This paper proposes a method for cooperative multi-static Multiple Input Multiple Output (MIMO) radar that can estimate the number of targets. The purpose of this system is to monitor humans in an indoor environment. First, target positions within the estimation range are roughly detected by the Capon method and the mode vector corresponding to the detected positions is calculated. The mode vector is multiplied by the eigenvector to eliminate the virtual image. The spectrum of the evaluation function is calculated from the remaining positions, and the number of peaks in the spectrum is defined as the number of targets. Experiments carried out in an indoor environment confirm that the proposed method can estimate the number of targets with high accuracy.

This paper presents a novel simultaneous decoupling and matching technique for transmitting (Tx) and receiving (Rx) ports in short-range multiple-input multiple-output (SR-MIMO) systems. The principal difference with conventional decoupling and matching network (DMN) approaches is that the proposed technique considers strong mutual coupling between closely-positioned Tx/Rx arrays, and the S-parameter variation due to the presence of each other's array. This technique has two stages; first, 180-degree hybrid couplers are connected to both Tx/Rx ports of a plane-symmetrical MIMO system. This decouples both Tx/Rx ports, and moreover, channels between them are orthogonalized. That is, the MIMO system is transformed into multi orthogonalized single-input single-output (SISO) systems. Second, Tx/Rx ports of each orthogonalized SISO system are simultaneously matched based on conjugate matching theory. Consequently, the transmission power of the short-range MIMO system is maximized. Numerical results show that the proposed technique realizes higher channel capacity than the conventional DMN; indeed it achieves the theoretically possible capacity. In addition to theoretical analyses, we provide an example for microstrip line (MSL) circuit implementation. This MSL model offers good simultaneous decoupling and matching performance yielding channel capacity comparable to that of an ideally-designed circuit model. This validates the implementation feasibility of the proposed technique.

An analog-beamforming-based eigenmode transmission technique is proposed that employs a network of interconnected 180-degree hybrid couplers at both transmitting and receiving sides of a plane-symmetrically configured short-range MIMO system. This technique can orthogonalize MIMO channels regardless of array parameters such as antenna spacing and Tx-Rx distance, provided the MIMO array is symmetric. For verifying the effectiveness of the proposed technique in channel orthogonalization, an experiment is conducted using a 4×4 MIMO array consisting of microstrip antennas and cascade-connected rat-race hybrid couplers. The results indicate a reduction in interference by approximately -28.3dB on average compared to desired signal power, and the ability to realize four-stream parallel MIMO transmission by using only analog passive networks. The proposed technique can achieve channel capacity almost equivalent to that of eigenbeam space division multiplexing with ideal digital beamforming.

In this paper, a method of calculating the mean channel capacity based on S-parameter of MIMO (Multiple-Input Multiple-Output) antenna is proposed. This method exploits the correlation matrix calculated from the antenna S-parameter matrix, and offers highly accurate estimates of the mean channel capacity without dependence on SNR (Signal-to-Noise Ratio). The numerical and experimental results revealed that the proposed method can calculate the channel capacity with fair accuracy independent of the number and spacing of the antenna elements if the radiation efficiency is sufficiently high.