Smart antenna system without adaptive algorithm

Figure 1 Schematic diagram of smart antenna beam control

Adaptive smart antenna system, iteratively obtains a set of weighted vectors, so that the ratio of the required signal in the array output signal S (t) to all other user signals is maximized, that is, the signal-to-interference ratio is maximum. With this implementation, the system It has the highest signal-to-interference ratio. However, due to the iterative method, the response speed of the system is limited. For high-speed mobile users, the performance of the system will also be affected.

(1)

When the weighted signal is known, the array antenna pattern can be calculated from equation (1). Where g (θ) is the array pattern; ωm is the weighted value of the m-th signal; φm (θ) is the signal reaching the array from a spatial angle Phase difference at unit m.
Figure 2 is the corresponding pattern of an 8-element circular adaptive array with half-wavelength array elements in a randomly distributed communication environment with 1 user and 10 interfering users. As can be seen from Figure 2, the system forms pits in the direction of interference , And form peaks in the direction of the desired signal. Usually adaptive algorithms may also have peaks at other angles without interference signals.

Figure 2 Adaptive direction diagram

Figure 3 is an isolateral lobe needle beam pattern. The isolateral lobe needle beam pattern can also be calculated by formula (1). The difference between the isolateral lobe pattern and the adaptive pattern is that the way in which the weighted signal is generated is different, etc. The weighting values ​​of the side lobe pattern are pre-calculated. When the side lobe smart antenna system is working, the signal arrival angle (DOA) needs to be determined by the direction finding algorithm first, and then the main lobe of the pattern is selected by selecting the appropriate weight. Point to the user's direction of arrival. This type of smart antenna is used to ensure the suppression of the interference in the non-main lobe area by the low equal side lobe level. For the interference in the main lobe area, the equal side lobe needle smart antenna system is used It cannot be suppressed. Since the width of the main lobe of the system pattern is determined by the aperture of the antenna array, the adaptive smart antenna's ability to suppress interference signals in the main lobe is also very limited. Compared with the adaptive smart antenna, the side-lobe intelligence The antenna does not need iteration, the response speed is fast, and the robustness of this scheme is better.

Figure 3 -15dB isolateral lobe needle beam pattern

3. Smart antenna system comparison simulation results This section gives the adaptive directional pattern (Applebaum algorithm [11]), -10dB equal side lobe needle beam pattern (as shown in Figure 4 dotted line), -15dB equal side lobe needle Simulation results of the performance of four smart antenna systems based on the shape of the beam pattern (Figure 3) and the -20dB equal side lobe needle beam pattern (Figure 4 solid line). The system used in the simulation uses the spacing of adjacent array elements at half wavelength The 8-element circular array assumes that the array uses isotropic elements. The iteration process is not considered for the adaptive smart antenna during simulation, which is the final steady-state result of the system.

Figure 4 -10dB, -20dB and other side lobe needle beam pattern

The simulation in this paper assumes that the CDMA system has ideal power control, the system has a spreading factor of 128, and no voice excitation. There is no interference other than users, no interference from neighboring cells, and no multipath interference in the cell. The system threshold Eb / N = 6dB. According to the above assumption, the maximum number of users that a base station using an omnidirectional antenna can support is 32.
Figure 5 shows the cumulative probability distribution of Eb / N in a 32-user CDMA system with a threshold of 6dB after the base station introduces four different smart antennas. Each curve in Figure 5 is a 10,000 random user distribution The statistical results can be seen from Figure 5, after using smart antennas, the system's Eb / N has been significantly improved. This shows that without increasing the number of users, the use of smart antennas can reduce the signal power required by the system, Increase the coverage area of ​​the base station. When the out-of-bounds probability is 0.01, four smart antennas with adaptive pattern, -10dB equal side lobe needle beam pattern, -15dB equal side lobe needle beam pattern, -20dB equal side lobe needle beam pattern are used The system is improved by 5.25dB, 4.75dB, 5.05dB, 4.45dB compared to the system with omnidirectional antenna.

Figure 5 Cumulative probability distribution of four smart antenna systems for 32 users

Figure 6 shows the out-of-bounds probability distribution of the system using four smart antennas for capacity expansion when the number of users is lower than the threshold (6dB). Each point in the curve in Figure 6 is 10,000 random users Statistical results of the distribution. With an out-of-bounds probability of 0.01, an adaptive pattern, a side lobe needle beam pattern of -10dB, a side lobe needle beam pattern of -15dB, a side lobe needle beam pattern of -20dB, etc. are used. The capacity expansion of the four smart antenna systems is 6.81, 4.81, 6.62, and 5.66 times that of the omnidirectional antenna system.

Figure 6: Out-of-bounds probability of the expansion of four smart antennas

Figure 7 shows an 8-unit equilateral side-lobe needle beam pattern smart antenna. When the out-of-bounds probability of Eb / N <6 is 0.01, the number of users that can be supported by a smart antenna system with different side-lobe level patterns Curve. The dotted line in the figure is the number of users that the adaptive smart antenna can support: 218. As can be seen from FIG. 7, when the side lobe level is -20dB, the number of users that the system can support is 181. With the needle beam pattern As the side lobe level increases, the system capacity increases. When the side lobe level is -15dB, the number of users that the system can support reaches a maximum: 213, which is only 5 fewer than the system using an adaptive smart antenna. After the level exceeds -15dB, the system capacity will decrease as the side lobe level of the needle beam pattern increases. When the side lobe level is -10dB, the system can support 153 users.

Figure 7 Smart antenna capacity expansion user number ratio side lobe level

It can also be found from Figures 6 and 7 that the smart antenna system using the -15dB equal side lobe pattern and the adaptive smart antenna have an approximate capacity expansion. In order to explain this phenomenon, the necessary requirements are given in Figure 8. When the signal origin is 180 degrees, and the other 200 interfering users are randomly distributed, the directional pattern obtained by the adaptive algorithm. As can be seen from FIG. 8, when the number of interferences is much greater than the number of array elements, the directional pattern obtained by the adaptive algorithm (Figure 8 ) And side lobe patterns such as -15 (Figure 3) have similar main lobe width and side lobe level. This phenomenon can be explained by the principle of the adaptive algorithm. When the number of interference is less than the number of array elements, the self-adaptive The algorithm can generate pits to completely suppress the interference. When the number of interference is much greater than the number of array elements, because the interference has spread in all directions of the circle, the adaptive algorithm can no longer suppress interference by forming pits, it can only pass The formation of lower side lobe levels to suppress interference. This conclusion is very important, which shows that when using smart antennas to expand capacity, it may not be necessary to use an adaptive algorithm, only need to select the appropriate side lobe pattern to achieve And expansion capability similar adaptive algorithm.

Figure 8: Adaptive direction with 200 users

It can also be seen in Figures 6 and 7 that the performance of smart antenna systems using the -15dB pattern and the -10dB and -20dB patterns is very different, which indicates that choosing different isolateral lobe patterns will significantly affect the expansion of the smart antenna. Capability. Comparing the three equal side lobe patterns in Figure 3 and Figure 4, it can be seen that when the array structure is fixed, the side lobe level is inversely proportional to the main lobe width. The -10dB reverse pattern has a narrow main lobe , But its side lobes are higher and the system performance is reduced; although the -20dB pattern has lower side lobes, because of its wider main lobe width, the system performance is also reduced. So there is no need to pursue too narrow in practical applications The main lobe or the side lobe that is too low should be selected to have an equal side lobe pattern with similar main lobe width and side lobe level as the pattern obtained by the adaptive algorithm. At this time, the performance of the needle beam smart antenna is close to the best.

4. Conclusion This paper studies two different smart antenna systems using adaptive pattern and equal side lobe needle beam pattern. The simulation results of the capacity expansion of these two types of smart antennas to the existing CDMA system are given. It shows that when using smart antennas to expand capacity, it may not be necessary to use an adaptive algorithm, and only need to select a suitable isolateral lobe pattern to achieve capacity expansion similar to that of the adaptive algorithm. The simulation results also show that different isosidelobe patterns are selected. , Will significantly affect the expansion capacity of the smart antenna. So in practical applications, the side lobe pattern with the main lobe width and side lobe level similar to the pattern obtained by the adaptive algorithm should be selected to make the performance of the smart antenna close to the best.