Smart Antenna, formerly known as Adaptive Array (AAA), was originally applied to radar, sonar, and military applications. It is mainly used for spatial filtering and positioning. The familiar phased array radar is a relatively simple adaptive. Antenna array. Mobile communication researchers have given the more attractive name to adaptive antenna arrays for mobile communications: smart antennas, known in English as smart antennas or intelligent antennas. ——As the name implies, the adaptive antenna array consists of multiple antenna elements, each of which is followed by a weighting device (that is, multiplied by a certain coefficient, which is usually a complex number, which adjusts the amplitude and adjusts the phase, and in the phased array radar Only the phase is adjustable), and finally merged with the adder. The smart antenna of this structure can only perform spatial domain processing, and the smart antenna with airspace and time domain processing capability is relatively complicated in structure. Each antenna is followed by a delay tap weighting network (structural and time domain FIR). The equalizer is the same). The main meaning of adaptive or intelligent means that these weighting coefficients can be appropriately changed and adaptively adjusted. The above is actually the structure when the smart antenna is used as the receiving antenna. When it is used for transmitting, the structure changes slightly. The weighting or weighting network is placed before the antenna, and there is no adding combiner. ——Assume that the narrowband condition of the antenna transmission is satisfied, that is, the response output of an incident signal in each antenna unit has only a phase difference without amplitude change, and these phase differences are determined by the difference in length of the path taken by the incident signal to each antenna. If the incident signal is a plane wave (only one incident direction), these phase differences are uniquely determined by the carrier wavelength, angle of incidence, and antenna position distribution. Given a set of weighting values, a certain incident signal strength, the signal of different incident angles will have different output signal strengths after the combiner due to the difference in phase difference between the antennas. ——With the incident angle as the abscissa, the corresponding smart antenna output gain (dB) is plotted as the ordinate (antenna term), and the smart antenna is different from the omni- antenna (ideal) The time is a straight line), and the direction of the antenna is closer to the direction (direcTIonal), that is, there are main lobe, side lobe, etc., but smart antennas usually have a narrow main lobe. More flexible primary and secondary lobes size, positional relationship, and larger antenna gain (antenna terminology, an important indicator of the antenna is the ratio of the gain in the strongest direction to the average gain in each direction), and the maximum of the fixed antenna The difference is that different weights usually correspond to different patterns. We can change the weight to select the appropriate pattern, ie the antenna pattern. - Let's explain what is the appropriate direction. In order to maximize the useful signal and suppress the interference signal, the most intuitive thing is that we can align the main lobe with the incident direction of the wanted signal and the lowest gain point in the pattern. (called a zero trap) aligns with the direction of the interfering signal. Of course, this is only an ideal situation. The actual wireless communication environment is very complicated. There are many interference signals, multipath transmission, and the number of antenna elements is not much (limited degrees of freedom). The difference between the useful signal and the interference signal may be in the incident direction. Not being able to make the previous solution is not feasible, but the pursuit of the maximum signal to noise ratio SINR is still the ultimate goal. The actual working principle of the smart antenna is more complicated than that described above, especially when the space and time joint processing is performed. At this time, it is better to explain from the perspective of signal processing, especially adaptive filtering. Since this requires a strong theoretical, Professional background knowledge, not introduced here. ——The mobile communication channel transmission environment is harsh, multi-path fading, inter-symbol interference (ISI) caused by delay spread, and FDMA TDMA system (such as GSM) co-channel interference introduced by frequency reuse (CCI, Co -Channel Interference), MAI (MulTIple Access Interference) in CDMA system, etc., all of which reduce link performance and system capacity. Our well-known equalization, code matching filtering, RAKE reception, channel coding and decoding techniques are all to counter or reduce Their impact. These techniques actually use time and frequency domain information, but in fact the useful signal, its delay version and the interference signal are different in time and frequency domain, in the airspace (incident angle DOA, DirecTIon Of Arrival There are also differences. The diversity antenna, especially the sector antenna, can be regarded as the initial utilization of this part of the resource, and the smart antenna technology should be used more fully. ——Smart antenna is a technology with better shrinkage. In the early days of mobile communication development, operators have always wanted to cover as large a region as possible with as few base stations as possible to save investment, which means that the user's signal may have experienced before reaching the BTS (base transceiver station). Long propagation path, with large path loss, so that the received useful signal does not fall below the threshold, or increase the transmit power of the mobile station, or increase the receiving gain of the base station antenna, due to the mobile station (especially The transmission power of a mobile phone is usually limited. It is really feasible to increase the antenna gain. Relatively speaking, it is easier to use a smart antenna to achieve a larger gain than to use a single antenna. ——In the middle and late stages of mobile communication development, in order to expand system capacity and support more users, it is necessary to shrink the cell range and reduce the frequency reuse coefficient to improve the frequency utilization. The commonly used method is cell splitting and sectorization. What comes up is that the interference increases, and the CCI and MAI that were originally reduced by the distance (in fact, by means of path loss) are increased by a large proportion. However, with the smart antenna, by using the difference of the incident angle between the useful signal and the interference signal, the appropriate combining weight is selected to form the correct antenna receiving mode, that is, the main lobe is aligned with the useful signal, and the low gain side lobe is aligned with the main interfering signal. Thus, interference can be suppressed more effectively, the frequency reuse factor can be reduced by a larger proportion (such as making multiplexing factor 3 possible in GSM), and more users (in CDMA) can be simultaneously supported. From a certain perspective, we can regard the smart antenna as a more flexible fan antenna with a narrower main lobe. —— Another advantage of smart antennas is that multipath effects can be reduced. In CDMA, RAKE receivers can separate and coherently combine multipaths with delay differences greater than one chip, while smart antennas can be inseparable for delays. The angularly separable multipath is further separated to more effectively reduce the multipath effect. The main purpose of using smart antenna technology is to more effectively combat the mobile communication channel, and the channel transmission environment of the time division and code division multiple access system is essentially the same, so in addition to the specific algorithm differences, the smart antenna Can be widely used in a variety of time division, code division multiple access systems, including the second generation system that has been commercialized, that is, a widely applicable system. -- Another possible use of smart antennas is to make emergency call location and provide higher positioning accuracy, as it also obtains angle of arrival information while obtaining information such as delay, strength, etc. that can be used for positioning. ——The research content of smart antenna can be divided according to its role in mobile communication. The limitation of volume and power of mobile station (especially mobile phone) makes smart antenna difficult to implement on mobile station (one exception is WLL wireless local Ring system), so the main research is at the base station side of the smart antenna receiving and transmitting, that is, uplink and downlink. ——It is relatively difficult to realize the downlink of the smart antenna. This is because it is difficult for the smart antenna to accurately know the characteristic information of the downlink channel (especially the exit angle of the main propagation path) when designing the transmitting beamforming. The antenna operating mode should be matched to the channel. One method is to make a closed-loop test structure like the IS-95 uplink power control, but it has the following disadvantages: waste of valuable system resources, additional delay, interference by uplink channels, and the like. Another method is to use the uplink channel information to estimate the downlink channel. This is obviously feasible in the TDD (Time Division Duplex) system. This is also the third generation proposal (TDD method) submitted by China for TD-SCDMA. main reason. However, this is not the case in FDD (Frequency Division Duplex) systems, because the uplink and downlink channels use different frequencies (the third generation system has larger uplink and downlink frequency differences than the second generation), and the uplink and downlink channels. The correlation is very weak, and many parameters are not the same. At present, most researchers believe that the incident and exit angles of the main propagation paths of the uplink and downlink channels are basically the same, so we can only obtain part of the information of the downlink channel. The hair beam will never be optimal. ——The downlink channel includes a control channel and a traffic channel. Since the control channel is shared by everyone, a wide beam should be formed, and the traffic channel corresponding to each user is applied with narrow beam transmission, that is, they have different weighting coefficients, so that the control channel (such as the pilot channel) and the traffic channel actually experience different transmission environments, there will be different fading, and the mobile station usually uses the pilot channel to estimate the amplitude and phase distortion of the channel when performing downlink reception, to perform on the traffic channel. Coherent reception, but based on the fact that the two channels have the same transmission environment, it is obvious that the former does not satisfy this condition, and the non-coherent reception has a large signal-to-noise ratio loss for coherent reception. Some suggestions (such as cdma-2000) have taken this into consideration. The downlink channel also has an auxiliary pilot channel, which can also be transmitted as a narrow beam. However, due to the limited number, it is more feasible to assign it to A group of users (the beam formed at this time should also be aligned to this group of users, which may occur in hotspots and intelligent sectorization based on the number of active users) or a user with higher link quality (such as When transferring high speed data to him). —— Another difficulty in implementing downlink with smart antennas is that the weighting is performed at the front end of the antenna (in practice, it is mostly implemented at baseband or intermediate frequency, because it is easier and more flexible), the filter of the latter stage, D/A digital mode The changes in the characteristics of the converter, the mixer, and the antenna elements (each way) necessarily change the resulting beam, and it is impossible or very difficult to adjust the weighting coefficients with common feedback methods to offset the change. A feasible but not very good method is to periodically test and adjust the characteristics of the latter stage. ——Because the current smart antenna technology is not very mature, the various post-selection schemes of the third generation mobile communication only use smart antennas as optional technologies except for China's TD-SCDMA, and have not been written into specific recommendations, the second generation. The system also generally does not adopt smart antenna technology. When the smart antenna is used for uplink and downlink, it does not propose new requirements for the mobile station. It is easy to use it as an upgraded version of the omnidirectional antenna and the fan antenna for the existing base station system. However, when the smart antenna is used for downlink transmission, it usually puts new requirements on the mobile station's reception, which involves large area and less flexibility. ——The current mobile communication system (mainly narrowband CDMA system) has the phenomenon that the downlink capacity exceeds the uplink. Even if the loss of soft handover is considered, it is urgent to improve the uplink capacity from the surface, but the high-speed data in the third generation system. Multimedia services are more likely to appear in the downlink channel. Considering this asymmetric demand, the bottleneck in the future may be downlink. Therefore, despite the above difficulties, it is still necessary and urgent to study the downlink of smart antennas. —— The difference between downlink and uplink processing in TDD mode is not significant, and is not discussed separately here. The uplink receiving technology of smart antennas is relatively mature. The earliest introduction of mobile antennas by adaptive antenna arrays is also to improve the quality and capacity of the uplink channels. There are two main ways for smart antenna uplink and reception: full adaptive mode and pre-multi-beam based beam switching. Theoretical workers are more interested in the former, and engineers and technicians prefer the latter. In the adaptive mode, the weights corresponding to the spatial or spatial and time domain processing can be arbitrarily adjusted according to a certain adaptive algorithm to perform the maximum possible matching on the current transmission environment, and the corresponding smart antenna receiving beam can be arbitrarily pointed. of. In the switching beam, the weights can only be selected from pre-calculated sets of values. The smart antenna working mode at a certain time can only be selected from several pre-designed beams, not arbitrarily pointed, so it is only possible Partial matching of the current transmission environment is theoretically not optimal. ——The core of the research of fully adaptive smart antenna is adaptive algorithm. At present, many well-known algorithms have been proposed. Generally speaking, there are two categories: non-blind algorithm and blind algorithm. The non-blind algorithm refers to an algorithm that needs to rely on a reference signal (pilot sequence or pilot channel). At this time, the receiving end knows what is sent, and when the algorithm is processed, the channel response is determined first and then according to certain criteria (such as optimal forced Zero forcing) Determine each weighted value, or directly determine or gradually adjust the perceptual value according to certain criteria, so that the smart antenna output is most correlated with the known input. The commonly used correlation criteria are MMSE (Minimum Mean Square Error), LMS ( Minimum mean square) and LS (least squares), etc. The blind algorithm does not need to send the known pilot signal at the origin, and the decision feedback algorithm (Decision Feedback) is a special type of blind algorithm. The receiver estimates the transmitted signal and uses this as the reference signal for the above processing, but it should be noted. It should be ensured that there is a small error between the decision signal and the actual transmitted signal. Blind algorithms generally use some features inherent in the modulated signal that are independent of the specific information bits, such as constant modulus CM, subspace Subspace, finite symbol set Finite Alphabet, cyclostationary Cycle-stationary, etc., and adjust the weight to make the output To satisfy this characteristic, various gradient-based algorithms using different amounts of constraints are common. Compared with the blind algorithm, the non-blind algorithm usually has smaller error and faster convergence speed, but it wastes certain system resources. There is a semi-blind algorithm that combines the two, that is, the initial weight is determined by the non-blind algorithm. The blind algorithm is used for tracking and adjustment. On the one hand, the advantages of the two can be combined. On the one hand, it is consistent with the actual communication system, because usually the pilot symbols are not transmitted from time to time but are time-divided with the corresponding traffic channels. use. —— Although the fully adaptive smart antenna can theoretically achieve optimality, relatively speaking, various algorithms have the required amount of data, large amount of calculation, simple channel model, slow convergence rate, and in some cases even There may be disadvantages such as error convergence. When there are many interferences and serious multipaths under actual channel conditions, especially when the channel is fast time-varying, it is difficult to track a user in real time. It is in this context that a pre-multi-beam based switched beam mode of operation is proposed. At this time, the entire airspace (various possible incident angles) is covered by some pre-computed beam splitting, and the beams corresponding to the weights of the groups have different main lobe pointing, and there is usually some overlap between the main lobes of adjacent beams. The main task of the time is to pick one (and possibly several, but need to merge and then output) as the working mode, which is obviously easier to implement than the adaptive method. In fact, we can think of it as a sector antenna. A technique between fully adaptive antennas. Among the beam-switched antennas, the following are worth studying: how to divide the airspace, that is, to determine the beam problem, including the number and shape; the criteria for selecting the beam; the implementation of the beam tracking mainly refers to the implementation of the fast search algorithm; The theoretical relationship of adaptive beamforming. ——As the basis of smart antenna research, it is meaningful to establish a more reasonable channel propagation model and study the optimal position distribution of antenna elements.
Ungrouped,High Quality Ungrouped,Ungrouped Details, CN Shenzhen Waweis Technology Co., Ltd. , https://www.waweisasdapter.com