RU2640030C1 - Method of adaptive distribution of bandwidth-time resource - Google Patents

Abstract

FIELD: radio engineering, communication.SUBSTANCE: in the method, for each user station, the angular parameter of arrival of a signal of the reverse channel is estimated by processing in the adaptive antenna system of the base station, a prohibiting graph is created for the set of said angular parameters at the base station, indicating for which user stations it is impossible to perform a full spatial selection of signals, and for the resulted prohibiting graph by the method of the regular colouring of the vertices produce a schedule for the distribution of the bandwidth-time resource for the forward and reverse channels with maximum density for the observed spatial-angular distribution of the user stations. Then, on the base station, the schedule of the distribution of the bandwidth-time resource of the forward and reverse channels of the generated frame by the positions of the vectors of weight processing is supplemented.EFFECT: increasing the capacity of the radio channel and the efficiency of use of the bandwidth-time resource.12 dwg

Description

The invention relates to the field of radio engineering, in particular to methods for transmitting and receiving a radio signal using an adaptive antenna array, and can be used in cellular communication systems.

A known method of adaptive distribution of the time-frequency resource [1], including determining for each generated frame the forward and reverse channels for each subscriber station, the required values of energy parameters for various types of coding and modulation depending on a given quality, measuring the value of energy parameters in the current frame direct and return channels, transmitting to the base station the values of energy parameters measured in the forward channel and the assignment at the base station of each ab a subscriber station of the type of coding and modulation, as well as a portion of the time-frequency resource in the generated frames of the forward and reverse channels.

The disadvantages of this method are not high enough bandwidth of the radio channel and the efficiency of using the time-frequency resource.

Closest to the proposed method is the adaptive distribution of the time-frequency resource [2], which consists in the fact that for each generated frame of the forward and reverse channels for each subscriber station, the required values of energy parameters for various types are determined coding and modulation, depending on the specified quality, measure the energy parameters in the current frame of the forward and reverse channels, for each subscriber station AI determine the value of the time-frequency resource necessary for transmitting the required amount of data in the generated frame of the forward and reverse channels, and assign each type of modulation and coding to each subscriber station, as well as the value of the time-frequency resource, while the required energy values are determined for each subscriber station parameters for various types of coding and modulation, depending on a given amount of data, predict the magnitude of the energy parameters of the generated frame forward and reverse data on the values of transmission powers and measured energy parameters of previous frames, determine the required values of transmission powers for various types of coding and modulation depending on the required values and the predicted value of energy parameters, exclude from further consideration the types of coding and modulation for which the required value of transmission powers is unattainable due to restrictions on the range of power adjustment, for all remaining types of coding and modulation, the values the other time-frequency resource necessary for transmitting the required amount of data in the generated frame of the forward and reverse channels, the necessary values of the time-frequency resource N of the subscriber stations, corresponding to the remaining types of encoding and modulation with the maximum data transfer rate, are summarized and the resulting amount is compared with the available frequency - a temporary resource of the generated frame, in case of non-exceeding, each subscriber station is assigned a type of modulation and coding, as well as the corresponding transmit power and the value of the time-frequency resource in such a way as to minimize the average power of the generated frame subject to the transfer of all necessary data, in case of exceeding the value of the time-frequency resource is distributed between subscriber stations in accordance with their priority and assign each subscriber station a modulation and coding mode with the maximum data rate and the corresponding transmit power.

The disadvantages of this method are the insufficiently high bandwidth of the radio channel and the efficiency of using the time-frequency resource due to the fact that the spatial-angular distribution of subscriber stations and the possibility of spatial-angular selection of signals in the antenna system of the base station are not taken into account.

The technical result consists in increasing the bandwidth of the radio channel and the efficiency of using the time-frequency resource by using the graph method to formulate a distribution schedule of the time-frequency resource having the maximum density for the observed spatial-angular distribution of subscriber stations.

The required technical result is achieved by the fact that in the method of adaptive distribution of the time-frequency resource, which consists in the fact that for each generated frame of the forward and reverse channels for each subscriber station, the required values of energy parameters for various types are determined coding and modulation, depending on a given quality, measure the values of energy parameters in the current frame of the forward and reverse channels, for each subscriber stations determine the value of the time-frequency resource necessary for transmitting the required amount of data in the generated frame of the forward and reverse channels, and assign each type of modulation and coding to each subscriber station, as well as the value of the time-frequency resource, according to the invention , for each subscriber station, evaluate the angular the parameter of the arrival of the signal of the return channel by processing in an adaptive antenna system, for a set of the indicated angular parameters at the base station, a prohibitory graph is generated showing for which subscriber stations it is impossible to perform spatial selection of signals in an adaptive antenna system with a given quality, for the obtained forbidding graph by the correct vertex coloring method, for the frame following the generated one, a time-frequency resource distribution schedule for the forward and reverse channels with the maximum density for the observed spatial-angular distribution of subscriber stations, after which, into the service control information, prepared for transmission to the tech In the frame, enter data on the time-frequency resource allocation schedule assigned for the generated frame, then group signals of the current frame for each antenna element of the adaptive antenna system are generated in the direct channel by weighted summation of the signals generated for each of the subscriber stations with weights specified by vectors Partial subscriber subscriber units are allocated in the schedule for the direct channel in the time-frequency positions of the current frame, as well as at the base station the signals of the current frame by weighting the output signals of the antenna elements of the adaptive antenna system using the weighting vectors specified in the schedule for the return channel in the time-frequency positions of the current frame, after which at the base station for each time-frequency fragment of the generated frame make a list of operating subscriber stations, for each of them the weighting vector for the forward channel in the generated frame and the weighting vector for the return channel in the forming emom frame and then to the base station schedules allocation complementary frequency-time resource forward and reverse channels formed by the frame positions of said weighting vectors.

In FIG. 1 is a structural diagram of an embodiment of a device by which the inventive method is carried out.

In FIG. 2 shows the personnel structure of the organization of the distribution of the time resource and elementary time intervals (EVI).

In FIG. Figure 3 shows an example of the organization of a spatial-frequency-time schedule.

In FIG. Figure 4 shows an example of constructing a “forbidding graph” on a set of N subscriber stations (AS) served by a base station (BS) at the first stage (without taking into account differences in requests for a radio channel resource).

In FIG. 5 shows an example of constructing a “prohibitory cone” in the direction of a given subscriber, with which ribs are formed in the “prohibitory graph”.

In FIG. Figure 6 shows an example of constructing a “prohibitory graph” at the last stage (making corrections related to various requests for a radio channel resource)

In FIG. 7 (a-d) shows a step-by-step example of the operation of the algorithm for the correct “coloring of nodes” of the “forbidding graph” constructed on the basis of the operation of “contracting” the nodes.

In FIG. 8 shows the final result obtained by the algorithm of the correct “coloring of nodes” of the “forbidding graph” for the example shown in FIG. 6.

In FIG. Figure 9 shows the optimal time-frequency resource allocation schedule, compiled in accordance with the obtained solution to the "coloring of nodes" task of the "prohibit graph".

Consider the implementation of the proposed method by the example of data transmission in a direct channel from a base station to one of N subscriber stations and in a reverse channel from a subscriber station to a base station.

Base station 1 contains an adaptive antenna system 2, a first coding and modulation unit 3, an adaptation unit 5, a direction estimation unit 6, a space-time-time schedule unit 7, and a first demodulation and decoding unit 4, the first output of which is connected to the corresponding control input of the unit 5 adaptation, the information input is connected to the corresponding output of the adaptive antenna system 2 and to the corresponding input of the direction estimation unit 6, the control input is connected to the corresponding first output of the adaptation unit 5, and from the information output the signal is fed to the higher levels of the base station. The group input of the direction estimation unit 6 is connected to the corresponding outputs of the antenna elements included in the adaptive antenna system 2, the output of the direction estimation unit 6 is connected to the input of the space-time-time schedule unit 7, the group output of which is connected to the group control input of the adaptive antenna system 2 and the output is connected to the input of the adaptation unit 5, while the control input of the space-time-frequency schedule unit 7 is connected to the second output of the adaptation unit 5. Information from the higher levels of the base station is received at the group information input of adaptation unit 5, and the group information output is connected to the group input of the coding and modulation unit 3, the group output of which is connected to the group information input of the adaptive antenna system 2, from the broadcast output of which the signal is transmitted through the output base station 1 to the input of the subscriber station 8. The subscriber station 8 includes a second block 9 of demodulation and decoding, the information input of which is the air input of the subscriber station 8, and the output is connected to the input of the schedule schedule support unit 10, the control input of the second demodulation and decoding block 9 is connected to the first output of the schedule schedule support block 10, the second output of the schedule schedule support block 10 is connected to the control input of the second coding and modulation block 11 . Information from the upper levels of the subscriber station 8 is fed to the information input of the second coding and modulation unit 11, from the output of the second coding and modulation unit 11, the signal is transmitted through the output from the subscriber station 8 to the input of the base station 1, which is also an adaptive antenna broadcast input system 2, and from the information output of the second demodulation and decoding unit 9, the information signal is transmitted to the higher levels of the subscriber station 8.

A communication system includes at least one base station and N subscriber stations served by a given base station, where N takes values 1, 2, etc.

The base station transmits data to subscriber stations on the forward channel, and subscriber stations transmit data to the base station on the reverse channel.

Direct channels of different base stations use the same frequency resource (frequency band). The return channels of different subscriber stations also use the same frequency resource.

The forward and reverse channels can use the same frequency resource for time duplex or different frequency resources for frequency duplex.

Data transmitted in a direct channel and intended for different subscriber stations, in the example of implementation of the proposed method, are divided among themselves in the direction space (using processing in an adaptive antenna system), and can also be divided in time and also in time and frequency.

The data transmitted in the return channel from different subscriber stations, in the example of the implementation of the proposed method, are divided among themselves in the direction space (using processing in an adaptive antenna system), and can also be separated in time and also in time and frequency.

The transmitter of the base station of the system has the ability to change the encoding and modulation of the transmitted data intended for individual subscriber stations once per frame, as well as redistribute the time-frequency resource of the direct channel between them.

The adaptive antenna system of the base station has the ability to change sets of weight processing vectors once a frame, designed to implement spatial selection in the transmission mode of signals to individual subscriber stations and designed to implement spatial selection in the reception mode of signals transmitted by individual subscriber stations.

The transmitter of any subscriber station in the system has the ability to change the encoding and modulation of the transmitted data once per frame, as well as transmit data in any part of the time-frequency resource of the reverse channel frame.

The system operates using time division into frames, as shown in FIG. 2. We will call the current frame in which the system is currently operating, and the frame immediately following the current frame to be formed. Each frame set and remain unchanged for each of the N subscribers Methods encoding / modulation allocation of time-frequency resource, weighting vector for the forward and reverse channels. Each frame is divided into elementary time intervals, on which, according to the schedule, there is information exchange over the air with a fixed active subset of subscribers. For the period of one frame, each of N subscribers receives at least one elementary time interval for transmission and reception. This means that active subscriber reception subsets and active subscriber transmission subsets are complete groups. On the interval of the current frame, the base station 1 prepares for the frame following the generated one and transmits control information already prepared for the formed frame containing data on the assigned time-frequency resource allocation schedule, the applied coding and modulation methods, as well as energy parameters in the form of levels broadcasts in the forward and reverse channels.

Adaptive antenna system 2, including M antenna elements, can be implemented using standard methods described in [3].

Adaptive antenna system 2 emits / receives signals to / from multiple subscriber sources divided into groups. In this case, a work schedule compiled for the current frame is used, based on the information contained in the service message of one of the previous frames. It is shown in FIG. 3. Specifically, information from the higher levels of the base station is fed to the group information input of adaptation block 5. Each line of the group information input receives a stream of data intended for one of the subscribers. At each elementary time interval in adaptation block 5, data is read from the distribution schedule of the time-frequency resource of the broadcast mode shown in FIG. 3. On their basis are determined:

1. The active positions of the time-frequency resource used for broadcast in a direct channel in a given elementary time interval (in the example shown in Fig. 3, they are displayed as shaded cells).

, где i - номер элементарного временного интервала (i=1,2, …, k), k - число элементарных временных интервалов в одном кадре, j - номер ячейки из диапазона частотного ресурса (j=1,2, …, l), l - число частотных ячеек, составляющих частотный ресурс радиоканала,

- число информационных потоков, которые должны предаваться в ресурсной ячейке (i, j) в прямом канале.2. The numbers of information flows that should be broadcast in a direct channel in each active time-frequency position. In FIG. 3 they are shown as a sequence of schedule cells

where i is the number of the elementary time interval (i = 1,2 , ..., k), k is the number of elementary time intervals in one frame, j is the number of the cell from the frequency resource range ( j = 1,2, ..., l ), l is the number of frequency cells making up the frequency resource of the radio channel,

- the number of information flows that must be transmitted in the resource cell ( i, j ) in the direct channel.

потоков через ресурсную ячейку (i, j). На фиг. 3 они показаны в виде последовательности ячеек расписания

соответственно.3. The types of modulation and coding used in the broadcast in the direct channel of each of

resource flows through the cell (i, j). In FIG. 3 they are shown as a sequence of schedule cells

respectively.

The adaptation unit 5 can be implemented on the basis of a digital processing program operating on the basis of a conventional processor or a specialized computing module.

The adaptation unit 5 supplements the information flows delivered over the elementary time interval with control overhead information and, in accordance with the schedule data, feeds them to the corresponding group input lines of the coding and modulation unit 3. Also, the adaptation unit 5 through the group output transmits to the encoding and modulation unit 3 control commands designed to activate the modulation and encoding procedures associated with the time-frequency distribution of the forward channel resource, prescribed in the schedule of the current frame.

From the corresponding first output of adaptation block 5, control commands for the activation of demodulation and decoding procedures are transmitted to the control input of the first demodulation and decoding unit 4, linked to the time-frequency distribution of the reverse channel resource, prescribed in the schedule of the current frame for subscriber station 8.

The adaptation unit 5, based on the data obtained in the current frame, calculates the energy parameters, as well as the types of modulation and coding that will be applied in the frame following the generated one. Methods for calculating these parameters are known and described in [2].

From the second output of adaptation block 5, information on energy parameters in the form of broadcast levels in the forward and reverse channels in the current, generated and following frames for the base station and for N subscriber stations is transmitted to the control input of block 7 of the spatial frequency-time schedule. Data on energy parameters are also transmitted in the form of results of measurements of received signal levels by base station 1 in the return channel and N subscriber stations in the forward channel, recorded in the current frame. In addition, information is transmitted about the requests of N subscriber stations for resources of the forward and reverse channels for the frame following the generated one.

Block 3 coding and modulation can be implemented on the basis of a digital processing program that runs on the basis of a conventional processor or specialized computing module.

In coding and modulation unit 3, each of the streams transmitted in the forward channel undergoes coding and modulation transformations specified in the time-frequency resource allocation schedule. As a result, from the group output of the coding and modulation unit to the group information input of the adaptive antenna system 2 in an elementary time intervali (i =1,2 ...k) served

сигналов ресурсной ячейки (i, j), j=1,2, …, l, задают квадратурные уровни, назначенные для трансляции через антенные элементы (m=1…М). На фиг. 3 указанные вектора показаны в виде множества последовательностей столбцов

encoded and modulated signals. Adaptive antenna system 2 at the beginning of each elementary interval (for definiteness we will denote it by number i ) through the group control input reads from the group output of block 7 of the spatial-frequency-time schedule the set of values of the weight processing vectors of dimension M prescribed in the current frame schedule for the direct channel on this elementary interval. These vectors for each of

the signals of the resource cell ( i, j ), j = 1,2, ..., l , specify the quadrature levels assigned for translation through the antenna elements ( m = 1 ... M ). In FIG. 3, these vectors are shown as multiple column sequences.

, для каждого из

сигналов. Далее для каждого антенного элемента m (m=1, …, М) адаптивная антенная система 2 выполняет взвешенное суммированиеIn the process, the adaptive antenna system 2 forms quadratures

for each of

signals. Further, for each antenna element m ( m = 1, ..., M ), the adaptive antenna system 2 performs a weighted summation

after which the generated group signal Si (m) is supplied to the input of the corresponding antenna element with the number m . Note that since the summation index r is uniquely related to the index of the frequency cell j , the summation in (2) leads to the fact that the result Si (m) does not depend on j .

As a result, broadcasting output adaptive antenna system 2 for each elementary timeslot (i) will be broadcasting a plurality of signals, each of which will form an individual directivity pattern with the maximum area in the direction of the respective subscriber, and minima zones in directions other subscribers.

Adaptive antenna system 2, working in the reception mode, at the beginning of each elementary time interval ( i ), reads from the group control input connected to the group output of the space-time-time schedule unit 7, the distribution of the frequency-time resource of the reverse channel mode of the current frame shown in FIG. 3. On their basis are determined:

1. The active positions of the time-frequency resource used to receive signals of the reverse channel in this elementary time interval (in the example shown in Fig. 3, they are displayed as shaded cells).

, где i - номер элементарного временного интервала (i=1,2, …, k), j - номер ячейки из диапазона частотного ресурса (j=1,2, …, l),

- число информационных потоков, которые должны быть приняты в ресурсной ячейке (i, j) обратного канала.2. The numbers of information flows that must be received at each active time-frequency position. In FIG. 3 they are shown as a sequence of schedule cells

where i is the number of the elementary time interval ( i = 1,2, ..., k ), j is the cell number from the frequency resource range ( j = 1,2, ..., l ),

- the number of information flows that must be received in the resource cell ( i, j ) of the reverse channel.

сигналов. На фиг 3 они показаны в виде множества последовательностей столбцов

с комплексными значениями.3. The vectors of weight processing of dimension M , used for adaptive spatial processing at the reception in the resource cell ( i, j ) of the return channel for each of

signals. In Fig. 3, they are shown as a plurality of column sequences.

with complex values.

After that, the adaptive antenna system 2 calculates the total number of signals received on the elementary time interval ( i ) of the return channel according to the formula

адаптивная антенная система 2 формирует соответствующий сигнал выхода на основе операции взвешенного суммирования квадратурных сигналов

и

, присутствующих на выходах М антенных элементов:Next, for each weighting vector from the set

adaptive antenna system 2 generates a corresponding output signal based on the operation of weighted summation of quadrature signals

and

present at the outputs of the M antenna elements:

) - номер информационного потока, передаваемого в ячейке обратного канала. Сигнал выхода Si(r, t) с индексом r, соответствующим абонентской станции 8, адаптивная антенная система 2 подключает к информационному входу первого блока 4 демодуляции и декодирования. Также сигнал выхода Si(r,t) с выхода адаптивной антенной системы 2 подается на соответствующий управляющий вход блока 6 оценки направлений, где используется в качестве обучающего, или эталонного, при работе в режиме слежения за угловыми перемещениями абонентской станции 8. Алгоритмы оценки и слежения за угловыми параметрами источников известны и описаны в [3], [4], [5]. Сформированные оценки направлений прихода абонентских сигналов через выход блока 6 оценки направлений передаются на вход блока 7 пространственно-частотно-временного расписания. Блок 6 оценки направлений может быть реализован с помощью программы, работающей на базе обычного процессора или специализированного вычислительного модуля.The indices r that specify the numbers of the output signals Si (r, t) generated at the output of the receive path of the adaptive antenna system 2 are uniquely associated with the set of indices i, j and q ( q = 1, 2, ...,

) is the number of the information stream transmitted in the reverse channel cell. The output signal Si (r, t) with the index r corresponding to the subscriber station 8, the adaptive antenna system 2 connects to the information input of the first block 4 of demodulation and decoding. Also, the output signal Si (r, t) from the output of the adaptive antenna system 2 is fed to the corresponding control input of the direction estimation unit 6, where it is used as a training or reference one when operating in the tracking mode of the angular movements of the subscriber station 8. Assessment and tracking algorithms behind the angular parameters of the sources are known and described in [3], [4], [5]. The generated estimates of the directions of arrival of subscriber signals through the output of block 6 of the direction estimation are transmitted to the input of block 7 of the spatial-frequency-time schedule. Block 6 assessment of directions can be implemented using a program that runs on the basis of a conventional processor or specialized computing module.

Block 7 of the spatial-frequency-time schedule can also be implemented using a program operating on the basis of a conventional processor or a specialized computing module.

Block 7 of the spatial-frequency-time schedule reads from the input information about the realized angular distribution of directions of arrival of subscriber signals. Based on this information, block 7 of the spatial-frequency-time schedule constructs a “forbidding graph”. For the explanation of FIG. 4 gives an example of such a “forbidding graph”. The process of constructing a “forbidding graph” is as follows. The position of the base station 1 is placed at the conditional origin. To each direction of arrival of the subscriber signal, we associate a node of the “inhibit graph”. The position of the location of the specified node in space is formed in two stages:

1) draw from a conditional coordinate origin a ray whose direction coincides with the direction of arrival of the subscriber signal indicated above;

2) we mark on the received ray a point that is distant from the conventional origin at a distance in the close range of a given value. (The choice of distance is arbitrary, since it does not affect further operations and calculations. Its choice serves only to illustrate the operation of the “prohibit graph” formation mode so that nodes of close directions do not merge).

Next, iterates over all directions of the arrival of subscribers' signals. At the same time, for each selected direction of the subscriber’s signal arrival, we construct a “prohibiting cone” with an angular size at the top of 2⋅As (see Fig. 5). And all the nodes of the “forbidding graph” that fall inside the specified “forbidding cone” are connected by edges with the node associated with the selected direction of the subscriber's signal arrival. The size of the forbidding sector 2⋅As, as shown by the results of the experimental verification, should be chosen within the range, [0.75-0.85] ⋅ (Ф ° / M ), where Ф ° is the total size of the angular sector of the adaptive antenna system 2, M is the number of antenna elements in the adaptive antenna system 2. When 2⋅As is selected from this range, the radio channel bandwidth as a result turns out to be either the same or slightly different from the maximum. After that, the space-frequency-time schedule unit 7 reads from the control input information on the resources of the communication channel requested by subscriber stations in the frame following the generated one. The values of the requested resources are subjected to a sampling procedure, as a result of which integer values are obtained D _a1 , D _a2 , ..., D _{a N} , showing how many “minimum portions” of the resource should be allocated to subscriber stations when scheduling. The practical experience of cellular networks has shown that to solve the problems of maximizing the capacity of the radio channel when scheduling, it is sufficient to use from 3 to 6 levels of discretization of the requested resources. Further, in accordance with the obtained values of D _a1 , D _a2 , ..., D _{a N} , in the “prohibitory graph”, the operation D _{a n} -fold duplication ( n = 1,2, ..., N ) of nodes mapped to subscriber stations is performed. In FIG. Figure 6 shows, for example, the situation of operations of 2-fold and 3-fold duplication of nodes in the “forbidden graph”. The indicated operation of duplicating graph nodes is known and described in [6]. It consists in introducing an additional node into the graph that copies all the edges of its inverse image node and introducing the edges between the additional node and the inverse image node. On this, the formation of the “prohibitory graph” ends.

After that, in block 7 of the spatial-frequency-time schedule, three processing steps are performed, as a result of which a spatial-frequency-time schedule is compiled for the frame following the generated one.

At the first processing step, the “correct coloring” of the nodes of the “forbidden graph” is performed. The correct coloring of graph nodes by the minimum number of colors is a classical problem of graph theory, which is described in detail in [6], [7]. It involves coloring all nodes of the graph so that there are no two nodes painted in the same color and connected by an edge. In our case, each color can be interpreted as the fragment number of the time-frequency channel resource. The “correct coloring” of the “prohibitory graph” ensures that for all pairs composed of subscribers for which the angular distance of the directions of arrival of the signals does not allow the adaptive antenna system 2 to spatially select signals with a given quality level, there will not be a single pair of subscribers “painted in one color ”, or (which is equivalent) there will be no pair in which the subscribers are allocated a common time-frequency resource in the generated schedule. Each subscriber group associated with a particular coloring color can use the total time-frequency resource of the channel, since all the signals of the subscribers that make up this group can be divided with a given quality level solely on the basis of spatial selection methods in the adaptive antenna system 2. Each group subscribers associated with a particular coloring color, it is necessary to allocate a separate time-frequency resource in the generated schedule, as otherwise it may turn out that in a combined frequency on-time resource will work subscribers whose signals can not be separated with a given level of quality in an adaptive antenna system 2. The minimum number of colors "proper coloring" provides a maximum time-frequency radio resource is allocated to each user group associated with a different color paint.

The algorithm of “correct coloring” of graph nodes is known and described in [6], [7]. It is based on the sequential execution of the operation of "tightening nodes". This operation involves the arbitrary selection of a pair of nodes in the graph that are not connected by an edge, and the subsequent replacement of the selected pair with one node that has connecting edges that at least one of the nodes of the contracted pair had. FIG. 7a-7g explain the indicated “contraction” operation. The process ends when the graph reaches a state of complete connectedness (Fig. 7d). The solution to the coloring problem is determined by the resulting fully connected graph with a minimum number of nodes. Since at each step of the algorithm it is necessary to consider all kinds of pairs of “contractible nodes”, the number of processed graphs increases exponentially with an increase in the number of steps. Therefore, the exact solution to the problem of coloring nodes in an acceptable time can only be obtained for graphs with a small number of nodes (up to 8-12). Since the number of subscribers served by base station 1 may turn out to be larger, other algorithms having less high complexity are required to solve the coloring problem. In the theory of graphs, a whole class of such algorithms called “greedy” has been developed, which at each step use an abbreviated enumeration of combinations of nodes to “contract”, and as a result have polynomial complexity. Moreover, with a probability close to 1, they give an exact solution to the problem of “correct coloring” of the nodes and only in some cases they are mistaken in the direction of increasing the minimum number of colors (usually by 1 position). These algorithms are known and their description can be found in [8], [9]. Specifically, test tests of the work of “greedy algorithms” were carried out, using at each step the choice of a single pair of “contractible nodes” using the predictive estimates of Turan, described in [6], and based on the Mullat method described in [10], [eleven]. In both cases, in samples of more than 2,000 tests, not a single error was found solving the problem of “correctly coloring” the nodes of the graph with the minimum number of colors. The “coloring” algorithm acquires a particularly simple and visual form in the case of using forecast estimates of Turan. These estimates allow us to find the upper and lower boundaries of the minimum number of colors based on analytical relationships of the number of nodes and the number of edges of the graph. According to them, for a given number of nodes, with an increase in the number of edges, only the growth of the indicated boundaries can occur. Therefore, at each step for the “contracting” operation, one should choose a pair of unconnected nodes that have the maximum number of common “neighbors” (neighboring nodes are the graph nodes connected by an edge), since with this choice, after the “contraction” the number of edges in the graph is reduced as much as possible. The example shown in FIG. 7a - FIG. 7d, demonstrates such a method of choosing pairs for “contraction”. In FIG. Figure 8 shows the final result of solving the problem of the correct “coloring of nodes” of the “forbidding graph”. The colors of the “coloring” are conventionally indicated by Roman numerals.

Also in FIG. Figure 8 shows a sector structure corresponding to a certain configuration of a base station with conventional (non-adaptive) antennas, with the help of which one can explain the physical meaning of the algorithm for compiling a spatial-frequency-time schedule constructed on the basis of the “coloring of nodes” of the “forbidden graph”. The size of an individual sector in FIG. 8 is given by the angular resolution of the adaptive antenna system 2. As you can see, the operation of the algorithm can be interpreted as adjusting the azimuths of sectors so that "dense" subscriber clusters fall in the vicinity of the central zones of the sectors (that is, they go beyond the zones of overlapping sectors). For other subscribers who are in the vicinity of overlapping zones, the algorithm, when creating the schedule, takes into account the possibility of working in a mode combining time-frequency resources with those subscribers who are on the opposite edge of the sector.

In FIG. Figure 9 shows an example of an optimal time-frequency resource allocation schedule, compiled in accordance with the obtained solution to the “coloring of nodes” task of a “forbidding graph”.

At the second step, the time-frequency resources of the radio channel allocated to base station 1 for broadcast and reception are divided into equal fragments by a number equal to the number of colors obtained when solving the problem of the “correct coloring” of nodes of the “forbidden graph”. After that, the distribution of the time-frequency resource of the radio channel for the schedules of the forward and reverse channels in the frame following the generated one is compiled. At the same time, each subscriber receives exactly that fragment of the time-frequency resource that corresponds to the group to which he fell when solving the problem of the “correct coloring” of nodes of the “forbidden graph”. Then, the received schedule for each subscriber is supplemented with information about the applied modulation, coding, and the required transmission level. After that, information about the compiled schedule of the frame following the generated one is transmitted through the output to the input of the adaptation unit 5, in which information flows of the higher levels of the base station 1 are supplemented with control overhead information containing data on the generated schedule. Further, in the generated frame it will be sequentially through the coding and modulation unit 3, through the adaptive antenna system unit 2 and through the direct channel ether, the compiled schedule will be broadcast to N subscriber stations (and to subscriber station 8, in particular).

At the third step, the time-frequency resource distribution schedule compiled for the generated frame is supplemented with information about the weight processing vectors that should be used in the forward and receive transmission modes on the return channels at base station 1. For this purpose, for each resource cell ( i , j) the schedules of the generated frame are compiled lists of numbers of subscriber stations:

A) which base station 1 will transmit on a direct channel:

,

,

B) from which base station 1 will receive a signal on the return channel:

и при приеме

, k - значение из списка, однозначно определяющее номер абонентской станции, с которой осуществляется радиосвязь.After that, for each resource cell ( i, j ) of the schedule of the generated frame, weight processing vectors are calculated, which will be used at base station 1 during broadcast

and when taking

, k is a value from the list that uniquely determines the number of the subscriber station with which radio communication is carried out.

Algorithms for calculating weight processing vectors are known and are given in [3], [12], [13].

After that, block 7 of the spatial-frequency-time schedule supplements each resource cell ( i, j ) of the schedule of the generated frame with the obtained values of the weighting vectors. On this, the process of scheduling the spatial-frequency-temporal distribution for the generated frame ends.

и декодирования по закону

, где индексы (i, j) определяются позицией ресурсной ячейки обратного канала в текущем кадре, а r задает номер информационного потока в указанной ресурсной ячейке (i, j), связанного с абонентской станцией 8.To the control input of the first block 4 of demodulation and decoding from the corresponding first output of block 5 adaptation is transmitted a command to set the desired mode of operation with demodulation according to the law

and decoding by law

where the indices ( i, j ) are determined by the position of the resource cell of the return channel in the current frame, and r sets the number of information flow in the specified resource cell ( i, j ) associated with the subscriber station 8.

The first block 4 of demodulation and decoding can be implemented using a digital processing program based on a conventional processor or specialized computing module. Base station 1 must support the operation of N such demodulation and decoding units 4, each of which organizes the processing of the reverse channel signal for one of the N subscriber stations.

The processing of Si (r, t) output signal with index r, a corresponding signal betray subscriber station 8 in the resource cell (i, j), the first block 4 the demodulation and decoding produces a data stream from which the allocated service information (reports measurements performed by subscriber station 8, resource requests for the forward and reverse channels for the frame following the generated one), and the data stream for higher levels of base station 1. The data stream for higher levels of base station 1 is fed to the information output of the first block 4 de modulation and decoding.

In this case, through the first output of the first block 4 of demodulation and decoding to the control input of adaptation block 5, dedicated service information is transmitted, supplemented by data on the energy parameter in the form of measurements of the level of the signal of the subscriber station 8 received in the reverse channel

At the subscriber station 8, at the beginning of each elementary time interval, the schedule schedule support unit 10, which can be implemented using a program operating on the basis of a conventional processor or a specialized computing module, supplies an on / off signal to the control input of the second demodulation and decoding unit 9 , depending on whether or not the schedule for this elementary time interval is ( i ) broadcast on a direct channel from base station 1 to a subscriber station tion 8.

The second block 9 of demodulation and decoding can be implemented similarly to the prototype [2].

,

, j=1, …, l, где (i, j) - номера ресурсных ячеек, в которых по расписанию текущего кадра происходит трансляция сигнала от базовой станции 1 на абонентскую станцию 8. Второй блок 9 демодуляции и декодирования производит обработку принятого сигнала, поступающего с информационного входа, в соответствии типами модуляции, кодирования и частотно-временным ресурсом, заданными расписанием прямого канала текущего кадра для абонентской станции 8. В результате второй блок 9 демодуляции и декодирования осуществляет выделение принятых данных и разделяет их на данные высших уровней абонентской станции 8 и служебную информацию. Указанная служебная информация содержит распределение частотно-временного ресурса, виды модуляции/кодирования, энергетические параметры в виде уровней передачи в прямом и обратном каналах формируемого кадра для абонентской станции 8. После этого второй блок 9 демодуляции и декодирования дополняет служебную информацию данными энергетического параметра в виде измерений уровня принятого по прямому каналу сигнала базовой станции 1. Данные высших уровней второй блок 9 демодуляции и декодирования передает на выход абонентской станции 8, а служебную информацию на управляющий выход, с которого она поступает на вход блока 10 поддержки распорядка расписания. Со второго выхода блок 10 поддержки распорядка расписания передает на управляющий вход второго блока 11 кодирования и модуляции данные об измеренном энергетическом параметре в виде уровня принятого в прямом канале текущего кадра сигнала базовой станции 1, а также сигнал включения/выключения, в зависимости от того, прописана или нет в расписании на данном элементарном временном интервале (i) трансляция по обратному каналу от абонентской станции 8 на базовую станцию 1.The work of the subscriber station 8 is similar to [2]. Namely, if according to the schedule of the current frame at an elementary time interval i ( i = 1, ..., k ), the operation of the second demodulation and decoding unit 9 is activated, then the schedule support unit 10 transmits data on the type of modulation and encoding used through the first output

,

, J = 1, ..., l, where (i, j) - number of resource of cells in which the schedule of the current frame occurs broadcast signal from the base station 1 to the subscriber station 8. The second unit 9 performs demodulation and decoding processing of the received signal coming from the information input, in accordance with the types of modulation, coding and time-frequency resource specified by the direct channel schedule of the current frame for the subscriber station 8. As a result, the second demodulation and decoding unit 9 extracts the received data and separates and x data higher levels of the subscriber station 8 and service information. The specified overhead information contains the distribution of the time-frequency resource, types of modulation / coding, energy parameters in the form of transmission levels in the forward and reverse channels of the generated frame for the subscriber station 8. After that, the second demodulation and decoding unit 9 supplements the overhead information with energy parameter data in the form of measurements the level of the signal received on the direct channel of the base station 1. The data of the higher levels of the second block 9 demodulation and decoding transmits to the output of the subscriber station 8, and ludicrous information to the control output, from which it enters the input of the schedule routine support unit 10. From the second output, the schedule schedule support unit 10 transfers to the control input of the second coding and modulation block 11 data on the measured energy parameter in the form of the level of the signal of base station 1 received in the direct channel of the current frame, as well as an on / off signal, depending on which or not in the schedule for this elementary time interval ( i ) broadcast on the return channel from the subscriber station 8 to the base station 1.

The second block 11 coding and modulation can be implemented similarly to the prototype [2].

,

, j=1, …, l, где (i, j) - номера ресурсных ячеек, в которых по расписанию текущего кадра происходит трансляция сигнала по обратному каналу от абонентской станции 8 на базовую станцию 1. Второй блок 11 кодирования и модуляции дополняет данные высших уровней абонентской станции 8 и запрос ресурсов прямого и обратного каналов для кадра, следующего за формируемым, поступающие к нему через информационный вход, данными рапорта произведенных измерений. После чего второй блок 11 кодирования и модуляции производит кодирование и модуляцию в соответствии с правилами, заданными расписанием обратного канала текущего кадра, усиливает полученный сигнал до уровня, заданного соответствующим энергетическим параметром в расписании и транслирует через выход сформированный сигнал по обратному каналу на эфирный вход базовой станции 1.Also similar to [2], if the schedule of the current frame on the elementary timeslot i (i = 1, ..., k) is an activation of the second encoding unit 11 and modulation, the block 10 supporting schedules through the second output transmits the type of data used modulation and coding

,

, j = 1, ..., l , where ( i, j ) are the numbers of resource cells in which according to the schedule of the current frame the signal is transmitted on the return channel from the subscriber station 8 to the base station 1. The second coding and modulation unit 11 supplements the data of higher levels of the subscriber station 8 and a request for resources of the forward and reverse channels for the frame following the generated one, coming to it through the information input, with the data of the report of the measurements taken. After that, the second coding and modulation unit 11 performs coding and modulation in accordance with the rules specified by the schedule of the reverse channel of the current frame, amplifies the received signal to the level specified by the corresponding energy parameter in the schedule, and transmits the generated signal via the return channel to the broadcast input of the base station through the output one.

Thus, by using the graph method to formulate a time-frequency resource distribution schedule having a maximum density for the observed spatial-angular distribution of subscriber stations, in combination with spatial signal selection performed in the adaptive antenna system of the base station, the radio channel capacity and efficiency are increased use of the time-frequency resource.

Information sources

1. Application for invention US 2002/0183010, H04B 17/00. "Wireless communication systems with adaptive channelization and link adaptation".

2. Patent RU No. 2267863, IPC H04B 17/00, priority 21.08.2003.

3. Monzingo RA, Haupt RL, Miller TW Introduction to Adaptive Arrays. - 2 ^nd Edition / Published by SciTech Publishing, Inc., 2011 .-- 510 p.

4. Abramovich Yu.I., Spencer H.K., Gorokhov A.Yu. Isolation of independent radiation sources on non-equidistant antenna arrays // Advances in modern radio electronics, 2001. - No. 12. - S. 3-17.

5. Roy R., Kailath T. ESPRIT - Estimation of Signal Parameters Via Rotational Invariance Techniques // IEEE Transaction on Acoustics Speech and Signal Processing, Vol. 37, No. 7, May 1989, p. 984-995.

6. Zykov A.A. Fundamentals of graph theory. - M .: Nauka, 1987 .-- 384 p.

7. Ore O. Graph theory. - M.: Nauka, Main Edition of Physics and Mathematics, 1980. - 336 p.

8. Curtis S.A. The Classification of Greedy Algorithms // Science of Computer Programming 49 (2003), p. 125-157, Elsevier.

9. Halldorsson M., Szegedy M. Lower bounds for on-line graph coloring // Theoretical Computer Science 130 (1994), p. 163-174, Elsevier.

10. Mullat I.E. Extreme subsystems of monotone systems. 1 // Automation and telemechanics, 1976. - №5. - S. 130.

11. Mullat I.E. Extreme subsystems of monotone systems. 2 // Automation and telemechanics, 1976. - No. 8. - S. 169.

12. Zhuravlev A.K., Lukoshkin A.P., Poddubny S.S. Signal processing in adaptive antenna arrays. - L.: Publishing House of Leningrad University, 1983. - 240 p.

13. Widrow B., Stearns S. Adaptive signal processing. - M .: Mir, 1989 .-- 440 p.

Claims (1)

The method of adaptive distribution of the time-frequency resource, which consists in the fact that at the base station including the adaptive antenna system, for each generated frame of the forward and reverse channels for each subscriber station, the required values of energy parameters for various types of coding and modulation depending on the specified quality are determined , measure the values of energy parameters in the current frame of the forward and reverse channels, for each subscriber station determine the value of the time-frequency p the resources necessary for transmitting the required amount of data in the generated frame of the forward and reverse channels, and assign each type of modulation and coding to each subscriber station, as well as the time-frequency resource, characterized in that for each subscriber station, the angular parameter of the return channel signal is estimated by processing in an adaptive antenna system, for a set of estimated angular parameters at the base station, a prohibitory graph is generated showing for which subscriber stations it is impossible to spatial selection of signals in an adaptive antenna system with a given quality, for the obtained forbidding graph by the correct vertex coloring method, for the frame following the generated one, the time-frequency resource distribution schedule for the forward and reverse channels with the maximum density for the observed spatial-angular distribution of subscriber stations is generated, then in the control overhead prepared for transmission in the current frame, enter the data assigned for the generated adra to the time-frequency resource distribution schedule, then group signals of the current frame for each antenna element of the adaptive antenna system are generated in the direct channel by weighted summation of the signals generated for each of the subscriber stations with weights specified by the weighting vectors specified in the schedule for the direct channel in the time-frequency positions of the current frame, as well as at the base station, partial subscriber signals of the current frame are extracted by weighting the signals moves of the antenna elements of the adaptive antenna system using the weighting vectors prescribed in the schedule for the return channel in the time-frequency positions of the current frame, after which at the base station for each time-frequency fragment of the generated frame a list of working subscriber stations is compiled, for each of them calculate the weighting vector for the forward channel in the generated frame and the weighting vector for the reverse channel in the formed frame, and then the schedule is supplemented at the base station the distribution of the time-frequency resource of the forward and reverse channels of the generated frame by the positions of the aforementioned weight processing vectors.

Images