CN115882912A - Self-adaptive millimeter wave beam searching method - Google Patents

Self-adaptive millimeter wave beam searching method Download PDF

Info

Publication number
CN115882912A
CN115882912A CN202211513538.3A CN202211513538A CN115882912A CN 115882912 A CN115882912 A CN 115882912A CN 202211513538 A CN202211513538 A CN 202211513538A CN 115882912 A CN115882912 A CN 115882912A
Authority
CN
China
Prior art keywords
search
codebook
layer
beam search
signal
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202211513538.3A
Other languages
Chinese (zh)
Inventor
刘春山
柴文琦
李松
赵楼
郭莽青
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Dianzi University
Original Assignee
Hangzhou Dianzi University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hangzhou Dianzi University filed Critical Hangzhou Dianzi University
Priority to CN202211513538.3A priority Critical patent/CN115882912A/en
Publication of CN115882912A publication Critical patent/CN115882912A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Radio Transmission System (AREA)

Abstract

The invention discloses a self-adaptive millimeter wave beam searching method. The method comprises the following steps: 1. a scene of beam search, preset layered beams and a narrow beam search codebook; 2. an overall flow of performing adaptive beam search using a hierarchical beam search codebook and a narrow beam search codebook; 3. when the layered beam search codebook is used for beam search, the generalized likelihood ratio test is adopted to judge whether to switch the beam search codebook and whether to stop the layered search of the current level and enter the next level; 4. when the hierarchical search starts a new layer of search, selecting an angle interval to be searched; 5. when a condition for switching from the hierarchical beam search codebook to the narrow beam search codebook is satisfied, performing beam search using the narrow beam search codebook; 6. when all beam searches are stopped, the final beamforming direction is selected. The invention can realize the switching from the layered beam codebook to the narrow beam codebook, thereby reducing the time cost of beam search and simultaneously ensuring the beam alignment precision.

Description

Self-adaptive millimeter wave beam searching method
Technical Field
The invention belongs to the technical field of millimeter wave communication, and provides a self-adaptive millimeter wave beam searching method, which can self-adaptively determine to continuously execute a multi-precision layered beam searching codebook for beam searching or switch to a narrow beam searching codebook for beam searching, thereby adapting to different signal-to-noise ratio conditions and reducing the beam searching time overhead.
Background
Millimeter wave communication is considered as one of the key technologies to realize high-speed data transmission. In a millimeter wave communication system, a multi-antenna matched beam forming technology is usually adopted to improve the directional gain of signals, so that the path loss in the transmission process is made up. In millimeter wave communication, the beam directions of the transmitting end and the receiving end should be aligned with the main path direction of the communication channel as much as possible to achieve beam alignment, so that the maximum gain can be obtained.
The beam search method based on spatial scanning is one of the important means for realizing beam alignment. According to the difference of the beam Search codebooks, the beam Search method based on space scanning can be divided into two categories, one is Exhaustive Search (Exhaustive Search), and the algorithm mainly uses narrow beams to perform traversal Search on each beam pair to find a receiving and transmitting beam which is most matched with the main path direction of a channel. The exhaustive search requires traversing all beam pairs, which results in longer beam scanning time and more resource consumption. The other is Hierarchical Search (Hierarchical Search), which uses Hierarchical multi-precision beam Search codebooks of different widths to Search beams layer by layer, firstly uses a wider beam Search codebook to determine a local angle interval of a channel main arrival path, and then reduces the beam width in the determined local angle interval to continue searching; by continuously reducing the local angle interval and the used beam width, the best narrow beam matched with the communication channel is found. The hierarchical search method performs well under high snr conditions, but generally requires longer beam search times than the exhaustive search method to achieve similar beam alignment accuracy under low snr conditions.
In conclusion, the existing two types of classical millimeter wave beam searching methods can obtain better beam alignment performance only under the condition of the applicable signal-to-noise ratio, and cannot adapt to complex and variable mobile millimeter wave communication scenes. In order to solve the problem, a novel beam search method needs to be designed, which not only can adaptively select a proper beam search codebook according to different communication scenes, but also can adjust the beam search time to achieve a satisfactory beam search effect with minimum resource consumption.
Disclosure of Invention
The invention discloses a self-adaptive millimeter wave beam searching method aiming at a complex millimeter wave communication system. The method disclosed by the invention adopts generalized likelihood ratio test to judge the signal-to-noise ratio of an unknown channel in real time, and can realize the switching from a layered search beam codebook to a narrow beam codebook, so that the proper beam search codebook can be selected in a self-adaptive manner under high and low signal-to-noise ratios, and the beam alignment precision can be ensured while the beam search time overhead is reduced.
The invention provides a self-adaptive millimeter wave beam searching method, which comprises the following steps:
step 1, a wave beam searching scene, a preset multi-precision layered wave beam searching codebook and a narrow wave beam searching codebook;
step 2, using the hierarchical beam search codebook and the narrow beam search codebook to execute a total flow of the adaptive beam search;
step 3, when the layered beam search codebook is used for beam search, the generalized likelihood ratio test is adopted to judge whether to switch the beam search codebook and whether to stop the layered search of the current level and enter the next level;
step 4, when the layered search starts a new layer of search, selecting an angle interval to be searched;
step 5, when the condition of switching from the layered beam search codebook to the narrow beam search codebook is met, using the narrow beam search codebook to perform beam search;
and 6, when all the beam searching is stopped, selecting the final beam forming direction.
Preferably, the beam search scenario in step 1 includes a transmitter and a receiver, wherein the transmitter continuously transmits a pilot signal to the whole angular space covered by the transmitter, and the receiver performs beam scanning using a preset beam search codebook, so as to find an optimal beam forming direction.
Preferably, the beam search codebook in step 1 is divided into two types, one is a layered beam search codebook W, and the other is a narrow beam search codebook
Figure BDA0003969916380000021
Preferably, the hierarchical beam search codebook in step 1 has a tree structure, as shown in fig. 1, that is, a coverage area of a beam with a wider upper layer is jointly covered by a plurality of narrower beams in a next layer. The hierarchical beam search codebook used in the present invention has a total of K layers, where K layers and 2 layers k A codeword, the set of codewords for the k-th layer being denoted W k And is and
Figure BDA0003969916380000022
denotes the ith codeword, N, in the k layer R ＝2 K Indicating the number of antennas with which the receiver is equipped. Code word w (k,i) CV for angular range of coverage w(k,i) Is shown, wherein CV is w(k,i) ∈[-1,+1]Is the result after taking the cosine of the coverage angle. In the hierarchical beam search codebook, the union of all codeword coverage ranges of any one layer should completely cover the whole antenna array angle domain; and the coverage of the codeword of any layer can be fully represented by the union of the coverage of the two codewords of the corresponding next layer.
Preferably, the narrow beam search codebook in step 1 is single-layered, and the whole angular space is jointly covered by the narrow beams. The present invention searches all codewords W in the last layer of a codebook using hierarchical beams K As a narrow beam search codebook, is represented as
Figure BDA0003969916380000031
Preferably, the beam search in step 2 starts from a hierarchical beam search codebook, and beam scanning is performed.
Preferably, the overall beam search process described in step 2 is shown in fig. 2. Specifically, when the hierarchical beam search codebook is used to perform beam scanning, the receiving end judges whether the condition of switching to beam searching using the narrow beam search codebook is satisfied after completing one cycle of pilot signal acquisition each time. And if the condition of switching to the narrow beam search codebook is met, stopping using the layered beam search codebook to perform beam search, and performing beam search by using the narrow beam search codebook until the beam search is stopped. If the condition of switching to the narrow beam search codebook is not met and the current hierarchical search level is not the last level, judging whether to enter the next hierarchical search level; and if the condition for entering the next-layer hierarchical search is met, performing the beam search by using the next-layer beam search codebook. And if the condition for switching to the narrow beam search codebook is not met and the current hierarchical search level is the last level, judging whether the stopping condition of the beam search is met. And if the conditions are not met, continuously executing the hierarchical search of the current level.
Preferably, the beam search process in step 2 selects a final beamforming direction when the beam search stop condition is satisfied.
Preferably, the step 2 of completing a cycle of pilot signal acquisition by the hierarchical search receiving end means that the receiving end sequentially acquires the pilot signal transmitted by the receiver once by using all the beams to be selected of the current level.
Preferably, the switching from the hierarchical beam search codebook to the narrow beam search codebook in step 2 is determined by detecting the snr of the unknown channel. The reason is that the hierarchical beam search codebook is suitable for high signal-to-noise ratio scenarios, whereas the narrow beam search codebook is suitable for low signal-to-noise ratio scenarios. The beam search codebook switching mechanism provided by the invention can perfectly make up for the defects brought by using a single beam search codebook, and can use the most appropriate beam search codebook to perform beam search under the conditions of different signal-to-noise ratios, thereby improving the accuracy of beam search and reducing the consumption of resources.
WhereinPreferably, when the hierarchical beam search codebook is used for beam search in step 3, the transmitting end uses the beam covering the interested angle interval
Figure BDA0003969916380000041
Transmitting signal, N T Indicating the number of antennas equipped at the transmitting end. The receiving end adopts the cyclic scanning mode to collect the pilot signal transmitted by the transmitting end, and in each cyclic scanning, the receiving end uses the code word w respectively (k,i) And w (k,i+1) And carrying out beamforming and receiving a pilot signal sent by a sending end. Setting the time of acquiring signals by a single wave beam in each cycle scanning of a receiving end as n 0 More than or equal to 1 symbol period, after m times of cyclic scanning, using jth, j ∈ { i, i +1} code word w of kth layer at receiving end (k,j) The signal received by the corresponding beam can be expressed as:
y (k,j) ＝h (k,j) s (k,j) +z (k,j) (1)
wherein
Figure BDA0003969916380000042
Represents an equivalent channel between the transmitting and receiving ends, based on the channel condition>
Figure BDA0003969916380000043
Represents a channel matrix between the sending end and the receiving end, and->
Figure BDA0003969916380000044
Represents the transmitted pilot signal and->
Figure BDA0003969916380000045
P T Indicates the transmitting power>
Figure BDA0003969916380000046
Representing circularly symmetric gaussian noise. The matched filtered signal can be expressed as
Figure BDA0003969916380000047
Wherein
Figure BDA0003969916380000048
Representing circularly symmetric Gaussian noise, the signal-to-noise ratio of the channel can be obtained as->
Figure BDA0003969916380000049
Preferably, in step 3, in order to detect the level of the current channel snr, a channel snr target value β is set in advance, and if the channel snr is greater than the target value, it is determined as a high channel snr, otherwise, it is determined as a low channel snr. The above settings can be written as the following assumptions:
Figure BDA00039699163800000410
however, η in the above formula (k,j) It cannot be calculated directly from the received signal and therefore cannot directly test the above assumptions. The invention uses generalized likelihood ratio test to make hypothesis test, and can obtain test statistic:
Figure BDA0003969916380000051
where γ ∈ (0, ∞) denotes the threshold value of the hypothesis test,
Figure BDA0003969916380000052
and &>
Figure BDA0003969916380000053
Respectively is represented at>
Figure BDA0003969916380000054
And &>
Figure BDA0003969916380000055
Likelihood function under the condition. Through calculation and derivation, theEquation (4) may be rewritten as:
Figure BDA0003969916380000056
in equation (5), the present invention assumes a noise variance σ 2 Is the only known channel information, and r (k,j) 、m、n 0 Can be obtained by measurement, so that the test statistic G (r) can be directly calculated at the receiving end (k,j) ). Therefore, the switching criteria of the beam search codebook are as follows:
Figure BDA0003969916380000057
after each cycle scanning of the receiving end, a channel signal-to-noise ratio detection is carried out. If test statistic G (r) (k,j) ) If the detection result of (2) satisfies the condition of the formula (6), the beam search codebook is switched, and the beam search is performed using the narrow beam search codebook. And if the detection result does not meet the condition of the formula (6), continuing to use the hierarchical beam search codebook to perform beam search.
Preferably, in step 3, when performing the hierarchical search, a generalized likelihood ratio test is used to detect whether the snr of the cumulative received signal satisfies a preset target value. When the current layer is the k-th layer, the signal-to-noise ratio of the received signal is as shown in formula (2) from the output signal of the matched filter
Figure BDA0003969916380000058
Let the preset target value of the signal-to-noise ratio of the accumulated received signal be ρ, then according to the generalized likelihood ratio test, the available test statistic is:
Figure BDA0003969916380000061
after each cyclic beam sweep, if the following condition is satisfied:
Figure BDA0003969916380000062
the beam search of the current level (k-th layer) is stopped, otherwise the beam search of the current level is continued, where γ' is the set likelihood threshold value.
Preferably, when the hierarchical search in step 4 enters a new hierarchical search, if k =1, that is, when the hierarchical search is initially entered, the angular interval to be searched is determined as CV w(1,1) UCV w(1,2) The beam to be searched is w (1,1) And w (1,2) The corresponding beam.
Preferably, when the hierarchical search in step 4 enters the (k + 1) th layer (k > = 1), the matched filtered signal r accumulated by the receiving end when the beam search of the k-th layer is stopped is used (k,i) And r (k,i+1) And determining the beam search angle interval of the (k + 1) th layer. The method specifically comprises the following steps: there are three possibilities for the beam search angle interval of the (k + 1) th layer to be determined, as shown in fig. 3. The specific rule determined is as follows: when the estimated value of the channel main path angle
Figure BDA0003969916380000063
When, CV is selected w(k,i) Searching an angle interval as a beam of the (k + 1) th layer; when the estimate of the channel main path angle->
Figure BDA0003969916380000064
When selecting CV w(k,i+1) Searching an angle interval as a beam of the (k + 1) th layer; when an estimate of the main path angle of the channel &>
Figure BDA0003969916380000065
When selecting CV w(k,*) As a beam search angle interval of the (k + 1) th layer. Wherein theta is (k,i) And theta (k,i+1) Respectively represent angle intervals CV w(k,i) And CV w(k,i+1) Is estimated, the estimate of the channel main path angle->
Figure BDA0003969916380000066
From equation (9):
Figure BDA0003969916380000067
wherein
Figure BDA0003969916380000068
Represents the value range of theta, is selected and is selected>
Figure BDA0003969916380000069
Represents CV w(k,i) And CV w(k,i+1) The angle of the interface. />
Figure BDA0003969916380000071
Wherein
Figure BDA0003969916380000072
Figure BDA0003969916380000073
Indicating the steering vector, g, at the receiving end * (k,i) (θ) represents g (k,i) (ii) the complex conjugate of (theta),
Figure BDA0003969916380000074
representing the beamforming gain at the receiving end. Therefore, the angle interval to be searched for at the k +1 th layer can be summarized as:
Figure BDA0003969916380000075
when in use
Figure BDA0003969916380000076
j e { i, i +1} selects ∈ @>
Figure BDA00039699163800000712
As the beam search angle interval of the (k + 1) th layer, all code word pairs in the intervalThe corresponding beam is the beam to be searched for in the K +1 < K layer. Otherwise it is determined in%>
Figure BDA0003969916380000077
Is the center, width and->
Figure BDA00039699163800000713
And taking the equal angle interval as a beam searching angle interval of the (K + 1) th layer, wherein the beams corresponding to all code words in the interval are the beams to be searched of the (K + 1) th layer less than the K layer.
Preferably, when the step 5 uses the narrow beam search codebook to perform the beam search, the active beam set is initialized first
Figure BDA0003969916380000078
And begins to perform a cyclic beam sweep.
Preferably, step 5 performs a cyclic scan of all beams in the active beam set, each time n beams are transmitted 0 A pilot signal. After the m-th cycle is completed, the beam w i Signals received from the beginning of the first scan
Figure BDA0003969916380000079
Can be expressed as:
y i ＝h i s+z i (11)
wherein the content of the first and second substances,
Figure BDA00039699163800000710
is an equivalent channel between the sending and receiving ends>
Figure BDA00039699163800000711
Is a pilot signal, and
Figure BDA0003969916380000081
wherein P is T For the antenna to transmit power, is>
Figure BDA0003969916380000082
Is a variance of σ 2 Is independent ofA circularly symmetric gaussian variable. w is a i After m times of cyclic scanning of the corresponding beam, the accumulated pilot signal is mn 0 It is possible to obtain:
Figure BDA0003969916380000083
wherein the content of the first and second substances,
Figure BDA0003969916380000084
is the signal after matched filtering, T i Obedience degree of freedom is 2, and non-central parameter is
Figure BDA0003969916380000085
Non-central chi-square distribution. />
Preferably, after the active beams are circularly scanned in step 5 each time, the active beam set is updated according to the posterior probability of pairwise intensity comparison of all the beam intensities. The specific rule is as follows: when f (T) i ,T j ) If the wave beam is more than gamma, the wave beam i is considered to be stronger than the wave beam j, and then the wave beam j is moved out of the active wave beam set; wherein f (T) i ,T j ) Indicating that when the accumulated pilot signal corresponding to beam i and beam j is T i And T j The equivalent channel of beam i has a higher a posteriori probability of f (T) than beam j (see equation 12) i ,T j )＝P r {h i ＞h j And f represents a posterior probability threshold value. The above rule applies to pairwise comparison of all beams in the active beam set, so after each cyclic scan, the updated active beam set can be expressed as:
Figure BDA0003969916380000086
wherein, the posterior probability function f (x, y) can be expressed as follows:
Figure BDA0003969916380000087
wherein L is l (. Cndot.) is a laguerre polynomial of order l. The posterior probability can be calculated approximately by summing finite terms of l according to the analytical expression of the formula (14); or the Monte Carlo method can be adopted to approximate the calculation, and the calculation result is stored in the lookup table to be called in real time.
Preferably, step 5 determines whether the following three stop conditions are met after the active beam set is updated each time. If either condition is met, the narrow beam search is stopped.
Stop condition 1. Only one beam codeword remains in the active beam set:
Figure BDA0003969916380000091
stop condition 2. Two adjacent codewords remain in the active beam set, and the following conditions are satisfied:
Figure BDA0003969916380000092
stop condition 3. Limitation of number of pilot signals:
Figure BDA0003969916380000093
wherein N is max Is the maximum number of pilot signals allowed by the whole beam search process.
Preferably, in step 6, when the hierarchical beam search codebook is used to search the last K-th layer, the obtained beamforming directions are as follows:
Figure BDA0003969916380000094
wherein, w (K,i) Represents any codeword in the last layer K-th layer beam search codebook, e represents a hadamard product,
Figure BDA0003969916380000095
represents->
Figure BDA0003969916380000096
And w (K,i) The difference of the angles in the central direction of the main lobe, d represents the antenna spacing, and λ represents the signal wavelength.
Preferably, in step 6, when the narrow beam search codebook is used for beam search, and when the stop condition 1 or the condition 3 is satisfied, the obtained beam forming direction is:
w opt ＝w i* (19)
wherein
Figure BDA0003969916380000097
Represents the last remaining set of active beams ≦>
Figure BDA0003969916380000098
The codeword corresponding to the strongest beam in (1).
When the stop condition 2 is satisfied, the beam forming direction is obtained as follows:
Figure BDA0003969916380000099
e represents the product of the Hadamard multiplication,
Figure BDA0003969916380000101
represents->
Figure BDA0003969916380000102
And w (K,i) The difference of the angles in the central direction of the main lobe, d represents the antenna spacing, and λ represents the signal wavelength.
Compared with the prior wave beam searching technology, the invention has the following beneficial effects:
the invention creatively designs a beam searching codebook switching mechanism, which can be switched to the most appropriate beam searching codebook in time according to the change of the signal-to-noise ratio of a channel, thereby realizing the self-adaptive beam searching, ensuring the precision of the beam searching and reducing the resource consumption of the beam searching. The technical scheme is suitable for various communication environments, and can ensure that a satisfactory beam searching effect can be achieved no matter the signal-to-noise ratio of a channel is high or low.
Drawings
Fig. 1 is a schematic diagram of a hierarchical beam search codebook structure.
Fig. 2 is a flowchart of an adaptive beam searching method according to the present invention.
Fig. 3 is a schematic diagram of a conventional method for dividing a hierarchical search angle interval.
FIG. 4 is a diagram illustrating a method for dividing a hierarchical search angle interval according to the present invention.
Fig. 5 is a comparison graph of the spectrum efficiency of the adaptive beam search method of the present invention.
Fig. 6 is a comparison diagram of the adaptive beam search method of the present invention over the beam search time.
Detailed Description
The technical content of the invention is described in detail and concretely with reference to the accompanying drawings.
Step 1, a beam searching scene, a used multi-precision layered beam searching codebook and a narrow beam searching codebook.
Consider a single-sided beam search scenario involving a transmitter that transmits a pilot signal using a fixed beam and a receiver that searches for the best beam direction by beam scanning. Make the receiving end equipped with N R ＝2 K A uniform linear array with antennas spaced at half wavelength and having multi-precision hierarchical beam search codebook W and narrow beam search codebook
Figure BDA0003969916380000103
Where K represents the number of layers of the hierarchical beam search codebook and the set of codewords at the K-th layer is represented as W k The ith codeword in the kth layer of the hierarchical beam search codebook is ≧ or>
Figure BDA0003969916380000104
Represents, and +>
Figure BDA0003969916380000105
w (k,i) For covered angular range interval
Figure BDA0003969916380000106
Indicate wherein>
Figure BDA0003969916380000107
∈[-1,+1]Is the result after taking the cosine of the angular range covered. Searching all codewords W in the last layer of a codebook using hierarchical beams k As a narrow beam search codebook, expressed as @>
Figure BDA0003969916380000111
Wherein w i Is a code word and->
Figure BDA0003969916380000112
The transmitting end is provided with N T Half-wavelength uniform linear array of antennas using fixed codewords->
Figure BDA0003969916380000113
The corresponding beam transmits a pilot signal. Assuming that the channel is a block fading model that remains the same during the beam search, the channel matrix is ≦>
Figure BDA0003969916380000114
/>
As shown in fig. 1, the hierarchical beam search codebook structure diagram satisfies the following two criteria:
criterion one is as follows: the coverage area of the codeword of any layer of the hierarchical beam search codebook can be completely represented by the union of the coverage areas of two codewords of the corresponding next layer:
Figure BDA0003969916380000115
the second criterion is as follows: the union of all codeword coverage for any one layer of the hierarchical beam search codebook should completely cover the angular domain of the entire antenna array:
Figure BDA0003969916380000116
and 2, performing the overall process of the adaptive beam search by using the hierarchical beam search codebook and the narrow beam search codebook.
The general flow chart of the adaptive beam search proposed by the present invention is shown in fig. 2. Specifically, when beam scanning is performed by using the hierarchical beam search codebook, the receiving end judges whether a condition for switching to beam search using the narrow beam search codebook is satisfied after completing one cycle of pilot signal acquisition each time. And if the condition of switching to the narrow beam search codebook is met, stopping using the layered beam search codebook to search, and performing beam search by using the narrow beam search codebook until the beam search is stopped. If the condition of switching to the narrow beam search codebook is not met and the current hierarchical search level is not the last level, judging whether to enter the next hierarchical search level; and if the condition for entering the next-level hierarchical search is met, performing the beam search by using the hierarchical beam search codebook of the next level. And if the condition for switching to the narrow beam search codebook is not met and the current hierarchical search level is the last level, judging whether the stopping condition of the beam search is met. And if the conditions are not met, continuously executing the hierarchical search of the current level.
Step 3, when the layered beam search codebook is used for beam search, the generalized likelihood ratio test is adopted to judge whether to switch the beam search codebook and whether to stop the layered search of the current level and enter the next level;
in the hierarchical search method, the transmitting end uses beams covering its angle interval of interest
Figure BDA0003969916380000121
Transmitting signal, N T Indicating the number of antennas equipped at the transmitting end. The receiving end adopts the cyclic scanning mode to collect the pilot signal transmitted by the transmitting end, and in each cyclic scanning, the receiving end uses the code word w respectively (k,i) And w (k,i+1) Performing beamformingAnd receives a pilot signal transmitted from the transmitting end. Setting the time of acquiring signal by single wave beam in each cycle scanning of the receiving end as n 0 More than or equal to 1 symbol period, after m times of cyclic scanning, using jth, j ∈ { i, i +1} code word w of kth layer at receiving end (k,j) The signals received by the corresponding beam can be expressed as:
y (k,j) ＝h (k,j) s (k,j) +z (k,j) (23)
wherein
Figure BDA0003969916380000122
Represents an equivalent channel between the sending end and the receiving end, and->
Figure BDA0003969916380000123
Pilot signal, -comprising all m-cycle scans>
Figure BDA0003969916380000124
Circularly symmetric Gaussian noise containing all m cyclic scans and having variance of sigma 2 . The matched filtered output signal may be expressed as:
Figure BDA0003969916380000125
wherein, P T Which represents the transmission power of the antenna,
Figure BDA0003969916380000126
still with variance σ 2 Is circularly symmetric gaussian noise. The channel signal-to-noise ratio can be expressed as->
Figure BDA0003969916380000127
The switching from the hierarchical beam search codebook to the narrow beam search codebook is determined by detecting the signal-to-noise ratio of the channel.
In order to detect the current channel snr, a target value β of the channel snr is set in advance, and if the channel snr is greater than the target value, it is determined as a high channel snr, otherwise it is determined as a low channel snr. The above settings can be written as the following assumptions:
Figure BDA0003969916380000131
however, η in the above formula (k,j) It cannot be calculated directly from the received signal and therefore cannot directly test the above assumptions. The invention uses generalized likelihood ratio test to make hypothesis test, and can obtain test statistic:
Figure BDA0003969916380000132
where γ ∈ (0, ∞) denotes the threshold value of the hypothesis test,
Figure BDA0003969916380000133
and &>
Figure BDA0003969916380000134
Respectively is represented at>
Figure BDA0003969916380000135
And &>
Figure BDA0003969916380000136
Likelihood function under the condition. By computational derivation, equation (26) can be rewritten as:
Figure BDA0003969916380000137
in equation (27), the present invention assumes a noise variance σ 2 Is the only known channel information, and r (k,j) 、m、n 0 Can be obtained by measurement, so that the test statistic G (r) can be directly calculated at the receiving end (k,j) ). Therefore, the switching criteria of the beam search codebook are as follows:
Figure BDA0003969916380000138
after each cycle scanning of the receiving end, a channel signal-to-noise ratio detection is carried out. If test statistic G (r) (k,j) ) If the detection result of (2) satisfies the formula (28), the beam search codebook is switched and the beam search is performed using the narrow beam search codebook. And if the detection result does not meet the formula (28), continuing to use the hierarchical beam search codebook for beam searching.
When judging whether to stop the layered search of the current level, the signal-to-noise ratio of the accumulated received signals is detected, and whether the accumulated received signals meet a preset signal-to-noise ratio target value is detected.
The matched filter output r is known from the matched filter output signal equation (24) (k,j) Has a signal-to-noise ratio of
Figure BDA0003969916380000141
The matched filtered output r is detected after each cyclic scan of any one beam in the hierarchical search (k,j) Whether the signal-to-noise ratio of->
Figure BDA0003969916380000142
Where ρ is a preset signal-to-noise ratio target value. As long as the detection result of any one beam satisfies->
Figure BDA0003969916380000143
The hierarchical search stops the beam search of the current level, and executes the beam search of the next layer until the last layer of the hierarchical beam search codebook is reached. To detect->
Figure BDA0003969916380000144
If the condition holds, the following assumptions can be made:
Figure BDA0003969916380000145
due to the fact that
Figure BDA0003969916380000146
H in (1) (k,j) Is unknown and therefore cannot be directly tested for simple hypotheses, which are similarly tested using the generalized likelihood ratio test, the test statistics can be written as:
Figure BDA0003969916380000147
in equation (30), the present invention assumes a noise variance σ 2 Is the only known channel information, and r (k,j) Can be obtained by measurement, so that the test statistic can be directly calculated at the receiving end
Figure BDA0003969916380000148
Obtaining a beam search stop condition of a k layer:
Figure BDA0003969916380000149
wherein γ' is a set likelihood threshold value, if formula (31) is satisfied, the beam search of the current level (k-th layer) is stopped, otherwise, the beam search of the current level is continued.
And 4, when the layered search starts a new layer of search, selecting an angle interval to be searched.
When the k-th layer of the hierarchical search satisfies the beam search stop condition, it is necessary to determine the beam search angle interval CV of the (k + 1) -th layer k+1 ，
Figure BDA0003969916380000151
As shown in fig. 3, the conventional method for dividing the layered search angle interval has only two results for the beam search angle interval at the (k + 1) th layer:
Figure BDA0003969916380000152
or->
Figure BDA0003969916380000153
The method for dividing the angle has certain problems in practical application: when the main path angle of the channel is located at the junction of two angle intervals, the error probability of the hierarchical search is higher.
Therefore, the present invention provides a method for dividing a beam search angle interval, as shown in fig. 4, the angle search interval of each layer is divided into three intervals, and the core of the method lies in that when the main path angle of the channel is located at
Figure BDA0003969916380000154
And &>
Figure BDA0003969916380000155
Is selected based on the angle range(s) around the boundary>
Figure BDA0003969916380000156
The beam search angle interval is regarded as the (k + 1) th layer. Therefore, after the beam search for each layer is stopped, the hierarchical search needs to determine the search interval of the next layer by determining which of the three interval centers the main path direction of the channel is closer to. />
To solve this problem, the present embodiment derives the basis for the angle interval determination from the single-path model. Assuming that the millimeter wave single path channel can be represented as
Figure BDA0003969916380000157
Wherein u (theta) and->
Figure BDA0003969916380000158
The steering vectors at the receiving end and the transmitting end are shown, alpha represents unknown path gain, and theta is an unknown main path angle of the channel.
Order to
Figure BDA0003969916380000159
And->
Figure BDA00039699163800001510
Therefore, after the receiving end performs m times of cyclic scanning, the matched filtering output signal is:
Figure BDA00039699163800001511
to derive the channel path angle theta, maximum likelihood detection is used to estimate an angle closest to theta
Figure BDA00039699163800001512
Figure BDA00039699163800001513
Wherein phi k Denotes the value range of theta, p (r) (k,i) ,r (k,i+1) L τ, θ) is a likelihood function. Expressed as:
Figure BDA0003969916380000161
let J (theta, tau) = | r (k,i) -τg (k,i) (θ)| 2 +|r (k,i+1) -τg (k,i+1) (θ)| 2 The above problem (33) can be equivalent to:
Figure BDA0003969916380000162
the minimum value of τ can be derived:
Figure BDA0003969916380000163
wherein g is * (k,i) (theta) represents g (k,i) (iii) a complex conjugate of (θ),
Figure BDA0003969916380000164
j e is equal to the value of i, i +1 represents the beamforming gain at the receiving end, will +>
Figure BDA0003969916380000165
Substituting equation (35) results in the question (35) being equivalent to:
Figure BDA0003969916380000166
the invention takes the value space phi of theta k Constructed as a set of three elements:
Figure BDA0003969916380000167
wherein theta is (k,i) And theta (k,i+1) Respectively indicate the angle interval
Figure BDA0003969916380000168
And &>
Figure BDA0003969916380000169
The angle of the center of (a) is,
Figure BDA00039699163800001610
indicating the angle at the intersection of two angular intervals. Is determined by the formula (37)>
Figure BDA00039699163800001611
The angle interval to be searched of the layer K +1 < K is obtained as follows: />
Figure BDA00039699163800001612
When in use
Figure BDA00039699163800001613
j e { i, i +1} selects ∈ @>
Figure BDA00039699163800001614
As beam search interval of the (k + 1) th layer, otherwise
Figure BDA0003969916380000171
Is the center, width and->
Figure BDA0003969916380000172
The equal angle interval is used as the beam search interval of the (k + 1) th layer.
And 5, when the condition of switching from the layered beam search codebook to the narrow beam search codebook is met, performing beam search by using the narrow beam search codebook. The method comprises the following specific steps:
5.1 Using narrow Beam search codebook for Beam search, the active Beam set is initialized first
Figure BDA0003969916380000173
And begins to perform a cyclic beam sweep.
5.2 cyclically scanning all beams in the active beam set.
Let n send n at a time 0 A pilot signal. After the m-th cycle is completed, the beam w i Signals received from the beginning of the first scan
Figure BDA0003969916380000174
Can be expressed as:
y i ＝h i s+z i (40)
wherein the content of the first and second substances,
Figure BDA0003969916380000175
is an equivalent channel between the sending end and the receiving end, is>
Figure BDA0003969916380000176
Is a pilot signal, and
Figure BDA0003969916380000177
wherein P is T For the antenna to transmit power, < >>
Figure BDA0003969916380000178
Is that the variance is σ 2 Independent circularly symmetric gaussian variables. w is a i After m times of cyclic scanning of the corresponding wave beam, the accumulated pilot signal is mn 0 It is possible to obtain:
Figure BDA0003969916380000179
wherein, the first and the second end of the pipe are connected with each other,
Figure BDA00039699163800001710
is the signal after matched filtering, T i Obedience degree of freedom is 2, and non-central parameter is
Figure BDA00039699163800001711
Non-central chi-square distribution.
And 5.3, after the active beams are scanned circularly each time, updating the active beam set according to the posterior probability of pairwise intensity comparison of all the beams.
The specific rule is as follows: when f (T) i ,T j ) If the wave beam is more than gamma, the wave beam i is considered to be stronger than the wave beam j, and then the wave beam j is moved out of the active wave beam set; wherein f (T) i ,T j ) Indicating that when the accumulated pilot signal corresponding to beam i and beam j is T i And T j The equivalent channel for beam i has a higher a posteriori probability of f (T) than beam j (see equation 41) i ,T j )＝P r {h i ＞h j And f represents a posterior probability threshold value. The above rule applies to pairwise comparison of all beams in the active beam set, so after each cyclic scan, the updated active beam set can be expressed as:
Figure BDA0003969916380000181
wherein, the posterior probability function f (x, y) can be expressed as follows:
Figure BDA0003969916380000182
wherein L is l (. Cndot.) is a laguerre polynomial of order l. The posterior probability can be calculated approximately by summing finite terms of l according to the analytical expression of the formula (13); or the Monte Carlo method can be adopted to approximate the calculation, and the calculation result is stored in the lookup table to be called in real time.
And 5.4, after updating the active beam set each time, judging whether the following three stopping conditions are met. If either condition is satisfied, the narrow beam search stops.
1. Only one beam codeword remains in the active beam set:
Figure BDA0003969916380000183
2. two adjacent codewords remain in the active beam set, and the following conditions are satisfied:
Figure BDA0003969916380000184
3. limitation of the number of pilot signals:
Figure BDA0003969916380000185
wherein N is max Is the maximum number of pilot signals allowed by the whole beam search process.
And 6, when all the beam searching is stopped, selecting the final beam forming direction.
6.1 when the last K layer is searched by using the layered beam search codebook, the next search interval does not need to be determined, but the direction used for beam forming at last needs to be determined, and then
Figure BDA0003969916380000191
Is the direction of beamforming. When/is>
Figure BDA0003969916380000192
j∈{ i, i +1}, the corresponding codeword w in the codebook can be searched directly using the layered beam (K,j) Performing beam forming when
Figure BDA0003969916380000193
In the time, no directly corresponding code word in the hierarchical beam search codebook can realize the beam forming direction->
Figure BDA0003969916380000194
Adjacent codewords need to be rotated by a certain angle, so that the beamforming direction can be changed>
Figure BDA0003969916380000195
Let w opt The code word for final beamforming is represented, and the above rule can be summarized as follows:
Figure BDA0003969916380000196
wherein, w (K,i) Represents any codeword in the last layer K-th layer beam search codebook, e represents a hadamard product,
Figure BDA0003969916380000197
d denotes the antenna spacing and λ denotes the signal wavelength.
6.2, using a narrow beam search codebook to search beams, and determining the final beam forming direction after the stop condition is met.
When beam search is stopped under condition 1 or 3, w is selected i* As the last selected beam codeword, namely:
w opt ＝w i* (48)
using w i* And carrying out beam forming.
When the condition 2 is used as the stop condition, w may be set to i* Moving half the beam width to get a better beam codeword than two codewords in the active beam set, the method of getting the best beam codeword is as follows:
Figure BDA0003969916380000198
e represents the product of the Hadamard-product,
Figure BDA0003969916380000199
d denotes the antenna spacing and λ denotes the signal wavelength. Using w opt And carrying out beam forming.
Simulation result and analysis:
example simulation parameter settings for the adaptive millimeter wave beam search method: the sending end is equipped with a single antenna (N) T = 1), pilot signals are transmitted using omnidirectional transmission, w T And =1. The receiving end is provided with 32 uniform linear antenna arrays (N) R = 32), the beam search codebook used is a DFT layered beam search codebook, the number of layers being K = log 2 32=5. And all code words of the last layer of the hierarchical beam search codebook are taken as a narrow beam search codebook. Setting a target value beta =0 of a signal-to-noise ratio of a channel, assuming that a check threshold gamma =4, and the number n of pilot signals transmitted by a single beam in one scanning 0 And =1. When a layered beam search codebook is used for beam search, setting a target value rho =4 of the signal-to-noise ratio of a received signal, and assuming that a detection threshold gamma' =2; when the beam search is performed using the narrow beam search codebook, the posterior probability threshold is set to Γ =0.95. The maximum number N of pilot signals allowed in the whole beam searching process max ＝2000。
Fig. 5 and 6 show the comparison between the spectrum efficiency and the Beam Search time of the adaptive Beam Search algorithm proposed by the present invention and three existing methods in LOS channel, including an adaptive Beam hierarchical Search algorithm (AHS, see the following patents: liuchun, li song, zhao, an adaptive Millimeter Wave Beam hierarchical Search method [ P ]. Zhejiang province: CN113225116B, 2022-05-31.), an adaptive narrow Beam Search algorithm (IDBS, see the following papers: c.liu, m.li, l.zhao, p.whiting, s.v. hand and i.b.collins, "Millimeter-Wave Beam Search With actuation and Beam Shifting," IEEE Transactions on Communications, vol.19, wir.518, pp.5117-5117.12, and conventional layered Search algorithms (tw2020.12/2020.12).
As can be seen from fig. 5 and fig. 6, in an LOS channel, the adaptive beam search algorithm provided by the present invention can achieve good spectral efficiency under all channel signal-to-noise ratio conditions, and can approach or even achieve the optimal spectral efficiency. In the search time, the beam search time can be adaptively adjusted according to the change of the signal-to-noise ratio of the channel, and the method has good adaptivity, and is specifically represented as follows: under the condition of low channel signal-to-noise ratio, the beam searching time is relatively long; under the condition of high channel signal-to-noise ratio, the beam searching time is shortened.
As can be seen from fig. 5 and 6, in LOS channel, compared with the beam search algorithm IDBS using only narrow beam search codebook, the adaptive beam search algorithm proposed in the present invention has close spectrum efficiency and the same beam search time under the condition of low channel snr, but the spectrum efficiency is much shorter than the beam search time of IDBS algorithm in the beam search time, which reduces resource consumption.
It can be seen from the above that, in the LOS channel, compared with the beam search algorithm AHS using only the hierarchical beam search codebook, the adaptive beam search algorithm proposed by the present invention has a slightly higher spectral efficiency than the AHS algorithm under the condition of high channel signal-to-noise ratio, and the beam search time is almost the same, and has a slightly higher spectral efficiency than the AHS algorithm under the condition of low channel signal-to-noise ratio, but the beam search time is much shorter than the AHS algorithm, which greatly reduces the resource consumption. Compared with the traditional hierarchical search algorithm, the self-adaptive beam search algorithm provided by the invention is far superior to the traditional hierarchical search algorithm in both spectrum efficiency and beam search time.
In summary, the adaptive millimeter wave beam searching method provided by the invention uses two codebooks, namely the multi-precision layered beam searching codebook and the narrow beam searching codebook to perform beam searching, can switch the beam searching codebook in real time according to the change of the signal-to-noise ratio of a channel, can ensure the effect of beam searching, can reduce the time of beam searching, and has good adaptivity. Compared with the beam searching method using a single beam searching codebook, the method provided by the invention can be suitable for more various communication scenes, and can use the most appropriate beam searching codebook to search beams no matter in low-channel signal-to-noise ratio or high-channel signal-to-noise ratio scenes, thereby achieving the satisfactory beam searching effect.

Claims (7)

1. A self-adaptive millimeter wave beam searching method is characterized by comprising the following steps:
step 1, a beam searching scene, a preset multi-precision layered beam searching codebook and a narrow beam searching codebook;
step 2, using the hierarchical beam search codebook and the narrow beam search codebook to execute a total flow of the adaptive beam search;
step 3, when the layered beam search codebook is used for beam search, the generalized likelihood ratio test is adopted to judge whether to switch the beam search codebook and whether to stop the layered search of the current level and enter the next level;
step 4, when the layered search starts a new layer of search, selecting an angle interval to be searched;
step 5, when the condition of switching from the layered beam search codebook to the narrow beam search codebook is met, using the narrow beam search codebook to perform beam search;
and 6, when all the beam searching is stopped, selecting the final beam forming direction.
2. The adaptive millimeter wave beam searching method of claim 1, wherein:
the beam searching scene in step 1 comprises a transmitter and a receiver, wherein the transmitter continuously transmits pilot signals to the whole angle space covered by the transmitter, and the receiver uses a preset beam searching codebook to scan beams so as to search the optimal beam forming direction;
step 1, the wave beam searching codebook is divided into two types, one type is a layered wave beam searching codebook W, and the other type is a narrow wave beam searching codebook
Figure FDA0003969916370000011
The hierarchical beam search codebook in the step 1 is in a tree structure, that is, the coverage area of a beam with a wider upper layer is jointly covered by a plurality of narrower beams in the next layer; the hierarchical beam search codebook used has a total of K layers, where the K layers have a total of 2 k A codeword, the set of codewords for the k-th layer being denoted W k And is and
Figure FDA0003969916370000012
denotes the ith codeword, N, in the kth layer R ＝2 K Indicating the number of antennas equipped in the receiver; code word w (k,i) The covered angle range is used>
Figure FDA0003969916370000013
Indicate wherein>
Figure FDA0003969916370000014
Is the result after taking the cosine of the coverage angle; in the hierarchical beam search codebook, the union of all codeword coverage ranges of any layer should completely cover the whole antenna array angle domain; the coverage range of the code word of any layer can be completely represented by the union of the coverage ranges of two corresponding code words of the next layer;
the narrow beam search codebook in the step 1 is single-layer, and the narrow beams jointly cover the whole angle space; searching all codewords W in the last layer of a codebook using hierarchical beams K As a narrow beam search codebook, expressed as
Figure FDA0003969916370000021
3. The adaptive millimeter wave beam searching method of claim 2, wherein:
the beam search in step 2 starts from the hierarchical beam search codebook to perform beam scanning; and the total flow of beam search is specifically realized as follows:
when the layered beam searching codebook is used for carrying out beam scanning, after a receiver finishes one-cycle pilot signal acquisition each time, whether the condition of switching to beam searching by using a narrow beam searching codebook is met is judged; if the condition of switching to the narrow beam search codebook is met, stopping using the layered beam search codebook to perform beam search, and turning to using the narrow beam search codebook to perform beam search until the beam search is stopped; if the condition of switching to the narrow beam search codebook is not met and the current hierarchical search level is not the last level, judging whether to enter the next hierarchical search level; if the condition of entering the next-layer hierarchical search is met, using a next-layer beam search codebook to execute beam search; if the condition for switching to the narrow beam search codebook is not met and the current hierarchical search level is the last level, judging whether the stopping condition of beam search is met; if the conditions are not met, continuing to execute the hierarchical search of the current level;
the beam searching process selects a final beam forming direction when the beam searching stopping condition is met;
the layered search receiver finishes one cycle of pilot signal acquisition, namely the receiver sequentially uses all beams to be selected of the current level to acquire the pilot signal transmitted by the receiver once;
the switching from the layered beam search codebook to the narrow beam search codebook is determined by detecting the signal-to-noise ratio of the unknown channel.
4. The adaptive millimeter wave beam searching method of claim 3, wherein:
when the hierarchical beam searching codebook is used for beam searching in the step 3, the transmitter uses the beam covering the interested angle interval
Figure FDA0003969916370000022
Transmitting signal, N T Indicating the number of antennas equipped to the transmitter; the receiver adopts a cyclic scanning mode to acquire the transmitterTransmitted pilot signal, the receiver using the code word w in each cyclic scan (k,i) And w (k,i+1) Carrying out wave beam shaping and receiving a pilot signal sent by a transmitter; setting the time for a receiver to acquire a signal by a single wave beam in each cyclic scanning as n 0 More than or equal to 1 symbol period, after m times of cyclic scanning, using jth, j epsilon { i, i +1} code word w in kth layer at receiver (k,j) The signals received by the corresponding beams are represented as:
y (k,j) ＝h (k,j) s (k,j) +z (k,j) (1) (1)
wherein
Figure FDA0003969916370000031
Represents an equivalent channel between the transmitter and receiver, and->
Figure FDA0003969916370000032
Representing a channel matrix between the transmitter and the receiver,
Figure FDA0003969916370000033
represents a transmitted pilot signal and
Figure FDA0003969916370000034
P T indicates the transmitting power>
Figure FDA0003969916370000035
Representing circularly symmetric gaussian noise; the matched filtered signal is represented as:
Figure FDA0003969916370000036
wherein, P T Which represents the transmission power of the antenna,
Figure FDA0003969916370000037
representing circularly symmetric Gaussian noiseAcoustic, obtaining a signal-to-noise ratio of the channel of
Figure FDA0003969916370000038
In step 3, in order to detect the signal-to-noise ratio of the current channel, a target value beta of the signal-to-noise ratio of the channel is set in advance, and the signal-to-noise ratio of the channel is set to be a high signal-to-noise ratio of the channel when the signal-to-noise ratio of the channel is larger than the target value, otherwise, the signal-to-noise ratio of the channel is set to be a low signal-to-noise ratio of the channel; the above settings can be written as the following assumptions:
Figure FDA0003969916370000039
however, η in the formula (k,j) It cannot be calculated directly from the received signal and therefore cannot be checked directly; using the generalized likelihood ratio test for hypothesis testing, test statistics can be obtained:
Figure FDA00039699163700000310
where γ ∈ (0, ∞) denotes the threshold value of the hypothesis test,
Figure FDA00039699163700000311
and &>
Figure FDA00039699163700000312
Respectively is represented at>
Figure FDA00039699163700000313
And &>
Figure FDA00039699163700000314
A likelihood function under the condition; through calculation derivation, equation (4) is rewritten as: />
Figure FDA0003969916370000041
In equation (5), the noise variance σ is assumed 2 Is the only known channel information, and r (k,j) 、m、n 0 Can be obtained by measurement, and thus the test statistic G (r) can be directly calculated at the receiver (k,j) )；
Therefore, the switching criteria of the beam search codebook are as follows:
Figure FDA0003969916370000042
after each cycle scanning of the receiver, one-time channel signal-to-noise ratio detection is carried out; if test statistic G (r) (k,j) ) If the detection result of (2) meets the condition of the formula (6), switching the beam search codebook, and performing beam search by using the narrow beam search codebook; if the detection result does not meet the condition of the formula (6), continuing to use the layered beam search codebook to perform beam search;
step 3, when the layered search is executed, detecting whether the signal-to-noise ratio of the accumulated received signals meets a preset target value by using generalized likelihood ratio test; when the current layer is the k-th layer, the signal-to-noise ratio of the received signal is as shown in formula (2) from the output signal of the matched filter
Figure FDA0003969916370000043
Let the preset target value of the signal-to-noise ratio of the accumulated received signal be ρ, then according to the generalized likelihood ratio test, the available test statistic is:
Figure FDA0003969916370000044
after each cyclic beam sweep, if the following condition is satisfied:
Figure FDA0003969916370000045
stopping the beam search of the current k level, otherwise continuing the beam search of the current level, wherein gamma' is the set likelihood threshold value.
5. The adaptive millimeter wave beam searching method of claim 4, wherein:
when the hierarchical search enters a new level search, if k =1, namely when the hierarchical search is initially entered, the angle interval to be searched is determined as
Figure FDA0003969916370000051
The beam to be searched is w (1,1) And w (1,2) A corresponding beam;
when the layered search enters a k +1 th layer and k is more than or equal to 1, searching matched filtering signals r accumulated by the receiver when the wave beam search of the k layer stops (k,i) And r (k,i+1) Determining a beam search angle interval of the (k + 1) th layer; the method specifically comprises the following steps: there are three possible beam search angle intervals of the (k + 1) th layer to be determined, and the specific rule of the determination is as follows: when the estimated value of the channel main path angle
Figure FDA0003969916370000052
When, CV is selected w(k,i) Searching an angle interval as a beam of the (k + 1) th layer; when the estimated value of the channel main path angle
Figure FDA0003969916370000053
When, CV is selected w(k,i+1) Searching an angle interval as a beam of the (k + 1) th layer; when the estimate of the channel main path angle->
Figure FDA0003969916370000054
When, CV is selected w(k,*) Searching an angle interval as a beam of the (k + 1) th layer; wherein, theta (k,i) And theta (k,i+1) Respectively represents an angle interval->
Figure FDA0003969916370000055
And &>
Figure FDA0003969916370000056
Is estimated, the estimate of the channel main path angle->
Figure FDA0003969916370000057
From equation (9):
Figure FDA0003969916370000058
wherein the content of the first and second substances,
Figure FDA0003969916370000059
represents the value range of theta, is selected and is selected>
Figure FDA00039699163700000510
Represents->
Figure FDA00039699163700000511
And &>
Figure FDA00039699163700000512
The angle of the interface; />
Figure FDA00039699163700000513
Wherein
Figure FDA00039699163700000514
Figure FDA00039699163700000515
Indicating the steering vector, g, at the receiver * (k,i) (theta) represents g (k,i) (ii) the complex conjugate of (theta),
Figure FDA00039699163700000516
represents the beamforming gain of the receiver;
therefore, the total of the angle intervals to be searched of the (k + 1) th layer is:
Figure FDA0003969916370000061
when in use
Figure FDA0003969916370000062
Select it>
Figure FDA0003969916370000063
As the wave beam searching angle interval of the (K + 1) th layer, the wave beams corresponding to all code words in the interval are the wave beams to be searched of the (K + 1) th layer; otherwise it is determined whether or not the signal is positive>
Figure FDA0003969916370000064
Is the center, width and->
Figure FDA0003969916370000065
And taking the equal angle interval as a beam searching angle interval of the (K + 1) th layer, wherein the beams corresponding to all code words in the interval are the beams to be searched of the (K + 1) th layer less than the K layer.
6. The adaptive millimeter wave beam searching method of claim 5, wherein:
step 5, when using narrow beam search codebook to search beams, firstly initializing active beam set
Figure FDA0003969916370000066
And begin to perform a cyclic beam sweep;
step 5, circularly scanning all beams in the active beam set, and transmitting n beams each time 0 A pilot signal; after the m-th cycle is completed, the beam w i Signals received from the beginning of the first scan
Figure FDA0003969916370000067
Can be expressed as:
y i ＝h i s+z i (11)
wherein the content of the first and second substances,
Figure FDA0003969916370000068
is an equivalent channel between the transmitter and the receiver, is>
Figure FDA0003969916370000069
Is a pilot signal, and
Figure FDA00039699163700000610
wherein P is T For the antenna to transmit power, is>
Figure FDA00039699163700000611
Is a variance of σ 2 Independent circularly symmetric gaussian variables of (1); w is a i After m times of cyclic scanning of the corresponding beam, the accumulated pilot signal is mn 0 It is possible to obtain:
Figure FDA00039699163700000612
wherein the content of the first and second substances,
Figure FDA00039699163700000613
is the signal after matched filtering, T i Obedience degree of freedom is 2, and non-central parameter is
Figure FDA0003969916370000071
Non-central chi-square distribution of (c);
step 5, after the active beams are circularly scanned each time, updating the active beam set according to the posterior probability of pairwise intensity comparison of all the beam intensities; the specific rule is as follows: when f (T) i ,T j ) If t, then beam i is considered stronger than beam jThen, the beam j is shifted out of the active beam set; wherein f (T) i ,T j ) Indicates that when the accumulated pilot signal corresponding to the beam i and the beam j is T i And T j The intensity of the equivalent channel of beam i is higher than that of beam j, i.e., f (T) i ,T j )＝P r {|h i |＞|h j L, wherein Γ represents a posterior probability threshold value; the above rule applies to pairwise comparison of all beams in the active beam set, so after each cyclic scan, the updated active beam set can be expressed as:
Figure FDA0003969916370000072
wherein, the posterior probability function f (x, y) can be expressed as follows:
Figure FDA0003969916370000073
wherein L is l (. H) is a laguerre polynomial of order l; the posterior probability can be calculated approximately by summing the finite terms of l according to the analytical expression of the formula (14); or approximate calculation can be carried out by adopting a Monte Carlo method, and the calculation result is stored in a lookup table to be called in real time;
step 5, after the active beam set is updated each time, judging whether the following three stop conditions are met; if any one condition is met, the narrow beam search is stopped;
stop condition 1. Only one beam codeword remains in the active beam set:
Figure FDA0003969916370000074
stop condition 2. Two adjacent codewords remain in the active beam set, and the following conditions are satisfied:
Figure FDA0003969916370000075
stop condition 3. Limitation of number of pilot signals:
Figure FDA0003969916370000081
wherein N is max Is the maximum number of pilot signals allowed by the whole beam search process.
7. The adaptive millimeter wave beam searching method of claim 6, wherein:
step 6, when the last K layer is searched by using the layered beam search codebook, the obtained beam forming directions are as follows:
Figure FDA0003969916370000082
wherein, w (K,i) Represents any codeword in the last layer K-th layer beam search codebook, e represents a hadamard product,
Figure FDA0003969916370000083
represents->
Figure FDA0003969916370000084
And w (K,i) The angle difference of the center direction of the main lobe, d represents the antenna spacing, and lambda represents the signal wavelength;
step 6, when the narrow beam search codebook is used for beam search, and when the stop condition 1 or the stop condition 3 is satisfied, the beam forming direction is obtained as follows:
w opt ＝w i* (19)
wherein
Figure FDA0003969916370000085
Represents the last remaining set of active beams ≦>
Figure FDA0003969916370000086
The codeword corresponding to the strongest beam in (1);
when the stop condition 2 is satisfied, the obtained beam forming direction is:
Figure FDA0003969916370000087
e represents the product of the Hadamard-product,
Figure FDA0003969916370000088
represents->
Figure FDA0003969916370000089
And w (K,i) The difference of the angles in the central direction of the main lobe, d represents the antenna spacing, and λ represents the signal wavelength. />
CN202211513538.3A 2022-11-29 2022-11-29 Self-adaptive millimeter wave beam searching method Pending CN115882912A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211513538.3A CN115882912A (en) 2022-11-29 2022-11-29 Self-adaptive millimeter wave beam searching method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211513538.3A CN115882912A (en) 2022-11-29 2022-11-29 Self-adaptive millimeter wave beam searching method

Publications (1)

Publication Number Publication Date
CN115882912A true CN115882912A (en) 2023-03-31

Family

ID=85764684

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211513538.3A Pending CN115882912A (en) 2022-11-29 2022-11-29 Self-adaptive millimeter wave beam searching method

Country Status (1)

Country Link
CN (1) CN115882912A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116701882A (en) * 2023-07-26 2023-09-05 北京航空航天大学 Self-adaptive multi-beam alignment method based on question-answer learning
CN117134811A (en) * 2023-10-26 2023-11-28 北京航空航天大学 Safe wave beam alignment algorithm based on privacy question-answer learning

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116701882A (en) * 2023-07-26 2023-09-05 北京航空航天大学 Self-adaptive multi-beam alignment method based on question-answer learning
CN116701882B (en) * 2023-07-26 2023-10-20 北京航空航天大学 Self-adaptive multi-beam alignment method based on question-answer learning
CN117134811A (en) * 2023-10-26 2023-11-28 北京航空航天大学 Safe wave beam alignment algorithm based on privacy question-answer learning

Similar Documents

Publication Publication Date Title
CN115882912A (en) Self-adaptive millimeter wave beam searching method
CN111954228B (en) Beam management method, beam management device, electronic equipment and computer readable storage medium
CN106027133B (en) Hierarchical beam searching method under multipath channel
CN112564752A (en) Intelligent surface auxiliary communication method for optimizing activation and reconfiguration of sparse antenna
CN110109051A (en) The array with mutual coupling DOA estimation method of battle array is controlled based on frequency
KR20090092696A (en) Method for generating soft output for received signal in mimo receiver and transceiver including transmitter and receiver
CN110045322B (en) Short wave automatic direction finding method based on high-resolution direction finding graph intelligent identification
CN108631851B (en) Self-adaptive beam forming method based on uniform linear array null deepening
CN107566305A (en) A kind of millimeter-wave systems channel estimation methods of low complex degree
US20120087450A1 (en) Wireless communication node connections
CN114844545A (en) Communication beam selection method based on sub6GHz channel and partial millimeter wave pilot frequency
CN109347532A (en) Improved GOMP detection algorithm in generalized spatial modulation system
CN111431566B (en) Multi-path optical signal fast search method applied to atmosphere laser communication spatial modulation system
CN113225116B (en) Self-adaptive millimeter wave beam layered search method
CN111273215A (en) Channel inconsistency error correction direction finding method of channel state information
CN111132007A (en) NB terminal high-precision positioning algorithm based on 5G/NB-IoT cluster node information fusion
CN114567525B (en) Channel estimation method and device
CN105301557A (en) Direction-of-arrival estimate configuration method
WO1999033141A1 (en) Discrimination procedure of a wanted signal from a plurality of cochannel interfering signals and receiver using this procedure
Marandi et al. Beam selection based on sequential competition
Wu et al. Joint channel estimation for RIS-assisted wireless communication system
Wang et al. DOA estimation of smart antenna signal based on MUSIC algorithm
Abohamra et al. Direction of Arrival algorithms for user identification in cellular networks
KR101524587B1 (en) System for search cell
Ando et al. A proposal of an end-to-end DoA estimation system aided by deep learning

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination