Transmission device and procedure

Abstract

The invention relates to a transmission device with a transmitter and receiver equipped with multibeam directional antennas, realizable through various independent configurations of passive feed networks. In a first independent embodiment, a two-input (12, 13) Nolen network (11) produces uniform and linear weight distributions. In a second independent embodiment, a combination of Nolen networks (11) feeds a 16x16 array, generating a 2D uniform weight distribution at the outputs and 2D uniform-linear weight distributions in orthogonal dimensions. In a third embodiment, 90° / 180° hybrids (8) and power dividers (9) generate uniform-phase and monopulse weight distributions.In a fourth independent embodiment, 90° / 180° hybrid-based networks (18) are combined to feed a 16x16 element array, generating a 2D uniform weight distribution at the outputs and 2D uniform-monopulse weight distributions in orthogonal dimensions.

Description

Transmission device and procedure TECHNICAL SECTOR The present invention relates to a multimode direct beam transmission device for improving point-to-point wireless channel capacity. It also relates to the method used with the device. It is applicable in the field of communication. STATE OF THE ART In the field of wireless communications, the demand for higher data rates and improvements in channel capacity is constant. Multiple Input Multiple Output (MIMO) technology and spatial diversity are fundamental to improving the capacity of wireless channels. By deploying a greater number of antennas at both the transmitting and receiving ends, MIMO introduces multiple physical paths for signal transmission. This method significantly increases the spectral efficiency of the network, allowing for the transmission of a larger volume of data within the same frequency band. In high-frequency bands, the propagation of electromagnetic waves between the transmitter and receiver can be accurately modeled using ray tracing techniques. This method implies that the number of physically independent channels connecting the transmitter and receiver is directly proportional to the number of identifiable rays. Under these conditions, the gain attributed to each channel is influenced by the cumulative distance the wave travels, as well as by the reflection and diffraction coefficients encountered at each boundary. The maximum capacity, denoted as Cmax, is obtained by distributing the total power Pt across N channels to maximize the sum of the base-2 logarithm of one plus the signal-to-noise ratio in each channel. This is mathematically represented as: In this context, Yn is defined as the ratio of the product of the channel gain (gn) and the power assigned to channel n (pn) , with respect to the noise level (No) , expressed as o1n n0 . The optimal power distribution can be obtained from the algorithm known as 'water-filling'. In general, the MIMO technique consists of generating different beams in certain directions in order to reach the receiver using all available physical paths. In the context of peer-to-peer (P2P) communication, a transmitter and receiver are located in fixed, known positions, with a direct line of sight between them. Furthermore, it is often advantageous for the system design to be independent of the environment, as this greatly simplifies and reduces development costs. This leads to the use of highly directional antennas (for both transmission and reception) that are perfectly aligned. Ideally, using a dual-polarization system results in two independent channels for MIMO (2x2 MIMO). In this particular case, MIMO is especially simple and does not require complex beamforming systems or digital processing; standard dual-polarization antennas can provide access to ports where the two channels are already decoupled. Dual-polarization antennas are generally simple to implement, passive, and do not require complex components such as DSPs, analog-to-digital converters, programmable phase shifters, etc. Therefore, a dual-polarization antenna system allows for good channel capacity at a very low cost and with a very simple design. With only two linearly independent polarizations, the channel's MIMO capacity increases, but is limited. The objective of the method described in this application is to increase the number of independent channels in a simple way (without requiring complex digital beamforming algorithms) so that the system is independent of the environment (possible additional reflections). Currently, several systems exist for improving the capacity of wireless channels using MIMO and similar technologies. Relevant prior art documents include: First is US2003031264A1, which describes an adaptive communication system that diagonalizes the channel matrix using a unit matrix filter. While it improves transmission efficiency, it does not address the implementation of asymptotic expansions or near-field optimization. This patent uses digital processing to diagonalize the channel matrix, which requires complex digitization and digital processing systems. Our invention performs the diagonalization of the channel matrix at the analog level (without the need for digital processing), greatly simplifying the system when the transmitter and receiver are aligned and have a direct line of sight. US11107455B1 mentions a constant beam pattern acoustic transducer; however, it does not address electromagnetic channel diagonalization. This patent develops an acoustic wave beamformer that creates an approximately constant far-field pattern with frequency over a very high frequency range. It also involves complex signal processing and is not geared toward achieving channel matrix diagonalization. Therefore, again, our invention performs channel matrix diagonalization at the analog level (without the need for digital processing), greatly simplifying an electromagnetic wave transmitter / receiver system when the transmitter and receiver are aligned and have a direct line of sight. The system in US2022094403A1 uses a distributed antenna network to improve communications, employing DIDO (Distributed Input Distributed Output) technology, but it does not use asymptotic modal expansion or orthogonal directional beamforming. This patent addresses a slightly different problem: an asymmetrical situation with a base station and one or more users not necessarily sharing a physical location. The beamforming treatment it performs uses extensive and complex digital processing; therefore, unlike our invention, it does not offer a simple analog-based solution for diagonalizing the channel matrix in cases of line of sight between the transmitter and receiver. The applicant is unaware of any device that could be considered similar to the invention. BRIEF EXPLANATION OF THE INVENTION The invention consists of a transmission device according to the independent claims and variants thereof that resolve problems of the prior art. The invention also consists of the method used. The present invention introduces a transmission device and associated method that enable the enhancement of point-to-point wireless channel capacity through an innovative multimodal direct beamforming (MDRT) approach. This device utilizes asymptotic modal expansion to generate orthogonal beams and a diagonal channel array, eliminating the need for complex beamforming algorithms. The device includes multi-port directional antennas capable of generating additional channels with power degradation proportional to 1 / r²m, where m is a positive integer. This design significantly increases spectral efficiency and transmission capacity while maintaining low cost and structural simplicity by avoiding complex digital processing. Furthermore, the system can be implemented with various types of antennas, passive feed networks, and directivity elements such as reflectors or lenses. The invention provides a versatile and efficient solution for improving the capacity of wireless channels, being applicable in the field of communications and other environments that require high-capacity, low-cost point-to-point links. The transmitting and receiving antennas are multiport and have passive feed networks capable of generating the aforementioned beams. The invention introduces an innovative approach to meeting bandwidth demands through Multimode Direct Ray Transmission (MDRT), a technique designed to significantly enhance wireless channel capacity. The essence of MDRT lies in a novel asymptotic expansion of the near field, resulting in the formation of orthogonal channels with power degradation described by 1 / r²ⁿ, where n = 1, 2, 3... Compared to existing solutions, such as those indicated in the previous section, the proposed invention allows a significant increase in the number of independent channels in point-to-point wireless systems without the need for complex beamforming algorithms. Furthermore, it performs diagonalization of the channel matrix through asymptotic expansion, which optimizes the interaction between transmitter and receiver in the near and far field. It improves channel capacity compared to traditional dual-polarization solutions, increasing spectral efficiency without requiring greater processing resources. The invention proposes a new type of highly directional antenna with n ports and n independent beams, such that, when facing each other (transmitting and receiving antennas aligned), a diagonal channel matrix is ​​obtained: The first two channels can be described by the usual Friis formula |hn|2=|h22|2=O (1 / r2) and correspond to the horizontal and vertical polarization of the main beam. The sequence continues with a behavior of the type O (1 / r2m) , for m=2, 3, 4, 5, ..., where m is a constant corresponding to a natural number greater than 2. That is, the next 4 channels (m=2) behave as |h33|2=...=|h66|2=O (1 / r4) . For each possible value of m, a new set of orthogonal channels is obtained that can be used according to the invention. For m=1, the two channels that appear correspond to horizontal and vertical polarization, which are already known. For m=2, four new channels appear. Their power balance is worse than that of the first two channels; however, this is a beneficial development as it increases communication capacity (as indicated by the water-filling algorithm). For each m, the number of channels increases, although for m greater than 2, the resulting power balance is lower. This provides additional channels that allow for increased channel capacity if power is properly distributed across each subchannel using the waterfill algorithm. With the exception of the first two main channels, h11 and h22, the remaining channels have nulls in the direction of data transmission. The invention introduces a novel approach by performing an asymptotic modal expansion of the Green's function, leading to a formulation that captures the interaction of currents within a specific range, bridging the gap between the far and near fields. Following this expansion, we obtain a generalized version of the Friis formula, which represents the source interaction as a series of terms decreasing in magnitude as 1 / r²m for m = 1, 2, 3.... The canonical form of the Helmholtz Green's function is: Point O1 is the central reference point of region Q1 where the sources are located. Point O is the central reference point of region Q2 where the targets are located. The vector r connects source i to target j. Using a comprehensive expansion established for the Green function, detailed as follows: After a long theoretical development, the following asymptotic expansion is obtained: This expression allows us to obtain the optimal currents for synthesizing the previously proposed beams. The system would consist of two of these antennas (passive, without the need for complex elements such as digitizers, DSPs, programmable phase shifters, etc.) aligned, with n ports at each end, and with a diagonal channel matrix that allows maximizing the transmitted bit rate using basic water-filling techniques. The asymptotic expansion method used allows obtaining simple expressions for the hj terms. These terms have been tested with complex, high-precision numerical simulations and with measurements made with prototypes in an anechoic chamber, and satisfactory results have been obtained. Other variations can be seen in the rest of the memory. DESCRIPTION OF THE FIGURES For a better understanding of the invention, a section of figures is included, showing exemplary embodiments. Figure 1: Shows an array (1) of Nx x Ny elements with three independent ports for each polarization. Figure 2: Shows basic elements of microwave circuits such as: directional couplers (7), 90-degree hybrid coupler (8), power divider (9), meandering line (10). Figure 3: Shows a feed network based on directional couplers (7) and meandering lines (10) called the Nolen network (11). Figure 4: Shows a combination of Nolen networks (11) to feed a 16x16 array based on 16+2 Nolen networks with 2 input ports and 16 output ports. Figure 5: Shows an array (1) of Nx x Ny elements with three independent ports for each polarization. Figure 6: Shows a power supply network (18) based on power dividers (9) and 90 / 180 degree hybrids (8). Figure 7: Shows a combination of feed networks based on 90-degree / 180-degree hybrids (18), lines of a certain length to feed an array of 16x16 elements based on 16+2 one-dimensional networks like the one in Figure 5 with 2 input ports and 16 output ports. Figure 8: Shows an array (24) of 4 speakers (2x2) that allows generating 6 beams in total counting both polarizations. Figure 9: Shows an antenna based on an array (24) of 2x2 dual polarization elements and reflectors (25). Figure 10: Shows an antenna based on an array (24) of 2x2 dual polarization elements and lenses (26). Figure 11: Shows a transmitter-receiver scheme, based on passive multiport antennas (27), such that the channel matrix is ​​diagonal and satisfies |h_11|2=O (1 / r2) . MODES OF REALIZING THE INVENTION Next, a brief description is given of one way of carrying out the invention, as an illustrative and non-limiting example thereof. To validate the practical feasibility of the concept, an experimental prototype was designed using two ad hoc antennas. These antennas operate in a regime where traditional interference and channel capacity models do not fully apply, allowing for a substantial increase in bit rate without requiring additional spectrum or increased power consumption. Based on the results used, the gain obtained in terms of channel capacity is calculated for different antenna sizes, distances between transmitter and receiver, and frequencies. The results are shown in Tables 1 and 2. Table 1 shows the capacity and capacity increase achieved with respect to a standard dual-channel link (dual polarization) for a transmitting and receiving antenna of size lx * k, where the distance between the transmitter and receiver is r0o = 50 m. Table 2 shows the capacity and capacity increase achieved with respect to a standard dual-channel link (dual polarization) for an array of / transmission and reception of N * N elements, of size lx * lx where (tx _ ~ ia VtX transm 2) and a distance between the transmitter and receiver of roo = 100m. As can be seen in the tables, the increase in channel capacity obtained can be very significant, up to +149% compared to the Shannon limit if only the two main channels (m=1, horizontal and vertical polarization) were used, that is, compared to the optimum in the current state of the art considering methods that do not use complex digital processing. Table I. Capacity and capacity increase achieved with respect to the standard dual channel (dual polarization) for a link with transmitter and receiver constituting an array of N x N elements whose total size is lx x lx, where lx = and the distance between transmitter and receiver is 50m Table II. Capacity and capacity increase achieved with respect to the standard dual channel (dual polarization) for a link with transmitter and receiver constituting an array of N x N elements whose total size is lx x lx, where lx = j, and the distance between transmitter and receiver is 100m The antennas that produce these beams can be implemented using array antennas, reflectors, lenses, horns, or a combination of the above. Figures 1-6 show possible antenna solutions based on two-dimensional arrays of a certain size, including their passive feed network for specific sizes (16x16). This can be generalized to any size. The aim is to protect any essentially equivalent feed network that produces the same or very similar effects and weights using minor variations compared to the examples described in Figures 1-7. Figures 3 and 4 show one way to implement the invention using Nolen networks. With the aim of simplifying the design, the use of Nolen networks or any other equivalent form of passive feed network that produces the same weight distribution (uniform and linear in X or Y) can be avoided by using suboptimal weight distributions such as the monopulse distribution (shown in Figure 5). To describe the dual polarization case more precisely, Figure 8 shows an antenna based on four radiating elements (four horns) with dual polarization independent of the other horns: The invention corresponds to any method of generating the previously described beams to generate 6 orthogonal channels, either by using array antennas or by using a reduced number of radiating elements and an element to increase directivity (such as a reflector or lens). Thus, Figure 1 shows an array (1) of Nx x N elements with three independent ports for each polarization. The first port (corresponding to subfigure a) generates a single beam (2) and corresponds to the m=1 channels with a power of |2 = 2212 = 0 (~^). The second and third ports (corresponding to subfigures b and c) generate two beams with a null in the horizontal and vertical planes (3) and (4) respectively and correspond to the m=2 channels with a power of |^3312 = 0 (2j). These beams can be generated with a weight distribution as shown in the figure. For subfigure a), the weight distribution is uniform (5) in xy and y. For subfigure b) the distribution is uniform (5) in X and linear (6) in Y. For subfigure c) the distribution is uniform (5) in Y and linear (6) in X. This produces maximum coupling for the secondary channels with m=2.This weight distribution can be achieved with a passive feed network like the one shown in Figures 2, 3 and 4. Figure 2 shows basic microwave circuit elements such as: directional couplers (7), 90-degree hybrid coupler (8), power divider (9), and meander line (10). These and similar elements (e.g., 180-degree hybrid coupler) can be used to design feed networks for arrays with specific weights. Figure 3 shows a Nolen network (11) which corresponds to a feed network based on directional couplers (7) and meandering lines (10). Based on this type of network, different weight distributions (5, 6) can be obtained at each output port (top) for different input ports (12, 13) (left). This network can be designed so that one input port (top left port, (12)) produces a uniform weight distribution, while another port (bottom left port, (13)) produces a linear weight distribution. Figure 4 shows a combination of Nolen networks (11) to feed a 16x16 array based on 16+2 Nolen networks with 2 input ports and 16 output ports. Input (14) produces a 2D uniform weight distribution on the 16x16 output ports (see Figure 1a). Input (15) produces a 2D uniform weight distribution in one dimension (X) and a linear distribution in the other dimension (Y) (see Figure 1b). Input (16) produces a 2D uniform weight distribution in one dimension (Y) and a linear distribution in the other dimension (X) (see Figure 1c). This network can be used to feed a 16x16 antenna array to radiate the primary (m=1) and secondary (m=2) channels for a given polarization. Two of these networks can be used to feed a dual-polarized antenna array, thus achieving a total of 6 channels (2 primary and 4 secondary). An array (1) of Nx x N elements with three independent ports for each polarization is shown in Figure 5. The first port (corresponding to subfigure a) generates a single beam (2) and corresponds to the m=1 channels with a power of |2 = 2212 = 0 (~^). The second and third ports (corresponding to subfigures b and c) generate two beams with a null in the horizontal (3) and vertical (4) planes respectively and correspond to the m=2 channels with a power of |^3312 = 0 (2j). These beams can be generated with a weight distribution as shown in the figure. For subfigure a), the weight distribution is uniform (5) in xy and y. For subfigure b) the distribution is uniform (5) in X and monopulse (17) in Y. For subfigure c) the distribution is uniform (5) in Y and monopulse (17) in X. This produces a suboptimal but functional maximum, as shown in the attached tables, for the secondary channels with m=2.This weight distribution can be achieved with a passive feed network like the one shown in figures 6 and 7. Figure 6 shows a feed network (18) based on power dividers (9) and 90 / 180 degree hybrids (8). Based on this type of network, different weight distributions can be achieved at each output port (top) for different input ports (bottom). This network can be designed so that one input port (19) produces a uniform weight distribution (5), while another port (20) produces a monopulse weight distribution (17) (180-degree phase shift between the right and left halves). Figure 7 shows a combination of feed networks based on 90-degree / 180-degree hybrids (18), lines of a certain length to feed a 16x16 array based on 16+2 one-dimensional networks like the one in Figure 5 with 2 input ports and 16 output ports. Input (21) produces a 2D uniform weight distribution on the 16x16 output ports (see Figure 1a). Input (22) produces a 2D uniform weight distribution in one dimension (X) and a monopulse distribution in the other dimension (Y) (see Figure 1b). Input (23) produces a 2D uniform weight distribution in one dimension (Y) and a monopulse distribution in the other dimension (X) (see Figure 1c). This network can be used to feed a 16x16 element antenna array to radiate the primary (m=1) and secondary (m=2) channels for a polarization.Two of these networks can be used to feed a dual-polarized antenna array, thus achieving a total of 6 channels (2 primary and 4 secondary). An example of a 2x2 array (24) with four horns, generating a total of six beams when considering both polarizations, is shown in Figure 8. Two single beams (horizontal and vertical polarization) (Figure 7a) and two dual beams with nulls in the X and Y planes and horizontal and vertical polarization (Figure 7bc). It is known how to feed a single square horn to generate two beams with dual polarization. This is the smallest array example that allows for the generation of the six channels mentioned above. This greatly simplifies the design. Furthermore, gain can be achieved using reflectors or lenses, as shown in the following series of figures. Figure 9 represents an antenna based on an array (24) of 2x2 dual-polarization elements and reflectors (25). With a total of 6 orthogonal channels (2 main and 4 secondary). The horns can be fed with a suitable passive feed network. Figure 10 shows an antenna based on a 2x2 dual-polarization and lens array (26) (24). It has a total of 6 orthogonal channels (2 main and 4 secondary). The horns can be fed with a suitable passive feed network (similar to Figure 6 but much simpler). The last figure, figure 11, shows a transmitter-receiver scheme, based on passive multiport antennas (27), such that the channel matrix is ​​diagonal and satisfies |h_11|2=O (1 / r2) . The four secondary beams (center and right) correspond to the channels |h_33|2=---=|h_66|2=O (1 / r4) . The multiport transmitting / receiving antenna can be one of the examples described in the figures (array of size N_x*N_y with Nolen feed network, array of size N_x*N_y with Hybrid feed network, 2x2 array with reflector or lens to increase directivity, or any possibility that generates that type of beam and that automatically decomposes the channel matrix into a diagonal matrix where the first two elements (28, 29) have a behavior like O (1 / r2) . And the following four have a behavior like O (1 / r4) (30, 31, 32, 33) .

Claims

1- Transmission device, with a transmitter and a receiver, both with directional antennas having n ports and n independent beams, characterized in that the channel matrix between both antennas is diagonal, where the first two channels can be described with the formula of and correspond to the horizontal and vertical polarization of the main beam, the following orthogonal channels behave as for m a natural number greater than 2, where the optimal currents are obtained by applying an asymptotic expansion of the Green's function such that where the point Oi is the central reference point of the region Qi where the sources are located; the point O is the central reference point of the region Q2 where the targets are located; and the vector j connects the source i with the target j.

Images