WO2012152167A1

WO2012152167A1 - Wave beam search processing method, device and system

Info

Publication number: WO2012152167A1
Application number: PCT/CN2012/073970
Authority: WO
Inventors: 刘培; 李斌; 邹卫霞
Original assignee: 华为技术有限公司; 北京邮电大学
Priority date: 2011-05-10
Filing date: 2012-04-13
Publication date: 2012-11-15
Also published as: CN102164374B; CN102164374A

Abstract

A wave beam search processing method, device and system. The method abstracts the wave beam search as a problem of searching for a globally optimal solution by way of two stages, namely rough search and fine search, realizing coordination search of two communication parties, with a good search performance, which can significantly improve the efficiency of the wave beam search, and shorten the time and power consumption required by the wave beam search.

Description

Beam search processing method, device and system

The present application claims priority to Chinese Patent Application No. 201110120022.8, entitled "Beam Search Processing Method, Apparatus, and System", which is incorporated herein by reference. in.

Technical field

The embodiments of the present invention relate to the field of wireless communications technologies, and in particular, to a beam search processing method, apparatus, and system.

Background technique

With the development of multimedia applications, the demand for transmission rate and signal bandwidth of wireless communication applications has increased, and the demand for high-speed data transmission such as online video streaming services has become increasingly prominent. The corresponding wireless personal area network (WPAN) It also places higher demands on data transmission rate and signal bandwidth. At present, the 100 Mbps transmission rate of the mobile network and Wireless Local Area Networks (WLAN) is difficult to meet the application requirements. On the other hand, there are many wireless communication services such as broadband mobile communication, satellite navigation communication, and local area network. And metropolitan area networks, etc., so that valuable spectrum resources are depleted, and finding new communication bands with good transmission characteristics has become an urgent problem to be solved. Therefore, 60 GHz wireless communication capable of achieving Gbps or even Gbps transmission rate has become a new hot spot in the field of wireless communication. 60 GHz wireless communication belongs to the field of millimeter wave communication. Millimeter wave usually refers to electromagnetic waves with a wavelength of 1-10 mm, and its corresponding frequency range is 30 GHz-300 GHz. It has wide applications in many fields such as communication, radar, navigation, remote sensing, radio astronomy, etc. 60GHz communication has the following advantages: large traffic, unlicensed bandwidth of more than 5GHz; good directionality, strong security and confidentiality, often suitable for point-to-point short-range communication; high transmission quality, all-weather communication; International versatility and exemption Can be characterized. The 60 GHz antenna can integrate multi-antenna technology and realize beamforming. The beamforming technology requires the communication parties to dynamically update the Antenna Weighting Vector (AWA) according to the Channel State Information (CSI) to achieve the optimal direction. Beam adjustment

(Beam Steering). Under the premise of a given beam pattern set, beamforming is turned into finding the best beam pair (Beam Pair) to optimize the communication link. In the prior art, the IEEE 802.15 3c standard and the IEEE 802.11 ad standard both propose corresponding beam search algorithms to find the best communication beam number based on the codebook space.

In the process of implementing the present invention, the inventors have found that at least the following problems exist in the prior art: The existing beam search algorithm is mainly based on traversal search, and the search complexity is high, and the search time required for beam alignment is unbearable, which severely limits The performance of beamforming in 60 GHz applications has the following disadvantages: Blindness: In the beam search process, the communication parties do not have a clear search target, and the optimal solution can only be found after one exhaustive search; redundancy : The search process has great redundancy. The search of many beam numbers has no practical beneficial effect on the discovery of the optimal solution. Lack of coordination: The lack of necessary coordination between the receiving end and the transmitting end will inevitably greatly affect the search efficiency. .

Summary of the invention

An embodiment of the present invention provides a beam search processing method, including:

A first terminal receiving a response request signal terminal ²¹ to the number of antennas transmitted, i is a natural number, determining a first signal corresponding to the maximum received signal power of a first beam of a beam current; a first search is repeatedly performed, if If the i reaches or is greater than the predetermined number of searches, the first search is terminated, and the current beam pair is used as an initial solution; the first search includes: the requesting end synchronizes with the responding end to update the value of i, and receives Responding to the second signal transmitted by the updated antenna number 2 ¹ , determining, according to the current beam pair, a second beam pair having the highest received signal power corresponding to the second signal, and updating the Describe that the current beam pair is the second beam pair;

The requesting end repeatedly performs a second search, and if the received signal reaches a predetermined precision, terminating the second search, using the initial solution as an optimal beam pair and notifying the responding end of the optimal beam pair; The second search includes: the requesting end and the responding end update the number of antennas to respective maximum antenna numbers; the requesting end receives a signal transmitted by the responding end, and based on the initial solution, based on a predetermined algorithm Determining a beam pair of the next search, generating an indication based on the beam pair of the next search, notifying the responding end to adjust a next transmit beam according to the indication, and updating the initial solution to the next search Beam pair.

An embodiment of the present invention provides a beam search processing apparatus, including:

A first search module, a first signal terminal for receiving a response to the transmitted number of antenna ^21, i is a natural number, determining a first signal corresponding to the maximum received signal power of a first beam of a beam current; repeatedly performed a first search, if the i reaches or is greater than a predetermined number of searches, the first search is terminated, and the current beam pair is used as an initial solution; the first search includes: the requesting end is updated synchronously with the responding end a value of i, receiving the second signal transmitted by the responding end with the updated number of antennas 2 ¹ , determining a second beam pair having the highest received signal power corresponding to the second signal according to the current beam pair, Describe that the current beam pair is the second beam pair;

a second search module, configured to repeatedly perform the second search, terminate the second search if the received signal reaches a predetermined accuracy, use the initial solution as an optimal beam pair, and notify the optimal beam pair of the The second search includes: the requesting end and the responding end update the number of antennas to a respective maximum number of antennas; the requesting end receives a signal transmitted by the responding end, and according to the initial solution, Determining a beam pair of the next search based on a predetermined algorithm, generating an indication based on the beam pair of the next search, notifying the responding end to adjust a next transmit beam according to the indication, and updating the The initial solution is the beam pair of the next search.

An embodiment of the present invention provides a beam search processing system, including a request end for performing beam search and a response end for transmitting a signal, wherein the request end includes the beam search processing device described above.

The beam search processing method, apparatus and system provided by the embodiments of the present invention abstract the beam search into a problem of finding a global optimal solution through rough search and fine search, and realize coordinated search of the communication parties, and have a good search. Performance can significantly improve the efficiency of beam search and shorten the time and power consumption required for beam search. DRAWINGS

1 is a flowchart of a beam search processing method according to an embodiment of the present invention;

2 is a flowchart of a beam search processing method according to another embodiment of the present invention;

3 is a schematic diagram of a beam pattern used in simulation according to an embodiment of the present invention;

4 is a schematic diagram of an objective function of a beam search algorithm according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of a search trajectory of a Resenbrock algorithm according to an embodiment of the present invention; FIG.

6 is a schematic diagram showing target optimal power and actual search power values in different implementations of a simulation embodiment of the present invention;

7 is a schematic diagram of search times in different implementations of a simulation embodiment of the present invention;

8 is a schematic diagram of target optimal power and actual search power value in different implementations of another simulation embodiment of the present invention;

9 is a schematic diagram of search times in different implementations of another simulation embodiment of the present invention;

FIG. 10 is a schematic diagram showing the complexity of a Resenbrock search algorithm according to an embodiment of the present invention; FIG.

FIG. 11 is a schematic structural diagram of a beam search processing apparatus according to an embodiment of the present invention; FIG. FIG. 12 is a schematic structural diagram of a beam search processing system according to an embodiment of the present invention. detailed description

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is a partial embodiment of the invention, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

Since 60 GHz wireless communication can achieve Gbps or even several Gbps transmission rates, it has become a new research topic in the field of wireless communication. Improving the efficiency of the beam alignment process not only saves a large amount of transmit power, but also provides the necessary preconditions for implementing low-power devices. At the same time, shortening the search time can also greatly reduce the time delay for the device to access the network, which is convenient. Develop real-time transmission services to improve the user experience, and on the other hand, increase network communication capacity. In summary, an efficient beam search algorithm is of great significance for 60 GHz communication systems. In order to meet the above requirements, various embodiments of the present invention propose a novel beam search scheme based on a pattern search idea, which is especially suitable for beam alignment in a large codebook space situation, and the algorithm realizes a single order, and the search efficiency is significantly improved, which is suitable for Low power, low complexity 60 GHz millimeter wave communication device.

Considering that when the receiver beam and the transmitter beam are strictly aligned on a two-dimensional plane, the received signal power will reach a maximum. Therefore, in the embodiment of the present invention, the receiver-transmitter beam number (p, q) can be used to establish a two-dimensional search plane, and the beam search target is: searching for a beam pair corresponding to maximizing the received signal power, that is, 2 dimensions. Plane optimization search problem.

In general, the use of objective function gradient information can greatly improve the search efficiency. However, for the optimal beam pair search problem modeled above, there may be two factors to consider: First, the receiving letter The analytical solution of the power value is related to the angle (pose) of the antenna array. In practice, it cannot be accurately known. Second, the objective function in the actual application has local optimal solution and saddle point, which makes the algorithm easy to fall into the local solution. The embodiment of the present invention can be used as a high-efficiency mode search algorithm in the case of no derivative information by using a suboptimal search method, such as an optimization algorithm such as Resenbrock, which can efficiently achieve the best without knowing the analytical value of the search target function derivative. The search for beam pairs.

If the number of antennas (ie, the number of antenna elements) is set to 32, and the number of beams is 64, then from the 100 independent beam search experiments, since the objective function has many local solutions, the Resenbrock algorithm alone cannot guarantee each The optimal beam pair can be found every time; from multiple simulations, the probability of successful search is about 32%, and the average number is about 25. Therefore, embodiments of the present invention can perform a rough search of the beam several times in advance before using an optimization algorithm such as Resenbrock to find the optimal beam pair, thereby providing a good initial search point for the Resenbrock search algorithm, so as to avoid the algorithm falling into a local minimum. Improve beam search efficiency.

In the embodiment of the present invention, the beam coarse search P segment needs to have the following two requirements: First, as a pre-search algorithm targeting initialization, the complexity is not too high; second, after the rough search is completed, the initial provided The point distance from the optimal target value is not 4 , to effectively avoid falling into the local optimal solution.

For the conventional beamforming algorithm, the following features are apparent: The optimal beam corresponding to the number of antennas is 2M, which is located within the range of the optimal beam main lobe corresponding to the number of antennas 2 (M-1), which is an embodiment of the present invention. Use the above features for rough search. For example, the number of effective antennas can be set to four, and the optimal beam position can be obtained by a few searches; then the number of effective antennas is set to eight, and the optimal beam position can be started by the optimal solution of four antennas. A few searches were obtained. In general, the above 2-3 times rough can reduce the distance between the initial solution and the optimal solution of the Resenbrock algorithm, and reduce the probability of falling into the local optimal solution, thereby improving the search success rate, and even reaching 100%. The inference that the optimal beam corresponding to the number of antenna elements is 2M is located within the range of the optimal beam main lobe corresponding to the number of antenna elements 2 (M-1) is mainly based on the fact that when the number of antennas is doubled, The effective beamwidth (the distance between the beam gain maximum direction and the angle at which the adjacent beam gain is zero) is reduced by half. Correspondingly, from the one-dimensional point of view, when the number of array elements is 2M, it is possible that the optimal beam j should exist in the neighborhood of the beam number i with the number of array elements 2 (M-1), that is, the possible value of j The set is {i+1, il }.

Extend the above inference to the 2-D search plane to be searched. If the current refined beam pair number is (p, q), the corresponding number of antennas is 2 (M-1); the number of antennas is 2M. The optimal beam pair value set is {(p+1 , q), (p+l,q+l), (p,q+l), (pl,q+l), (pl,q), ( Pl, ql), (p, ql), (p+l, ql)}. The above conclusions are obvious. By continuously reducing the beamwidth resolution, the initial solution of the Rosenbrock algorithm search is efficiently found.

Based on the above analysis, the beam search processing method in the embodiment of the present invention may be performed in two stages of a rough search phase and a fine search phase. 1 is a flow chart of a beam search processing method according to an embodiment of the present invention. As shown in FIG. 1, the method includes:

Step 100, the requester receives a response to a first terminal number signal emitted by the antenna ^21, i is a natural number, determining a first signal corresponding to the maximum received signal power of a first beam of a beam current; a first repeatedly performed Searching, if the updated i is greater than a predetermined number of searches, terminating the first search, and using the current beam pair as an initial solution; the first search includes: the requesting end updates with the responding end i a value of the second signal transmitted by the responding end with the updated number of antennas 2 ¹ , determining a second beam pair having the highest received signal power corresponding to the second signal according to the current beam pair, and updating the The current beam pair is the second beam pair;

The signal transmitting end and the signal receiving end are optimal in order to satisfy the strength of the received signal, and may be before communication Find the best communication beam number by beam search. Of course, the signal transmission and reception are mutual, the signal receiving end also transmits a signal, and the signal transmitting end also receives the signal correspondingly. Therefore, for convenience of description, the embodiment of the present invention only considers the case where one end transmits a signal and the other end receives a signal. The responding end is used as a signal transmitting end, and the requesting end is used as a signal receiving end. After receiving the signal transmitted by the responding end, the requesting end finds a suboptimal solution approaching the optimal beam pair by using the beam searching method provided in this embodiment. You can even find the optimal beam pair for communication.

This step is a rough search phase in the beam search method provided in this embodiment. In this phase, the number of antennas of the requesting end and the responding end should be kept the same number. After performing a rough search, the requesting end and the responding end respectively The number of antennas has doubled. The last rough search result is used as the initial solution for this rough search.

Specifically, the requesting end and the responding end first perform a session request/confirmation process. After the process is completed, the requesting end requests the responding end to transmit a signal, and the responding end transmits the first signal to the requesting end according to the current number of antennas, and the current number of antennas of the responding end. For example, 2 ¹ , where i is a natural number. Initially, the number of antennas of the requesting end and the responding end may both be 2 (ie, i=l), and the requesting end receives the first signal transmitted by the responding end with 2 antennas, and then determines the beam with the largest received signal power corresponding to the first signal. Yes, initially (the number of antennas of both the requester and the response segment is 2) The requesting end can traverse all possible beams in the 2-D plane to determine the beam pair with the largest receiving energy, for example, called the current beam pair.

After the requesting end determines the current beam pair, the requesting end repeats the step of performing the first search, and if the updated i is greater than the predetermined number of searches, terminates the step of the first search, and uses the current beam pair as an initial solution for the fine search P. The segment performs a further fine search. The first search described in the embodiment of the present invention includes the following steps:

After determining the current beam pair, the requesting end synchronizes the value of i with the responding end to update each The number of antennas, for example, can be set to 1 = 1 +1 in synchronization, which means that the number of antennas is doubled by synchronization. Then, the requesting end requests the responding end to transmit a signal again, and the responding end continues to transmit the second signal with the updated 2 ² antennas. After receiving the second signal, the requesting end uses the first beam pair as an initial solution to determine a beam pair with the largest received energy corresponding to the signal, for example, a second beam pair, and updates the current beam pair to a second beam pair. That is, the second beam pair is used as the current beam pair, and the second beam pair is correspondingly the initial solution of the fine search P segment. The specific process may be that the requesting end sequentially traverses the eight beam pairs around the first beam pair on the 2-D plane, and obtains the second beam pair with the largest received signal power corresponding to the second signal.

The requesting end and the responding end continue the above cyclic process, and the number of antennas is set to 2 ³ , and the requesting end uses the second beam pair as an initial solution to determine the beam with the largest receiving energy corresponding to the signal transmitted by the responding end by the ²³ antennas. Correct. It can be seen from the above that when determining the optimal beam pair corresponding to the number of antennas being 2 ¹⁺¹ , the optimal beam pair corresponding to the number of antennas is 2 ¹ as the initial solution of the search, and the above steps are sequentially performed cyclically. Until the updated i is greater than the predetermined number of searches.

Step 101: The requesting end repeatedly performs a second search, and if the received signal reaches a predetermined precision, terminating the second search, using the initial solution as an optimal beam pair and notifying the optimal beam pair The second search includes: the requesting end and the responding end update the number of antennas to a respective maximum number of antennas; the requesting end receives a signal transmitted by the responding end, and according to the initial solution, Determining, according to a predetermined algorithm, a beam pair of the next search, generating an indication based on the beam pair of the next search, notifying the responding end to adjust a next transmit beam according to the indication, and updating the initial solution to the next Beam pairs for one search.

The steps of the above rough search are sequentially executed in a loop until the updated i is greater than the predetermined number of searches. The number of searches in the embodiment of the present invention is set according to the maximum number of antennas at the requesting end. After the rough search process of the search times, the rough search method in the beam search method provided in this embodiment is completed. The search phase provides a good initial solution for the next step of the fine search phase. The following is the fine search phase. Specifically, after determining the corresponding optimal beam pair in each rough search step, the requesting end performs an operation of setting i=i+i, and when it is determined that the updated i is less than or equal to the number of searches, the user continues to perform the operation. A rough search step; if it is judged that the updated i is greater than the number of searches, the rough search phase is ended and the fine search phase is entered. Specifically, the method includes the step of the requester repeatedly performing the second search, and the step of terminating the second search if the received signal reaches a predetermined precision, using the initial solution as the optimal beam pair and notifying the responding end of the optimal beam pair. The second search includes the following steps:

The requesting end and the responding end synchronize to update the number of antennas to the respective maximum number of antennas, that is, the requesting end sets its own working antenna to its maximum number of antennas, and the responding end also sets its own working antenna to its maximum number of antennas. Then, the requesting end continues to request the responding end to transmit a signal, and after receiving the request, the responding end sends a signal to the requesting end with its maximum number of antennas. After receiving the signal, the requesting end uses the current beam pair as an initial solution of the predetermined algorithm, and determines a beam pair for the next search based on a predetermined algorithm. The current beam pair described herein is a beam pair having the largest received signal power determined by the requesting end when the requesting end determines that the updated i is greater than the predetermined number of searches, that is, the requesting end performs a rough search. The optimal beam pair that is finalized by the segment.

Of course, this embodiment can also set the condition for terminating the first search to be: i reaches a predetermined number of searches (that is, i is equal to a predetermined number of searches), and is not necessarily greater than the number of times. .

During the fine search phase of the requesting end, the predetermined algorithm may be used for beam searching, and the result of approaching the optimal beam pair is obtained by a predetermined algorithm. The requesting end determines the beam pair of the next beam search by using the optimal beam pair finally determined by the rough search phase as the initial solution of the predetermined algorithm. After determining, the requesting end generates an indication based on the determined beam pair of the next search, and the notification sounds The receiver adjusts the next transmit beam according to the indication, and then the requester updates the initial solution to the next searched beam pair to determine the next searched beam pair as the initial solution for the next search. Specifically, the requesting end notifies the responding end of the next beam number according to the result of the algorithm, and simultaneously adjusts its own receiving beam. After receiving the notification, the responder can adjust the transmit beam of the next transmitted signal according to the indication therein.

After several times of the fine search adjustment process, the requesting end determines that the received signal reaches a predetermined accuracy in a certain fine search process, for example, by determining whether the received power of the signal reaches the requirement, etc., the requesting end can terminate the fine search. The process terminates the second search step. The optimal beam pair corresponding to the fine search is used as the final result of the beam search, and is notified to the responding end, and then the requesting end and the responding end facilitate the communication with the optimal beam.

The beam search processing method provided by the embodiment of the present invention abstracts the beam search into a problem of finding a global optimal solution through the two stages of rough search and fine search, and realizes coordinated search of both communication parties, and has good search performance and can be remarkable Improve the efficiency of beam search and shorten the time and power consumption required for beam search.

The predetermined algorithm used in the fine search phase in the embodiment of the present invention may adopt the foregoing

The Resenbrock algorithm, of course, can also use other algorithms, as long as it can satisfy the stepwise approximation of the optimal solution.

In the foregoing embodiment, the step of updating the number of antennas by the responding end may be performed in the following two manners. One way is that both the requesting end and the responding end automatically update the number of antennas based on a predetermined update period, that is, after one period T , the operation of i=i+l is performed synchronously. Another way is that the requesting end sends an indication message for updating the number of antennas to the responding end, and after performing the rough search step every time, the requesting end notifies the responding end to perform the operation of #i=i+l, and the self-synchronization is performed 1= 1+1 operation. In the foregoing embodiment, in the process of performing the rough search phase, if the requesting end determines that the received signal has reached the predetermined precision, the requesting end may directly use the current beam pair determined at this time as the optimal beam. For the notification response end, the requesting end uses the current beam as the final beam search result, and directly notifies the responding end of the optimal beam pair for communication for subsequent communication. Of course, the fine search P segment can also be performed to more approximate the optimal solution. The specific can be set according to actual needs, the former has a small delay, and the latter has a higher precision.

In general, the beam search method provided by the embodiment of the present invention has good compatibility with the existing 802.11ad draft standard, and only the working device includes a specific computational search algorithm engine. The design signal interaction and the existing standard defined beam training frame structure can be mutually compatible, that is, there is a search process in which the feedback channel is coordinated and unified in the existing standard. The requesting end may send a feedback frame to the responding end through a feedback channel, where the feedback frame includes at least a field for indicating a beam adjustment direction, and a field for indicating a beam adjustment length. The feedback frame structure adopted by this search algorithm is shown in Table 1.

As shown in the above table, the feedback frame can be composed of one byte length. Bit 8: Beam number adjustment direction: 0 is to reduce the beam number direction, 1 is to increase the beam number direction (the beam number is sequentially numbered from -90 degrees to 90 degrees); 1st-7: in binary form The relative length of the beam adjustment. Example: If the current beam number of the responder is i, after it receives the feedback frame, adjust the next beam number to: i+3.

The beam search processing method provided by the embodiment of the present invention, by means of the Resenbrock mode search mechanism, It realizes the coordinated search of both communication parties, has good search performance, can significantly improve the efficiency of beam search, shorten the time and power consumption required for beam search, and is extremely important for 60GHz WPANs millimeter wave communication system.

2 is a flowchart of a beam search processing method according to another embodiment of the present invention. As shown in FIG. 2, the method includes:

Step la, the requesting end initiates a beam alignment session request, and the requesting end starts the search engine in the subsequent search;

Step 2a, the response end returns a confirmation request;

Step 3a, the requesting end determines the number of beams according to the maximum number of antennas, sets the number of times of the rough search process, that is, the number of searches m, and sets the current coarse search counter i=l;

For example, if the number of antennas on the request side is 32 and the number of antennas on the receiving side is 64, m can be set to 5. In addition, the initial value of i can be set according to the actual situation (that is, i may not be set to 1 at the beginning, but start at a predetermined initial value).

Step 4a, the requesting end and the responding end set the number of transmitting working antennas to 2 ¹ , and the responding end transmits a signal to the requesting end;

Step 5a, the requesting end traverses all possible beams on the 2-D plane, and determines an optimal beam label pair with the largest received energy, and sets 1=1+1;

Step 6a, the requesting end determines whether i is greater than m; if yes, then jumps to step 10a, otherwise continues to the next step;

Step 7a, the requesting end uses the current optimal solution as an initial solution, and sequentially traverses eight beam pairs around the initial solution on the 2-D plane; if the current optimal solution is (p, q), the eight around the current optimal solution The beam pairs are {(p+l, q), (p+l,q+l), (p,q+l), (pl,q+l), (pl,q), (pl,ql) , (p,ql), (p+l,ql)} ; Step 8a, the requesting end determines and updates the maximum beam pair of the received signal energy, and sets i=i+l;

Step 9a, the requesting end and the responding end set the number of transmitting working antennas to 2 ¹ , the responding end transmits a signal to the requesting end, and jumps to step 6a;

Step 10a, the transmitting end and the receiving end set the number of working antennas to the respective maximum antenna numbers; the subsequent responding end uses the corresponding beam transmitting signal in the next search according to the beam adjustment indication received by the feedback channel;

Step 11a, using the received signal power, the requesting end runs the Rosenbrock algorithm, and determines the beam pair for the next search;

Step 12a, determine if the Rosenbrock algorithm is terminated? If yes, go to step 15a; otherwise, proceed to the next step;

Step 13a, the requesting end uses the feedback channel to notify the responding end of the next beam number, and adjusts the receiving beam at the same time;

Step 14a, the responding end adjusts the next transmit beam according to the received beam number information; and jumps to step 11a;

Step 15a, the requesting end terminates the beam search algorithm, and uses the feedback channel to notify the response segment of the optimal beam number;

In step 16a, both the requesting end and the responding end start communication by using the optimal beam.

In the beam search method provided by the embodiment of the present invention, if the channel has reciprocity, the search engine (algorithm) can run simultaneously on the receiving end and the transmitting end, that is, the same input will necessarily produce a collaborative search result, thereby enabling reception. The end and the transmitting end are synchronized in the 2-D plane to advance the search process, and the beam adjustment action can be coordinated without feedback signaling. In a wider application scenario, the external interference power received by the receiving end and the transmitting end may be inconsistent. At this time, the search results at both ends are lost. Deteriorating search performance. Therefore, the search engine (algorithm) can also be run at either end, and the search result is sent back to the other end through the feedback channel to overcome the search failure in the case of noise asymmetry.

The beam search method provided by the embodiment of the present invention has the following beneficial effects:

First, the beam search is abstracted into a high-dimensional space (the above embodiment is a two-dimensional space as an example) to find the global optimal solution, the clever mathematical abstraction is conducive to the development of a more efficient beam search algorithm;

Second, based on the optimization model modeling, the transmitter and receiver achieve maximum synergy search. This synergy will undoubtedly improve search efficiency and shorten search time. This cooperative policy is centrally controlled by the Initiator: The Initiator uses the relevant scheduling signaling (preamble frame) to notify the responding end Responder of the next expected beam pair generated by the current decision, and the Responder according to the received adjustment instruction (including The next beam numbering information), the beam pair is adjusted to the corresponding beam number, in order to cooperate with the Initiator's next search operation, thereby avoiding blind beam number adjustment, thereby achieving efficient collaborative search;

Third, the optimized search mechanism is adopted to avoid redundant search in the blind (traversal) search. Each step of the search provides a positive and heuristic push for the final optimization solution, thereby effectively reducing the search complexity and reducing overhead (Overhead). Save energy. Compared with the existing beam search, the algorithm complexity is greatly reduced, and the beam alignment process is shortened, which is low-complexity, low power consumption, low access delay 60 GHz application equipment. It has positive and important theoretical and practical significance. 4. As the number of beams increases, the beneficial effects of beamforming technology (increased network capacity) will be further enhanced, and the corresponding cost of existing beam search is severe. This situation is limited. However, the complexity of the method of the present embodiment approximates the result of 0 (Klog ₂ N ) (where 0 represents a high-order exponential function, K is a constant, and N is the number of antennas), which provides a high-precision beamforming technique. Necessary Prerequisites to further improve the communication performance of the 60 GHz system.

The beam search solution provided by the embodiment of the present invention is suitable for a beamforming method based on a codebook space in a 60 GHz millimeter wave communication system (ie, beam search), which can significantly reduce the search complexity of the algorithm compared to the existing search scheme. Degree, reduce the transmission header overhead and reduce power consumption. Even in the case of large code space, the existing algorithm can achieve high-efficiency beam search because of its high complexity and difficulty in application. In fact, for other systems using beamforming technology, this algorithm can also be used to improve efficiency and reduce the power consumption of the process; on the other hand, it allows for finer beams to further improve system communication performance.

The following is a combination of simulation results to further illustrate the significant beneficial effects of the method of the present invention. In the simulation, the 1-D uniform linear array has the number of beams set to twice the number of antennas (this setting is also mainly based on the design of the IEEE 802.15.3c related standard). Without loss of generality, the number of antennas at both the transmitting end and the receiving end is very obvious. The search algorithm is still applicable for different number of transmitting-receiving antennas. Accordingly, the pre-designed beam pattern (i.e., the codebook space) is as shown in FIG.

4 is a schematic diagram of an objective function of a beam search algorithm according to an embodiment of the present invention. As shown in FIG. 4, if the receiving end and the transmitting end beam number are the same, the signal receiving power reaches a maximum value; at the same time, the target search function There is a unique optimal value on the two-dimensional plane composed of the connected-to-beam numbers. FIG. 5 is a schematic diagram of a search trajectory of the Resenbrock algorithm in the embodiment of the present invention. As shown in Figure 5, when the random initial point distance is within a certain range (after 3 refinement searches), the Rosenbrock optimization search algorithm can be used to find the optimal solution within a limited number of times (ie, the optimal beam number is found). Combination).

6 is a schematic diagram of target optimal power and actual search power value in different implementations of a simulation embodiment of the present invention, and FIG. 7 is a schematic diagram of search times in different implementations of a simulation embodiment of the present invention, as shown in the figure, the simulation of the embodiment The number of multiple antennas in the scene is 32, the number of beams is 64, and the search space is 64 x 64. Two-dimensional beam numbering plane. From the perspective of 200 search implementations, this search algorithm finds the optimal solution in most cases; in rare cases, it can find a suboptimal solution that is not more than 1.5dB (intended to reduce the search complexity). The number of individual searches can exceed 40 times, but from the 200 independent implementations, the average number of searches is about 33, which reduces the search complexity (beam tracking time) by 50% compared to linear exhaustive search. Note: The target optimal power is the optimal reception performance of the direction that the optimal beam pair obtained by the traversal search can provide. 200 independents respectively correspond to different random independent antenna placement postures (ie, line array normal direction Θ).

FIG. 8 is a schematic diagram of target optimal power and actual search power value in different implementations of another simulation embodiment of the present invention. FIG. 9 is a schematic diagram of search times in different implementations of another simulation embodiment of the present invention. In the simulation scenario, the number of multiple antennas is 16 and 32, the number of beams is 32 and 64, respectively, and the search space is 32 x 64 two-dimensional beam numbering plane. Algorithm expansion: When the number of antennas is larger, the party with less array elements is no longer further refined. After the number of antennas is larger, the algorithm is searched by Rosenbrock. Performance does not produce much impact (the difference between the best beam and the beam is reduced); more importantly, the number of searches is reduced to 27, which is reduced by 6 times compared to the case where the number of beams is 64 x 64.

FIG. 10 is a schematic diagram showing the complexity of the Resenbrock search algorithm according to an embodiment of the present invention. As shown in the figure, compared with the existing exhaustive-linear search (the curve labeled "Γ" in the figure, the algorithm of this embodiment follows the beam thinning. The number of elements and array elements is further increased, and the search performance advantage is more obvious (the curve labeled "2" in the figure). When the number of antennas is 32, the optimization algorithm is reduced by 50% compared to the linear search complexity; For 64 o'clock, the search algorithm complexity of the optimization algorithm is reduced by 67%; (Note that in the simulation, the existing scheme search complexity of IEEE 802.11.ad is set to 0 (2N), and the actual complexity is greater than 0 (2N), which is 0. (2N+49) even 0(N ² ), so the contrast performance is only its lower bound; and the IEEE 802.15.3c scheme is progressive The noise is about 0 (N ² )).

A person skilled in the art can understand that all or part of the steps of implementing the above method embodiments may be completed by using hardware related to the program instructions. The foregoing program may be stored in a computer readable storage medium, and the program is executed when executed. The foregoing steps include the steps of the foregoing method embodiments; and the foregoing storage medium includes: a medium that can store program codes, such as a ROM, a RAM, a magnetic disk, or an optical disk.

FIG. 11 is a schematic structural diagram of a beam search processing apparatus according to an embodiment of the present invention. As shown in FIG. 11, the beam search processing apparatus includes a first search module 11 and a second search module 12, wherein the first search module 11 is configured to receive a response. The first signal transmitted by the antenna number 2 ¹ is a natural number, and the first beam pair having the highest received signal power corresponding to the first signal is determined as the current beam pair; the first search is repeatedly performed, if i is equal to or greater than The predetermined number of searches terminates the first search, and the current beam pair is used as an initial solution; the first search includes: the requesting end synchronizes with the responding end to update the value of i, and receives the response end. Determining, by the current beam pair, a second beam pair having the highest received signal power corresponding to the second signal, and updating the current beam pair to the second signal, according to the second signal transmitted by the updated antenna number 2 ¹ a pair of beams; the second search module 12 is configured to repeatedly perform the second search, and terminate the second search if the received signal reaches a predetermined accuracy, and the initial solution is An optimal beam pair and the optimal beam pair are notified to the responding end; the second search includes: the requesting end and the responding end updating the number of antennas to respective maximum antenna numbers; the requesting end Receiving a signal transmitted by the responding end, and determining, according to the initial solution, a beam pair for the next search based on a predetermined algorithm, generating an indication based on the beam pair of the next search, and notifying the responding end to adjust according to the indication The next time the beam is transmitted, and the initial solution is updated to the beam pair for the next search.

In the beam search processing apparatus provided in this embodiment, the P-segment is searched roughly by the first search module, and the fine search phase is performed by the second search module, and the two stages are roughly searched and refined. The beam search is abstracted into a problem of finding the global optimal solution, which realizes the coordinated search of both communication parties, has good search performance, can significantly improve the efficiency of beam search, and shorten the time and power consumption required for beam search.

Further, the first search module 11 determines the operation of the second beam pair with the highest received signal power corresponding to the second signal according to the current beam pair. Specifically, the first search module 11 traverses the eight surrounding the current beam pair on the 2-D plane. The pair of beams acquires a second beam pair having the highest received signal power corresponding to the second signal. Moreover, the first search module 11 is further configured to terminate the first search when the received signal reaches a predetermined accuracy, directly use the current beam pair as an optimal beam pair, and notify the responding end of the optimal beam pair. The second search module 12 notifies the responding end to adjust the operation of the next transmit beam according to the indication. Specifically, the second search module 12 sends a feedback frame to the responding end through the feedback channel, where the feedback frame includes at least a field for indicating a beam adjustment direction. And a field for indicating the length of the beam adjustment.

For detailed functions of the beam search processing apparatus provided in this embodiment, refer to the processing flow of the requesting end in the foregoing method embodiments, which is not mentioned here.

The beam search processing device provided by the embodiment of the present invention avoids redundant search in blind (traversal) search by optimizing the search mechanism, and each step search provides a positive and heuristic push for the final optimization solution, thereby effectively reducing search complexity. , saving energy consumption. Compared with the existing beam search, the algorithm complexity is greatly reduced, and the beam alignment process is shortened, which is low complexity, low power consumption, low access delay 60 GHz application equipment. It has positive and important theoretical and practical significance.

12 is a schematic diagram of a composition of a beam search processing system according to an embodiment of the present invention. As shown in FIG. 12, the beam search processing system includes a requesting end 1 for performing beam search and a response end 2 for transmitting a signal, wherein the requesting end 1 The beam search processing device provided by the foregoing embodiments of the beam search processing device is described. For the function and the beam search process, refer to the foregoing embodiments, and details are not described herein. In the beam search processing system provided by this embodiment, the beam search processing device abstracts the beam search into a problem of finding a global optimal solution through the rough search and the fine search, and realizes the coordinated search of the communication parties. The search performance can significantly improve the efficiency of beam search and shorten the time and power consumption required for beam search.

It should be noted that the above embodiments are only for explaining the technical solutions of the present invention, and are not intended to be limiting; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that: The technical solutions described in the foregoing embodiments are modified, or some of the technical features are equivalently replaced. The modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

Rights request

A beam search processing method, comprising:

A first terminal receiving a response request signal terminal ²¹ to the number of antennas transmitted, i is a natural number, determining a first signal corresponding to the maximum received signal power of a first beam of a beam current; a first search is repeatedly performed, if Terminating the first search by using the updated i to be greater than or equal to the predetermined number of searches, and using the current beam pair as an initial solution; the first search includes: the requesting end synchronizes with the responding end to update i And receiving, by the responding end, the second signal transmitted by the updated antenna number 2 ¹ , determining a second beam pair having the largest received signal power corresponding to the second signal according to the current beam pair, and updating the current a beam pair is the second beam pair;

The beam search processing method according to claim 1, wherein the requesting end determines, according to the current beam pair, a second beam pair having a maximum received signal power corresponding to the second signal, including:

The requesting end traverses eight beam pairs around the current beam pair on the 2-D plane, and acquires a second beam pair with the highest received signal power corresponding to the second signal.

3. The beam search processing method according to claim 2, wherein if said current wave The beam pair is ( p,q ), then the eight beam pairs around the current beam pair are {(p+1 , q), (p+l,q+l), (p,q+l), ( Pl,q+l), (pl,q), (pl,ql), (p,ql), (p+l,ql)}.

The beam search processing method according to claim 1, wherein the updating the number of antennas at the responding end comprises:

The requesting end and the responding end automatically update the number of antennas based on a predetermined update period; or the requesting end sends an indication message of updating the number of antennas to the responding end.

The beam search processing method according to any one of claims 1 to 4, wherein the predetermined algorithm is a Resenbrock algorithm.

The beam search processing method according to any one of claims 1 to 4, wherein the method further comprises:

The first search is terminated if the received signal reaches a predetermined accuracy, the current beam pair is directly used as an optimal beam pair, and the optimal beam pair is notified to the responding end.

The beam search processing method according to any one of claims 1 to 4, wherein the requesting end notifying the responding end to adjust the next transmit beam according to the indication comprises:

The requesting end sends a feedback frame to the responding end through a feedback channel, where the feedback frame includes at least a field for indicating a beam adjustment direction, and a field for indicating a beam adjustment length.

A beam search processing device, comprising:

A first search module, a first signal terminal for receiving a response to the transmitted number of antenna ^21, i is a natural number, determining a first signal corresponding to the maximum received signal power of a first beam of a beam current; repeatedly performed a first search, if the i reaches or is greater than a predetermined number of searches, the first search is terminated, and the current beam pair is used as an initial solution; the first search includes: the requesting end is updated synchronously with the responding end a value of i, receiving the second end of the responding end with the updated number of antennas 2 ¹ a signal, determining, according to the current beam pair, a second beam pair having a maximum received signal power corresponding to the second signal, and updating the current beam pair to the second beam pair;

a second search module, configured to repeatedly perform the second search, terminate the second search if the received signal reaches a predetermined accuracy, use the initial solution as an optimal beam pair, and notify the optimal beam pair of the The second search includes: the requesting end and the responding end update the number of antennas to a respective maximum number of antennas; the requesting end receives a signal transmitted by the responding end, and according to the initial solution, Determining, according to a predetermined algorithm, a beam pair of the next search, generating an indication based on the beam pair of the next search, notifying the responding end to adjust a next transmit beam according to the indication, and updating the initial solution to the next Beam pairs for one search.

The beam search processing device according to claim 8, wherein the first search module is further configured to:

And traversing the eight beam pairs around the current beam pair on the 2-D plane to obtain a second beam pair having the largest received signal power corresponding to the second signal.

The beam search processing device according to claim 8 or 9, wherein the first search module is further configured to:

The beam search processing device according to claim 8 or 9, wherein the second search module is further configured to:

A feedback frame is transmitted to the responding end through a feedback channel, the feedback frame including at least a field for indicating a direction of beam adjustment, and a field for indicating a length of beam adjustment.

12. A beam search processing system comprising a requester for performing a beam search and for transmitting A response end of the signal, characterized in that the requesting end comprises the beam search processing device according to any one of claims 9 to 11.