that Q is the received signal strength requirement. β is the Confidence level, be satisfied where the received signal strength requirement should α_mj is the signal attenuation from the handset location m to base station j.

one solves for the following unknown variables: p_j the power level the base station jp_MAX the worst-case power level x_mj a binary variable equal to 1, if the mobile device location should be covered by base station j; otherwise 0th

the Self-calibration problem is provided as follows: <st32:df xmlns:st32="http://lighthouseip.com/">min p<st32:sub>MAX</st32:sub> (26)</st32:df><st32:df xmlns:st32="http://lighthouseip.com/">St pMAX ≥ pj j for all (27)</st32:df><st32:df xmlns:st32="http://lighthouseip.com/">amj pj ≥ Q XMJ for all m and j (28)</st32:df><img img-content="mf" img-format="tif" he="22" wi="96" file="00310001.tif" />&lt;?page 16?&gt;<st32:df xmlns:st32="http://lighthouseip.com/">x<st32:sub>mj</st32:sub> = 0.1 (31)</st32:df>where M_k Amounts of mobile sites are responsible for different represent k, different sub-areas of the service area, whose union gives the system coverage area. M_k is the cardinality of the set M_k, In the simplest implementation of the Model there is one sub-area of ​​the entire coverage area includes.

By Setting the power level so that they these constraints fulfill, be the power level P_j to a minimum useful value forced down and still be a satisfactory the proper signal supplied which is to be supplied to the mobile area with a probability β covers.

Of the Expression in (26) is the objective of the optimization model; the expressions (27) to (31) are the constraints. The constraints in (27) organize p_MAX its value. By minimizing of p_MAX while the constraints fulfilled in (27) are, you force all power level p_j after below. The constraints in (28) ensure that the desired signal is delivered from the serving base station. The constraints in (29) and (30) put together a cover with a probability of β safely. The constraints in (30) are necessary to prevent Mobile sites, which can be covered by more than one base station, in Expression (29) counted twice will. The expression in (31) specifies the values ​​obtained by the Decision variables x_mj get accepted can.

More Subsidiary conditions may added be to provide additional Requirements or targets reproduce. If no additional added constraints be the solution in this mathematical program based on the below in Table 1 method illustrated can be obtained.

Table 1. An algorithm for the solution the self-calibration problem

step 1 It represents the RF power levels of all base stations on the same the permissible Minimum value.

step 2 It assigns each mobile device location the base station to which the strongest Signal.

step 3 One finds among the covered mobile locations to or the location (s) m 'with the weakest reception signal.

step 4 If the weakest Received signal is not less than that set Q, is terminated.

step 5 Otherwise you identify all k = k 'such that M_K'The mobile device location m 'contains. Man removed the mobile device location m 'from the cover by its assigned base station if the coverage requirements in the constraint (29) can still be fulfilled. [This implies that constraint (29) with strict inequality for all were k 'met.]

step 6 If possible is the mobile site m 'of the coverage area to remove its associated base station, it proceeds to step 3 over.

step 7 Otherwise, one increments the power level of the base station Cover mobile site m 'and proceeds to step 2 about the end

Of the Self-calibration process in the flow chart of <figref idrefs="S49">10</figref> shown. The process begins at start terminal <figref>1001</figref> and as the instruction block <figref>1003</figref> reliant, is the transmission power the base stations on a permitted Minimum value set. subsequent Instructions of instruction block <figref>1005</figref> specify, that mobile sites temporarily Base stations are assigned to them the strongest signal supply.

The Received Signal Strength is the evaluation block <figref>1007</figref> evaluated to determine whether they needed the Received Signal Strength reach or exceed. If this is the case, terminates the self-calibration process, such as of block <figref>1017</figref> reliant. If the decision, that the received signal strength is not sufficient, then the instructions of decision block <figref>1011</figref> certainly, whether the specified coverage probability after removal the mobile device m 'of the coverage area its associated base station is satisfied. If the specified fulfills cover is, the mobile device is m 'from &lt;?page 17?&gt;the coverage area of its assigned base station, as indicated by the instructions in block <figref>1013</figref> specified. The process continues and returns to block <figref>1007</figref> back. If the cover probability requirement is not met, is the power level of the base station by the instructions the block <figref>1015</figref> incremented and the flow returns to block <figref>1005</figref> back

The in Table 1 above-described procedure can with minor performed modifications be if simple constraints of the formulation of the problem added will. For example, if there is a maximum allowable RF power levels, which must be respected, Step 7 must verify that the recalculated Power level meets this constraint. If not, has the problem not a viable solution and the algorithm ends without finding a. Otherwise, step 2 executed.

If complex constraints are added, can the resulting "Volume Cover" problem by using Existing algorithms for the solution solve such problems.

On self-configurable wireless system according to the invention to be used Prime simple approximate data, to start up and run, and then uses signal strength measurements, collected, determine both the RF power level better to be able to as well as Channel assignment purposes. This initial data is entered by a technician are or may be determined from measurements that are collected, when mobile devices be moved under the supervision of a technician in the service area. When the system is up and running, to collect data on be and the Selbstkonfigurations- and self-calibration processes refined.

conclusion

It herein is a novel system and method for automated Even configuration of channel assignments and self-calibration described by RF power levels of a wireless communication system. Thus have Self-configuration and self-calibration no longer be an activity, the installation of a new system or its extension must precede. The implementation of such systems is well within the skills current technology and air interface standards. The existing wireless system architecture can be obtained in a manner which enables where appropriate, between the present "manual" and an autonomous mode switch. It should be noted that although the described embodiments of the invention in general with regard to frequency allocations were characterized in analog systems, it is understood that various changes, amendments and substitutions may be made therein without departing from the area of departing invention, as defined in the appended claims. especially it should be noted that the embodiments described Although generally based on channel assignment and RF power transmission levels has been described in TDMA systems, the same procedure, to the power setting in CDMA systems and analog systems can be applied as well as GSM systems. They are also on hybrids be applicable of two or more such systems. Furthermore, appreciated that the methodology of the invention also microcellular Systems, including building internal Systems, will be applicable.

E. channel allocation algorithm

at a preferred embodiment, the invention is implemented by the method of the invention Channel allocation algorithm from the class adaptivdynamischer channel allocation algorithms selected as described above and in more detail is explained in the designated as M. Benveniste-6 cited application. This ADCA algorithm achieves the following goals: It provides resistant more than regular fixed channel allocation and adapts fully to traffic variations at. Therefore, it minimizes the number of cell sites, the necessary are to meet a traffic load with variable distribution.

If one realizes that Kanalumordnungen should be used sparingly, ADCA may implement such Kanalumordnungen, relies but not limited to these. [For a meeting of Kanalumordnungsbetrachtungen see as Benveniste-6 designated cited application.] In addition, the used