This paper presents mobile-to-mobile (M-to-M) channel measurements and models at 1.85 GHz in dense-scattering suburban areas. Three test scenarios were considered for different transmitter (Tx) and receiver (Rx) antenna placements. Scenario 1: inside-vehicle-to-inside-vehicle (IV2IV), where both Tx and Rx antennas were placed inside the test vehicles; Scenario 2: inside-vehicle-to-walk (IV2W), where the Tx antenna was placed inside the vehicle and the Rx antenna was inside the pocket or next to the face of the tester; and Scenario 3: walk-to-walk (W2W), where the Tx and Rx were moving at walking speeds, the Tx antenna was placed on a test cart, and the Rx was inside the pocket or next to the face of the tester. To mimic more typical M-to-M channels in practice, eight sub-scenarios were configured under the availability of the line-of-sight (LoS) between the Tx and Rx. Detailed channel models are proposed for path loss, shadowing, small scale, and multi-scale fading based on the collected measurements. The level crossing rate (LCR) and average fading duration (AFD) are studied from the measurements and compared with the theoretical expressions.

This paper presents an optimal amplify-and-forward (AF) relay scheme for a multiple-input-multiple-output (MIMO) system consisting of M-mobile sources, M-mobile destinations, and N-cooperative distributed mobile relay nodes. A wireless mobile relaying network with channel state information is investigated. The received signals from the mobile sources are ex- changed between mobile relays to achieve optimal performance. The main contribution in this paper is the derivation of mobile relaying amplifying matrices (MARMs) designed for cooperative MIMO networks based on the minimum mean square error (MMSE) criterion. By adopting the new proposed MRAMs, the bite error rate (BER) performance of the system is evaluated using Monte-Carlo simulations.